EVOLVING PERCEPTIONS OF

ASSORTMENTS, IMAGE AND CUSTOMER

EXPERIENCES IN RETAIL ENVIRONMENTS

by

Cathy Hart

Doctoral Thesis

Submitted in partial fulfilment of the requirements

for the award of Doctor of Philosophy in Retail and Marketing of
Loughborough University

School of Business and Economics

Loughborough University

July 2016

i
 ABSTRACT

Consumers’ perceptions of retail stores, the assortment of products and the

locations in which they are situated, all influence consumers’ decisions about
where, when and how to shop. In the highly competitive arena of retailing, this
knowledge greatly affects retailer performance and marketing strategy. Despite
this strategic importance, the literature lacks empirical research into the
assortment of categories offered in retail stores and their influence on store image.
Moreover, image conceptualisation for different retail contexts is fragmented and
the consumer image perceptions of categories are unexplored. Recent attention
has shifted to customer experience research, yet this literature refers to the
consumers’ interaction with one retailer, when consumer shopping journeys may
include many retail interactions.

The research presented in this thesis addresses the above gaps, contributing to
theory and knowledge in the area of retail marketing. The thesis comprises eight
papers published over a 17-year period. It examines the perceptions and roles of
retail assortments; category perceptions and management in retail stores;
customer perceptions of image and experiences of town centres.

The thesis draws from a range of empirical studies documented within the eight
papers. These include: a comparative study of ten major retailers’ assortment
strategies; a survey of 504 consumers’ perceptions of supermarket assortments; a
qualitative study involving six national focus groups of apparel consumers; and a
survey of 536 consumers across five regional city centres. The studies are
informed by two literature reviews documenting retail assortment research and
town centre research.

Collectively, the thesis delivers a significant research contribution to knowledge

and theory. The first four thesis papers clarify the strategic relationships between
retail assortment and image. The following two papers identify new perspectives of
image congruence and relationships between category and store image. The final
two papers consolidate all previous contexts in a multi-faceted retail image
ii
construct, emphasising the previously neglected psychological image elements.
Fundamentally, consumers’ experiences in town centre retail environments
contribute to the ongoing process of “holistic” image creation.

The findings demonstrate that as retail environments change, consumers’

perceptions are also evolving from a functional, static image perspective to a
dynamic, experience driven concept. The research commentary concludes by
outlining future research agenda.

Keywords
Retail Assortment, Retail Image, Customer Experience, Town Centres, Consumer
Perceptions, Consumer Behaviour.

iii
ACKNOWLEDGEMENTS

This PhD research journey has been a rewarding, and sometimes challenging
experience, which was achieved with the help and encouragement of many people
along the way.

I would like to thank all of the co-authors who contributed to shaping the research,
particularly John Cadogan, Belinda Dewsnap and Mohammed Rafiq for their
valuable insights. I am also grateful to colleagues at the School of Business and
Economics, Loughborough University for their advice and assistance throughout.

To the various industry partners and retail organisations that supported, and in
many cases, funded the reported research studies, a big thank you. Particular
mention goes to Jim Murphy for his faith in this PhD; David Fuhr, Andy Godfrey
and Tom Ironside for steering me into town centre research; to the Loughborough
BID team, whose enthusiasm kept me there.

A special mention goes to Grazyna Stachow, researcher and friend, for paving the
way and converting ancient files into a thesis document. I am also grateful to Kay
Harris for her patience getting the thesis over the final hurdle.

I am particularly indebted to my supervisor Neil Doherty, for his continued support,

valuable feedback and guidance.

Finally, a special thanks to my family for their belief in my work. To my parents

who first set me on my education journey; my husband, Havilland, for his endless
support and encouragement; to my children Tim and Zoe, who make me very
proud. I dedicate this thesis to all of them.

iv
TABLE OF CONTENTS
ABSTRACT ....................................................................................................... ii

ACKNOWLEDGEMENTS ................................................................................. iv

TABLE OF CONTENTS .................................................................................... v

LIST OF TABLES ............................................................................................. xi

LIST OF FIGURES .......................................................................................... xii

Commentary: Overview, theoretical context and structure of the thesis.... 1

1. Introduction ........................................................................................... 1

2. Background to the research areas ........................................................ 3

2.1 Retail assortment: Non-food diversification........................................ 4

2.2 Retail image: Fashion categories....................................................... 5

2.3 The customer experience and town centre image ............................. 6

3. Literature and theoretical context for the portfolio ................................. 7

3.1 The changing role of assortments in retail strategy ........................... 7

3.2 Retail Image: Conceptualization and contexts ................................... 9

3.3 Customer experience ....................................................................... 12

4. The portfolio of papers: Research approaches and contributions ....... 14

4.1 Retail assortment: Perceptions and roles ........................................ 16

4.2 Retail image: Category behaviour and management ....................... 21

4.3 Customer experience of retail environments.................................... 24

5. Discussion of the Portfolio Papers: Contributions, limitations and future

research agenda........................................................................................... 27

5.1 Summary of research contributions ................................................. 27

5.2 Research limitations ........................................................................ 34

5.3 Impact and research agenda ........................................................... 35

5.4 Concluding remarks ......................................................................... 38

References: Hart ............................................................................................ 39

References: Other .......................................................................................... 42

v
Paper 1. The Retail Accordion and Assortment Strategies: An Exploratory
Study ............................................................................................................... 52

Introduction ................................................................................................... 52

Retailer merchandise strategies ................................................................... 53

Specialisation Vs Diversification ................................................................... 54

The Retail Accordion .................................................................................... 56

The Study ..................................................................................................... 58

Results and Discussion ................................................................................ 60

Range Diversification .................................................................................... 62

Specialisation ............................................................................................... 68

Conclusions .................................................................................................. 71

Paper 2. Consumer Perceptions of Non-Food Assortments: An Empirical

Study ............................................................................................................... 77

Introduction ................................................................................................... 77

Background .................................................................................................. 78

Objectives .................................................................................................. 81

Methodology ................................................................................................. 81

Store Loyal Consumers ............................................................................. 83

Importance of Congruity and Range Fit ..................................................... 84

Results and Discussion ................................................................................ 85

Perceptions of Range Fit ........................................................................... 85

Dimensions of Choice ................................................................................ 86

Scope for Expansion .................................................................................. 87

Store Specific Perceptions ......................................................................... 91

Implications and Conclusions ....................................................................... 95

Paper 3. The Location and Merchandising of Non-Food in Supermarkets

....................................................................................................................... 103

Introduction ................................................................................................. 103

vi
 The store environment ................................................................................ 104

Store layout.............................................................................................. 105

In store behaviour .................................................................................... 106

The study .................................................................................................... 108

The display locations ............................................................................... 109

Consumer preferences ............................................................................... 110

Store-specific preferences ....................................................................... 114

Conclusions and implications ..................................................................... 116

Paper 4. The Dimensions of Assortment: A Proposed Hierarchy of

Assortment Decision Making ...................................................................... 121

Introduction ................................................................................................. 121

Assortment Definitions and Roles............................................................... 122

Assortment definitions ............................................................................. 122

Assortment roles ...................................................................................... 123

Levels and Dimensions of Assortment ....................................................... 125

A Framework for Assortment Decision Making ........................................... 132

Assortment Research Focus ...................................................................... 134

Discussion and Directions for Future Research.......................................... 139

Paper 5. An Exploratory Study of the Consumer Decision Process for

Intimate Apparel ........................................................................................... 151

Introduction ................................................................................................. 151

Background and Research Objectives ....................................................... 152

Methodology ............................................................................................... 156

Results and Discussion .............................................................................. 158

Problem Recognition ............................................................................... 158

Information Search .................................................................................. 159

Pre-purchase Evaluation.......................................................................... 161

Purchase and Post-purchase Behaviour ................................................. 163

vii
 Conclusions, Future Research and Managerial Implications ...................... 164

Paper 6. Category Management: A New Approach for Fashion Marketing?

....................................................................................................................... 172

Introduction ................................................................................................. 172

Literature Review........................................................................................ 174

Introduction .............................................................................................. 174

Category Management as the next generation QR.................................. 175

Research Opportunities: The importance of consumer intelligence and joint

category planning.................................................................................... 183

Methodology ............................................................................................... 185

Research design ...................................................................................... 185

Data collection methods........................................................................... 186

Sample..................................................................................................... 189

Data management ................................................................................... 189

Findings ...................................................................................................... 190

The pre-purchase environment ................................................................ 190

The consumer purchase process............................................................. 193

The in-store purchase environment ......................................................... 196

Discussion and Managerial Implications..................................................... 197

Category definition ................................................................................... 198

Category role and category strategy ........................................................ 198

Category tactics ....................................................................................... 199

Optimal assortment.........................................................................................199
Optimal new products .....................................................................................200
Optimal promotion ..........................................................................................201
Optimal replenishment ....................................................................................202
Conclusions and Research Next Steps ...................................................... 203

Paper 7. Enjoyment of the Shopping Experience: Impact on Customers'

Repatronage Intentions and Gender Influence.......................................... 212
viii
 Introduction ................................................................................................. 212

Theoretical Background .............................................................................. 214

Image ....................................................................................................... 214

Enjoyment ................................................................................................ 215

Repatronage ............................................................................................ 216

Gender ..................................................................................................... 216

Conceptual Development ........................................................................... 217

Dimensions of enjoyment of the shopping experience: Accessibility ....... 218

Dimensions of enjoyment of the shopping experience: Atmosphere ....... 219

Dimensions of enjoyment of the shopping experience: Environment....... 220

Dimensions of enjoyment of the shopping experience: Service Personnel221

Outcomes of Enjoyment........................................................................... 223

The Influence of Gender .......................................................................... 224

Methodology ............................................................................................... 226

Sample..................................................................................................... 226

Measurement ........................................................................................... 227

Scale properties ....................................................................................... 228

Analysis ...................................................................................................... 229

Results and Discussion .............................................................................. 230

Conclusions and Managerial Implications................................................... 233

Limitations and Future Research Directions ............................................... 234

Paper 8. Conceptualising Town Centre Image and the Customer Experience

....................................................................................................................... 245

Introduction ................................................................................................. 245

Literature review ......................................................................................... 246

Customer experience ............................................................................... 253

Method ....................................................................................................... 256

Sample and data collection ...................................................................... 257

ix
 Content analysis ...................................................................................... 258

Results........................................................................................................ 261

Discussion .................................................................................................. 266

A conceptual model of town centre image and experience ...................... 267

Cognitive / physical .................................................................................. 268

Sensory / affective ................................................................................... 269

Social ....................................................................................................... 270

Global impressions .................................................................................. 271

Categorisation and image of other locations ............................................ 271

Consumer and situation influences .......................................................... 272

Conclusions and implications ..................................................................... 273

Managing image and the customer experience in the town centre .......... 274

x
LIST OF TABLES
Table 1 Portfolio of papers by literature themes 2
Table 2 Perceptions of range “fit” in each non-food category 85
Table 3 Perceptions of degree of choice for each non-food area 87
Table 4 Importance for expansion in non-food categories 88
Table 5 Relationships between perceptions of choice and fit with perceived
importance of expansion per category 90
Table 6 Store specific perceptions: (a) range fit, clothing and entertainment; (b)
perceptions of “too little” choice for clothing, home and entertainment; (c)
importance for expansion in clothing 91
Table 7 Relationship between frequency of purchase and perceived importance of
expansion per category 94
Table 8 Level of importance attached to each display location per non-food item
(by percentage frequency) 110
Table 9 Degree of association between display location preferences per non-food
item and store patronage 113
Table 10 Most preferred display location by retailer 115
Table 11 Dimensions of assortment in retail literature 128
Table 12 Dimensions of Product Mix/Assortment used in marketing literature
13 0
Table 13 Category management (CM) literature: Theoretical and empirical
evidence 177
Table 14 Model Fit Measures 229
Table 15 Correlations and Measurement Properties 229
Table 16 Town centre attributes in the literature 249
Table 17 Dimensions of town centre image identified in the literature 251
Table 18 Dimensions of town centre image and experience following Csv tests
26 3

xi
LIST OF FIGURES
Figure 1 Interlinking Research Themes 34
Figure 2 Food Retailer Growth in Product Lines, 1984-1994 61
Figure 3 Mixed Retailer Contraction in Product Lines, 1984-1994 62
Figure 4 Retailer Reasoning behind Range Diversification 64
Figure 5 Reasons for Deleting a Product Range 69
Figure 6 Reasons for expansion preferences 89
Figure 7 Most important (best) and least important display location for different
non-food items 111
Figure 8 A hierarchy of assortment decisions 133
Figure 9 Assortment level research matrix 138
Figure 10 The Category Management Process 174
Figure 11 Data collection methods 188
Figure 12 Consumer Brand Perceptions: The difference between positive and
negative responses 192
Figure 13 Conceptual Model Of Shopping Enjoyment 230
Figure 14 Generic model of retail customer experience - Adapted from Grewal et
al., (2009) and Verhoef et al., (2009). 256
Figure 15 Contextualised model of town centre image and experience 268

xii
 Commentary: Overview, theoretical context and structure
of the thesis.

1. Introduction
The following thesis documents a portfolio of eight published papers with their
preceding commentary, in submission for PhD by publications. The research
presented in the thesis portfolio contributes to theory, knowledge and discussion in
the area of retail marketing. Drawing from a range of empirical studies, it provides
a new understanding of perceptions of retail assortment, retail image and
customer experience in different retail contexts. Table 1 (page 2) outlines the eight
papers and their corresponding research themes. The first four papers comprise a
comparative study of retailers’ assortment diversification strategies; a study of
consumer perceptions of supermarket non-food assortments, investigating both
category expansion and merchandising preferences; a review of the retail
assortment literature and research. The two following papers explore in-depth the
consumer behaviour responses and category management issues for a fashion
retail sector. The final two papers examine consumer images and customer
experiences of town centres.

Whilst individually, these studies make important contributions to their host

literatures, namely retail assortment, retail image and customer experience, it is
the concept of retail image, which permeates all eight papers and in so doing it
provides supplementary contributions relating to: (i) Evaluating the evolving role of
the retail assortment in defining image; (ii) synthesising the different contexts of
retail image, and (iii) conceptualising the link between image and customer
experience.

This chapter encompasses the supporting commentary to the thesis as

follows. It continues with a summary of the author’s personal research journey
outlining the rationale for three research themes, whilst the third section critically
reviews the literature and theoretical context underpinning the research themes.
Next, the research approaches employed in the thesis are introduced, before
presenting the portfolio of papers. The final section discusses the overall
contribution of the thesis and linkages across the three research themes.
1
Cognizant of the limitations experienced during this research, the commentary
concludes by summarising these, before outlining the agenda for future research.

Table 1: Portfolio of papers by literature themes

Retail assortment: Perceptions and roles
Hart, C.A. (1999) The Retail Accordion and Assortment Strategies: An Exploratory Study,
The International Review of Retail, Distribution and Consumer Research, 9(2),
pp.111-126, ISSN: 0959-3969. Hart contribution: 100%.
Hart, C.A. and Davies, M.A. (1996a) Consumer Perceptions of Non-Food Assortments: An
Empirical Study, Journal of Marketing Management, 12(4), pp. 297-312, ISSN:
0267-257X. Hart contribution: 50%.
Hart, C.A. and Davies, M.A. (1996b) The Location and Merchandising of Non-Food in
Supermarkets, International Journal of Retail and Distribution Management, 24(3),
pp. 17-25, ISSN: 0959-0552. Hart contribution: 50%.
Hart, C.A. and Rafiq, M. (2006) The Dimensions of Assortment: A Proposed Hierarchy of
Assortment Decision Making, The International Review of Retail, Distribution and
Consumer Research, 16(3), pp. 333-351, ISSN: 0959-3969. DOI:
10.1080/09593960600697063. Hart contribution: 70%

Retail image: Category behaviour and management

Hart, C.A. and Dewsnap, B. (2001) An Exploratory Study of the Consumer Decision
Process for Intimate Apparel, Journal of Fashion Marketing and Management,
5(2), pp. 108-119, ISSN: 1361-2026. DOI:10.1108/EUM0000000007282. Hart
contribution: 50%.
Dewsnap, B. and Hart, C.A. (2004) Category Management: A New Approach for Fashion
Marketing? European Journal of Marketing, 38(7), pp. 809-834, ISSN: 0309-0566.
DOI: 10.1108/03090560410539267. Hart contribution: 50%

Customer experience of retail environments

Hart, C.A., Farrell, A.M., Stachow, G.B., Reed, G.M. and Cadogan, J.W. (2007a)
Enjoyment of the Shopping Experience: Impact on Customers' Repatronage
Intentions and Gender Influence, The Service Industries Journal, 27(5), pp. 583-
604, ISSN: 0264-2069. DOI: 10.1080/02642060701411757. Hart contribution: 40%
Hart, C.A., Stachow, G.B. and Cadogan, J.W. (2013) Conceptualising town centre image
and the customer experience, Journal of Marketing Management, 29(15-16),
pp.1753-1781, DOI: 10.1080/0267257X.2013.800900. Hart contribution: 50%
2
 It should be noted that the portfolio of eight papers is a focused selection
from a far wider corpus of research output, conducted over a period of 23 years.
Conference presentations, collaborations and other published outputs, while
excluded from the core portfolio, provide relevant evidence along the research
journey and are referenced accordingly. The individual portfolio papers are
distinguished throughout by bold type.

2. Background to the research areas

My first and second degrees in textile design and fashion established an early
graduate career in the textiles industry. I progressed from design to sales and
marketing management roles in the UK East Midlands textiles industry, liaising
with buyers and designers for major retailers such as Marks and Spencer and the
Arcadia Group. This was a logical skills fit as I had also worked in retailing for
seven years, prior to and during my undergraduate studies, followed by 4 months
working in a US retail fashion store on graduation.

By the early 1980s, UK retailers were growing in scale and power within
their channels of distribution. My next career move combined manufacturing with
retail, working for a Leicester knitwear manufacturer diversifying into retailing.
Within three years, the Sweater Shop expanded rapidly from three factory outlets
to 55 branded stores. My interest in retailing renewed, I sought a new challenge
beyond independent retailing. Researching my masters’ dissertation in 1983 and
an external examining position at De Montfort University had developed an interest
in academia. Joining the Business School at Loughborough University in 1992 was
a leap of faith, moving from a creative practical industry role, with little formal
research training. It has been a long but interesting journey, and the following
thesis aims to synthesize the core focus of my research during these 23 years.

Academic research provided an exciting opportunity to broaden my retail

knowledge through researching different retail sectors and organizations. In
particular, as the Sainsbury Retail Fellow, to expand my hitherto focus on the
fashion industry. My initial mantra was to pursue research that would have value to
both practitioner and academic audiences. This philosophy remained throughout

3
the subsequent research projects, as illustrated in the following section, which
outlines the backdrop and rationale for three research areas of retail assortment,
retail image and customer experience.

2.1 Retail assortment: Non-food diversification

At the time of embarking on this body of research, in about 1990 UK grocery

retailing was enjoying a “Golden age” (Wrigley 1991; p.1537), as UK food retail
assortments expanded rapidly during the mid-1980s to 1990s. This expansion was
attributed to rapid market growth by the major grocery retailers, increasing retail
floor space and high competition together with the consumers’ desire for a one-
stop-shop (Wrigley 1991; Wrigley 1994). However, a static grocery market meant
that to fill this floor space, retailers sought new profitable growth areas with non-
food diversifications offering most scope. The negative implications of these non-
food brand extensions were grocery retailers had less authority and expertise in
new general merchandise categories, and space constraints limited the depth of
product variety offered. Retailers also risked alienating loyal customers by diluting
their beliefs associated with the retailer brand and its image as a food retailer
(Roedder et al., 1998). Hence, the role of the retail assortment becomes integral to
the retailer image as perceived by its customers.

Against this backdrop, as the recently appointed Sainsbury Retail Fellow at

the Business School, my induction included meetings with Sainsbury’s senior
management to set a research agenda. Ivor Hunt, Marketing Director, proposed
the initial research question; “What is the limit for non-food diversification in
supermarket assortments?” This line of enquiry commenced with an exploratory
study of 10 retailers (Hart 1993; Hart 1994; Hart 1999 1) followed by an empirical
0F

study of consumers (Hart and Davies 1994; Hart and Davies 1995a; Hart and
Davies 1995b, Hart and Davies 1996a, Hart and Davies 1996b). A project report
was also produced for Sainsbury’s (Hart and Davies, 1995c). It is interesting to
note the conclusions of the report, which indicated that according to their loyal
customers, Sainsbury’s strengths lay in food assortments and any subsequent

1
Emboldened references denote the portfolio papers. See Table 1 for further details.

4
diversification into non-food ranges required careful positioning. Three years later,
Sainsbury’s market leadership was overtaken by Tesco, largely due to the latter’s
aggressive superstore and non-food growth strategy, and their pioneering
Clubcard loyalty scheme. In contrast, Sainsbury’s made three further failed
attempts to introduce fashion into their supermarkets but struggled to find the right
positioning to fit their customer base and store environment. Together, these early
papers presented important new insights into the role of assortments for retail
image and strategy.

2.2 Retail image: Fashion categories

Subsequent research revealed that the assortment strategy problem was not
unique to food retailers. Having previously worked in the textiles and fashion retail
industries provided a valuable contrast to the grocery sector, developing a new
research stream. In 1996, Triumph International, a global brand and major supplier
to Marks and Spencer’s flagship intimate apparel category, commissioned me to
research the consumer perceptions and attitudes of brand images. Marks and
Spencer’s category performance suggested a declining consumer interest and in
response, M&S implemented assortment reduction within the bra category. Yet
conversely, the brand supplier, Triumph, reported steady sales performance from
other retail customers for the same product category.

This line of research investigation developed by exploring in-depth the

consumer perceptions of specific categories and brands within the relatively
neglected intimate apparel research context. Based on a series of consumer focus
groups across the UK, I delivered a confidential research report to the client
company (Hart 1996). Collaborating with Dr Belinda Dewsnap in 1999, we then
analysed the data further, to develop a conference paper and two journal papers in
the areas of consumer buyer behaviour and category management (Hart and
Dewsnap, 2000; Hart and Dewsnap, 2001; Dewsnap and Hart, 2004). These
papers thus filled a gap in understanding the consumer perceptions and behaviour
for an under-researched apparel category.

5
2.3 The customer experience and town centre image

As my research and knowledge of the literature evolved, it became clear that the
assortment is a crucial attribute of store image and a key attractor to retail
patronage (Lindquist, 1974). It was also becoming clear that retail image involved
wider contexts within and beyond the retail store (Stern et al., 2001). Yet by 1996,
very few academics had researched consumer perceptions of individual
departments or categories within stores and the retailer implications for managing
the increasing scale and diversity of assortments (McGoldrick, 2002). The limited
research could be attributed to the speed of retail development at the time, driving
new retail formats and strategy. In particular, the rise of the shopping mall was
attracting a new generation of retail researchers, keen to understand consumer
responses to this emerging format (c.f. Finn and Louviere, 1996; Chebat, Sirgy
and St-James, 2006). This academic attention was in stark contrast to the
relatively neglected area of town centre image research, arguably the earliest retail
location. The latter research gap offered a new context of image study, which
provided a logical affinity with store and category image research, due to the
concentration of retail outlets located in town centre agglomerations.

The Retail Sector Skills Council and the Learning and Skills Council funded
my third industry supported research study in 2004. As part of a wider study into
the skills, training and business issues in the retail industry, this research explored
consumer perceptions of five city centres within the East Midlands region. The
overall study generated a project report, three conference papers and three journal
papers (Hart, Stachow and Reed, 2005; Hart, Stachow and Reed, 2006a; Hart,
Stachow and Reed 2006b; Hart, Dennis and Stachow, 2008; Hart et al. 2007a;
Hart et al. 2007b; Hart, Stachow and Cadogan, 2013). Collectively, this body of
work found that consumers form distinct images of an urban centre, with the
assortment of stores being the primary image attribute. Collaborating with Dr
Andrew Farrell and Professor John Cadogan, we also examined the shopping
experience. Interestingly, the town centre research revealed that consumers’
experiences in diverse town centre environments contribute to the ongoing
process of a “holistic” image creation. Consequently, this thesis illustrates how
consumer perceptions of retail environments evolved from a largely functional,

6
static image perspective, to a more dynamic, experiential conceptualisation,
encompassing the customer experience.

The most recent phase of consumer-oriented research addresses the

holistic customer experience. Funded by the ESRC and six industry partners, my
fourth industry supported study builds on the research foundations developed over
the previous 20 years. Thus, the research included in this portfolio sought not only
to advance academic theory, but also to contribute research to bridge the gap
between academic and practitioner understanding.

This section has documented the evolving rationale for researching the
specific themes of retail assortments, retail image and customer experience. In the
next section, the underpinning literature and the associated linkages between
these three areas are discussed.

3. Literature and theoretical context for the portfolio

The purpose of this section is to provide an overview of the relevant themes from
the literature, which underpin this thesis, and in so doing, the important gaps within
this body of literature will be highlighted. The portfolio of papers included in this
thesis resides in the broad area of retail marketing. Specifically the first literature
theme (i) introduces assortment as a key element of retail strategy and reveals a
lack of empirical research to support strategic assortment decisions. (ii) The next
theme considers assortment within the concept of retail image and identifies the
literature gaps pertaining to image conceptualisation and retail contexts. (iii) The
emerging literature relevant to the customer experience is then introduced. It
should be noted in the portfolio that each individual paper discusses the theoretical
rationale and specific literature supporting the reported research.

3.1 The changing role of assortments in retail strategy

The assortment serves an important function in retailing linking supply with

demand by providing a selective range of merchandise for consumers to
purchase. As a key element of the retail marketing mix, merchandise is the
7
primary reason for a consumer to visit a store, browse or purchase. Within the
retail mix, the assortment operationalises the retailer’s strategy becoming a
strategic positioning tool to attract and retain core customers (Grewal et al., 1999).
Furthermore it is used as a classifier of store formats, creating a “dynamic
dimension of retailing” (Brown, 1990, p.146), as shifting patterns of assortment
shape retail evolutionary theory (Hollander, 1966). It is unsurprising therefore, that
the product assortment is acknowledged as a key attribute of retail image
(Martineau, 1958; Lindquist, 1974; Samli, 1989).

At the time of the reported accelerated expansion of retail assortments in

the 1980s, the academic literature primarily focused on the assortment selection
problem i.e. which specific portfolio of products should a retailer carry within a
finite retail space (Rajaram, 2001). This literature can be sub-categorized under
three broad research streams; retail space management (c.f. Corstjens and Doyle,
1981; Urban, 1998); assortment variety (e.g. Kahn and Lehmann 1991, Boatwright
and Nunes, 2001); and consumer driven assortment research (e.g. Simonson,
1999, Leszczyc and Timmermans, 2001). The common thread linking these three
research streams was the overarching operational focus on managing individual
products within an assortment. Prior to 2001, the literature treated assortment as
one collection of products, with limited research of sub-assortments such as
categories, or departments and their optimization to maximize retailer
performance. Indeed, given the dramatic changes in retailer assortments there is a
noticeable lack of empirical or theoretical support for strategic assortment
decisions such as what mix or width of categories to offer in a retail store. For
example, while McNair (1958) and Hollander (1966) note the strategic relevance of
assortments in repositioning stores over time, 30 years on, very little, subsequent
attention had focused on understanding or describing assortment movement. This
is a critical strategic shortcoming of the retail literature which formed the rationale
for paper 1 (Hart 1999) in the thesis and later, for paper 4 (Hart and Rafiq 2006).

Different levels of assortment exist in retailing with individual products

organized into complementary groups to facilitate management. Category
Management (CM) developed in the 1990s as a retailer-supplier initiative for
managing increasingly complex groups of assortments to achieve enhanced
8
business performance and consumer value. The CM concept began as part of the
Efficient Consumer Response (ECR) strategies “Efficient store assortment” and
“Efficient store merchandising” (Kotzab, 1999). As such, CM provides a
collaborative demand management approach to achieve “Collaborative shopper
value creation” by ignoring traditional product groupings and focusing on how
consumers shop related products in a category (ECR Europe, 2000). CM therefore
provides a more coherent approach to optimizing assortments and controlling the
visual impact of product categories through space allocation to satisfy consumer
needs (Kent and Omar, 2003; Varley, 2006). Thus, the CM process may in turn,
shape consumer perceptions of the assortment. If departments, product groups
and categories exert their own image then Category Management provided a
means for operationalising this. From 2000, the use of CM was more prevalent,
but applied principally to grocery retailing. At that time, few researchers had
studied consumers’ image perceptions of different non-food categories. This
emerging literature gap identified two potential areas of research enquiry, (i) the
scope for category management in a non-food context and (ii) the implications of
category management for consumers. These research areas are pursued further
in papers 5 (Hart and Dewsnap 2001) and 6 (Dewsnap and Hart 2004) of the
thesis.

3.2 Retail Image: Conceptualization and contexts

A considerable body of literature specific to retail image has emerged in the 58-
year period since Martineau’s (1958) seminal paper on store image. The
importance of image is clear; the successful performance of stores is “explicitly or
implicitly related to image management” (Samli, Kelly and Hunt, 1998, p. 27).
Consequently, image and its constituents are used to evaluate retailers’ marketing
performance from the customers’ perspective (McGoldrick 2002, p. 180).
Moreover, image operates as a predictive variable in assessing consumer
patronage to a retail store and is an important component of store choice decision
(Malhotra, 1983). Accordingly, an associated literature stream explicitly examines
the various image elements that influence retail patronage (c.f. Osman, 1993; Pan
and Zinkhan, 2006).

9
 An image is an overall impression of a store as perceived by consumers
(Keaveney and Hunt, 1992) and thus various approaches for analysing, monitoring
and measuring consumers’ images are documented in the literature (Oxenfeldt,
1974; Chowdhury, Reardon and Srivastava, 1998). Specifically, studies of
consumers’ image formation suggest store image is an attitude or set of attitudes
based upon evaluation of salient store attributes (Doyle and Fenwick, 1974;
James, Durand and Dreves, 1976). Store image is deconstructed further into
dimensions, components and attributes, with the latter referring to “the narrowest
most specific constructs” (Hansen and Deutscher, 1977 p. 59). Consequently an
early stream of image research identified image determinants and investigated
store image as a multi-attribute construct (Lindquist, 1974).

Despite increasing academic attention, there is limited consensus about the

salient image attributes and dimensions with researchers disagreeing as to which
specific image variables are most relevant (Samli, Kelly and Hunt, 1998). Notably,
attributes constituting store image often vary in importance and mix depending on
the location, sector, market and context for the shopping trip (Hansen and
Deutscher, 1977; Schiffman, Dash and Dillon, 1977; Davies 1992; Birtwistle,
Clarke and Freathy, 1999). Yet, the retail assortment features consistently as a
key image variable throughout the literature (Hart, 2009).

Image conceptualisation reveals contrasting interpretations in the literature.

Stern, Zinkhan and Jaju, (2001) categorise store image definitions into three
conceptual groups. (i) Studies focusing on functional qualities such as assortment,
price, service and layout (ii) Research emphasising consumers psychological
orientation and cognitive processes, for example in image formation; (iii) Complex
configuration combining functional attributes, consumer perceptions and attitudes
of a retail environment. Notably the first two conceptualisations interpret image as
a static entity and multidimensional construct combining “tangible or intangible
elements that stimulate consumer perceptions” (Stern, Zinkhan and Jaju, 2001, p.
214). Whereas the latter conceptualisation interprets image as a dynamic,
transactional process whereby individuals form images and integrate perceptions
and information about an object or store into a gestalt (MacInnis and Price, 1987).

10
 However, the measurement and operationalisation of image lags behind its
conceptualisation (Hartman and Spiro, 2005). The key issue is there are no
standardized tools or techniques capable of capturing and measuring the image
construct or gestalt (Zimmer and Golden, 1988). While conceptualizations tend to
emphasize the holistic nature of image, measures usually adopt a “piecemeal”,
attribute-by-attribute approach: “in short, an image was believed to be a picture but
was measured with a list” (Keaveney and Hunt, 1992, p. 167).

Later, researchers identify retail image as a more complex paradigm

encompassing the brand, store and corporate images (Burt, 2000; Stern, Zinkhan
and Jaju, 2001). Examining the literature suggests up to six different levels or
contexts for retail image; the individual product brand image, department or
category image, store image, retail chain image, shopping mall image and more
recently, town centre image. Considering that retailing involves managing multiple
image contexts, this suggests two further research gaps; 1) to understand the
potential interrelationships such as congruity between different image contexts,
and 2) the influence of brands upon these image contexts (Grewal and Levy,
2009). These linkages are pursued further in paper 2 (Hart and Davies 1996a).

While store image attracted most academic attention, conceptualisation of

other retail image constructs is less developed. Town centre image research is
particularly sparse when compared with more recent image contexts such as
shopping malls. The image of a town centre exhibits similar characteristics to
store image i.e. a multidimensional concept comprising both tangible / functional
attributes relating to physical features (Bell, 1999); with more intangible attributes
representing its atmospheric qualities (Van Raaij, 1983). It is unsurprising
therefore, that researchers frequently transfer store image measures to the mall or
town centre context (e.g. Gentry and Burns, 1977; Nevin and Houston, 1980).
Nevertheless town centres contain community spaces, public facilities and
atmospheric concerns such as safety that may influence consumer perceptions
and patronage intentions (Oppewal and Timmermans, 1999). The image literature
thus reveals a vital research gap; the lack of empirically established image
variables pertinent to the town centre retail environment (Gautschi, 1981). Without
this, it is difficult to measure, operationalise or manage the consumer’s perceived
11
image of a town centre. This conceptual issue is developed further within paper 8
(Hart, Stachow and Cadogan, 2013) of the thesis.

3.3 Customer experience

The third body of literature underpinning the portfolio of papers is an evolving area
within retail marketing, namely the customer experience. The customer experience
literature originated from aspects of consumption and services marketing
(Holbrook and Hirschman, 1982; Carbone and Haeckel, 1994; Schmitt, 1999),
expanded to branding (Berry and Lampo, 2004; Brakus, Schmitt and Zarantonello,
2009), consumer behaviour research (Pullman and Gross, 2004; Gentile, Spiller
and Noci, 2007) and retail environments (Arnold et al. 2005; Haytko and Baker,
2004; Grewal, Levy and Kumar, 2009). Conceptualised as a psychological
construct, the customer experience is defined as the consumer’s “internal and
subjective response” to any direct / indirect contact with a retailer (Meyer and
Schwager, 2007 p. 118). In a retail setting, direct contact typically entails physical
interactions with retail stores, products or services. The indirect contact may
involve unplanned encounters with representations of the company, products and
services such as advertising or product reviews, which occur increasingly online
(Novak, Hoffman and Yung, 2000; Rose et al. 2012).

Similar to image, customer experience contact results in a “take-away

impression”, formed by the consumer and stored in their long-term memory
(Carbone and Haeckel, 1994). Thus, customer experience overlaps with the image
concept through the customer involvement with the retail environment. Customer
experience is also considered holistic in nature, as it may begin with anticipation
and journey to the store through to the completion of the shopping trip (Gilmore
and Pine, 2002; Haytko and Baker, 2004). Yet, the holistic aspect potentially
delimits the customer experience leading to conceptual and measurement issues,
for example identifying which part of experience to measure; perceptions of
specific elements or the total experience? (Palmer, 2010).

Nevertheless, the two concepts differ. Experience suggests a dynamic,

interactive construct that develops over time; in contrast to the static “overall
12
picture” concept of image in the consumer’s mind (Oxenfeldt, 1974, p.8). The
customer experience is interpreted as a journey involving a series of memorable
interactions or touch points, which combine to form the overall experience (Lemke,
Clark and Wilson, 2011). Notably, consumers “co-create” their unique experiences
due to their active involvement with the environment (Gentile et al. 2007, p. 396).

The customer experience concept is growing in importance and relevance,

due to its reported influence on customer satisfaction and future consumer
behaviour. Prior to 2009, the academic coverage of customer experience literature
was sporadic. However, a special issue of the Journal of Retailing (Verhoef et al.,
2009) sought to remedy this gap by conceptualizing the customer experience in a
retail setting. Other articles develop key characteristics of customer experience,
emotions and particularly the relevance of the customer journey (Pullman and
Gross, 2004) incorporating cues or touch points (Berry, Carbone and Haeckel,
2002).

Despite an increasing attention from academic researchers and

practitioners, the term customer experience remains a nebulous and unclear
concept, having been “widely used and abused”, in its ambiguous application
(Palmer, 2010). Definitions vary from a hedonistic to a utilitarian viewpoint.
Moreover, from a retail environment perspective, there are two important gaps. To
date the customer experience literature refers to the consumer’s interaction with
one brand, retailer or service. While a customer’s experience extends across
multiple channels, the nature of their experience within an agglomeration of retail
stores such as a town centre is unknown. Furthermore, there is limited empirical
evidence regarding the composition of customer experience, or how this may be
captured and measured in a retail setting. These areas for research enquiry were
subsequently developed in paper 7 (Hart et al. 2007a) and paper 8 (Hart,
Stachow and Cadogan, 2013) of the thesis.

The above section summarised the theoretical context for three connected
literature themes and in so doing, it identified a number of research gaps residing
in each literature area.

13
4. The portfolio of papers: Research approaches and contributions
This section introduces the portfolio of eight published papers of which the thesis
is comprised, which were originally written to address the research gaps identified
in the previous section. Before introducing each individual paper, a broad overview
of the research approaches is provided below. It does not describe the
methodologies in detail, as this is discussed as appropriate in the relevant papers.

There was no overarching research philosophy adopted over the 17-year

research period, but rather each study was tailored in its research approach to
match the specific research “problem”. As the studies progressed, the knowledge
and confidence to incorporate more innovative research designs also increased. I
particularly enjoyed the challenge of creating a new research solution to fit the
problem and circumstances. As a result, the eight papers encompass mixed
method approaches to research design that draw on a variety of data collection
methods using a diverse range of data sources as follows.

The findings from two exploratory, qualitative studies are presented

comprising:

1. Depth interviews conducted with key stakeholders in 10 major UK retail

companies, including 13 senior managers from the food and mixed retail
sector (paper 1);
2. Six national focus groups comprising 48 intimate apparel consumers and
using three different data collection methods; a word association exercise,
semi-structured open discussion, and a “blind” brand experiment (papers 5
and 6).

Detailed literature reviews were also undertaken during the research period.
Two of these are reported in the portfolio, covering retail assortment research
(paper 4) and town centre research (paper 8).

The findings from the earlier qualitative papers, informed two larger
quantitative consumer surveys:

14
 3. A consumer intercept survey conducted in 1995, involving 504 respondents
recruited across six major UK grocery retailers (papers 2 and 3).
4. A consumer intercept survey conducted in 2005, involving 536 respondents
in five East Midlands city centres (papers 6 and 7).

This mixed approach demonstrates a rigorous theoretical development from

exploratory through empirical studies, which are advanced in subsequent papers.
Moreover, the innovative approaches adopted in the above studies, deliver richer
research insights, enhancing the wider contribution to the image literature. For
example, in study (2) above, Hart and Dewsnap (2001) and Dewsnap and Hart
(2004) developed a creative qualitative research design incorporating blind tests
and projective techniques to uncover in-depth consumer behaviour and attitudes
towards brand images in an underexplored category. In study 4, Hart et al.
(2007a) and Hart, Stachow and Cadogan (2013) combined quantitative and
qualitative approaches in a consumer image study, to elicit and capture overall
impressions of town centre image. The rationale for the chosen research design
and methodology is specific to the respective study, and this is reported in detail
within the individual paper chapters.

The portfolio of papers, presented in the following sub-sections, showcases

my academic career to date, demonstrating many significant, theoretical
contributions to the retail marketing literature. Evaluating eight published papers,
this research commentary will next summarise the key findings and the
contribution for the individual papers. While each of the papers makes an
individual contribution to the literature, collectively, the thesis provides a significant
body of research output, that contributes to the following three parallel yet highly
connected literature streams, and their associated research gaps, as identified in
the previous section;

1. Retail assortment
2. Retail image
3. Customer experience

15
4.1 Retail assortment: Perceptions and roles
Paper 1: Hart, C.A. (1999) The Retail Accordion and Assortment Strategies: An
Exploratory Study. The International Review of Retail, Distribution and Consumer
Research, 9(2), pp. 111-126, ISSN: 0959-3969.

In the late 1990s, the critical strategic question of which categories a retailer
should carry in a particular store, had not been addressed in the literature,
consequently it provided a point of departure for my assortment research agenda.
Hart (1999) set an important context for the ensuing assortment research by
comparing and contrasting the assortment strategies of food multiples with mixed
retailers by recording the relative change in assortment movement i.e. expansion
and contraction. The data were gathered in 1993 and results initially presented at
UK and International conferences (Hart, 1993; Hart, 1994). The ensuing feedback
then helped shape the findings into a journal paper, which was subsequently
published (Hart, 1999).

The paper’s key contribution is in providing new evidence to support the

Accordion theory (Hollander, 1966), which compared with other cyclical theories in
the literature, such as the Wheel of Retailing, had been neglected and limited to
anecdotal evidence (Brown, 1987). Additionally, it identifies the key factors and
issues associated with diversifying or contracting retailer assortments. Taking the
retail organizational perspective, the exploratory study involved in-depth interviews
with key stakeholders in 10 major retail companies, including 13 senior managers
from the food and mixed retail sectors.

Using the “number of lines” as a proxy measure of assortment, the

assortment movements for a 10-year period were synthesized and contrasted in
the food and mixed retail sectors. Significantly, the findings demonstrated that the
five major food retailers had all diversified and expanded their assortment by 7% -
400%. In contrast, the mixed retailers had all rationalized their assortments during
the same period, deleting between 5%-65% of their product lines. The latter, for
example reducing a store assortment from 75,000 lines to 25,000 lines. The
rationale for these strategies revealed factors critical to successful assortment
diversification i.e. retailers ranked the congruence of a new diversified range with
16
the existing assortment and the retailer image as most important, but omitted to
conduct any image research with consumers at this level. The implications of this
were clear; most of the sample retailers had exited diversified market areas, with
two major retailers returning to the same ‘failed’ market area within two years of
deleting the range.

The findings still have value today as retailer assortment strategies suggest
major shifts with online players such as Amazon demonstrating extreme
diversification by introducing a fresh food offer. In contrast, the post-recession
growth of convenience formats indicate major grocery retailers are contracting
their assortments and space, by exploiting new formats and channels. The latter
reinforces the conclusion that consumer behaviour “will force the supermarkets to
revert to their core business” (Hart, 1999, p. 125).

Notably, the findings distinguished between core and non-core retail

assortments, with the latter offering most scope for non-food assortment
diversification. However, as supermarket food assortments reached saturation, a
recurrent question was which non-food non-core assortments should complement
the core food?

Paper 2: Hart, C.A. and Davies, M.A. (1996a) Consumer Perceptions of Non-Food
Assortments: An Empirical Study. Journal of Marketing Management, 12(4), pp.
297-312, ISSN: 0267-257X.

Hart and Davies (1996a) sought to provide answers to the strategic challenges
facing retailers, relating to assortment diversification, but from from the consumer
perspective. Exploring the retailers’ perceptions of assortments set the scene for a
subsequent quantitative survey into consumer perceptions of non-food
assortments in supermarkets. The key link between this study and the previous
paper is the image congruity of non-food ranges with the core assortment.
Specifically, do consumers distinguish separate images for non-core, non-food
assortments in a supermarket, compared to those of core food? If retailers need to

17
project a synergistic image, how far can supermarket assortments be diversified,
before the store image, as perceived by the consumer, is seriously diluted?

This empirical study focused on supermarkets in part due to the grocers’

non-food diversification strategies, but also the absence of image research for
categories in grocery stores. A structured questionnaire was designed to measure
consumer perceptions of three key variables; degree of fit or congruity, choice or
depth available in each category; and importance for expansion across three
established and three diversified non-food categories. A 5-point semantic
differential scale measured the intensity of perceptions, combined with open-
ended qualitative questions and rankings. Market research interviewers recruited
504 consumer respondents in an intercept survey across six major food retailers.

Capturing for the first time, consumers’ images of supermarket product

categories, the findings reported in Hart and Davies (1996a) showed variation in
the seemingly uniform one-stop-shop supermarket environment. Significantly, the
findings show that consumers perceive different images for core merchandise
assortments compared with diversified assortments in the same store. Specifically,
clothing and entertainment categories exhibited least fit with supermarket core
assortments, due to offering insufficient depth of choice in a supermarket
environment. The paper fills important gaps in the store image literature by (i)
examining image at the product group or category level in supermarkets and (ii)
capturing consumer perceptions of range ‘fit’ or compatibility with the retailer
image.

Another useful contribution of this paper is that consumers appear

subconsciously to divide supermarket assortments into three groups of core, non-
core and peripheral merchandise; the latter group is considered least congruous
with the store image. In an era of expanding assortments, the results suggest
store image is over simplified and a new approach for measuring store image by
core, non-core and peripheral product categories is recommended.

18
Paper 3: Hart, C.A. and Davies, M.A. (1996b) The Location and Merchandising of
Non-Food in Supermarkets. International Journal of Retail and Distribution
Management, 24(3), pp. 17-25, ISSN: 0959-0552.

Diversifying into new product categories provided additional challenges for

retailers within the confined space limitations of supermarkets. The retailers had
confirmed in earlier exploratory research that success of new range
diversifications often depended on appropriate visual merchandising and location
of new products (Hart 1994). Thus, a subsequent paper (Hart and Davies 1996b)
based on the same consumer survey, examined how best to locate and display
non-food assortments in supermarkets. Here consumer respondents evaluated a
variety of different layout and display approaches for different non-food categories.

The existing retailer approach was to include non-food categories within the
main shopping aisles of the supermarkets. However, this reduced customer
experience by discouraging browsing. Locating comparison goods such as
clothing, next to convenience goods within the same format, meant consumers
were less likely to switch to browsing behaviour, due to their habitual shopping
behaviour in a familiar supermarket environment.

The in-store locations for three established product categories, namely,

household, health and beauty aids, and home products, were compared with
display perceptions for three new non-food categories, that is; clothing, stationery
and entertainment. The findings suggested that diversified non-food assortments
dictate different merchandising needs in supermarkets. Notably, 1 in 5 shoppers
preferred entertainment and clothing ranges displayed separately to the traditional
supermarket grid layout. Four of the six categories are considered best displayed
away from the conventional aisle locations suggesting that certain non-food
diversifications do not mix well with food, requiring a more appropriate ambience
conducive to browsing. Providing empirical evidence that diversified non-food
assortments dictate different layout and merchandising approaches in
supermarkets, this paper contributes to the debate emerging in the two previous
papers. As supermarkets expanded in scale and generalized their assortments,
offering a one-stop shop, consumers preferred the variety of shopping formats
19
similar to that offered by the high street. This is perhaps due to supermarket
shopping being traditionally viewed as a necessary but not always pleasurable
experience. Overall, this study made an important contribution to the literature and
highlighted the considerable scope for retail managers to develop more enjoyable
shopping experiences through the creative merchandising of non-food items.

Paper 4: Hart, C.A. and Rafiq, M. (2006) The Dimensions of Assortment: A

Proposed Hierarchy of Assortment Decision Making. The International Review of
Retail, Distribution and Consumer Research, 16(3), pp. 333-351, ISSN: 0959-
3969. DOI: 10.1080/09593960600697063.

As the assortment research progressed, it uncovered a number of theoretical

inconsistencies. Hart (1999) highlighted the limited measures for categorizing and
measuring assortment movement and consistency. Hart and Davies (1996 a & b)
identified that assortments exist at a number of levels within and across stores yet
the terminology and understanding of these within the literature were confused.
The expanding assortments in the grocery and department store sector had
exposed a conceptual gap regarding assortment composition. This ambiguity
constrained any future application of assortment terminology or measurement,
signifying a need to revisit the literature to clarify the strategic role of the
assortment, and to conceptualise the relationship between assortment and image.

Hart (1999) called for a clearer categorization of ranges within the width of
the assortment. Hart and Rafiq (2006) responded by reviewing 50 years of
assortment literature, examining the composition and roles of assortment, while
clarifying assortment terminology. Furthermore, it identifies an imbalance in
operational versus strategic assortment research revealing a “lack of empirical and
theoretical support for the strategic role of assortment” (p. 344). Crucially, the
paper differentiates between the levels of category assortment and store
assortment and their respective dimensions, proposing a hierarchy for assortment
strategy decision making.

Drawing from the two key areas of assortment literature focus, it

recommends assortment strategy is categorized into macro and micro levels using
20
an Assortment level research matrix. The Macro level represents the strategic
store assortment decision regarding the mix of departments or merchandise
classes to offer. The second, Micro level represents the department or category
assortment decision of the number of product lines to carry. Classifying past
research, the matrix illustrates the micro (product-unit) focus attracts most
research attention with only a handful of papers acknowledging the existence and
importance of the macro-level of assortment (Cardozo, 1974; McGoldrick, 1979;
Stewart and Hood, 1983; Hart & Davies, 1996b; Cadeaux, 1997; Hart, 1999).
Through defining assortment, the composition and the dimensions by which it is
described, the paper reveals that a fourth dimension, consistency, is neglected in
the literature. In so doing, this extends the previous notion of image congruence
through the elusive dimension of assortment consistency. Here the consistency of
assortment i.e. the relatedness of merchandise categories from the consumer
perspective is a potential indicator for controlling store image via a retailer’s ability
to match their assortments with the target consumer.

The paper contributes to retail theory by providing a comprehensive

literature review, highlighting assortment ambiguity and confusion due to
inconsistent use of the terms of width, breadth and variety within multiple layers of
assortments. These terms and dimensions are clarified and defined for each
assortment context. Notably, the paper presents a clear route for describing and
quantifying the size and composition of a specific assortment, whether at the store
or category level.

4.2 Retail image: Category behaviour and management

Paper 5: Hart, C.A. and Dewsnap, B. (2001) An Exploratory Study of the

Consumer Decision Process for Intimate Apparel. Journal of Fashion Marketing
and Management, 5(2), pp.108-119, ISSN: 1361-2026.
DOI:10.1108/EUM0000000007282.

The next two papers in the portfolio expand the non-food context, probing deeper
into category image via consumer behaviour and Category Management in a

21
fashion apparel context. Specifically, Hart and Dewsnap (2001) explore in-depth
the in-store consumer decision process for the intimate apparel (IA) category. The
rationale for this study was to investigate consumer behaviour for a complex, yet
highly personal category of IA, the bra. Surprisingly, only two previous studies
(Richards and Sturman, 1977; Laurent and Kapferer, 1985) had focused explicitly
on this category, yet industry and secondary literature suggested that IA presents
considerable challenges to the consumer and retailer alike, warranting its own
research agenda. An exploratory study was designed based on the high
involvement decision process (Engel and Blackwell, 1982), in order to gain a
‘holistic’ and in-depth perspective of bra buyer behaviour and implications for
retailers. Six focus groups involving 48 UK consumers utilized projective
techniques, open discussion and blind test experiment were conducted nationally
in the UK. The findings revealed that while retailers consider the product class to
provide hedonic pleasure value, the consumer’s purchase decision process is the
reverse, even unpleasant. This is in stark contrast to the grocery retailers seeking
to create a more pleasurable shopping experience in a convenience environment.
Notably, the study finds that due to high involvement, IA behaviour relates to
perceived risk, self-image and self-concept. For example, consumers seek
congruence between their self-image and image of store staff and other
consumers. This extends the previous notion of (retailer driven) image congruence
between department image and the retailer brand image (Hart 1999; Hart and
Davies 1996a), to the individual consumer in the retail store environment.
Moreover, the personal nature of the bra suggested the majority of women prefer
to shop alone for bras unlike the social interaction associated with outer fashion.

Another key contribution is this is the first study to examine the complete
decision process of the bra buyer and the associated implications for retail
management. Given the importance of function and fit, the findings show that
assortment depth and stock availability are crucial factors influencing consumer
purchase (e.g. style and size) along with brand image and style loyalty. Thus,
previous insights from retailer assortment strategy (Hart 1999; Hart and Davies
1996a; Hart and Rafiq 2006) provide useful implications for assortment depth
carried by retailers that will help strengthen the category and store image.

22
Paper 6: Dewsnap, B. and Hart, C.A. (2004) Category Management: A New
Approach for Fashion Marketing? European Journal of Marketing, 38(7), pp. 809-
834, ISSN: 0309-0566. DOI: 10.1108/03090560410539267.

Hart and Rafiq (2006) highlighted the assortment’s emerging role in the practice
of Category Management (CM). With continued range diversifications, there is a
point at which assortments reach saturation and consumer perceptions of
assortment variety then become more important than actual variety. For instance,
Broniarczyk, Hoyer and McAlister (1998) found reducing the number of items
could simplify consumer choice and increase sales for the category. CM uses in-
depth information on the consumer to support joint marketing decisions for the
category. While electronic point of sale (EPoS) systems provide volume sales data
that is the “what” of category sales, CM depends on understanding the “why” and
“how” of consumer behaviour. CM had been applied predominantly to grocery
retailing, thus Dewsnap and Hart (2004) examined whether this established CM
approach could be successfully transferred to fashion marketing. A key
contribution of Dewsnap and Hart (2004) is demonstrating how consumer
research findings can inform the CM process for a product category in the fashion
sector. Fundamentally, CM Informs retailer approaches to managing assortments
via usage-related categories, which offers significant benefits to fashion marketing
by influencing consumer behaviour in-store. Deploying an in-depth understanding
of the consumer for a particular category should therefore enable suppliers and
retailers to move beyond “improved efficiency to improved consumer experience”,
through inspired consumer solutions Gruen (2002, p.18). Drawing from the same
IA data set, the consumer responses were analysed and structured according to
the category management research framework (Johnson, 1999). The objective
was to understand customer motivations and the impact of point of purchase
marketing mix variables.

The findings indicated that consumers typically sought a garment based on

usage, or wearing occasion, in contrast to the retailer’s product based view of
categories - focusing on size and construction. A key challenge for IA retailers is
managing an assortment of up to 60 sizes across a variety of styles in a range of
23
colours. Thus, an important finding of this study is that by reducing the variety and
number of SKUs within the functional bra category, the retailer is perceived to be
less effective in meeting customer’s needs. Surprisingly for an involved product,
consumers undertook extensive information searches even when purchasing a
replacement garment, largely due to discontinuation of styles and lack of size
consistency. Limited availability in the right style / size then led to consumers
convenience shopping and bulk buying when their chosen style was available.

4.3 Customer experience of retail environments

Paper 7: Hart, C.A., Farrell, A.M., Stachow, G.B., Reed, G.M. and Cadogan, J.W.
(2007a) Enjoyment of the Shopping Experience: Impact on Customers'
Repatronage Intentions and Gender Influence. The Service Industries Journal,
27(5), pp. 583-604, ISSN: 0264-2069. DOI: 10.1080/02642060701411757.

A common thread emerges in the previous two papers i.e. the need for and
application of in-depth consumer insights to deliver a better offer and an improved
customer experience, whether in store or another retail environment. Arguably, in
order to provide a focused offer, town centres similarly need to address the same
needs if they are to attract and retain customer patronage. Hart et al. (2007a)
specifically focuses on how attractiveness of the town centre and enjoyment of the
shopping experience can influence consumer behaviour and ultimately their re-
patronage of that location. The study involved a comprehensive consumer
intercept survey of 536 respondents in five regional centres, assessing four key
dimensions of the shopping experience; accessibility, environment, atmosphere
and service personnel. Forming part of a wider funded study into the East
Midlands retail sector, this paper launched a new and very productive stream of
town centre research.

The findings showed that enjoyment of the shopping experience is reflected

in the above stated dimensions and has a significant, positive influence on
consumers’ repatronage intentions to town centres. However, contrary to previous
gender based studies of shopping behaviour (Fischer and Arnold, 1994), we found
a stronger relationship between shopping enjoyment and patronage for males,

24
than for females. Indeed, we suggested that males’ propensity to be decisive,
efficient and time saving in their shopping behaviour, leads to greater loyalty to a
shopping location. For women, enjoyment appears related to shopping as a leisure
activity, including browsing, shopping around and availability of choice, which may
reduce their loyalty to a specific retail location. The paper delivered important
theoretical contributions to the literatures relating to gender differences in
shopping and repatronage intentions. Further, it revealed early indicators that
consumers behave differently in a town centre compared with, for example
shopping malls. Accordingly, consumers’ perceptions of town centre image will
also differ to other retail locations.

Moreover, the findings suggested that shopping enjoyment may influence

the time and money consumers are willing to spend in a shopping visit. The paper
offers implications for managing consumers’ experiences in a town centre with a
need for town management partnerships “to coordinate the shopping experience
...so that customers’ enjoyment is maximised” (p. 599). Notably, targeted
marketing activity could build on shopping enjoyment to encourage greater
patronage to town centres. These findings and recommendations were later
substantiated with evidence from a subsequent ESRC funded town centre study
(2014). Indeed, this paper provided the early foundation for three funded research
awards and an impact case study (Hart and Stachow, 2013).

Paper 8: Hart, C.A., Stachow, G.B., and Cadogan, J.W. (2013) Conceptualising
town centre image and the customer experience. Journal of Marketing
Management, 29(15-16), pp. 1753-1781, DOI: 10.1080/0267257X.2013.800900.

Hart et al. (2007a) raised additional questions about how consumers perceive the
image of town centres, considering the wider aspects of the public domain,
physical environment and diverse mix of leisure, retail and services. Hart,
Stachow and Cadogan (2013) aimed to examine the composition of town centre
image and the constituent parts identified in the literature. Providing the first
comprehensive review of the published town centre image literature, the meta-
analysis evaluated the image attributes and dimensions employed. This initial
25
contribution highlighted that reported numbers of town centre dimensions varied
considerably from three to nine dimensions, with limited consistency between
dimensions and their interpretation. Importantly, studies using store image
measures to assess town centre patronage provided conflicting results.
Concurring with Gautschi (1981), we argued that measures based on pre-specified
attributes from other image contexts may lead to attribute omission by
representing the researcher’s assumptions, as opposed to the consumer’s image
perceptions. The image literature confirmed the need to measure image specific to
the retail context and in so doing, exposed a crucial theoretical gap – the lack of
empirically established image variables pertinent to the town centre environment.

Indeed, criticisms of “pre-selected, imposed image attributes” suggest that

structured measures fail to capture the gestalt of image. The paper’s next
contribution to theory is proposing a model of town centre image, to facilitate
image conceptualisation. In one of the few studies to elicit empirically the
consumer descriptors of store image, Zimmer and Golden (1988) found that
consumers view store image in terms of both attributes and “global impressions”.
Adapting the Zimmer and Golden methodology, our study then elicited 2013 open-
ended descriptors of town centre image from the same data set of 536
respondents. Extensive content analysis and replication by six independent coders
generated a final set of 38 categories contained within eight dimensions of town
centre image. Significantly, we identified new, consumer-generated dimensions
and attributes that underpin a broader image construct than previously assumed.
Interestingly, our results revealed the largest, new dimension of “global
impressions”, which reinforced Zimmer and Golden’s (1988) finding, and
confirmed the gestalt nature of town centre image. This represented a first attempt
to conceptualise a more holistic image, in a town centre context, supporting future
image operationalisation.

Yet, customer perceptions of a location image do not operate in isolation

and develop through experience of that location (MacInnis and Price, 1987). The
emerging customer experience concept (Verhoef et al. 2009) suggested an
overlap with image, and a secondary objective was thus to explore customer
experience as a guiding framework for conceptualising the wider town centre
26
context of image. In summary, the paper delivered important new contributions to
research, theory and practice by: (i) Proposing a model of town centre image; (ii)
Identifying empirically the consumer-generated dimensions and attributes driving
town centre image; (iii) Revealing the neglected psychological, intangible elements
that reinforce the holistic, gestalt of town centre image; and, (iv) Confirming town
centre image as a broader construct and the need to measure image specific to
the retail context. Fundamentally, the image construct provides the basis for
understanding and managing the customer experience in the town centre.

5. Discussion of the Portfolio Papers: Contributions, limitations and

future research agenda.

Each of the research themes discussed above has vital relevance to retail
marketing strategy and collectively the three themes interlink at different levels in
the retail environment. This section summarizes the overall contribution of the
portfolio, highlighting the linkages between the papers and research themes,
including associated follow on work. The section concludes with a summary of the
limitations inherent in this body of work, and an agenda for future research.

5.1 Summary of research contributions

When commencing the assortment research in this thesis, the existing literature
focused on analysing and managing individual products within the overall
assortment, for example which SKUs should be carried within a store (Simonson,
1999; Rajaram, 2001). At that time, there was limited empirical support for
assortment strategy decisions; understanding of consumers’ perceptions of the
mix of categories carried in a store; or any consistent approach to describing
assortment composition and movement.

Spanning a 10-year period from 1996-2006, the assortment research

included in this thesis sought to address these gaps. By capturing retail
assortment movement, Hart (1999) demonstrated new evidence for the retail
accordion theory by analysing the retailers’ “accordion-like expansions and
withdrawals” (McGoldrick, 2002, p. 159). While the retail accordion may be
27
considered inexact as an explanatory model, factors indicating the limits of
assortment diversification identified by Hart (1999), provide clear insights to the
boundaries and risks of merchandise scrambling, “where continual expansion of
new and unrelated lines may result in eventual assortment contraction” (Fernie,
Fernie and Moore, 2015, p. 35).

Recent literature and texts explain practitioner buying policy and practice,
yet offer few new theoretical insights for guiding assortment strategy. Thus, Hart
and Rafiq (2006) fills a crucial theoretical gap in the literature, by clarifying
assortment theory (Varley and Rafiq 2014; Goworek and McGoldrick 2015, p. 91)
and for categorising the macro and micro levels of assortment research. From a
macro assortment perspective, Samli (1989) recommends internal consistency
across the width of an assortment is essential for the retailer to project a uniform,
synergistic image. Yet the means for measuring and managing assortment
consistency to match existing or new assortments with the target consumer
remains unresolved in the literature.

In addition to contributing to the literature gaps identified in Section 3.1, the

first four papers clarify the strategic relationship between retail assortment and
image. Specifically, Hart (1999) lends support to assortment decisions by
demonstrating the need for image congruence of a new diversified range with the
existing assortment and the retailer image. Hart and Davies (1996a and 1996b)
develop the link by identifying that consumers form perceptions of non-food
categories in supermarket environments. For example, Hart and Davies (1996a)
find that consumers perceive different images for non-food categories within
supermarkets, despite being within the same supermarket format. Yet the way in
which the non-core categories are located and merchandised in a supermarket
can affect how consumers perceive the category (Hart and Davies 1996b).
Finally, as noted above, Hart and Rafiq (2006) recognize that as assortments
diversify and shift over time, this signifies a need to manage the consistency of
different category images within the width of the assortment.

Despite the increasing availability of retail data, assortment planning

remains a challenging task for retailers (Mantrala et al. 2009). Indeed, the growth
28
of online assortments adds more complexity, particularly for multi-channel
retailers. Recent literature therefore extends the stream of micro level research
(Kahn, Weingarten and Townsend, 2013) as researchers investigate perceived
variety in an online store and consumer centric perspectives of assortment
(Chang, 2011). Indeed, responding to the call by Hart and Rafiq (2006) for a more
sensitive form of assortment measurement, Bauer, Kotouc and Rudolf (2012) build
on our assortment research to develop a grocery assortment perception scale
(GAP) incorporating the consumer perspective in an assortment measure.

By 2000, category assortments were playing an increasing role within store

image (Ailawadi and Keller, 2004). The portfolio theme of “Retail image: Category
behaviour and management”, shifts the focus from the retailer-supply perspective
to the consumer-demand perspective, examining consumer behaviour for a high
involvement product category and the reciprocal influence of image (Hart and
Dewsnap, 2001). This in-depth approach captured for the first time, unique
insights of consumer behaviour and attitudes to the intimate apparel (IA) bra
category, previously overlooked by the retailer and academics. The findings
contrasted particularly with the hitherto image focus on supermarket categories,
illustrating a new application of image congruence i.e. between customers’ self-
image, image of store staff and other consumers. Thus, the paper contributed to
addressing the literature gaps identified in Section 3.2, regarding understanding
the relationships between different image contexts such as category image and
store image, and the influence of brands to category image. Summarising the most
significant contribution, Hart and Dewsnap (2001) were; “the first to explore the
consumer decision process for the bra and its relation to women's involvement and
self-concept gave very important insights to the psychological elements acting as
mediators throughout the process of bra purchasing” (Yip et al. 2011, p. 1106).
Notably, the paper stimulated a new stream of literature [36 citations 2] that
1F

included consumer behaviour research for IA and fashion categories in USA and
Australia, Hong Kong and Korea, to South Africa and Portugal. A number of these
studies applied aspects of the Hart and Dewsnap (2001) research approach (c.f.
Risius et al. 2012; Robinson and Doss, 2011), and others were based explicitly on

2
Per Google Scholar

29
the Hart and Dewsnap (2001) IA qualitative research framework (Law and Yip
2010; Law, Wong and Yip, 2012).

The previous paper suggested implications for retail management, which

are explored further in Dewsnap and Hart (2004). Collecting in-depth consumer
data for the intimate apparel category provided ideal opportunity to analyse the
potential for transferring the category management approach to the fashion sector.
The study findings established that consumer usage and behaviour for the mostly
functional, IA category differed to outer fashion categories. Indeed, it showed that
category management has significant value to usage related fashion categories to
facilitate i) optimizing assortments, sales space and display; ii) extending time for
wearer evaluation of new products, iii) stock management responses to maximize
product availability. Fundamentally, the study demonstrated the importance of
consumer research to inform the effective management of fashion merchandise
categories and to improve customer experience in-store. The contribution of these
two papers was three fold; delivering consumer behaviour insights for the brand
supplier, improved category understanding for the retailer, combined with impact
on theory (Varley, 2006; Goworek and McGoldrick, 2015).

Reviewing the extant literature, Dewsnap and Hart (2004) noted limited
empirical research studies on CM. Since then, the number of published articles
has increased, yet with few additional empirical contributions. Holweg, Schnedlitz
and Teller (2009) categorises the existing CM literature into seven research
streams, noting the generic CM approach may not be applicable to all retail
product categories. By presenting one of the first empirical consumer studies to
apply CM to fashion categories, Dewsnap and Hart (2004) encouraged a new
literature stream of CM research in the fashion and grocery sector [60 citations].
Specifically, Sheridan, Moore and Nobbs (2006) develops the above application of
CM by examining its role and application in the Fast Fashion retail sector.
Furthermore, the brand manufacturer-retailer dynamic and power in CM
relationships are explored by various authors in Finland, Sweden and Australia
(Lindblom and Olkkonen, 2008; Lindblom et al. 2009; Glynn, 2007).

30
 The thesis examines four different image contexts. Starting with store
image, it probes deeper into category and department image (see also Hart and
Rafiq, 2004), before, finally, expanding to town centre image. As noted in Section
3.2, the town centre is the most complex yet underexplored image context as it
encompasses all other levels of retail image, from product brands, categories,
retail stores, to in-town shopping malls. Hart et al. (2007a) therefore filled a crucial
literature gap by examining the factors contributing to consumers’ town centre
image and its influence on consumer behaviour and patronage decisions [80
citations]. The paper particularly demonstrates the emerging importance of
experiential factors such as enjoyment of the shopping experience.

The portfolio papers illustrate the two most cited characteristics of retail
image; comprising tangible or functional attributes relating to physical features,
and intangible attributes representing atmospheric or psychological features
(Martineau 1958). Papers 1-4 demonstrate the assortment’s functional role in
image, fulfilling utility of the shopping experience. Yet as Hart and Rafiq (2006)
suggest, retailers also need to focus on improving the experiential aspects of store
visits, to increase the entertainment value of shopping. Papers 5-8 explore further
these intangible, more experiential factors. The existing image literature
emphasizes the effects of physical design elements or atmospherics on
customers’ behavioural responses in the retail location (Hart et al. 2007a).
However, experience is a broader construct and prior to 2013, lacked
conceptualization in the wider retail location.

The findings by Hart, Stachow and Cadogan (2013) reveal a multi-faceted

image construct and the neglected psychological, intangible elements of retail
image. By eliciting empirically the image variables pertinent to the town centre, it
addresses a key literature gap and shows that the literature emphasis on
measuring the largely tangible, functional dimensions of town centre image is
overstated. Indeed, since functional measures do not sufficiently capture the
emotional responses leading to image perceptions (Bell, 1999), these findings
have important implications for measuring and operationalising all retail image
contexts. As such, this paper builds upon and forms a ‘capstone’ to the previous
seven papers in the portfolio.
31
 The portfolio research theme of customer experience, emerged from the
image research and is still evolving, through the town centre context. Hart,
Stachow and Cadogan (2013) conceptualises town centre image as involving the
customer’s cognitive / physical, sensory / affective and social image responses to
the totality of the town centre experience. Papers 7 and 8 thus develop and
contribute to this emerging body of literature. Hart et al. (2007a) introduces the
notion of the customer shopping experience in town centres, finding a stronger
relationship between shopping enjoyment and patronage for males than females.
Subsequent literature that builds on the Hart et al. (2007a) findings, investigates
the impact of gender differences in shopping enjoyment (Brosdahl and Carpenter,
2011; Kotzé et al. 2012) and heterogeneity of gender perceptions of shopping
environments (Teller and Thompson, 2012). More recent authors study the
antecedents of retail customer experience (Bagdore, 2013), and experience
perceptions and outcomes (Foster and McLelland, 2015).

Examining the town centre context reveals potential relationships across

other image contexts. Figure 1 illustrates visually the highly interconnected nature
of the research themes, which permeate the eight papers presented within this
thesis. Store image is the link between the customer experience and assortment.
The retail store environment is where the experience takes place (Burt and
Davies, 2010) and includes different levels of assortment and categories.
Importantly, as identified by Hart, Stachow and Cadogan (2013), store images
transfer and influence consumers’ experiences and perceptions of the town centre.

Research supplementary to the portfolio papers, confirms the assortment’s

pivotal role in retail image, underpinning retail strategy. Hart (2009) demonstrates
that assortment is the most consistent factor across all contexts of retail image
including the category, department, store, mall and town centre images. Further,
the paper proposes that different image contexts may be interlinked through the
assortment dimension to help achieve image congruence.

The customer experience research theme integrates the assortment and

store image research. The consumers’ perceptions of stores and assortments
translate into memorable touch points in their experience journeys. This research
32
theme thus complements the overtly functional perspective, with the experiential
approach. As retailers and town centres compete increasingly with other locations,
formats and online shopping, the experiential dimension offers most potential for
competitive advantage (Anteblian, Filser and Roederer, 2014).

Cutting across the three themes is the need for retailers to manage and
control image, assortments and the customer experience, represented in the
overlapping segment of “management”. Finally, the notion of consumer
perceptions, underpins the three themes implicitly. The unifying characteristic
throughout the research is how consumers perceive different retail environments
and their assortments.

This section has discussed the linkages across the portfolio and
summarised the collective contribution of the eight papers. In so doing, it delivers
the final contribution of the thesis; the connectivity across what were previously
assumed to be separate and distinct research domains, namely retail image, retail
assortment and customer experience

33
 Retail
Categories Image
Town
Hart & Davies Hart, Stachow & centres
1996a Cadogan 2013

Hart & Davies

Hart 1999 1996b
Hart et
Dewsnap
al. 2007a
Hart & Rafiq & Hart 2004
2006 Customer
Retail Experience
Assortment Hart &
Dewsnap
2001

Management

Figure 1 The Interlinking Research Themes

5.2 Research limitations

Methodologically, the earlier thesis papers were limited by the available

conceptualisation and measures at that time. Tracking or synthesizing assortment
movement is a fine art, at best inaccurate due to the inexact measure of “number
of lines” and sample of assortments and retailers studied in Hart (1999). Capturing
and measuring image is similarly imprecise and the early reported studies (Hart
and Davies 1996a; Hart and Davies 1996b) adopted existing image techniques
such as Semantic Differential Scaling (SDS) to rate consumers’ image perceptions
for different categories across retail stores. While considered the most popular
image scaling technique, SDS is also criticised for only measuring the “objective
reality” represented by image attributes (Keaveney and Hunt, 1992 p.167). Indeed,
many shortcomings of image measurement are due to ill-conceived attitude scales
that force responses to items that do not necessarily represent the image being
captured (Gautschi,1981). After 60 years of image studies, Hart, Stachow and
Cadogan (2013) questioned the traditional retail image approaches, thus setting
34
the scene for a new phase of image research – a dynamic, experiential approach
that encompasses perceptions of the total experience.

This thesis has focused on image perceptions of different retail

environments and it is not within its scope to discuss the role or perceptions of
brands within retail image or the consumer formation of image. However, work
additional to the portfolio examines these and other related areas (Hart, 2009;
Stachow and Hart, 2010; Hart, Smith and French, 2011).

Throughout, the emphasis has remained on perceptions associated with

physical retail environments, as opposed to online retailing or virtual experiences.
Clearly, the rapid growth of all forms of digital media and online consumer
behaviour will have implications for retail assortments, image and customer
experience. These aspects are discussed further in the next section.

5.3 Impact and research agenda

The town centre customer experience research was timely due to the post-
recession impact of a declining UK high street and the urgent call to understand
changing consumer behaviour in town centres (Portas, 2011). While the retail
industry generates considerable market research data at an aggregate level,
detailed and individualised customer data is not gathered or applied at the town
centre level. Hence, understanding of consumer behaviour in town centres is
lacking and with this, the means for capturing and measuring consistently the
consumers’ perceptions, in order to increase town centre attraction, patronage and
performance (Wrigley and Lambiri, 2015).

The last two thesis papers formed the basis for a REF impact case (Hart
and Stachow, 2013), and three successful grant applications focusing on the
“Customer experience of town centres”. The first two grants were funded by the
ESRC and supported by six industry co-investment partners (Argos, Association of
Convenience Stores, Boots, British Retail Consortium, Action for Market Towns
and the Loughborough BID). The Loughborough University Enterprise Project

35
Group funded the third study involving a location analytics company collaborator.
In total, my town centre experience research generated £235,575 of research
funding in the three years 2012-2015. The subsequent output (c.f. Hart et al.,
2014; Hart, 2015) created wide impact amongst Government Departments, Local
Authorities, Industry bodies, major retailers, local town centre partnerships and
BID teams.

The above funding resulted in three research studies involving:

1. An exploratory study investigating how 180 consumer respondents

experienced six different town centres during a 4-week in-depth “tracking
study” across six UK locations.
2. Developing a scale to capture and measure the functional and experiential
touch points of the customer experience using a confirmatory study of 1530
consumer respondents, across 10 town centres.
3. Testing and implementing a practical town centre consumer “tool” to capture
and track customer experience consistently on a large scale i.e. using a
mobile app and cloud based impact dashboard for easy administration and
reporting.

Thus, as my research evolved from the hitherto functional to experiential

perspective of image and the customer experience, it filled an immediate, real-
world research need, generating greater interest, attracting the wider, practitioner
audience. From a theoretical perspective, the above recent research contributes to
the final two literature gaps identified in Section 3.3. Hart et al. (2014) draws
explicitly from the findings of Hart, Stachow and Cadogan (2013) to investigate
the nature of the customer experience within and across an agglomeration of retail
stores in a town centre. Notably, the three studies confirm that the customer
experience journey comprises both functional and experiential touch points. This
combined with current town centre research activity and emerging outputs provide
new, empirical evidence regarding the composition of customer experience and
how this may be captured and measured in a retail setting.

36
 Each of the three research themes offers potential for further research and
it is worth noting particularly interesting areas emerging from the thesis, pertinent
to current retail marketing research. The 17 years of the research period
witnessed the fastest growing phase of online assortments, benefitting from
seemingly unlimited physical shelf space combined with lower operating costs and
barriers to entry. Here store image plays an important role differentiating a
physical retailer offer from its online competition. For multichannel retailers, the
need for image congruence between a retailer’s online and offline assortments
becomes more relevant. An earlier parallel collaboration applied my research
insights in an online context (c.f. Hart, Doherty and Ellis-Chadwick, 2000; Doherty,
Ellis-Chadwick and Hart, 2003) which suggests future research direction. For
example; in the current environment, with contracting retail space and competitive
pressure to diversify into profitable markets, at what point should physical
merchandise assortments transfer online? Under what circumstances do
consumers consider online diversifications incongruent to the core assortment and
retailer image?

The customer experience journey involves both physical and virtual

interactions across multiple channels. The ESRC research explored the role of
digital in the town centre and the extent to which online and digital activity is
replacing or enhancing the physical shopping experience. The findings revealed
the town centre digital experience is underdeveloped and as a result, the town
centre customer experience is fragmented, perceived as a collection of separate
experiences with individual retailers (Hart et al. 2014; Hart and Laing, 2014). How
can digital shape future consumer journeys in town centres?

Echoing the academic literature, retailers and high streets typically focus on
consumer interactions with the physical, tangible aspects such as parking and
access. Yet consumers’ reasons for visiting retailers and town centres extend
beyond purely functional outcomes, to include satisfaction, emotions and
enjoyment. This offers further potential for customer experience research.
Specifically, what is the relationship between touch points and how can the
customer experience journey be measured? There is need to quantify consumers’
perceived importance of each touch point for different types of experiences. This
37
knowledge would then inform retailers and town centre stakeholders of priority
touch points, guide their strategy at a local level and facilitate measuring change in
consumers’ perceptions over time. Understanding the value of experiential touch
points to specific market segments, would then help create distinctive customer
experiences.

Finally, digital provides the means for linking the touch points of the
customer experience. It also offers scope to capture, measure and influence
positively, the consumers perceptions and images of retail environments and
further work is underway to investigate how consumers’ combine their physical
and virtual experiences.

5.4 Concluding remarks

This commentary provided an overview of the theoretical context for the thesis. It
evaluated three interlinking research areas of retail assortment, retail image and
customer experience and the associated gaps within each literature stream. Eight
published papers comprising the thesis were then introduced. This followed with a
summary of the rationale, key findings and contribution for each individual paper.

Collectively, the thesis delivers a significant research contribution. It

focuses on a core area of research activity, investigating perceptions of
assortments and image in different retail contexts and how this contributes new
understanding to knowledge and theory. The first four papers clarify strategic
relationships between retail assortment and image. Examining consumer
behaviour and perceptions for an unexplored apparel category, the following two
papers identify new perspectives of image congruence, relationships between
category and store image and implications for category management. The final
two papers consolidate all previous contexts in a multi-faceted retail image
construct, emphasising previously neglected psychological image elements.
Fundamentally, consumers’ experiences in town centre retail environments
contribute to the ongoing process of “holistic” image creation.

38
 As retail environments change, consumers’ perceptions of these have also
changed. Consequently, this thesis highlighted particularly how consumer
perceptions of retail environments evolved from the largely functional, static image
perspective, to a more dynamic, experiential conceptualisation, encompassing the
customer experience. It is vital that future research identifies how to capture and
understand how these perceptions translate into actual behaviour, in order to
create appropriate marketing strategies to maximise customer value.

References: Hart

Dewsnap, B. and Hart, C.A. (2004) Category Management: A new approach

for fashion marketing? European Journal of Marketing, 38(7), pp. 809-
834, ISSN: 0309-0566. DOI: 10.1108/03090560410539267.
Doherty, N., Ellis-Chadwick, F. and Hart, C. (2003) An analysis of the factors
affecting the adoption of the Internet in the UK retail sector. Journal of
Business Research, 56(11), pp. 887-897, ISSN: 0148-2963. DOI:
10.1016/S0148-2963(01)00275-2.
Hart, C.A. (1993) The determination of retail product range boundaries. In Davies,
A. M., Finlay, N. P., Hart, C., Kirkup, M., Rafiq, M., Saker, M.J., Saunders,
A. J., Smith, G. I., Wright, L.T. (eds.) Emerging Issues in Marketing,
Proceedings of the 1993 Marketing in Education Group Conference,
Loughborough University, p.483.
Hart, C.A. (1994) A study of merchandise assortment strategies within major
multiple retailers. In Recent Advances in Retailing and Services Science
Conference, Book of Abstracts from the 1994 Canadian and European
Institutes of Retailing and Services Studies Conference, Banff, Canada,
p.41.
Hart, C.A. (1996) A study of consumer perceptions of bras. October. Confidential
client report for Triumph International UK. 25 pages.
Hart, C.A. (1999) The retail accordion and assortment strategies: An
exploratory study. The International Review of Retail, Distribution and
Consumer Research, 9(2), pp. 111-126, ISSN: 0959-3969.

39
Hart, C.A. (2009) The Role of the Assortment in Retail Image. In The Academy of
Marketing Science /ACRA Proceedings of the Twelfth Triennial National
Retailing Conference, New Orleans, USA, ISBN: 0-939783-90-8.
Hart, C.A. (2015) Chapter 8: In Wrigley, N. and Lambiri, D. (eds.) British High
Streets: from crisis to recovery? A comprehensive review of the evidence.
Southampton University.
Hart, C.A. and Davies, M.A. (1994) Consumer perceptions of non-food ranges in
multiples. In Proceedings of the Marketing Education Group 1994 Annual
Conference, University of Ulster, p.466.
Hart, C.A. and Davies, M.A. (1995a) Supermarket product location: Are the right
products in the right place? In Proceedings of the Eighth International
Conference on Research in the Distributive Trades, Bocconi University,
Milan, pp. A4.1-A4.7.
Hart, C.A. and Davies, M.A. (1995b) Spatial perceptions of retail stores: Are
retailers being left on the shelf? In Jobber, D. and Uncles, M.E. (eds.)
Proceedings of the Marketing Education Group Conference, University of
Bradford, p.939, ISBN: 1 85143 131 4.
Hart, C.A. and Davies, M.A. (1995c) Consumer perceptions of non-food ranges in
supermarkets. September. Confidential client report for J. Sainsbury plc, 19
pages.
Hart, C.A. and Davies, M.A. (1996a) Consumer perceptions of non-food
assortments: An empirical study. Journal of Marketing Management,
12(4), pp. 297-312, ISSN: 0267-257X.
Hart, C.A. and Davies, M.A. (1996b) The location and merchandising of non-
food in supermarkets. International Journal of Retail and Distribution
Management, 24(3), pp. 17-25, ISSN: 0959-0552.
Hart, C.A. and Dewsnap, B. (2000) Understanding the bra consumer: Implications
for retailer product positioning. The European Institute of Retailing and
Services Studies, Seventh International Conference on Recent Advances in
Retailing and Services Science, Sintra, Portugal, July.
Hart, C.A. and Dewsnap, B. (2001) An exploratory study of the consumer
decision process for intimate apparel. Journal of Fashion Marketing
and Management, 5(2), pp.108-119, ISSN: 1361-2026. DOI:
10.1108/EUM0000000007282.
40
Hart, C.A, Dennis, C. and Stachow, G.B. (2008) Defining and measuring the
attractiveness of town and city centres. In Fifteenth International
Conference on Recent Advances in Retailing and Services Science, The
European Institute of Retailing and Services Studies, Zagreb, ISBN: 978-
90-6814-172-6
Hart, C.A, Doherty, N.F. and Ellis-Chadwick, F.E. (2000) Retailer adoption of the
Internet: Implications for retail marketing. European Journal of Marketing,
34(8), pp. 954-974, ISSN: 0309-0566.
Hart, C.A., Farrell, A.M., Stachow, G.B., Reed, G.M. and Cadogan, J.W.
(2007a) Enjoyment of the shopping experience: Impact on customers'
repatronage intentions and gender influence. The Service Industries
Journal, 27(5), pp. 583-604, ISSN: 0264-2069. DOI:
10.1080/02642060701411757.
Hart, C.A., and Laing, A. (2014) The consumer journey through the high street in
the digital era. In Wrigley, N. and Brookes, E. (eds.) 2014. Evolving high
streets: resilience and reinvention - perspectives from social science. ESRC
/ Univ of Southampton, ISBN. 9780854329809, 52 pp.
Hart, C.A. and Rafiq, M. (2004) Department image attributes and measurement.
Paper presented to the British Academy of Management Annual
Conference. University of St Andrews, Scotland, 30 August-1st September.
ISBN: 0-9549608-1-5.
Hart, C.A. and Rafiq, M. (2006) The dimensions of assortment: A proposed
hierarchy of assortment decision making. The International Review of
Retail, Distribution and Consumer Research, 16(3), pp. 333-351, ISSN:
0959-3969. DOI: 10.1080/09593960600697063.
Hart, C.A., Smith, I.G. and French, A.P. (2011) Exploring retail brand image
associations. In 16th Conference of the European Association of Education
and Research in the Commercial Distribution (EAERCD), Parma, Italy,
p.41, ISBN: 978-88-906195-0-2.
Hart, C.A. and Stachow, G.B. (2013) Driving high street recovery through
consumer based research. Impact case study. Loughborough University.
Hart, C.A., Stachow, G.B. and Cadogan, J.W. (2013) Conceptualising town
centre image and the customer experience, Journal of Marketing

41
 Management, 29(15-16), pp. 1753-1781, DOI:
10.1080/0267257X.2013.800900.
Hart, C.A., Stachow, G.B., Farrell, A.M. and Reed, G.M. (2007b) ‘Employer
Perceptions of Skills Gaps in Retail: Issues and Implications for UK
Retailers’, International Journal of Retail and Distribution Management,
35(4), pp 271-288, 0959-0552.
Hart, C.A, Stachow, G.B, Rafiq, M. and Laing, A. (2014) The customer experience
of town centres. ESRC Project Report. Published March 17. Loughborough
University, 36 pp. Available at:
[Link]
s/request-form/
Hart, C.A., Stachow, G.B. and Reed, G.M. (2005) Research into the East Midlands
retail industry: Skills training and business issues. Project report, Skillsmart
Retail.
Hart, C.A., Stachow, G.B. and Reed, G.M. (2006a) Exploring consumers' images
of town and city centres as shopping destinations. 13th International
Conference on Recent Advances in Retailing and Services Science,
EIRASS. Budapest, July, p. 91.
Hart, C.A., Stachow, G.B. and Reed, G.M. (2006b) Shopping in the East Midlands:
Consumer perceptions of town and city centres. Contemporary Issues in
Retail Marketing. Manchester Metropolitan University, September, p. 66.
Stachow, G.B. and Hart, C.A. (2010) Exploring place image: Formation and
measurement. Place Branding and Public Diplomacy, 6(2), pp. 145-155,
DOI: 10.1057/pb.2010.13.

References: Other

Ailawadi, K.L. and Keller, K.L. (2004) Understanding retail branding: Conceptual
insights and research priorities. Journal of Retailing, 80(4), pp. 331-342.
Antéblian, B., Filser, M. and Roederer, C. (2013) Consumption experience in retail
environments: A literature review. Recherche et Applications en Marketing
(English Edition), 28(3), pp. 82-109.

42
Arnold, M.J., Reynolds, K.E., Ponder, N. and Lueg, J.E. (2005) Customer delight
in a retail context: Investigating delightful and terrible shopping experiences.
Journal of Business Research, 58(8), pp. 1132-1145.
Bagdare, S. (2013) Antecedents of retail customer experience. Journal of
Marketing and Communication, 8(3), pp. 45-51.
Bauer, J.C., Kotouc, A.J. and Rudolph, T. (2012) What constitutes a “good
assortment”? A scale for measuring consumers' perceptions of an
assortment offered in a grocery category. Journal of Retailing and
Consumer Services, 19(1), pp.11-26.
Bell, S.J. (1999) Image and consumer attraction to intraurban retail areas: An
environmental psychology approach. Journal of Retailing and Consumer
Services, 6(2), pp. 67-78.
Berry, L.L., Carbone, L.P. and Haeckel, S.H. (2002) Managing the total customer
experience. MIT Sloan Management Review, 43(3), p. 85.
Berry, L.L. and Lampo, S.S. (2004) Brand in labour‐intensive services. Business
Strategy Review, 15(1), pp. 18-25.
Birtwistle, G., Clarke, I. and Freathy, P. (1999) Store image in the UK fashion
sector: Consumer versus retailer perceptions. The International Review of
Retail, Distribution and Consumer Research, 9(1), pp. 1-16.
Boatwright, P. and Nunes, J.C. (2001). Reducing assortment: An attribute-based
approach. Journal of Marketing, 65 (3), pp. 50-63.
Brakus, J.J., Schmitt, B.H. and Zarantonello, L. (2009) Brand experience: What is
it? How is it measured? Does it affect loyalty? Journal of marketing, 73(3),
pp. 52-68.
Broniarczyk, S.M., Hoyer, W.D. and McAlister, L. (1998) Consumers’ perceptions
of the assortment offered in a grocery category: The impact of item
reduction. Journal of Marketing Research, 35 (May), pp. 166-176.
Brosdahl, D.J. and Carpenter, J.M. (2011) Shopping orientations of US males: A
generational cohort comparison. Journal of Retailing and Consumer
Services, 18(6), pp. 548-554.
Brown, S. (1987) Institutional change in retailing: A review and synthesis.
European Journal of Marketing, 21(6), pp. 5-36.
Brown, S. (1990) The wheel of retailing: Past and future. Journal of Retailing, 66
(2), pp. 143-149.
43
Burt, S. (2000) The strategic role of retail brands in British grocery retailing.
European Journal of marketing, 34(8), pp. 875-890.
Burt, S. and Davies, K. (2010) From the retail brand to the retail-er as a brand:
Themes and issues in retail branding research. International Journal of
Retail & Distribution Management, 38(11/12), pp. 865-878.
Cadeaux, J.M. (1997) A closer look at the interface between the product lines of
manufacturers and the assortments of retailers. International Journal of
Retail & Distribution Management, 25(6), pp.197-203.
Carbone, L.P. and Haeckel, S.H. (1994) Engineering customer experiences.
Marketing Management, 3(3), p. 8.
Cardozo, R.N. (1974). How images vary by product class. Journal of Retailing,
50(4), pp. 85-98.
Chang, C. (2011) The effect of the number of product subcategories on perceived
variety and shopping experience in an online store. Journal of Interactive
Marketing, 25 (3), pp. 159-168.
Chebat, J.C. Sirgy, M.J. and St-James, V. (2006) Upscale image transfer from
malls to stores: A self-image congruence explanation. Journal of Business
Research, 59 (12) November, pp. 1288-1296.
Chowdhury, J., Reardon, J. and Srivastava, R. (1998) Alternative modes of
measuring store image: an empirical assessment of structured versus
unstructured measures. Journal of Marketing Theory and Practice, 6(2),
pp.72-86.
Corstjens, M. and Doyle, P. (1981). A model for optimising retail space allocations.
Journal of Marketing, 27 (July) pp. 822-833.
Davies, G. (1992) Positioning, image and the marketing of multiple retailers.
International Review of Retail, Distribution and Consumer Research, 2(1),
pp.13-34.
Doyle, P. and Fenwick, I (1974) How store image affects shopping habits in
grocery chains. Journal of Retailing, 50(4), pp. 39-52.
ECR Europe (2000) The Essential Guide to Day-to-Day Category Management.
Andersen Consulting.
Engel, James F. and Blackwell, Roger D. (1982) Consumer Behaviour, 4th Edition,
New York, The Dryden Press.
Fernie, J., Fernie, S. and Moore, C. (2015) Principles of retailing. Routledge.
44
Finn, A. and Louviere, J.J. (1996) Shopping center image, consideration and
choice: Anchor store contribution. Journal of Business Research, 35 (3) pp.
241-251.
Fischer, E. and Arnold, S.J. (1994) Sex, gender identity, gender role attitudes and
consumer behaviour. Psychology and Marketing, 11(2), pp. 163–82.
Foster, J. and McLelland, M.A. (2015) Retail atmospherics: The impact of a brand
dictated theme. Journal of Retailing and Consumer Services, 22, pp.195-
205.
Gautschi, D.A. (1981) Specification of patronage models for retail center choice.
Journal of Marketing Research, 18(2), pp. 162-174.
Gentile, C., Spiller, N. and Noci, G. (2007) How to sustain the customer
experience: an overview of experience components that co-create value with
the customer. European Management Journal, 25(5), pp. 395-410.
Gentry, J. W. and Burns, A. C. (1977) How ‘important’ are evaluative criteria in
shopping center patronage? Journal of Retailing, 53(2), pp. 73–84.
Gilmore, J.H. and Pine, B.J. (2002) Customer experience places: The new offering
frontier. Strategy and Leadership, 30(4), pp .4-11.
Glynn, M.S. (2007) How retail category differences moderate retailer perceptions
of manufacturer brands. Australasian Marketing Journal (AMJ), 15(2), pp.
55-67.
Goworek, H. and McGoldrick, P. (2015) Retail Marketing Management: Principles
and Practice. Pearson Higher Ed.
Grewal, D., Levy, M., Mehrotra, A. and Sharma, A. (1999) Planning merchandising
decisions to account for regional and product assortment differences.
Journal of Retailing, 75 (3), pp. 405-424.
Grewal, D. and Levy, M. (2009) Emerging issues in retailing research. Journal of
Retailing, 85(4), pp. 522-526.
Grewal, D., Levy, M. and Kumar, V. (2009) Customer experience management in
retailing: An organizing framework. Journal of Retailing, 85(1), pp. 1-14.
Gruen, T.W. (2002) The evolution of category management. International
Commerce Review: ECR Journal, 2(1), p.17.
Hansen, R.A. and Deutscher, T. (1977) An empirical investigation of attribute
importance in retail store selection. Journal of Retailing, 53(4), pp. 59-72.

45
Hartman, K.B. and Spiro, R.L. (2005) Recapturing store image in customer-based
store equity: A construct conceptualization. Journal of Business research,
58(8), pp. 1112-1120.
Haytko, D.L. and Baker, J. (2004) It’s all at the mall: Exploring adolescent girls’
experiences. Journal of Retailing, 80(1), pp. 67-83.
Holbrook, M. B., and Hirschman, E. C. (1982) The experiential aspects of
consumption: Consumer fantasies, feelings, and fun. Journal of Consumer
Research, 9(2), pp. 132-140.
Hollander, S.C. (1966) Notes on the Retail Accordion. Journal of Retailing, 42, pp.
29-40, 54.
Holweg, C., Schnedlitz, P. and Teller, C. (2009) The drivers of consumer value in
the ECR Category Management model. The International Review of Retail,
Distribution and Consumer Research, 19(3), pp. 199-218.
James, D.L., Durand, R.M. and Dreves, R.A. (1976) Use of a multi-attribute
attitude model in a store image study. Journal of Retailing, 52(2), pp. 23-32.
Johnson, M. (1999) From understanding consumer behaviour to testing category
strategies. Journal of the Market Research Society, 41(3), July, pp. 259-
288.
Kahn, B., and Lehman, D. (1991) Modelling choice among assortments. Journal of
Retailing, 67 (Fall), pp. 274-299.
Kahn, B.E., Weingarten, E. and Townsend, C. (2013) Assortment variety: Too
much of a good thing. Review of Marketing Research, 10, pp.1-23.
Keaveney, S.M., and Hunt, K.A. (1992) Conceptualization and operationalization
of retail store image: A case of rival middle-level theories. Journal of the
Academy of Marketing Science, 20(2), 165-175.
Kent, T. and Omar, O. (2003) Retailing. Palgrave Macmillan.
Kotzab, H. (1999) Improving supply chain performance by Efficient Consumer
Response? A critical comparison of existing ECR approaches. Journal of
Business and Industrial Marketing, 14(5/6), pp. 364-377.
Kotzé, T., North, E., Stols, M. and Venter, L. (2012) Gender differences in sources
of shopping enjoyment. International Journal of Consumer Studies, 36(4),
pp. 416-424.
Laurent, G. and Kapferer, J.N. (1985) Measuring consumer involvement profiles.
Journal of Marketing Research, 22, pp. 41-53.
46
Law, D. and Yip, J. (2010) Profiling Chinese Hong Kong consumers' attitudes on
the atmospheric management strategies adopted by intimate apparel
retailers. The Journal of the Textile Institute, 101(2), pp. 173-181.
Law, D., Wong, C. and Yip, J. (2012) How does visual merchandising affect
consumer affective response? An intimate apparel experience. European
Journal of Marketing, 46(1/2), pp. 112-133.
Lemke, F., Clark, M. and Wilson, H. (2011) Customer experience quality: An
exploration in business and consumer contexts using repertory grid
technique. Journal of the Academy of Marketing Science, 39(6), pp. 846-
869.
Leszczyc, P.T.P., and Timmermans, H. (2001) Experimental choice analysis of
shopping strategies. Journal of Retailing, 77 (4), pp. 493-509.
Lindblom, A. and Olkkonen, R. (2008) An analysis of suppliers’ roles in category
management collaboration. Journal of Retailing and Consumer Services,
15(1), pp. 1-8.
Lindblom, A., Olkkonen, R., Ollila, P. and Hyvönen, S. (2009) Suppliers' roles in
category management: A study of supplier–retailer relationships in Finland
and Sweden. Industrial Marketing Management, 38(8), pp. 1006-1013.
Lindquist, J.D. (1974) Meanings of Image (A Survey of Empirical and Hypothetical
Evidence). Journal of Retailing, 50(4), pp. 29–38. 116.
Malhotra, N.K. (1983) A threshold model of store choice. Journal of Retailing. 59
(2), pp. 3–21.
Martineau, P. (1958) The Personality of the Retail Store, Harvard Business
Review, 36 (January-February), pp. 47-55.
MacInnis, D.J., and Price, L.L. (1987) The role of imagery in information
processing: Review and extensions. Journal of Consumer Research, 13(4),
pp. 473-491.
Mantrala, M.K., Levy, M., Kahn, B.E., Fox, E.J., Gaidarev, P., Dankworth, B. and
Shah, D. (2009), Why is assortment planning so difficult for retailers? A
framework and research agenda. Journal of Retailing, 85 (1) pp. 71-83.
McGoldrick, P.J. (1979) Store image: How departmental images differ in a variety
chain. Retail and Distribution Management, (Sept / Oct), pp. 21-24.
McGoldrick, P.J. (2002) Retail Marketing, (2nd ed.) Berkshire, UK: McGraw-Hill Inc.

47
McNair, M.P. (1958). Significant trends and developments in the post-war period.
In Smith, A.B. (ed.), Competitive Distribution in a Free High Level Economy
and its Implications for the University, University of Pittsburgh Press,
Pittsburgh, PA, pp. 1-25.
Novak, T.P., Hoffman, D.L. and Yung, Y.F. (2000) Measuring the customer
experience in online environments: A structural modeling approach.
Marketing science, 19(1), pp.22-42.
Meyer, C. and Schwager, A. (2007) Understanding customer experience. Harvard
Business Review, 85(2), p.116.
Nevin, J.R. and Houston, M.J. (1980) Image as a component of attraction to
intraurban shopping areas. Journal of Retailing, 56(1), pp.77-93.
Oppewal, H. and Timmermans, H. (1999) Modeling consumer perception of public
space in shopping centers. Environment and Behavior, 31(1), pp. 45-65.
Osman, M.Z. (1993) A conceptual model of retail image influences on loyalty
patronage behaviour. International review of retail, distribution and
consumer research, 3(2), pp. 133-148.
Oxenfeldt, A.R. (1974) Developing a favorable price‐quality image. Journal of
Retailing, 50(4), pp. 8‐14, 115.
Palmer, A. (2010) Customer experience management: A critical review of an
emerging idea. Journal of Services Marketing, 24(3), pp. 196 – 208.
Pan, Y. and Zinkhan, G.M. (2006) Determinants of retail patronage: A meta-
analytical perspective. Journal of Retailing, 82(3), pp. 229-243.
Portas, M. (2011) The Portas Review: An independent review into the future of our
high streets. London: Department for Business, Innovation and Skills.
Pullman, M.E. and Gross, M.A. (2004) Ability of experience design elements to
elicit emotions and loyalty behaviors. Decision Sciences, 35(3), pp. 551-
578.
Rajaram, K. (2001) Assortment planning in fashion retailing: Methodology,
application and analysis. European Journal of Operational Research, 129,
pp. 186-208.
Richards, E.A. and Sturman, S.S. (1977) Life-style segmentation in apparel
marketing, Journal of Marketing, 41 (4), pp. 89-91.

48
Risius, D., Thelwell, R., Wagstaff, C. and Scurr, J. (2012) Influential factors of bra
purchasing in older women. Journal of Fashion Marketing and
Management: An International Journal, 16(3), pp. 366-380.
Robinson, T. and Doss, F. (2011) Pre-purchase alternative evaluation: Prestige
and imitation fashion products. Journal of Fashion Marketing and
Management: An International Journal, 15(3), pp. 278-290.
Roedder John, D., Loken, B. and Joiner, C. (1998) The negative impact of
extensions: Can flagship products be diluted? The Journal of Marketing, 62,
pp. 19-32.
Rose, S., Clark, M., Samouel, P. and Hair, N. (2012) Online customer experience
in e-retailing: An empirical model of antecedents and outcomes. Journal of
Retailing, 88(2), pp. 308-322.
Samli, A.C. (1989) Retail marketing strategy: Planning, implementation, and
control. New York: Quorum Books.
Samli, A.C., Kelly, J.P. and Hunt, H.K. (1998) Improving the retail performance by
contrasting management and customer-perceived store images: A
diagnostic tool for corrective action. Journal of Business Research, 43(1),
pp. 27-38.
Schiffman, L.G., Dash, J.F. and Dillon, W.R. (1977) Contribution of store-image
characteristics to store-type choice. Journal of Retailing, 53(2), p. 3.
Schmitt, B. (1999) Experiential marketing. Journal of Marketing Management,
15(1), pp. 53-67.
Sheridan, M., Moore, C. and Nobbs, K. (2006) Fast fashion requires fast
marketing: The role of category management in fast fashion positioning.
Journal of Fashion Marketing and Management: An International Journal,
10(3), pp. 301-315.
Simonson, I. (1999) The effect of product assortment on buyer preferences.
Journal of Retailing, 75 (3), pp. 347-370.
Stern, B., Zinkhan, G. M. and Jaju, A. (2001) Marketing images: Construct
definition, measurement issues, and theory development. Marketing
Theory, 1(2), pp. 201-224.
Stewart, D. and Hood, N. (1983) An empirical examination of customer store
image components in three UK retail groups. European Journal of
Marketing, 17 (4), pp. 50-62.
49
Teller, C. and Thomson, J.A. (2012) Gender differences of shoppers in the
marketing and management of retail agglomerations. The Service
Industries Journal, 32(6), pp. 961-980.
Urban, T. (1998) An inventory-theoretic approach to product assortment and shelf-
space allocation. Journal of Retailing, 74(1), pp. 15-35.
Van Raaij, W. F. (1983) Shopping center evaluation and patronage in the city of
Rotterdam. Papers on Economic Psychology, 27. Rotterdam, the
Netherlands: Erasmus University Press.
Varley, R. (2006) Retail Product Management, 2nd edn., Routledge, Oxford.
Varley, R. and Rafiq, M. (2014) Principles of Retailing, 2nd edn. Hampshire:
Palgrave Macmillan.
Verhoef, P.C., Lemon, K.N., Parasuraman, A., Roggeveen, A., Tsiros, M. and
Schlesinger, L.A. (2009). Customer experience creation: Determinants,
dynamics and management strategies. Journal of Retailing, 85(1), pp. 31-
41.
Wrigley, N. (1991) Is the ‘‘golden age’’ of British grocery retailing at a watershed?
Environment and Planning A, 23(11), pp. 1537–44.
Wrigley, N. (1994) After the store wars: Towards a new era of competition in UK
food retailing? Journal of Retailing and Consumer Services, 1(1), pp. 5–20.
Wrigley, N. and Lambiri, D. (2015) British High Streets: from Crisis to Recovery? A
Comprehensive Review of the Evidence. Southampton University.
Yip, J., Chan, H.H., Kwan, B. and Law, D. (2011) Influence of appearance
orientation, BI and purchase intention on customer expectations of service
quality in Hong Kong intimate apparel retailing. Total Quality Management
and Business Excellence, 22(10), pp. 1105-1118.
Zimmer, M.R and Golden, L.L (1988) Impressions of Retail Stores: A Content
Analysis of Consumer Images. Journal of Retailing, 64(3), pp 265-293.

50
 The Retail Accordion and Assortment Strategies: An
Exploratory Study

Cathy Hart

The International Review of Retail, Distribution and Consumer Research,

Vol. 9 (April 1999), 111 - 126.

Abstract
Retailer merchandise assortments need to be constantly reviewed and developed
to retain existing and attract new customers. Diversification and specialisation are
two extreme assortment strategies which are periodically pursued by retailers,
forming the basis of the Retail Accordion Theory. While this theory attempts to
synthesise the patterns of diversification and specialisation, empirical research into
the factors explaining variations in diversification and specialisation decisions
tends to be sparse and anecdotal. This paper focuses on the rationale behind the
contrasting assortment strategies of two retail sectors. It presents an exploratory
study of ten UK retail multiples. The results of in-depth interviews with senior
managers are discussed, the possible constraints on portfolio innovation are
determined and the retailer perceptions of the extent of diversification are
identified.

KEY WORDS
Assortment, merchandise strategy, Retail Accordion.

51
Paper 1. The Retail Accordion and Assortment Strategies: An
Exploratory Study

Introduction
As UK retailers battle with an increasingly competitive retail environment,
marketing and merchandising strategies are experiencing constant change. In
particular, range assortments are continually under review in order to identify an
optimum selection - not only to attract shoppers and build loyalty, but also to
maximise profits from frequently limited space.

Observation of the UK retail sector has highlighted conflicting directions

within merchandising strategies. On the one hand major multiple food retailers are
expanding ranges into more non-food areas, including clothing, entertainment,
stationery and pharmacy. Other large multiples appear, however, to be moving in
the opposite direction with major range rationalisation programmes.

Observations of similar historical trends in specialisation vs. diversification

have been made before, and literature on the Accordion Theory has attempted to
synthesise such patterns. However, empirical research into the factors explaining
variations in diversification and specialisation decisions tends to be sparse and
anecdotal. This is partly due to the lack of historical statistics on merchandise
assortment movement, and partly the result of practitioners’ reluctance to divulge
company policy.

This paper focuses on attempting to understand more about the dynamics

behind the current directions in merchandise strategy. In particular, the criteria
behind range introductions and deletions, the specific reasons for introducing new
lines, and the constraints within which decision-makers operate, are explored. A
key aim of this paper is to compare and contrast the strategic assortment
approaches of food multiples against mixed businesses, and to identify the
different factors at play.

52
 Documentation of previous assortment strategies within the sectors are
reviewed in the next section. This is followed by sections which explain the
methodology used and the results obtained. The final section concludes by
offering implications and directions for future research.

Retailer merchandise strategies

The major food multiples have, during the mid 1980’s to 1990’s, experienced a
period of rapid market growth and profitability, described as the ‘golden age’ of
food retailing (Wrigley 1991). This was followed by an era of ‘store wars’in which
the major food multiples first competed for development sites (Wrigley 1994, Guy
1994) and then consumer loyalty through price, service, and promotion. This
accelerated expansion fostered criticisms that the supermarket offerings had
converged with a lack of differentiation between the major multiples (Drummond,
1995).

One way in which some food multiples have sought to develop competitive
advantage is through product diversification. The "George" clothing range at Asda
was identified as a key area of growth for Asda (Bagnall 1995); Tesco deepened
their non-food range in 1995 to include entertainment and toy "shops"; In the same
year, Sainsbury's introduced the Delia Smith speciality cookware range and
Safeway developed a baby wear range. Other supermarket assortments
organically expanded to offer consumers a one-stop-shop and value added
ranges. More recently, Sainsbury’s have expanded into financial services and
banking. It appeared that most major food multiples strategically shifted the
product mix to include higher margin non-food items and services when consumer
expenditure on food was currently static.

In the early 1990s the number of non-food lines increased as a direct result
of the increase in store sizes. Typically, the ratio of multiples food: non-food by
sales area was approximately 90:10 for 25,000 square feet of sales area,
compared to 55:45 for a store >50,000 square feet (source IGD). The combined
average selling area devoted to non-food has been estimated at 19% and is
predicted to rise to 29% by 2001 (Davidson, 1995).

53
 The strategic decision of which non-food area(s) to expand may depend on
a number of factors. Traditionally, the success factors for diversification have
been determined by those areas which could provide volume, linked sales with the
core business, combined with potential for market growth. Petrol retailing
represented a successful diversification which resulted in the food multiples
securing 25% of the petrol market. While fast moving lines such as toiletries,
stationery and news offered success through cross selling and scrambled
merchandising, high ticket non-food lines such as electrical goods represented
greater risks and were deleted from some superstore assortments.

Earlier, during the late 1970s' to mid 1980's, mixed retail businesses
followed a similar route. WH Smiths for example, developed a logical progression
from newspapers to periodicals and books, and from stationery to leisure goods
including home computers. Boots introduced entertainment, a cookshop and
home products. Woolworths was possibly the most documented case of
diversification, whose merchandise assortment had continually expanded in order
to remain competitive, with the inclusion of furniture, toys, entertainment, clothing
and cosmetics.

In the late 1980s to mid 1990s, many of the mixed retailers followed a
period of rationalisation within their assortments by deleting a number of product
ranges (Brown, 1988, McGoldrick 1990, Retail Week 1993). These strategies
were in direct contrast with the assortment strategies of the food multiples during
the same period. The reasoning behind these range rationalisations may be
attributed to a combination of the changing economic and competitive environment
and the shifting of emphasis from the high street to out of town shopping. Retail
theorists may suggest otherwise, that these assortment movements are part of a
cyclical pattern repeated over a period of time (Hollander1966, Brown 1987).

Specialisation Vs Diversification
McGoldrick (1990) summarised the two main alternatives in product assortment
strategy to be increased diversification or greater specialisation which aim to
"provide consumers with a greater choice in breadth or in depth". For some

54
retailers, specialisation is seen as the main means of survival whereas for others,
diversification is seen as the main form of growth.

A retailer assortment profile comprising width, depth, coherence, availability

and quality, is initially designed to achieve the desired market positioning strategy
(Knee and Walters 1988). Over time, however, the assortment may no longer
meet the changing needs of the target market and it may be necessary to adopt
one of four merchandising strategies of range rationalisation, repositioning, growth
or diversification (Cook and Walters 1991). Range rationalisation is the common
route followed by most retailers, and includes the ongoing review of assortments
and minor adjustments necessary to meet the retailer’s market position and profit
targets. It may include eliminating and even replacing entire product ranges.
Repositioning involves adjusting the product mix to "address the specific needs of
a market segment more effectively" (Ibid 1991). Three levels of repositioning have
been identified, zero, gradual, and radical repositioning (Corstjens and
Doyle1989). Zero repositioning is where the retailer maintains continued focus on
its initial target segment despite a changing retail environment. Gradual
repositioning includes the small merchandise adjustments required to maintain a
continuous match between the market and its own offer. Radical repositioning
occurs when the retailer makes a major shift into new target markets. It can be the
result of a shift in the marketing environment or because "its management have
failed to effectively gradually reposition over a long period".

A growth merchandise strategy is achieved by adding ranges to the existing

assortment and may be product or market led. By drawing upon the company’s
existing strengths it offers the lowest risk strategy. In contrast, diversification is the
most extreme strategy involving totally new merchandise ranges and markets. In
evaluating the available strategic choices, a retailer has to consider its existing
expertise in the area and the level of service support required for the merchandise.
Omura (1986) suggested a 20 celled matrix in which each cell offered varying
levels of risk. Cell1, was a low risk greater penetration of its current market with
existing merchandise, leading to extreme diversification in cell 20, in which new
merchandise requiring new service requirements, appealed to different market
needs. Two main forms of diversification have been noted. A form of skimming or
55
“Creaming”(Gist 1968), adopts the best high volume brands from other retailers
lines with the result of generalising the assortment. “Scrambling” by contrast, is
when the “quest for goods carrying better than average margins has led retailers
into product fields far away from their customary lines” (Gist 1968). The skimming
process adds small but sure profits, whereas scrambling involves large but unsure
profits.

It is noticeable that the product-mission matrix originally developed by

Ansoff (1965), has formed the basis for further frameworks applied to assortment
strategy (Kristenson 1983, Omura 1986, Cook and Walters 1991) with the added
ingredient of risk. Although risk, combined with existing skills and expertise, is a
high factor in evaluating a merchandise strategy, it has not proved sufficient to
deter retailers from attempting extreme merchandise diversifications such as
retailing cars through department stores (Hollander 1966) or supermarkets
(Poynor 1986).

The Retail Accordion

The extent to which some retailers have pursued diversification and specialisation
assortment strategies was originally observed by Hower (1943) and developed by
Hollander (1966) in relation to the Accordion Theory. Hollander noted a number of
factors that encouraged specialisation, using the orchestra analogy whereby "at
any one moment some players are retiring from the orchestra while others (mainly
with compressed instruments) are joining the band”.

Furthermore, he identified three phenomena occurring in the process of

specialisation; unsuccessful merchandise mixes which have encountered
"insurmountable limitations", the deletion of conventional lines from the retail
assortment and the market growth of a product area is adopted by new specialists.
These phenomena have also been observed in the 1980s - 1990s (McGoldrick
1990, Brown 1988).

More specifically, Hollander noted five possible forces which may influence
the contraction of an assortment constraining the opportunities for diversification.

56
i) Non-economic - from personal interest, expertise and enthusiasm
ii) Restraints - whereby retailers avoid raiding each other's assortments from fear
of retaliation or legal limitations.
iii) Capacity - the available capital and resources do not permit merchandise
expansion.
iv) Costs - adding classifications to an assortment may cause costs to grow at a
faster rate than revenue.
v) The market - consumers behaviour may encourage specialisation.

Hollander also suggested that certain combinations may be incongruous or

incompatible with each other.

While other similar examples of specialisation and diversification have been

observed occurring either simultaneously or sequentially, there is a lack of any
empirical evidence to support these observations. Indeed, the hypothesised
perpetual alternation between generalist and specialist outlets has yet to be
proved. Brown (1988), argued that the most significant applications of these
conceptualisations is to individual retail firms but the value of the theory lies rooted
in relation to a number of, or sector of retailers.

It is interesting to observe that in the 32 years since Hollander's Notes on

the Accordion (1966), few attempts have been made to synthesise retail
assortment movements. Indeed, compared with other extensively argued cyclical
theories (Brown 1991) the Accordion has been somewhat neglected. It is also
evident from recent strategies that retailers continue in their struggle with
assortment diversifications. For some companies, considering the levels of risk
involved, the initial dilemma lies in identifying the direction and extent of product
diversification.

These factors have prompted investigation into the retailer reasoning

behind expansions and contractions within assortments. Exploration of the
rationale behind some of the moves may determine possible success criteria for
new ranges, and conversely the criteria for deleting failed ranges. Furthermore,

57
an examination of the constraints to diversification within different sectors, might
also provide a valuable insight into the limitations of assortment diversification.

The Study
Two key objectives formed the basis of this exploratory study. The first objective
was to examine the movement of assortments over a given period of time. The
second objective was to determine the reasoning behind some of the major
diversification decisions to expand into new market areas, and conversely, the
specialisation decisions for exiting other areas.

In order to track the movements of retail assortments, a form of

measurement was required. According to the literature, the dimensions of product
assortments included “width, depth and coherence” (Knee and Walters 1988).
The latter dimension being the most intangible and difficult to measure, the first
two offered greater scope in translation to number of categories (width), and
product lines (depth). Category expansion and contraction might also be
measured via the different ‘vectors’ or variety of product groups introduced and
deleted. In evaluating these dimensions, pilot interviews identified that few
retailers held detailed product data beyond the previous two or three years.
Furthermore, this data was usually held at the category level and would thus need
to be collated aggregately to provide a picture of the total assortment. Moreover,
it was necessary to find a unit of measurement which could be comparative across
retailers and sectors. Previous studies in this area (Hollander, 1966) resorted to
the "number of lines" to describe the size of the assortment, without actually
recording the measurement. While limited, this measure provided a transferable
unit which could be used on two dimensions; first, to record the relative change in
number of product lines / size of the assortment by each retailer, over a ten year
period. Second, to compare assortment movement across sectors. The relative
change in product lines was expressed as a ratio.

The study concentrated on two broad retail sectors, food and mixed
retailing. The earlier literature identified in a previous section, suggested that a
number of mixed retailers had rationalised their ranges. Whereas the food

58
retailers appeared to be diversifying. The study sought to test whether these
movements were happening in isolation, or whether any noticeable trends were
discernible. These sectors were also selected for maximum comparison; the UK
food retailers comprise the largest and most concentrated retail sector, which by
definition could be termed specialists due to their depth of food ranges. In
contrast, the mixed retail sector, includes wide but shallower assortments and
variety stores, experiencing considerable movement of ranges. Within these two
sectors, a number of retailers were contacted with the final sample comprising ten
major UK multiple retailers including three food retailers, one food discount
operation, one hypermarket, two mixed retailers, a variety store, a stationer and a
chemist. The sample represented 34 per cent of the UK grocery market
(Henderson Crosthwaite,1993). The representative market share of the mixed
retailers was more difficult to define given their involvement in a variety of markets.
The sample included established retailers who had been active in their respective
sectors and could thus provide historical evidence of assortment development.

Due to the limitations experienced in measuring category movement, an

alternative method of researching product diversification and specialization was
explored. In-depth analysis on the behaviour of a single firm combined with
comparative studies across different retail firms has been described as the "best
option" for collecting empirical evidence of retailer criteria (Pellegrini and
Zanderighi 1990). Thus, depth interviews were chosen as the most effective
means of achieving an insight into the retailer reasoning behind the diversification /
specialisation decisions. Personal interviews were conducted with senior
personnel or individuals who were directly involved in, or responsible for selecting
or deleting product ranges. A variety of respondents were selected who could
offer a broad perspective including trading directors, marketing managers, senior
buyers and merchandisers.

The interviews were semi-structured including 20 key open-ended

questions to facilitate comparisons. The questions were broadly categorised
under four themes:

59
i) Key assortment movements pursued during the previous 5-10 year period,
including introductions and deletion of product ranges;
ii) The responsibility and rationale for these range decisions;
iii) Nature and extent of research prior to entering new market areas; and
iii) Perception of constraints to diversified ranges.

The questions were first pilot tested on a small representative sample, then
modified for the final interviews. Where possible more than one individual was
interviewed from each organisation. Thirteen interviews produced 16 hours of
taped discussion for transcription and content analysis. The interviews were
carried out during May and June 1993.

Results and Discussion

In respect of the requests for anonymity, the retailers will be referred to as "Food
A,B,C,D, or E" and "Mixed A,B,C,D or E" in the following sections.

Each retailer gave a total figure for current product lines compared with the
total number of lines 10 years previously. Examining the percentage change in
product lines across both retail sectors there is seen to be a clear difference in
trends between the sample of food and mixed retail businesses in the sample.
Figure 2 illustrates that the food retailers had all expanded their lines from a
minimum of 7% to a maximum of 400%. The relative number of lines also varied.
Food A demonstrated 120% growth from 9,000 lines in 1984 to 20,000 lines in
1994; Food B, indicating a 10% growth had expanded from 22,500 lines to 25,000
lines. In the most extreme case, Food D had expanded 400% from 600 lines to
3000 lines.

60
Figure 2 Food Retailer Growth in Product Lines, 1984 - 1994

450
400
350
Percentage Change

300
250
200
150
100
50
0
1 2 3 4 5
Food Retailers

Interestingly, an optimum limit of 25,000 lines appeared to have been

reached by the food retailers generally, although not all stores carried the full
range. These figures may be related to the growth in store sales area which
peaked during the time of the study. Furthermore, the overall increase in lines
during this period could be attributed to the more sophisticated space modelling
and sales based ordering systems which have enabled a higher concentration of
lines per square foot resulting in more effective space utilisation.

In contrast, all of the mixed retailers had contracted their lines by a

minimum of 5% to a maximum of 65%. (Figure 3) As an indication of the relative
number of lines carried, Mixed A reduced its assortment from 70,000 to 25,000
lines and Mixed C from 50,000 to 20,000 lines. It is interesting to note that for both
the mixed retail and food sectors, 20-25,000 lines appeared to be an optimum size
of assortment to offer. A future examination of developing store sizes for each
sector might provide interesting data for comparison.

There is clearly a strong indication that during the same 10 year period, the
food sector has been diversifying, while the mixed retailers have become more
specialised in their assortment. In discussing the reasoning behind these

61
decisions, both sectors are compared and contrasted within the following two
sections on range diversification and specialisation.

Figure 3 Mixed Retailer Contraction in Product Lines, 1984 - 1994

-10

-20
Percentage Change

-30

-40

-50

-60

-70
A B C D E
Mixed Retailers

Range Diversification
In examining the rationale behind range diversification decisions, the findings
indicated strong similarities between the two retail sectors. Thus the most
frequently cited responses have been combined into Figure 4 to illustrate the key
criteria influencing the introduction of a new range. Each criterion will be
discussed sequentially. The initial information source influencing which
prospective markets to enter resulted primarily from supplier contact which then
had to be "translated from supplier propaganda into objective information" (Food
C).

The suppliers were relied upon to research and identify mutual market
opportunities. Secondary data and market intelligence were also a key source of
information for the retailers, being used predominantly as a tool to guide and
support new range proposals. Given that the suppliers may also provide similar
information to other retailers, competitive movements and successes in new
market areas were keenly tracked, both for the retailers following "me too" reactive

62
merchandise strategies (Pellegrini and Zanderighi 1990) but also for those
defending market share. The fourth source for generating new ideas for
diversification derived from trade press, or visiting trade shows and exhibitions. On
individual occasions ideas were generated downwards by the board if there was a
market niche identified.

The responsibility for creating a new range depended on whether it was

considered a strategic change of direction requiring board approval, or simply a
development of existing product areas. In the latter case it would be the
responsibility of the buyer or category manager. In implementing the new range
most of the retailers followed a team approach, involving the marketing, buying,
merchandising, finance and logistics departments.

The majority of respondents agreed on the use of secondary market

research data as a means of identifying new market opportunities. However, it
was noticeable that the more senior the respondent the least importance was
attached to market research information. In contrast, the buyers' attitudes to the
extent of market research conducted prior to a diversification was revealing; most
recognised a lack of it but due to commercial pressure often had to enter a new
market area with limited information. "There's not enough (market research) really,
if you're going to spend a million on a new (range) launch you want to be
absolutely certain it’s not going to go wrong." (Mixed A)

63
Figure 4 Retailer Reasoning behind Range Diversification

Information Source

64
 A proposal for a new range subsequently emerged from one or more of the
sources of information. In evaluating the new range, a number of criteria had to be
fulfilled. Generally, each respondent gave three criteria or reasons for entering a
new market area (listed in Figure 4). Over half of the respondents used the
adjective "fit" as an interpretation of the new range congruity with the merchandise
mix, target segment and image. "Fit" with the existing ranges was interpreted in
two ways, operationally and strategically. The operational consideration would be
whether it would cohere with an existing merchandise area e.g. water filters with
kitchen accessories, "In the case of computer games there wasn't a fit with the
other ranges, they are odd things to merchandise and there isn't a natural
adjacency in which to fit the products" (Food C). A strategic consideration would
be if the retailer was already selling associated products and the new range was
viewed as a natural extension to the existing portfolio.

The second criterion was if an available market opportunity offered

sufficient potential to support the new range. All of the retailer buying departments
closely monitored their market areas to identify potential for growth. However,
identifying an opportunity would not necessarily result in a new range being
introduced. Three of the retailers saw themselves as followers, not innovators and
would rather miss the opportunity than risk a failed range. "We would see if the
opportunity already exists elsewhere in the market place. We see ourselves as
fast followers not leaders." (Food E). This risk avoidance strategy also linked in
with the sixth criteria for introducing a new range; competition was already active
in the proposed market area and some retailers felt they had a major gap
compared with competitors.

As expected, the "fit" with the target customer profile was given more
frequently than "fit" with the corporate mission. It was obviously of greater
importance to the respondents that the target customer would want to buy these
products otherwise the range would fail. "Authority doesn't mean everything under
the sun in that product group. It means everything that our customers want in that
product group, sold in a way that is suitable to their shopping trip." (Food C)
However, lack of congruency with the corporate mission could still prevent a new
range gaining final approval at board level: "Some products don't fit with where we
65
want to be and no matter how saleable the product was we might feel it could
damage our long term objectives of trying to build up a name in a certain market."
(Food C).

Finally, three of the ten retailers stated that a new range would only be
introduced if it was felt it could offer real choice and not duplication or
cannibalisation of existing ranges. Here, the food retailers and mixed retailers
were in agreement that the core ranges should remain untouched reinforcing the
view that the potential for diversification lay in the non-core area: "80% of range
stays the same and there's only about 20% change in any one year". (Mixed E)
"75% of the range selects itself. It's those range periphery or range extensions
where you have to sit back and look at it." (Food A) Furthermore, the food
retailers saw most potential for diversification in this 25% non-core non-food area:
"non-core non-foods is an important area, a large proportion of the packaged food
and drink grocery market is controlled by five retailers, whereas non-foods going
through the same five constitutes less than half. We see it as half the market for us
to still go for" (Food B). The decision of which non-food ranges to stock was
critical. While the primary reason for shopping in a supermarket is for food, Food
D and Food E felt that certain non-food ranges could complement the core,
creating a one-stop shop. However, there was a limit to the type of shopping
products which could be successfully stocked in a supermarket; items were
required to be self-selecting, easily transportable, high volume, regularly
purchased lines.

Following evaluation of the range “fit”, all of the retailers were asked to
consider the major constraints to the introduction of a new range. The most
frequent responses are again listed in Figure 4. 40% of respondents felt that the
opportunity cost of space was the main constraint (mixed C,D,E and Food A).
Losing shelf space from another established merchandise range to a new, untried
area was always a difficult decision, particularly for the mixed retailers. However,
two stores (Food B and Mixed A), felt that space was not an issue as it would be
managed operationally on a one-in-one-out basis; "if there's a chunky market to be
in space would not be a restriction. If we believe in it then that should run and
something else should be trimmed back" (Food B). Generally, it appeared that the
66
food retailers were managing to control the space issue through targeting the
ranges more effectively to the local demographics by branch profiling.

Shrinkage was given as the next constraint, particularly where high value
items in a predominantly self-service environment were concerned. The decision
of whether to carry computer games "live" (the complete product) or "dead" (empty
packaging on display) posed operational issues for the food retailers who could
not provide the necessary service support to this growing non-food area. Many
mixed retail diversifications were also constrained by shrinkage due to their high
street locations.

Surprisingly, having suggested that market intelligence aided the

introduction of a new range, two mixed retailers then considered "insufficient
market knowledge" to be a constraint. Specific questions such as ‘What was the
size of the market and was it big enough to take another player?’ were, if left
unanswered, likely to pose insurmountable problems post-entry. Their difficulty lay
in anticipating the customer's desires when there were no historical sales or
market data on which to base the decision. Perhaps the lack of clear coherence of
some mixed retail assortments held one of the answers; if retailers were not sure
of the positioning of their core offer, then how could they confidently select a new
diversification. "What to include and not - what is a core product? Is it deepening
the existing inventory rather than adding to it? If it's deepening it, it might have a
chance, if it’s adding to it, it won't." (Mixed D) By comparison, two food retailers
(A and B) believed that packaging and merchandise quality of new diversified
ranges were crucial factors to maintain the same positioning of non-core ranges as
core ranges, whereas Food E felt that the risk of cannibalising existing sales was
its only constraint.

To summarise, prior to introducing a new range, the constraints were often

traded off against the range fit, with a slightly heavier weighting towards the latter.
Ultimately, if there was sufficient scope, then the range would be trialled to verify
its market potential.

67
Specialisation
Having identified the retailer rationale and constraints for entering a new market
area, the respondents were questioned on recent deletions from the assortment
and their reasoning behind these. As illustrated in Figure 3, the mixed retailer
sample had deleted a large number of lines during the past 10 years. In general,
there was a refocusing on the stronger ranges where it was felt they could achieve
some "authority" in the market. This resulted in an assortment strategy of buying
narrower and deeper within the ranges compared with ten years ago.

The reasoning for deletion of ranges was often dependent on the product or
market area, and its positioning. (Figure 5 lists some of the key range deletions
and the typical reasons for their deletion). The key reason some mixed retailers
moved out of electrical appliances, cameras, watches and furniture was
insufficient service support. The retailers could not offer the selling expertise,
delivery and after sales service required for these shopping products. Their
inability to compete with specialist retailers also became evident with the deletion
of home furnishings, computer hardware, audio, and food due to the limited choice
and depth of range available. Two mixed retailers had removed food ranges due
to their lack of depth of assortment compared with the larger multiples, however, a
lunchtime snacks and drinks range had been added (by two different retailers).
Furthermore, despite having previously removed computer hardware from their
assortments, some mixed retailers also saw computer games as the current
growth area due to a market opportunity, and greater synergy with other
entertainment ranges e.g. music, (CDs) or books.

Travel, flowers, drinks and confectionery were all removed due to a "lack of
fit with the core business", echoing the same phrase previously given as a criterion
for new range introduction by three of the mixed retailers. "We were very much in
the leisure industry, we sold travel books, maps and guides, and it was felt that the
travel service was a natural extension, but it isn't, it's a specialist service". (Mixed
E)

68
Figure 5 Reasons for Deleting a Product Range

Product Area Reason

electrical appliances, cameras, inability to offer service support

watches, furniture
Mixed home furnishing, electrical, lack of consumer choice/depth
audio, computer hardware, food
Retail travel, flowers, drinks, lack of “fit” with core business
confectionery
jewellery incorrect pricing and packaging

Food ambient ready meals consumer resistance to

products
mens and children’s underwear poor (merchandise) quality “fit”

Range fit also appeared to be closely linked in with the retailer's positioning.
This was illustrated by the failure of a jewellery range which was blamed on
incorrect positioning. "We could have done more with it, could have got keener
prices, better packaging and displayed it better" (Mixed A). Moreover, two major
retailers were returning to the same "failed" market area within two years of
deleting the range. An elusive market niche justified their reasoning. "We got our
fingers burnt but we know there's still that big niche in the market we ought to be
after if we can get the right formula". (Mixed A)

It was clear that the mixed retail sector had more assortment positioning
problems compared with the food sector. Two of the mixed retailers had
expressed difficulty in identifying exactly what constituted their standard core
ranges and therefore struggled in suggesting directions for non-core ranges.
Unlike food retailers selling staple products, the mixed retailer core range was
more susceptible to changes in fashion. The high street location of most of the
mixed retailers had influenced the diversification into snack foods but limited space
determined the deletion of general food ranges. However, having vacated a
number of gaps in the assortment, seasonal one-off buys were considered the low
risk alternative to create variety and motivate impulse purchasing. "We would do a
one off purchase, and that would be it. If we had anything really strong we might

69
consider it, but I'm always nervous, it never works a second time around." (Mixed
E)

Perhaps the “confused positioning” (Kotler1988) experienced by some of

the retailers resulted from a lack of coherence across the assortment. Historically,
mixed retail assortments appeared to have evolved organically, rather than
through strategic planning. "Buyers were actively encouraged to think laterally and
seize opportunities for diversification and be entrepreneurs. We now have a ban
on thinking laterally because that leads you into disasters. We should not think
width but think depth. This is what our buyers are told. We're not interested in new
concepts but in deepening the concepts we're in." (Mixed C) To a certain extent,
specialisation in the core ranges could possibly be the safest strategy given the
growth in competition from other players, such as the food retailers currently
encroaching the same markets by diversification, "We chose to concentrate on our
core market, do what we know and do it well but if you take your eye off the ball,
competitors will take you on. It is happening, supermarkets are selling magazines
and books". (Mixed E)

In the previous five years the emphasis had shifted from range extension to
range engineering as the mixed retail stores realised there was a saturation point
in ranges. One important factor for mixed retailers was Direct Product Profitability
(DPP) which had been the main driving force behind some of the remodelling and
refocusing of ranges, "You can keep adding ranges but must come to a point of
saturation in ranges as well". (Mixed D) Poor sales performance and margins
were obvious reasons given for deleting a number of products, and DPP had been
instrumental in the decision making. However, the majority of the retailers felt that
range performance could not be evaluated purely on financial return and it was
important to view the total retail offer.

In contrast, the food retailers had deleted far fewer ranges as evidenced by
the trends illustrated in Figure 2, but three (A, B and E) agreed that ambient ready
meals were recent failures although all evidence had indicated a good growth
area. The market for frozen, delicatessen and fresh convenience meals was
expanding to fit in with an "on the move society" and its democratisation of tastes.
70
However, it appeared that the flavour and quality were perceived to be inferior to
chilled and frozen versions. Although consumer resistance to long life meals
showed no signs of lifting, it was unlikely these ranges would be removed
completely. By definition this could be termed a range extension, but the retailers
considered it to be diversification.

The high risk of diversification became apparent as new range introductions

were equally prone to failure as established ranges. A recent non-food
diversification of underwear had to be deleted due to the supply of poor quality
merchandise. "We thought we had the right supplier base and let them get on with
it but they delivered a very mediocre and average product….it didn't measure up to
our image" (Food A).

While the rate of diversification and specialisation differed greatly between

both sectors, the reasoning behind the moves held shared similarities. The
degree of “fit” between the range and the retailer image and positioning was in
essence a key reason for introducing or deleting a range from the assortment. It
was also noticeable that the food retailers appeared to be pursuing more of a
“skimming” diversification strategy compared with the more “scrambled” approach
adopted by the mixed retailers.

Conclusions
A number of issues have emerged from this exploratory study. First, many retailer
assortment diversification decisions appeared to be ad hoc and unplanned.
According to the respondents, there was insufficient market research conducted
prior to assortment diversification, with the buyers preferring to rely upon
"management judgement" (Greenley and Shipley 1989). Often, the same criteria
were used for entering or exiting a market area. This was reflected in some
companies returning to the same market area within a short period with
repositioned ranges, thus incurring high development costs. Second, these
inconsistent assortment additions and deletions may also have a long term impact
on the retailer's image and positioning with their target market. The findings

71
suggested that retail assortments may require closer monitoring to avoid the
necessity for radical repositioning (Corstjens and Doyle 1989).

Third, the study lends support to Hollander's comments on the Retail

Accordion theory. The emphasis from specialisation to diversification can still be
observed over 30 years on, but with less definition between the dominating forms.
While it can be argued that the food retailers have taken on the generalist role,
and the mixed retailers have reverted to more of a specialist role, the clear
dominance of either form is less marked than in historical trends. Furthermore, the
original Accordion Theory appeared grounded at the store level, with dominance
from evolving store formats rather than assortment formats. This study has
identified that the general-specific-general cycle (Gist 1968) currently appears to
be more applicable to changing assortments rather than formats.

Recent developments suggest that the UK food retailers may be

approaching an end to their diversification dominance, but with no sign yet of an
emerging alternating form. In terms of assortment strategy, the food retailers
adopted a higher risk form of diversification pursuing “new merchandise with
unrelated service requirements” (Omura 1986), compared with the mixed retailers
who, more wary of diversification adopted a low risk strategy, actively specialising
into their core business. However, it is interesting to note that these strategies are
not always in response to consumer requirements. One wonders if the
consumers, tired of the diversified, scrambled and shallow non-food assortments,
will force the supermarkets to revert back to their core business, echoing the
mixed retailer strategy. There is evidence to suggest that environmental forces
are already influencing this with recent developments in city centre convenience
stores and home shopping formats emerging.

Fourth, the study also highlighted some difficulty in proving or disproving

the Retail Accordion theory. The “number of lines” is clearly an inadequate
measure of assortment movement. Ideally, an additional dimension is required to
measure and track the “coherence” of the assortment, or the degree of scrambling.
This may then be evaluated with the retailer positioning. Perhaps a clearer
categorisation of ranges within the width of the assortment holds the answer. A
72
measure of the extent to which product ranges are related or unrelated to the core
offer, might provide more realistic returns.

Finally, a number of implications can be drawn from the Accordion Theory.

In modifying Hollander’s notes, additional forces of competition, space and market
positioning can be seen to influence the contraction of an assortment. Indeed,
tougher competition in more saturated markets may have resulted in other sectors
raiding alternative assortments e.g. DIY. Some lessons become evident such as
the importance of clear positioning for diversification, but equally, the study also
suggests that all assortments risk being raided and only those with a strong
customer proposition will survive the threat. The study has also identified a
number of new research questions such as, ‘how far should assortments
diversify?’ and ‘at what point could assortments become saturated?’ Does this
then signify the transition into specialisation, or are there other phases of the
Accordion?

Although the evidence for the Accordion Theory remains to a large extent
anecdotal, on reflection the accordion, or rather the orchestra analogy (Hollander
1966) may require further refinement. An additional dimension of the score, or
music being played could be interpreted as the assortment movement. As more
musicians with groups of instruments, (representing the stores and formats) enter
the orchestra, there will be point of saturation when there are too many playing
certain parts of the score, resulting in an imbalance and a confused audience.
Some musicians may then decide to play parts designed for other instruments.
The success of this switching will depend on the orchestrator’s (retailer’s) ability to
convince the audience that their deployment of instruments is compatible with the
score.

73
References

Ansoff, H. I. (1957) “Strategies for Diversification”, Harvard Business Review, 35,

5, pp. 113-124.
Bagnall, S (1995) "Asda Rises in Food Stores Ratings", The Times, June 30, p.
26.
Brown, S. J. (1987) "Institutional Change in Retailing: A Review and Synthesis",
European Journal of Marketing, 21, 6, pp. 1-36.
Brown, S.J (1988) “Retailing Change: Cycles and Strategy”, The Quarterly Review
of Marketing, 13, 3, pp. 8-12.
Brown, S.J. (1991) “Variations on a Marketing Enigma: The Wheel of Retailing
Theory”, Journal of Marketing Management, 7, 2, pp. 131-155.
Cook, David and David Walters, (1991) Retail Marketing Theory and Practice,
Prentice Hall, London.
Corstjens, Marcel and Peter Doyle (1989) "Evaluating Alternative Repositioning
Strategies", Marketing Science, 8, 2, pp. 170-180.
Davidson, F (1995) "Hungry for Growth", Report by James Capel.
Davies, G.J. and J.M. Brooks (1989) Positioning Strategy in Retailing, Paul
Chapman Publishing.
Drummond, G (1995) "The Big Squeeze", Supermarketing, 10 March 1995 pp. 17-
18.
Gist, R., (1968) Retailing: Concepts and Decisions, John Wiley and Sons, New
York.
Greenley, G. and Shipley, D (1989), “Dimensions of Product Planning in Retail
Marketing”, Irish Marketing Review, 4, 1, pp. 53-62.
Guy, C. M (1994) Grocery Store Saturation Has it Arrived yet?, International
Journal of Retail and Distribution Management, 22, 1, pp. 3-11.
Henderson Crosthwaite (1993), Food Retail Perception and Reality, London:
Henderson Crosthwaite.
Hollander, S. C. (1966), Notes on The Retail Accordion, Journal of Retailing,
Summer, pp. 29-40.
Hower, R. (1943), History of Macy’s of New York, Harvard University Press,
Cambridge, MA.

74
Knee, D. and Walters, D (1988), Strategy in Retailing, Theory and Application,
Philip Allen, Oxford, UK.
Kotler, P. (1988) Marketing Management: Analysis, Planning, Implementation and
Control, Prentice-Hall, Englewood Cliffs, NJ.
Kristenson, L (1983), “Strategic Planning in Retailing”, European Journal of
Marketing, 17 (2), pp. 43-59.
McGoldrick, P. (1990) Retail Marketing, McGraw Hill, Maidenhead, UK, p. 93.
Omura, G. S, (1986), "Developing Retail Strategy", International Journal of
Retailing, 1 part 3, pp. 17-32.
Pellegrini, L and Zanderighi, L. (1990) New Products: Manufacturers' Versus
Retailers' Decision Criteria, The International Review of Retail, Distribution
and Consumer Research, 1, pp. 149-174.
Poyner, M (1986), “There’s No Business Like Shop Business”, Retail and
Distribution Management, 14, 4, pp. 6-9.
Wrigley, N (1991) “Is the ‘Golden Age’ of British Grocery Retailing at a
Watershed?” Environment and Planning A, 23, 11, pp. 1537-1544.
Wrigley, N (1994) “After the Store Wars. Towards a New Era of Competition in UK
Food Retailing?”, Journal of Retailing and Consumer Services, 1, 1, pp. 5-
20.

75
 Consumer Perceptions of Non-Food Assortments: An
Empirical Study

Cathy A. Hart
and
Mark A. P. Davies

Journal of Marketing Management, 1996, Vol. 12, 297 - 312.

Abstract
Previously, retail image research has been conducted predominantly at the store
level. This paper examines consumer perceptions at the product group level to
determine if assortments enhance or detract from the retailer image. It presents
the results of an empirical study of 504 grocery shoppers and their attitudes
towards six non-food categories. It finds that consumer perceptions are both
category- and store-specific, and suggests that future assortment image research
should be measured on three levels, notably core, non-core and peripheral
product groups.

76
Paper 2. Consumer Perceptions of Non-Food Assortments: An
Empirical Study

Introduction
The key to long-term sustainable, competitive advantage for any organisation is a
comprehensive strategy requiring a clear positioning for maximising customer
satisfaction. One strategy currently pre-occupying a number of major UK food
retailers is diversification into further non-food ranges to enhance and strengthen
their market positioning and profit potential. A dilemma, however, is emerging.
How far can diversification proceed without weakening the store image among
customers or even alienating existing customers? This study contributes to the
debate by examining consumer perceptions towards expansion of non-food
ranges within supermarkets. A contention of this paper is that the traditional
emphasis on store image may be oversimplified. Important variations in image
across merchandise assortments at the product category level are discovered.
The findings distinguish between key or core merchandise assortments from those
considered non-core, with implications for the management of diversification
strategies.

During the 1980's and early 1990's, the "Big Three" UK food retailers,
Sainsbury, Tesco and Argyll, followed accelerated store expansion programmes,
each opening, on average 20 stores per year. This increased space enabled
organic expansion of the product assortments. Non-food ranges, in particular,
were targeted with diversifications into areas such as stationery, clothing, audio
and video. The aim of the superstores was to attract a wider market with the dual
objectives in offering both the convenience of one-stop shopping and value for
money (Duke, 1991). Although this diversification offered the attraction of higher
margins for the retailer, care was needed to ensure it would not be detrimental to
the retailer's image. Did these ranges enhance the image of the store in the
consumers mind, leading towards the ultimate one-stop shop? Or alternatively, to
use branding terms offered by Kotler (1988), did they detract from the image
because of “overpositioning” (in which consumers hold a narrow view of what
constitutes a stores’s range of merchandise) or “confused positioning”, resulting
77
from the new ranges adversely positioning the store? Retailers need to determine
if these additional ranges to their traditional categories offer genuine opportunities
for added differentiation. Within recent academic and trade literature, most
attention has been directed toward store expansion and the growing threat of
supermarket saturation, but the threat of merchandise saturation and its impact on
store image has been rather neglected. It is suggested that assortments, similar
to stores, can reach an optimum level beyond which further expansion would be
considered inappropriate by their main patrons. In this paper, differences in
consumer perceptions towards product categories between stores is examined,
providing valuable data on store image and opportunities for differentiation. In the
next three sections the background to the study and methodology are outlined,
followed by a detailed examination of the survey results. The implications for retail
assortment strategy are offered, together with suggestions for further research.

Background
Retailers need to get the right combination of products or assortment in their
stores for a variety of reasons. Firstly, assortment is a key determinant for store
choice and retaining customer loyalty (Greenley and Shipley, 1992). Various
authors have acknowledged that merchandise assortments form a major
component of a store's image. For example, Lindquist (1974) concluded that
"merchandise assortment" was mentioned most frequently for positive
consideration when isolating the components of a retail store's image. Out of the
various store image components, merchandise dimensions (in terms of suitability,
variety or assortment choice) are rated either second (Fisk, 1961, Louviere and
Johnson, 1990, East et al., 1992) or third (Berry, 1969) in importance to locational
convenience, price or quality. Whereas the primary determinant for store choice
varies for different studies, merchandise assortment remains constantly important.

Secondly, the implications of having a particular merchandise assortment

can either work negatively, alienating consumers, as with a limited or adverse
range; or positively, encouraging consumer loyalty (Berry, 1969). Internal
consistency across an assortment is essential to project a uniform, synergistic

78
image and one alien range may reduce a retailer's chance of developing this effect
(Samli, 1989).

Furthermore, an assortment can provide a distinctive edge in the weaponry

of competitive advantage (Knee and Walters, 1985). Porter (1985) asserted that
one of two generic strategies - differentiation or cost leadership - is the best way of
achieving competitive advantage. Of the two strategies, differentiation has been
suggested as being the more sustainable for retailers (Warnaby and Woodruffe
1993). According to Cadeaux (1992), assortments play a key role in a
differentiation strategy, in which retailers can pursue advantages with their
selections in order to attract one or more segments and build patronage loyalty.
This is further supported by Samli (1989) who considered a highly desirable
merchandise mix as vital for the survival of the retail store. He recommended that
the mix must be different to competitors' mixes and congruent to the retail image,
to establish differential advantage.

Despite the acknowledged importance of location, price and quality as

determinants of store choice listed earlier, they are likely to become less influential
in distinguishing between store images in the future. As the grocery market
saturates and the concentration of stores increases, maintaining store
differentiation through locational convenience is likely to become more difficult.
With the market for the core ranges maturing, there is a greater tendency for price
imitation, which is relatively easy to match at short notice. According to
McGoldrick (1990), perceptions of store differentials will tend to erode over time
with imitation - real or superficial. Compared to speciality retailers, the food
retailers have less experience in developing and managing non-food categories,
and may have more difficulty in imitating a successful expansion strategy adopted
by a competitor. This could enable innovative retailers to exploit merchandising
variety and assortment as the most salient determinant of store choice and to
possibly sustain a differential advantage.

Assortment decisions require both a balance in depth and width across a

range of products and congruity with the target market positioning (Walters and
White, 1987). This balance should be sufficient to allow for growth, but not too
79
wide to tie up valuable space in unwanted stocks. McGoldrick (1990) suggests
there are no simple guidelines in making strategic decisions regarding the width or
depth of product assortment, resorting to what Greenley and Shipley (1989) refer
to as "management judgement" replacing market research. Findings by Hart
(1994) confirm a lack of research in to assortment congruity with the target
positioning. Retailers ranked the fit or congruency of a new range with the existing
assortment as the most important criterion for selection, but omitted to conduct
any image research at this level.

The importance of range congruency or compatibility with the retailer image

is further argued by Davies and Brooks (1989) who proposed a model of
interaction between department, store and chain images in establishing an overall
image for the retailer. In the event of a poor relationship between department and
store image, customers would leave without purchasing, due to their expectations
being unfulfilled. While offering a new perspective to the image debate, the study
lacked the empirical evidence to support these observations. Research into retail
image has focused on either store image determinants or measurement ( e.g.,
Zimmer and Golden, 1988), or product / brand image (e.g., Jacoby and Mazursky,
1984). What appears to be lacking between store and product image is the
empirical evidence into the assortment and its impact on the consumer. Indeed,
Berry (1969) recommended that analysis of specific image determinants, such as
assortment, was warranted.

According to Samli (1989) image is a complex combination of a number of

components which is different and greater from the sum of its individual
components. Reinforcing this, Jolson and Spath (1973) suggested that consumer
attitudes toward each component do not combine as expected to produce an
overall attitude toward the retailer. Additionally, each determinant of patronage
varies by the product group, so each category has to be handled separately. A
study by Cardozo (1974) of two product classes (housewares and health and
beauty aids) concluded that store image varies by product class, and each of
these may have individual images. Cardozo's research, like previous image
studies (Berry, 1969) focused on the department store. The supermarkets, by
contrast, have been the subject of image studies at the store level alone (Doyle
80
and Fenwick 1974; Davies and Brooks, 1989), with little empirical work existing at
the product group level. This is surprising considering the recent retailer
diversification strategies which suggest that the supermarkets are now faced with
competition not only from other supermarket chains but also from speciality
retailers. Although these competitive forces were previously observed by
Hollander (1966), market saturation in the food retail environment is a recent
development posing new threats. The previous image literature has identified that
categories need to be treated separately, and the current situation implies that
supermarkets are probably most at risk.

Objectives

This research study focuses on consumer perceptions of merchandise assortment

by product category. Specifically, it seeks to determine consumer perceptions of
the compatibility or "fit" of the range with the retailer image. Earlier research has
identified that the retailers believed the degree of range fit with other existing
ranges to be the major criterion in the introduction of a new range (Hart 1994).
This study also seeks to identify the extent to which range diversifications (such as
clothing and entertainment) are offering a suitable choice consistent with the other
established ranges. Thus, consumer perceptions toward choice in the non-food
ranges are examined. With few clear guidelines for retailer assortment planning,
these results may help build a picture of assortment depth and consistency from
the consumer’s view. Furthermore, a particular aim is to identify expansion
potential for existing non-food categories as a suggestion for future range
directions for the retailers. Finally, an underlying objective is to discover if the
consumer perceptions towards non-food ranges vary by store, according to the
image of the specific retailer.

Methodology
Merchandise assortment has many definitions, but there are limitations to their
usage with consumers; they can only be questioned on available ranges in familiar
stores and cannot accurately judge a prospective range without experience.
Consequently, only existing non-food categories were used for the subject of this
study. Three key variables describing characteristics of the range were selected
81
for analysis. These were: degree of fit, extent of choice available and importance
for expansion. The degree of range "fit" has been identified in Hart's earlier work
(1994). Secondly, consumer perceptions of "choice" or depth available within
each category was considered pertinent in identifying both merchandise saturation
and scope for expansion. Finally, consumer perceptions of each category's
potential for expansion were used to identify which categories were preferred, and
as an indicator of its relative importance.

The semantic differential scale (Mindak, 1961) is a popular measurement

technique, and was felt to be an appropriate means of measuring the intensity of
perceptions toward the various components of range expansion and store image.
This was supplemented by open-ended questions (McDougall and Fry 1974),
allowing further qualitative explanation. These two measures were combined with
additional ranking to form a structured questionnaire.

Six categories or product classes were used to investigate differences in

perceptions from one product class to another. Cardozo (1974) investigated the
differences in store image from health and beauty aids to housewares. These two
categories are currently stocked within supermarkets, and after consultation with
retail management, it was felt necessary to include all non-food product classes or
categories likely to be stocked in these superstores to compensate for the low
presence of certain categories in some chains. This resulted in three established
categories of household (including detergents and cleaning products), health and
beauty aids - HBA (cosmetics, toiletries and pharmacy), and home products
(kitchen accessories, tableware and small electrical appliances). The remaining
three were more recent diversifications of clothing (mens, ladies or children’s
outerwear, underwear, socks and tights) stationery (books, greetings cards and
newspapers) and entertainment (audio, video and toys).

A quota sample of 504 supermarket shoppers was selected, drawn from the
six largest food retailers in the East Midlands area, UK. The sample for each store
was stratified according to how each retailer was represented as part of the
region’s market. It comprised of 28% Sainsbury, 19% Tesco, 14% Safeway, 16%
Asda, 13% Co-op, and 10% Morrisons’ consumers which represented the region's
82
grocery market shares. Professionals trained by the Market Research Society
interviewed consumers in two major city centres in the East Midlands area during
January and February 1994. The response rate varied by each question
according to whether all six ranges were stocked. Discount stores were excluded
due to the low presence of non-food ranges, and by the same respect, smaller
town centre and edge of town stores were also excluded.

Store Loyal Consumers

The study sample was chosen from consumers who were store loyal to
supermarkets of approximately 25,000 square feet and over, which, by definition
qualify as superstores (Duke, 1991). Store loyal subjects were chosen because,
in the current competitive climate it is considered more appropriate to sustain
margins by nurturing relationships with existing customers than to build volume
growth (Denison and Knox, 1992). Secondly, a pilot test involving interviews with
loyal and non-loyal consumers showed that regular or loyal consumers to each
store held more accurate image perceptions than non-loyal consumers. It
appeared that non-loyal consumers had difficulty in recalling what ranges existed
in specific stores. As Berry (1969) confirmed, an image is acquired and learned
through direct experience with the store.

Using terminology from brand loyalty, store loyalty may be defined as

repeat patronage at a particular store for any products or brands (Osman, 1993).
To measure repeat patronage, it was necessary to decide the extent of frequency
of prior store patronage. In this study, loyal consumers are classified as those who
have purchased from the same store at least four times in the last six shopping
trips. Thus this classification determines whether a relatively regular pattern of
behaviour exists. Merely asking consumers for their store patronage behaviour
over a specified time period (say, 1 month) was considered inappropriate because
the frequency of supermarket shopping varies with the task situation (e.g., the
ownership of freezers reduces the necessity of weekly visits). The pilot test
suggested that consumers had no difficulty in naming the store to which they were
most loyal. Although several consumers tended to use one or two stores most
frequently, one was preferred, based on this classification.

83
Importance of Congruity and Range Fit

One dimension of image congruity (and incongruity) has been defined as the
match (or mismatch) between the store attributes and a consumer's perceived
important attributes (Anderson, 1973). Spiggle and Sewall (1987) have suggested
that when there is incongruity, there is likely to be store switching. This study uses
a different definition of congruity or fit, in the context of extension of product
categories. Since merchandise assortment is one of the key components of store
image (Lindquist, 1974), congruity is defined as the match between category
extension and the existing store image. It is suggested that incongruity between
new categories and store image, like incongruency between other components
(Spiggle and Sewall, 1987) could lead to failure of the category. As a measure of
category fit, the respondents were asked to indicate on a five-point scale the
compatibility of each category with the store. This was expressed as the
appropriateness for the store to sell each of the six categories, labelled as “range
fit”. The anchors were 1= inappropriate (these ranges shouldn't be stocked) and 5
= totally appropriate.

Perceptions of range fit, degree of range choice and importance of

expansion for each non- food category were cross-tabulated (Tables 2-4). For
each category, analysis was conducted by X2–tests for measuring any significant
differences between stores. The Pearson coefficient was used to test for any
association between perceptions of range fit and choice for each category, with
perceived importance of expansion (Table 5). Where the results for categories
showed a significant dependence on stores, frequency distributions by stores are
tabulated (Table 6). In the case of importance for expansion, clothing was rated
most important (from Table 4) and is examined further by store. Finally, any
relationship between frequency of purchase and perceived importance of
expansion is examined (Table 7).

84
Results and Discussion

Perceptions of Range Fit

Whereas over 75% of consumers considered the household and HBA ranges to
be totally appropriate, only just over 50% felt that clothing and entertainment were
totally appropriate. In other words, more consumers considered household and
HBA to hold a good fit with that store’s image (Table 2). Interestingly, over 15% of
consumers felt the clothing and entertainment ranges to be totally inappropriate
within their regular stores. The X2–test was applied to the crosstabs by store and
perception of range fit for each non-food category. To avoid small expected
values, the responses “totally inappropriate” and “tend toward inappropriate” were
combined, as were “totally appropriate” and “tend toward appropriate”, giving
tables with 10 degrees of freedom. The X2-tests indicated that there were no
significant differences in perceptions of range fit across stores except for clothes
(X2 = 32.84, P < 0.001) and entertainment (X2 = 28.6, P < 0.005). Therefore
perceptions of range fit may be dependent on the choice of store for clothes and
entertainment, but not for the other categories. The differences between specific
stores are examined in a later section.

Table 2 Perceptions of range “fit” in each non-food category

Clothing House Home HBA Stationery Enter’t
(n=481) (n=504) (n=488) (n=504) (n=495) (n=476)
Totally inappropriate 15.4 0.2 4.9 0.8 5.3 15.8
Tend towards 9.6 0.8 4.1 1.8 3.4 8.2
inappropriate
Feelings neither way 15.0 6.7 15.0 10.3 15.2 14.7
Tend towards 7.9 6.7 13.1 11.9 16.4 9.5
appropriate
Totally appropriate 52.2 85.5 62.9 75.2 59.7 51.9
All values are given as percentages.

The more frequently purchased categories of household and HBA appeared

to be what the consumers expected to find in a supermarket core range. The
slightly lower ratings of 75% fit for HBA may suggest that this category is not
85
perceived as a total core range by everyone, possibly due to pharmacies retaining
an image of specialism in this category, compared to household products being
predominantly stocked by supermarkets. The depth within the range may also
influence how it is perceived by consumers, e.g. more “health” items such as cold
remedies may be considered more core than cosmetics. Home and stationery
ranges were considered to have a level of around 60% fit, which meant that
consumers were more equally divided as to whether these constitute core or non-
core ranges. As discussed earlier, clothing and entertainment achieved an even
lower status. These four categories may be interpreted to have a lower degree of
fit for two reasons: (1) consumers consider the category to be incongruous with
the core ranges of food and household; and (2) the range is incorrectly positioned,
and thus the retailer is perceived to have a different image for core products
compared to non-core products. According to earlier research by Hart (1994),
these categories of clothing and entertainment were selected for their fit with
existing ranges or fit with the customer profile. These results indicate that
approximately only 50% of consumers appear to agree with this.

Dimensions of Choice

The perceptions of choice available within each category were measured on a

five-point semantic differential scale, in which 1= far too much choice and 5 = far
too little choice. At an aggregate level across all stores, very few consumers (≤
4%) considered the stores to have too much choice within any of the categories.
Indeed, more than 60% of consumers felt the ranges were "about right" across all
categories, particularly in household products, in which 91% of consumers
considered the range to be adequate (See Table 3). Both HBA and stationery
(83.6% and 81.9% respectively) were also considered to have sufficient choice. In
contrast, the clothing ranges were considered to have less than enough choice by
over 35% of consumers. Home and entertainment categories were also perceived
to offer insufficient choice by 25% and 21% of consumers, respectively.

86
Table 3 Perceptions of degree of choice for each non-food area
Clothing House Home HBA Stationery Enter’t
Degree of choice (n=425) (n=501) (n=458) (n=495) (n=470) (n=424)
Too much 0.7 0.2 0.4 0.4 0.8
More than enough 2.8 3.0 2.0 1.8 1.7 1.7
About adequate 61.2 91.0 72.7 83.6 81.9 76.6
Less than enough 12.2 2.0 12.7 8.7 9.1 9.9
Too little 23.0 4.0 12.4 5.5 6.8 11.0
All values are given as percentages.

These results reflect the perceived depth of assortment, with household and
HBA scoring highly as the more established core non-food ranges. Considering its
more recent addition to supermarkets, stationery achieved a high rating for
adequacy of choice. Any further specialization in this category could prove
unproductive resulting in saturation within the range. The perceived lack of choice
in the clothing, home and entertainment ranges may have resulted from
consumers comparing the supermarket ranges with the depth of range carried by
the specialists. Clearly, the key difficulties in carrying limited stocks of these
ranges lies in identifying the optimum product mix to suit the profile of consumers
on an average shopping trip.

Scope for Expansion

Having considered the fit and choice within each of the non-food categories,
consumers were then questioned on their attitudes toward expansion in these
ranges. Results were expressed in terms of each category's relative importance
for expansion. The six categories were ranked according to preference, in which
the anchors were 1 = most important and 6 = least important. The frequencies
reported in Table 4 indicate that at the aggregate level, across all stores, there is a
marked difference in levels of importance between clothes, household, home and
HBA on the one hand, compared to stationery and entertainment on the other.
Nearly 50% of the sample felt that the clothing, household and home ranges were
most/very important for expansion, compared to 53% of consumers feeling
stationery was least/less important and nearly 70% ranking entertainment as
least/less important for expansion. The differences attributable to the “most
87
important” label were tested using the X2-test for independence. The level of
importance attributable to the ranges is significant with X2 = 168.6, P <0.001,
indicating that clothing, household, home and HBA could be priority areas for
expansion, whereas stationery and entertainment are a low priority for expansion.

Table 4 Importance for expansion in non-food categories

Clothing House Home HBA Stationery Enter’t
(n=397) (n=410) (n=404) (n=402) (n=396) (n=389)
Most important* 34.5 28.7 17.8 9.7 6.3 6.7
Very important 12.3 18.8 29.2 26.6 6.1 7.2
About right† 23.4 32.9 40.6 49.0 34.4 18.3
Less important 14.9 9.8 8.5 9.2 33.8 22.6
Least important 14.9 9.8 4.0 5.5 19.4 45.2
* Difference between ranges is significant, P 0.001, Χ = 168.6
2
†
Label is a combined score taken from two responses
All values are given as percentages.

The consumers were then asked to indicate their reason for their expansion
preference in an open-ended question. The four most frequent responses for the
highest ranked categories are listed in Figure 6. The convenience and one-stop
shop factor featured strongly, with 21% of consumers feeling it was a strong
motivator for expansion in four categories - consumers would prefer to purchase
clothing, household, home or HBA items during their normal food shop rather than
visit a separate specialist retailer. Location was a key factor in this, with many of
the superstores being located on edge of town or out of town sites, offering limited
local specialist choice in these categories. This situation offers the retailers an
opportunity for non-food range profiling at the store level according to the depth of
assortment mix locally available. This is particularly relevant in the clothing
category where more choice or depth could motivate greater purchasing.

88
Figure 6 Reasons for expansion preferences

%
Clothing 22 Its more convenient to buy clothes while food shopping.
(n = 136) 20 More choice in clothing would make me buy more.
12 I’d like more variety in childrenswear/babywear.
9 There’s not enough clothes choice locally.

Household 38 Every home needs these items every week.

(n = 105) 24 There’s not enough choice in household goods.
13 I don’t want to shop elsewhere when I use these items every week.
11 Supermarkets are for household goods, food and convenience items.

Home 28 I want to buy home items while food shopping.

(n = 72) 15 I want a better selection all year round.
13 I like to buy new products for the home.
10 Home products would be better value in supermarkets.

Health/Beauty 42 There’s not enough choice available.

Aids 22 It saves going to the chemist.
(n = 36) 11 I buy these items every week.
11 I’d like more (manufacturer) brands as well as own brand.

This qualitative data also reinforced which of the categories the consumers
perceived to be core ranges for supermarkets. Household products were felt to be
an essential range, in part attributable to their high usage, and were subsequently
considered high potential for expansion. Health and beauty aids was also
considered a core range, but with less expansion potential - nearly 50% of
consumers felt that the currently stocked ranges were "about right". Lack of
variety/choice in the specific category appears to be an important factor for
expansion. This was most evident in the clothes and home sector where many
consumers felt that more choice in supermarkets would motivate them to buy more
of these products. Specifically for clothes, some consumers felt that
childrenswear/babywear should be expanded within their regular supermarkets.

According to Table 4, consumers rated home products highly in terms of

importance for expansion. But further analyses of qualitative responses indicated

89
that some consumers preferred these ranges for browsing and had a low
inclination to buy on a frequent basis. This would suggest that many of these
items may not be purchased on an average food shopping trip. Many home items
may be more suitable for the hypermarket or larger superstore of over 50,000
square feet where additional shelf space is available to offer an appropriate
choice. As a corollary, not everyone will shop in the same way, possibly creating
opportunities for specific home products to be sold by smaller stores. This
illustrates the need to investigate each range separately. A further element to
emerge in the qualitative results was the supermarket image for value. Many
consumers felt that specific ranges should be expanded because they would offer
better value in a supermarket than a speciality store.

The responses in Figure 6 implied that more choice may encourage more
purchasing, and this prompted analysis into whether correlations existed between
the consumers' perceived importance for expansion and their perceptions of
choice in that range. The results are given in columns 2 and 3 of Table 5, and
show clothing and entertainment categories to be significant at the 1% level.
Secondly, a correlation between range fit and importance for expansion was
tested. The results are given in columns 4 and 5 of Table 5, and again, clothing
and entertainment show significant correlations at the 1% level. These findings
show that for clothes and entertainment only, what is considered important for
expansion is significantly influenced by the current range in the patronised store.

Table 5 Relationships between perceptions of choice and fit with perceived

importance of expansion per category
Choice Fit
Pearson Pearson
Category Coefficient Significance Coefficient Significance
Clothing 0.37 0.01 0.48 0.01
Household -0.01 NS -0.10 NS
Home 0.10 NS 0.07 NS
HBA 0.08 NS -0.01 NS
Stationery 0.08 NS -0.01 NS
Entertainment 0.35 0.01 0.29 0.01
NS = not significant

90
Store Specific Perceptions

Since the clothing and entertainment categories were found to be highly significant
for range fit, further analysis was conducted into range fit for these categories for
specific stores. Table 6 gives the results for each of the six retailers. An
immediate observation is that some consumers are most enthusiastic about the
range fit, with 80% suggesting clothing ranges are most appropriate in Asda
stores.

Table 6 Store specific perceptions: (a) range fit, clothing and entertainment;
(b) perceptions of “too little” choice for clothing, home and entertainment;
(c) importance for expansion in clothing
Morrisons Tesco Asda Sainsbury Co-op Safeway
I* A* I A I A I A I A I A
(a) Clothing 39 55 21 63 8 80 32 52 21 64 31 45
(n = 479)
Entert’t 24 59 14 64 15 72 39 52 24 63 22 60
(n = 473)

(b) Clothing 40 32 28 34 39 45

(n = 424)
Home 19 23 20 25 17 49
(n = 456 )

Entert’t 22 21 8 27 25 24
(n = 421)

(c) Most/much 59 44 61 36 52 37
important
(n = 395)
About right 27 22 18 27 17 27
(n = 395)
Less/least 14 33 21 36 31 36
important
*Inappropriate (I) and (A) appropriate scores per store given only. A third neutral label is excluded from
this table. All values are given as percentages.

91
This could be attributable to the more extensive and established clothing range
stocked in the Asda stores, resulting in a greater awareness. However, Tesco
also carried a greater depth in their clothing range with 63% of shoppers
perceiving it most appropriate to the Tesco image. Morrisons, Sainsbury and
Safeway, who at the time of the study carried the most limited clothing ranges
scored 55%, 52% and 45% respectively on the appropriateness scale. These
results suggest that range fit may be related to its allocated space in proportion to
the other ranges stocked.

In examining the perceptions of the entertainment ranges, Asda consumers

again appeared most enthusiastic with 72% of consumers considering it most
appropriate, followed by Tesco (64%) and the Co-op (63%). Nearly 40% of
Sainsbury consumers felt entertainment was not appropriate within their regular
store compared to only 15% of Asda and Tesco consumers. These results
indicated that although a new category may have high market growth potential, it
could also be perceived as inappropriate for specific retailers. Clearly, clothes and
entertainment categories require some selectivity and sensitivity across stores,
with clothing considered inappropriate by nearly 40% of Morrisons’ consumers and
entertainment by 40% Sainsbury consumers.

The results suggested that the clothing, home and entertainment categories
were considered to have insufficient choice. Further analysis was undertaken to
determine how this perception of choice might vary by store as well as category.
The feelings of "too little choice" are given in Table 6(b). The X2-test was applied
to the crosstabs by store and perceptions of choice for each non-food category.
To avoid small expected values, perceptions of choice was grouped into three
labels : “too little”, “about right” and “too much”, giving tables with 10 degrees of
freedom. There were no significant differences at the 5% level except in the case
of the home category (X2 = 31.9, P < 0.005). An immediate observation is that for
all stores a large proportion of consumers considered the clothing ranges to be too
limited in choice and this would clearly present an opportunity for expansion.
Safeway consumers, on average, felt more strongly about the limited choice in the
home category with 50% of consumers feeling the home range is given too little
attention. Contrastingly, the other five retailers were considered to have sufficient
92
depth. This suggests that further investigation into the depth of specific retailers'
home ranges might be revealing. While 20% of consumers in most stores felt
there was too little choice in entertainment, only about 10% believed this to be the
case at Asda, with 87% feeling that the current Asda entertainments range was
"about right". Taken together, this indicates that perceptions towards category
expansion are store specific, in some cases contributing to merchandise
saturation.

The X2-test was applied to the crosstabulated results by store and

importance attributable to each non-food category. To avoid small expected
values, importance was grouped into three labels: “most/much important”, “about
right” and “less/least important”, giving tables with 10 degrees of freedom. There
were no significant differences at the 5% level except in the case of the clothing
category (X2 = 19.3, P < 0.05). Evidently, the clothing category emerges as the
strongest priority for expansion, but within which stores? The consumer
perceptions of the importance of expansion in clothing for their regular
supermarket is given by store type in Table 6 (c). Range perceptions divided into
two store groups with Morrisons, Asda and Co-op consumers appearing most
positive towards clothing expansion whereas Tesco, Sainsbury and Safeway
consumers were somewhat more conservative in their views, with a similar
percentage believing that clothing should hold less or least importance for
expansion. Interestingly, Sainsbury and Safeway consumers both exhibited an
identical profile with respect to their attitudes towards expansion in the clothing
ranges. This finding prompted the question whether Sainsbury and Safeway's
limited clothing ranges restricted purchasing, thus influencing the degree of
importance for expansion. Thus the relationship between purchase frequency and
perceived importance for expansion across the categories was tested using
Pearson's correlation analysis. The results are presented in Table 7 and show
that clothing and entertainment are significantly correlated at the 1% level with
coefficients of 0.30 and 0.28, respectively.

93
Table 7 Relationship between frequency of purchase and perceived
importance of expansion per category
Pearson Correlation Coefficient Significance
†
Clothing 0.30
Household 0.05 NS
Home 0.12 *
Health and Beauty 0.06 NS
Stationery 0.07 NS
†
Entertainment 0.28
*Significant at 5% level
†
Significant at 1% level
NS=not significant

In examining the store profiles for non-food categories, the Sainsbury

consumer emerged as more of a supermarket purist, preferring to concentrate on
core food and core non-food ranges. This may be due to their perceived image of
the non-core, non-food ranges of home, stationery, clothing and entertainment as
having a lower concentration of own brand products, thus diluting the total
Sainsbury image. In comparison, although Safeway consumers exhibited a
similar profile to Sainsbury consumers in entertainment choice and clothing
expansion, they were unique in their attitudes toward home products. The
Safeway consumer appeared marginally more amenable to non-food expansion
than the Sainsbury consumer. Further examination of perceptions of non-food fit
with the store image revealed four distinct groups: Sainsbury and Safeway
achieved the lowest aggregate compatibility rating for non-food fit with store
image, followed by Morrison which fell almost halfway between the
Sainsbury/Safeway and the Tesco/Co-op group. This latter group had an identical
aggregate range fit, but were still well behind Asda which appeared to have the
highest overall rating of non-food fit with store image.

It appeared that the majority of food consumers were favourable to the

expansion of non-food categories but out of the 349 qualitative responses to the
open ended question why should their chosen non-food category be expanded, 25
were neutral, preferring the ranges to remain the same, or were not interested.
Twelve responses were negative which were revealing in their feelings; six
responses preferred food only supermarkets, wanting clothing, entertainment and

94
stationery to be excluded to make way for more food, and at the extreme, "I avoid
Asda because it's got too many non-food items". Overall, these results have
shown that consumer perceptions are not only store specific but also category by
store specific implying that retailers must be aware, as concluded by Jacoby and
Mazursky (1984), that incongruous product groups may affect their image
negatively. Interpretion of the results need to be treated with some caution. The
findings of category expansion are based on perceptions to hypothetical
questions, and may require further studies to assess their reliability. Additionally,
the relationship between perceptions and actual behaviour towards range
expansion may not always appear concordant. This is due, in part, to the
complexity of consumer characteristics, in which mood changes and situational
variables may influence the responses at the time of questioning (Gardner, 1985,
Foxall, 1983).

Implications and Conclusions

A major problem facing the larger format superstores is how far to extend beyond
core ranges. With more non-food diversifications, the assortment could become
fragmented, eroding a sharply defined store image. This may result in consumers
having a confused perception of what a store stands for, perhaps conflicting with
their expectations (arising from overpositioning). The results demonstrate that
retail image is a more complex paradigm rather than simply a unified concept at
the store level. By focusing on merchandise assortment this study has shown how
consumer perceptions can vary at the product category level between stores,
displaying a secondary level of image. It appears that consumers subconsciously
divide the supermarket assortment into three groups: core, non-core and
peripheral merchandise. It is suggested that consumers might associate their
perceptions of each store with a unique set of categories representing core, non-
core and peripheral ranges. Core merchandise are those product ranges they
expect to find in a supermarket, including food and some non-food ranges. Non-
core merchandise are those ranges which are considered less essential to the
average supermarket visit (i.e. home and stationery), but are still perceived to be
enhancing the image. Peripheral merchandise are the ranges which are
considered least congruous with the store and with other ranges (i.e. clothes and

95
entertainment), and could potentially detract from the image. They may also be
the most diversified product groups. Identification of this grouping may enable
management to better understand how to influence consumer acceptability of their
ranges within stores and provide guidance in their assortment expansion.

It is suggested that variations in range incongruity between product groups

may result from differences in perceived purchase risk. It is likely that for higher
outlay items, such as some items of clothing and entertainment products, there
may be more to lose in making a poor decision, with higher associated perceived
risk. This may explain why fewer shoppers feel clothing and entertainment are
totally appropriate with the existing store image relative to other categories,
particularly if the current ranges of these two categories are considered restricted.
It might also help to explain why perceived importance of expansion of these two
categories is significantly correlated with frequency of purchase. In general,
consumers who have little experience of buying particular categories within their
usual store may share more anxieties in their expansion. Conversely, the
consumer is able to feel safer with an existing and familiar category. In this
context, categorization of items as either non-core or peripheral merchandise may
be influenced as much by the experience of the store profile as by the
characteristics of the merchandise.

One implication of this is the strategic need for retailers to adapt to

changing consumer preferences, (Corstjens and Doyle, 1983). In practical terms,
a decision not to stock a new range based on initial hostile consumer perceptions
may be premature. Retailers may need to use a combination of cues to evaluate
the feasibility of a new range such as targeted promotional activity to encourage
trial purchase. As the new ranges become more established, purchases, habits
and preferences may modify in favour of the new ranges. Thus, peripheral
categories might change to non-core, and non-core to core over time, requiring
retailers to monitor the stability of their consumer expectations and perceptions.
Core, non-core and peripheral items may need to be re-categorised according to
their specific store life cycles with consequences for merchandise selection and
space allocation decisions. Without tracking for these changes, future growth
areas are likely to be under-invested.
96
 Diversification into non-core and peripheral ranges should be treated with
care. Firstly, the need for expansion of non-food items naturally begs the question
of space management of food items. The identified core ranges of both food and
non-food should have an assured allocation, but the non-core food ranges may,
over time, be competing for space with non-core and peripheral non-food ranges.
Whilst there is always room for strong brands whose images compliment that of
the store, or departmental level, the vulnerability of weaker brands may result in
reduced facings, and unless stock turn can be improved, de-listing. Secondly, the
decision to extend the range further into non-food categories (say, with own labels)
may be influenced by both the stimulus generalization of the store image as a
whole, the perceived fit of the new range to other established ranges and the
current choice of ranges offered. The strength of the store name and reputation of
specific ranges within categories could be used effectively to reduce the higher
perceived risk associated with non-core and peripheral merchandise. Thirdly,
extending ranges in non-core and peripheral areas such as clothing and
entertainment may lead to more comparison shopping. With the customer
spending more time in the store, the opportunity for increasing more unplanned
purchases of other ranges within the store could be encouraged with creative
merchandising and display, contributing to added value.

It is suggested that for future retail research, image should be measured

not only on the store level, nor as totally separate product groups as suggested by
Cardozo (1974), but by core, non-core and peripheral product groups. The
implications of this for retail management are widespread; analyses of the target
market's perceptions of "core" and "non-core" non-food categories may result in
supermarkets within the same chain stocking clothes and not entertainments, or
vice versa, or only in seasonal periods. Alternatively, as Cardozo (1974)
concludes, "analyses of store image may be undertaken to explore opportunities
for repositioning various portions of a store's offering, to identify dimensions along
which such re-positioning might be accomplished.." The major retailers are
becoming more aware of the importance of individual store profiling, particularly in
the local marketing of food ranges and this study has identified that it has scope in
the non-food categories.

97
 Further research is warranted in three specific areas: (1) Analysis within the
category, at the product group level, to identify specific directions for expansion.
For example, within the HBA category, there is an enormous variety of purchasing
experience, in which the motives and context for buying up-market beauty
products may have different implications for store positioning than buying
unsought goods such as a range of headache remedies. Perceived risk may also
be reduced at the level of ranges within categories or assortments, and between
brands. (2) Examination of image differences between food and non-food
categories. To identify if core, non-core and peripheral product groups exist
across food categories and their relative compatibility with the store image. With
this information retail management may then choose to contract, reposition, or
expand ranges to strengthen their store image. (3) Non-loyal shopper perceptions.
This study has focused on loyal shoppers, but how do their category perceptions
differ to those of the non-loyal shopper? If retail management seek to convert
these into loyal shoppers, they need to discover the extent of importance they
attach to these categories.

98
References

Anderson, R.E (1973) Consumer Dissatisfaction: The Effect of Disconfirmed

Expectancy on Perceived Product Performance”, Journal of Marketing
Research, 10, February, pp. 38-44.
Berry, L.L. (1969), "The Components of Department Store Image: A Theoretical
and Empirical Analysis", Journal of Retailing, 45, 1, pp. 3-20.
Cadeaux, J. M. (1992) "Industry Product Volatility and Retailer Assortments",
Journal of Macromarketing, Fall, pp. 28-37.
Cardozo, R.N. (1974), "How Images vary by Product Class, a Psycholinguistic
Approach to Store Image Measurement", Journal of Retailing, 50, 4, pp. 85-
98.
Corstjens, M. and Doyle, P. (1989), "A Dynamic Model for Strategically Allocating
Retail Space", Journal of Operational Research, 34, 10, pp. 943-951.
Davies, G.J and Brooks, J. M. (1989), Positioning Strategy in Retailing, Paul
Chapman Publishing, London.
Denison, T. and Knox, S., (1992), “Profiling the Promiscuous Shopper”, A report
commissioned and published by Air Miles Travel Promotions Ltd, Crawley,
UK.
Doyle, P. and I. Fenwick (1974), "How Store Image Affects Shopping Habits in
Grocery Chains", Journal of Retailing, 50, 4, pp. 39-52.
Duke, R. (1991), "Post Saturation Competition in UK Grocery Retailing", Journal of
Marketing Management, 7, pp. 63-75.
East, R., Lomax, W., Wilson, J. and Harris, P. (1992), "Patterns of Supermarket
Use", Report by The Consumer Research Unit, Kingston Business School,
Kingston, Surrey.
Fisk, G. (1961), "A Conceptual Model for Studying Consumer Image," Journal of
Retailing, 37 (Winter), pp. 1-8, 54.
Foxall, G.R. (1983), Consumer Choice, Macmillan Press, London, pp. 86-97.
Gardner, M.P, (1985), “Mood states and Consumer Behaviour: A Critical Review”,
Journal of Consumer Research, 12, December, pp. 281-300.
Greenley, G.E and Shipley, D. (1989), "Dimensions of Product Planning in Retail
Marketing", Irish Marketing Review, 4, 1, pp. 53-62.

99
Greenley, G.E. and Shipley, D. (1992), "A Comparative Study of Operational
Practices among British Department Stores and Supermarkets", European
Journal of Marketing, 26, 5, pp. 22-35.
Hart, C. A (1994), "A Study of Merchandise Assortment Strategies within Major
Multiple Retailers", Retailing and Services Science Conference, Lake
Louise, Alberta, Canada.
Hollander, S.C. (1966), "Notes on the Retail Accordion", Journal of Retailing, 42,
pp. 29-40, 54.
Jacoby, J. and Mazursky, D. (1984), "Linking Brand and Retailer Images - Do the
Potential Risks Outweigh the Potential Benefits? Journal of Retailing, 60, 2,
pp. 105-122.
Jolson, M.A. and Spath, W.F. (1973), "Understanding and Fulfilling Shoppers'
Requirements: An Anomaly in Retailing?", Journal of Retailing, 49, 2, pp.
38-50.
Knee, D. and Walters, D. (1985), Strategy in Retailing, Theory and Application,
Phillip Allan, Oxford.
Kotler, P. (1988) Marketing Management: Analysis, Planning, Implementation and
Control, Prentice-Hall, Englewood Cliffs, NJ.
Lindquist, J.D. (1974), "Meaning of Image (a survey of Empirical and Hypothetical
evidence)", Journal of Retailing, 50, 4, pp. 29-38, 116.
Louviere, J.J. and Johnson, R. D. (1990), "Reliability and Validity of the Brand
Anchored Conjoint Approach to Measuring Retailer Images", Journal of
Retailing, 66, 4, pp. 359-382.
McDougall, G.H.G and Fry, J. N. (1974), "Combining Two Methods of Image
Measurement", Journal of Retailing, 50, 4, pp. 53-61.
McGoldrick, P. J. (1990), Retail Marketing, McGraw Hill, Maidenhead, Berkshire.
Mindak, W. A. (1961), "Fitting the Semantic Differential to the Marketing Problem,"
Journal of Marketing, 25, April, pp. 28-33.
Osman, M.Z. (1993), “A Conceptual Model of Retail Image Influences on Loyalty
Patronage Behavior”, International Review of Retail and Distribution and
Consumer Research, 3, pp. 133-148.
Porter, M. E, (1985) Competitive Advantage: Creating and Sustaining Superior
Performance, Free Press, New York.

100
Samli, A. C. (1989) Retail Marketing Strategy, Planning, Implementation and
Control, Greenwood Press, pp. 279-305.
Spiggle, S and Sewall, M.A (1987), A Choice Sets Model of Retail Selection, The
Journal of Marketing, 51, April, pp. 97-111.
Walters, D and White, D. (1987), Retail Marketing Management, The Macmillan
Press Ltd.
Warnaby, G and Woodruffe, H. (1993), “Business Strategy in Retailing: Towards a
Cost Effective Differentiation Model”, Proceedings of the 1993 Annual
Marketing Education Group Conference, published by Loughborough
University Business School, pp. 968 - 977
Zimmer, M.R and Golden, L.L (1988) "Impressions of Retail Stores: A Content
Analysis of Consumer Images", Journal of Retailing, 64, 3, pp. 265-293.

101
 The location and merchandising of non-food in supermarkets

Cathy Hart
and
Mark Davies

International Journal of Retail and Distribution Management,

1996, Vol. 24 (3), 17 – 25.

Abstract
Refers to the tendency in UK supermarkets for diversification into non-food areas
which has emphasized the need to improve how and where the products are
presented to the customers. Service diversifications such as dry cleaners or
opticians, have traditionally been situated as separate "shops" behind the
checkouts. Contrastingly, product range diversifications such as stationery,
entertainment and clothing have tended to be incorporated into the main shopping
aisles. Attempts to identify whether consumers would prefer these non-food
ranges to be differently merchandised and more specifically, whether consumer
preferences are different between retailers? Asks what alternative display formats
may be used and which specific ranges are considered appropriate for these
formats. Using six different retail multiples, examines the perceptions and
preferences of 500 shoppers from the results of an attitude survey carried out in
two UK cities. Finally, provides recommendations for retail management.

Keywords
Supermarkets, merchandising, layout, display, consumer behaviour.

102
Paper 3. The Location and Merchandising of Non-Food in
Supermarkets

Introduction
Over the past ten years supermarkets have expanded in numbers, size and
product range, while store layouts have remained virtually the same [1]. For
traditional grocery products, the retailer’s need to maximise space and profitability
was combined with the consumer's need for convenience. More recent
expansions of clothing, entertainment and stationery products to the supermarket
merchandise mix have meant that consumers are now buying a much greater
range of goods in the supermarket environment. Many of these items involve
relatively high outlay above that of traditional grocery products, with an expected
increase in perceived risk in purchasing.

This poses a dilemma for retail management; where should these non-food
ranges be located within the store? The initial tendency was to locate these
products in the main shopping area to generate impulse purchasing. This resulted
in ranges not achieving their full potential due to some consumers avoiding,
ignoring or not perceiving the products as intended. One suggested solution is to
offer “stores within stores” [1], whereby the merchandising of separate sections,
aisles or departments is handled uniquely and provide distinctive images. By this
route, shoppers can quickly locate specific ranges of products.

If the supermarkets intend pursuing the more profitable routes of non-food

diversification to augment their core food ranges, they must consider the
advantages of modifying the established supermarket store layouts. A greater
emphasis on specific ranges with an appealing layout could provide more
attraction to the store resulting in an overall enjoyable food shopping experience.
For example, Tauber [2] suggests shoppers might be stimulated by sensory inputs
(such as the texture and visuals of display). Shoppers might also develop unique
relationships with a store’s ambience, partly contributed by the display and
location of merchandise. A more conducive location of certain ranges could

103
encourage greater browsing and merchandise displays may dispose consumers to
buy on impulse [3].

Little accessible empirical evidence exists in this area, as most research

tends to be conducted by the practitioners on an ad hoc basis. In this paper we
examine consumer preferences for merchandise location to identify if they are
similar or different to current retailer practice. We are particularly interested in
consumer attitudes toward non-food ranges, and if and how these differ between
store profiles. The paper is presented in four sections. The first gives a
background to the evolution of supermarket layouts. The second gives the
methodology of the survey. The results and discussion are presented in the third
section. Section four concludes by offering recommendations and implications for
retail management.

The store environment

The physical nature of the store, its layout, appearance and physical presentation
of goods forms a major supporting role in the retail product offering [4]. The
purpose of in-store merchandising is to make products more attractive and visible
and thus motivate purchase. Within department, variety and speciality stores,
display and lifestyle merchandising can offer additional attractions. The nature
and homogeneity of food stores has a more limiting role for retailers with food
shopping viewed as something that needs to be done. In the self-service
supermarket environment, merchandising can be a vital but more subtle marketing
tool whereby shopper behaviour is modified and manipulated. Cook and Walters
[5] advise that supermarket merchandisers "should not experiment....but rather
stay with store layout patterns with which the customer has become familiar and
which do at least respond to the need for convenience and a rapid shopping visit".
They add "Conversely, ladies wear shopping is seen as leisure shopping and
enjoyed. It follows that visual merchandising can be creative in this situation".
The influx of new grocery competition [6,7], all offering "convenience and rapid
shopping" has meant that the food multiples can no longer afford to follow these
traditional principles and multiple food retailers are seeking new alternatives to
make food shopping a more pleasurable experience. The growth of the one-stop

104
shop concept combined with the need for higher margins has encouraged the
introduction of service and product diversifications in food stores. Preoccupied,
merchandising considerations have been overlooked. Many supermarkets have
expanded into the specialist ranges but have not carried through all of their
merchandising techniques. The effect can be somewhat confusing. Diverse
products such as CDs, clothing and stationery are sometimes displayed in
identical fixtures next to fast moving consumer goods. Non-food categories are
rotated, as retail management grapple to find the optimum store location. There
are a variety of merchandising techniques to aid the food retailer. This study
focuses on two of the most critical for supermarkets: layout and product category
location.

Store layout

There are three main layout options available to retailers, the grid, free flow and
boutique [8]. The straight parallel aisles of the grid pattern have been commonly
adopted by supermarkets because management is able to utilise space effectively
and gain maximum exposure to merchandise by steering customers to walk
around the whole store. The weaknesses of the grid include the
monotony/uniformity of straight continuous gondolas, particularly in larger
superstores, and the psychological effect on customers of feeling constrained and
rushed, thus reducing the time they spend browsing. The free flow layout in
comparison, encourages an unstructured traffic flow and greater browsing by
arranging fixtures in more irregular patterns. Its weaknesses include inefficient
space utilisation and customer disorientation. The boutique layout divides the
selling floor (or part of it) into individual speciality shops which may be
distinguished by changes in flooring, fixturing or displays. This style of layout may
be a result of using concessions (named shops within shops), but its main
advantage lies in a "more complete orientation of design towards the target group"
[9].

Left to their own devices consumers would prefer to shop around the
perimeter of stores and, consequently, these areas tend to be more congested
than central aisles. Congestion is intensified at major centres of attraction e.g.

105
delicatessen and fresh foods, which are placed around the store perimeter to
enable easy servicing [9]. Aisles are up to three times wider than they were ten
years ago to encourage more browsing, although this requires careful control
otherwise customers shop only one side of the aisle. A more recent introduction
has been a zigzagged wines and spirits aisle to reduce walking to a browsing pace
[10]. There is growing evidence of some UK food retailers using variations of the
three layout options; such as pursuing the free flow concept combined with the
grid, a boutique approach for concessions behind checkouts, or more recently
relocating some ranges such as clothes, as a fashion "shop" within the main
central shopping area. While it appears that the supermarkets are moving towards
the multi-layout approach, the grid still dominates the majority of the shopping
area.

A number of guidelines have been developed by researchers and

practitioners alike to aid merchandise location, but the underlying principle is to
position high-demand items to "pull" the consumer through the store past other
low-demand lines. The choice of location is dependent on consumer behaviour,
volume of traffic and product type. Buttle [11] suggested that items requiring
"deliberation" before purchase should be located away from the main arteries to
avoid congestion. Conversely, he advised that, "impulse lines should be where
the population is densest". In measuring performance, location quality has been
found to have a larger effect on sales of low volume than high volume products
[12]. "Prime" locations have been identified through tracking studies and
controlled experiments. Magazines sold at checkouts generate double the sales
per square foot compared to those in a main aisle display [13]. Merchandise
location can be an effective tool for boosting slow selling lines if used carefully, but
there is a tacit agreement between retailers that relocating products too often
frustrates customers and may result in the opposite effect.

In store behaviour

From 50 per cent of unplanned purchasing activity which takes place in self-
service foodstores, about 25 per cent is influenced by the external environment
[14]. Within the store environment, this may include the power of branding, point

106
of sale stimuli, personal selling and store atmospherics. The opportunity for using
these factors to best advantage will be influenced significantly by the style of
layout and display location.

Rows of aisles or similar large quantities of display may be less effective in

triggering a more thorough information search and browsing compared to varying
the layout format frequently. This is because our senses tend to be dulled with
exposure to continuous stimuli, referred to as “adaptation theory” [15]. We
suggest that changes in shape, size or direction may also provide anchor points
for learning, so encouraging more detailed information search. Lack of continuity,
through changes, may be likened to the interruptions in advertising that have been
suggested by Lewin [16] for improving memorability. Furthermore, unique
positions such as end of aisles provide greater visibility and, hence, a greater
opportunity to view products.

Faced with a continuity in layout, shoppers will rely more on quick decision
rules to make the shopping task easier. Selective perception enables shoppers to
screen out much visual clutter which is considered incompatible with their existing
value systems. When consumers are faced with time-poverty there is a greater
tendency to rely on peripheral cues as a means of assessing the worth of a
product compared to relying on the detailed quality of the arguments presented by
any in-store marketing stimuli [17]. Under peripheral processing, more unplanned
purchases are likely [17]. Peripheral cues, according to Petty and Cacioppo [18],
include visual stimuli like branding and packaging, but display aesthetics and
location would also be expected to be a part of this. Source effects are also
considered to be an important peripheral cue. A store name might have some
effect, as a source in the evaluation of products or brands. In some cases, shops
within shops emphasising the brand name may be influential. We consider that
display locations emphasizing branding may be more important for items which
are less frequently purchased, or for new category items, because of their
associated perceived purchase risk. This would obviously include many of the
new ranges expanded in supermarkets in recent years.

107
 The more extensive supermarket layouts do not appear to cater well for the
time poverty shopping segment. This can be improved by encouraging, "sufficient
browsing so as to select additional products" [19]. Davies and Bell [20] suggest
the male grocery shopper in particular, while seeking convenience, could be
tempted to spend more time and money in store if "a more interesting, less clinical
layout" is introduced by the food retailer.

Of course, retailers can experiment by testing consumer opinion with small

changes in the layout. However, this method is limiting in the styles of different
layout which can be tested due to the costs involved. This study is concerned with
understanding consumer preferences for alternative merchandise layout and
locations in store. It is hoped this will encourage practitioners and academics alike
to develop additional research in identifying the reasoning behind these actions.

The study
The initial aim of this study was to examine whether consumers are content with
the current in-store display locations on offer for different merchandise categories.
Six non-food merchandise categories are examined here. After consultation with
retail management, it was considered appropriate to examine the established
convenience goods categories of household, health and beauty aids and home
products, together with the newer categories of diversifications into clothing,
stationery and entertainment.

To examine the satisfaction with current in-store locations, we investigated

whether new categories of merchandise require different in-store display locations
from established categories, and if so, what they were. We felt this was important
to examine, due to the increasing variety of new categories being considered by
many supermarket retailers, which at one time were only stocked by the traditional
specialists. The study focuses on six major food retailers (identified as stores A-
F), excluding those with a low presence of non-food ranges (i.e. discounters and
small town centre and edge of town stores). The second aim was to examine how
any preferences for display locations might vary across specific stores.

108
 A quota of 504 supermarket shoppers was selected from the East Midlands
area. The quota for each store approximated to their respective market shares for
the region, with the largest retailer comprising of 28 per cent of the quota and the
smallest comprising of only 10 per cent. Since this study is concerned with how
in-store display locations may vary across different categories, it was felt most
practical to focus on large format stores, which would have greatest scope for
diversification into new categories. Hence, only superstores (above 25,000 ft2)
were examined. Data collection involved a structured questionnaire, designed to
test the above research questions. Professional interviewers trained by the
Market Research Society interviewed shoppers in two major cities in the East
Midlands during 1994. Results were analysed using SPSS.

The display locations

Prior observation of the display locations in the leading supermarkets indicated

that four major types of display locations were commonly featured which
potentially might be important for each merchandise category. It was considered
important that only major forms were analysed because these would be most
familiar to the sample respondents. If respondents are unfamiliar with a display
location format, they are unlikely to be able to express their true feelings, so
increasing respondent bias. This study restricts the options to four types of display
locations and so the relative importance of these is the aim of the study. The four
types are:

(1) main aisle displays within the main shop (similar to those normally used for
food);
(2) displays separate from food on different flooring or fixturing, but within the
main shop;
(3) displays outside the main shop floor in separate boutique format; and
(4) displays as separate shops located within the supermarket, but with a
different name or concession.

These four display locations were used for testing each non-food category:
clothing, household, home, health and beauty aids, stationery and entertainment

109
 products. Consumers ranked their relative preferences for each display location
against each non-food category. A four-point scale was used for measurement,
ranging from 1 = most important to 4 = least important. These preferences do not
indicate whether they are store specific. This could be significant because
preferences might be influenced by familiarity with specific kinds of display location
as a result of patronizing only certain stores (i.e. the store loyal consumer).
Therefore, display location preferences above were cross-tabulated with typical
store patronage to identify any store specific preferences.

Consumer preferences
The relative preferences for each display location option for each non-food
category are recorded in Table 8.

Table 8 Level of importance attached to each display location per non food
item (by percentage frequency).
Display location
Non food item Aisle Fixturing Boutique Name

a
Clothes (A) 30.0 59.7 63.4 48.3*
b
(B) 70.0 40.3 36.6 51.7*

a
Household (A) 93.2 94.6 8.1 4.2
b
(B) 6.8 5.4 91.9 95.8

a
Home (A) 58.4 71.0 44.1 27.0
b
(B) 41.6 29.0 55.9 73.0

a
Health and (A) 87.5 91.8 13.2 7.8
b
Beauty (B) 12.5 8.2 86.8 92.2

a
Stationery (A) 54.9 66.1 45.8 34.4
b
(B) 45.1 33.9 54.2 65.6

a
Entertainment (A) 32.7 51.8 64.8 51.6
b
(B) 67.3 48.2 35.2 48.4
Notes: (A) Labels ‘most’ and ‘of much importance’ combined
(B) Labels ‘least’ and ‘of little importance’ combined
a
highest percentage for item under (A)
b
highest percentage for item under (B)
c
Figures across approximate 200, not 100 due to combined responses above.
Figures do not tally exactly to 200 due to rounding.
110
With the exception of stationery, categories appear to be clustered for preferred
display location according to whether they are traditional or newer categories.
Thus clothes and entertainment, the most recent introductions for many food
retailers, are considered to be best located separately from the shop floor of the
traditional categories. Since this is a popular location choice of stores, this might
infer that category familiarity has a mediating effect on preferences of display
locations. What is evident is that four out of the six categories are considered best
displayed away from or different to the conventional aisle locations reserved for
foods, with the exceptions of household and health and beauty aid (HBA) items
that tend to be more established ranges for many stores. This, perhaps, conveys
a perception by patrons that many recently introduced categories do not mix well
with food. It is also suggested that perceptions of congruity between categories
may be used as a cue of category acceptability. Inappropriately mixing some
categories with food may also be considered unhygienic.

Figure 7 Most important (best) and least important display location for
different non-food items

Display location Most important Least important

(i) Aisle _______ Clothes and

entertainment

(ii) Fixturing Household, home,

health and beauty _________
and stationery

(iii) Boutique Clothes and _________

entertainment

(iv) Name _________ Household, home,

health and beauty
and stationery.

111
There are a number of other reasons which may account for some of the results.
Although "aisles" is a popular choice of display location for HBA, fixturing is most
preferred. Health and beauty products are often purchased not by the end user,
but as a gift, and categories located in special positions away from conventional
aisles may appear more expedient and perhaps easier to find for shoppers buying
unusual purchases or in a hurry. Locations would be easier to identify by being
more visible to store traffic. Alternatively, these “privileged” positions may convey
an appropriate ambience and atmosphere conducive for browsing.

One reason for the popularity of using separate boutique style shops
outside the shop floor for clothing and entertainment is that they are relatively high
outlay items, with an associated security risk. If carried around in a supermarket
trolley along with other less important purchases, there is a greater threat of
pilferage, with associated consumer unease. Conversely, consumers prefer
stationery, as a relatively low ticket item, to be located in alternative formats. The
preference in fixturing over aisles for stationery may reflect consumer desires for:
ease of selection, e.g. separate fixturing may provide the opportunity for
merchandise co-ordination across stationery ranges; and ensuring that fragile
items (e.g. envelopes) do not become soiled or creased. If they were located with
food items, there would be a danger they would become damaged in the bottom of
a full trolley of goods.

The relative importance attributable to each display location by non-food

categories was cross-tabulated and analysed for particular store patronage.
2
Simple association tests were conducted using the X statistic (Table 9 shows the
results). There is no association between preferences of display locations and
stores patronized for all home products. Perhaps such goods are expected to be
located in similar places across all stores. Hence, it was felt there was less need
to analyse the data for home products by store type. One explanation might be
that the range of home items is relatively standardized between stores, so
requiring similar display formats. Since results are significantly different between
stores patronized for clothes, household, health and beauty, stationery and
entertainment goods, these were further analysed by store-level.

112
Table 9 Degree of association between display location preferences per non
food item and store patronage
2
Sample (n) Display location X calc statistic P Significance

Clothes
464 Aisle 12.57 0.028 *
464 Fixturing 12.98 0.024 *
465 Boutique 4.36 0.499 N/S
466 Name 22.49 0.000 **
Household
482 Aisle 36.33 0.000 **
480 Fixturing 39.82 0.000 **
479 Boutique 39.56 0.000 **
478 Name 38.58 0.000 **
Home
474 Aisle 3.33 0.649 N/S
473 Fixturing 7.18 0.207 N/S
472 Boutique 7.21 0.206 N/S
471 Name 1.93 0.858 N/S
Health and beauty
479 Aisle 15.19 0.100 N/S
478 Fixturing 22.31 0.0005 **
477 Boutique 17.47 0.004 **
466 Name 18.45 0.002 **
Stationery
468 Aisle 12.77 0.026 *
466 Fixturing 18.45 0.002 **
467 Boutique 16.75 0.005 **
468 Name 14.20 0.014 *
Entertainment
456 Aisle 4.90 0.428 N/S
456 Fixturing 13.96 0.016 *
457 Boutique 10.54 0.061 N/Sa
459 Name 11.28 0.046 *b
Notes: * Significant at 0.05
** Significant at 0.005
a
Marginal, but not significant at 0.05.
b
Marginal, just significant at 0.05.

113
Store-specific preferences

Table 10 shows how store patrons are grouped as regards preferences for each
display location of each non-food category. In examining clothes, the two most
preferred display locations were separate fixtures or boutiques, with aisles
considered the least important display location for most stores. Separate shops
appear to be considered a popular display location by one in five of all patrons,
with the exception of store E. Clearly, most shoppers believe clothes should have
a distinctive identity, physically separate from food counters, with fixtures and
boutiques as the most popular choices.

In examining the household category, the most important display locations

were considered equally to be fixtures and aisles, with little variation between most
stores. Generally, boutiques and within-store shops are very unpopular choices.
In examining health and beauty, aisles and fixturing are considered virtually equal
in terms of importance among four stores, except for stores A and D (in which
fixtures are preferred). In examining stationery, while most stores prefer different
fixturing, closely followed by aisles, marginally more store E shoppers feel the
boutique format is more important than fixtures. Store D shoppers consider
boutique to be most important, followed by name. Feelings towards the most
important display format for stationery appears to be very store specific. In
examining entertainment, boutique is considered most important for shoppers from
four of the six stores, in which fixturing is equally important with boutique for the
remaining two.

Clearly the results indicate that some store D and store E shoppers are
different from the norm. This could be attributed to the unique layout style of store
E, which combines the free flow concept together with the traditional grid pattern.
Further complexities arise with stationery and entertainment. The danger facing
store management is that not only should display formats and locations be
selected for particular non-food categories, but that patrons may vary according to
their store patronage. Hence, merchandisers should not consider a general
display location for all categories but use research within each non-food category
to identify specific requirements. Additionally what suits one store profile may not

114
be appropriate for another. Clearly, the typical patrons of store D and store E hold
different sets of perceptions from other shoppers.

Table 10 Most preferred display location by retailer

Non-food category
Retailer House HBA Home Stationery Enter’t Clothing
A F F F F FB FB
B FA FA F F B B
C AF FA F F B F
D AF F F B B BN
E FA FA A B B F
F FA FA F F FB FB
Key: A = Aisles, F = Fixturing, B = Boutique, N = Name

Since clothes and entertainment can be relatively large outlay products there is an
increase in perceived performance risk. It is for this reason they might be
expected to be shopping goods, and in some cases speciality goods, in which the
brand name is sought before shopping, and the name becomes more influential in
determining what is purchased. It was anticipated that shops within shops might
offer more assurances to the regular shopper as to where specific brands can be
obtained within the confines of a large spacious superstore (e.g. consider the
focused range of Burtons within a Debenhams store). Results indicated that one
in five respondents expressed a desire to have shops within shops for
entertainment and clothes. This was a significant increase on that scored for other
non-food categories. Although this might suggest shops within shops could be
designed for a niche market, named shops were not the strongest preference
indicated for either of these categories. For clothes, there was relatively less
preference indicated for named shops, with most requiring either fixturing or
boutiques. This would suggest that the importance of using separate shops within
a supermarket may be overstated for many consumers and that the power of
providing a separate identity within a large store may be less important than other
factors.

115
Conclusions and implications
Traditionally, food shopping has been viewed by many retailers and consumers
alike as a necessary but unpleasurable experience, compared to other types of
shopping which are viewed more as “leisure” or recreational shopping. The
diversification of supermarket ranges into non-food areas such as clothing,
entertainment and stationery provides retail management with a new opportunity
for developing a more pleasant shopping experience through creative
merchandising.

The survey results indicated that contrary to current retailer practice, most
shoppers preferred to see non-food categories displayed in formats other than the
main shopping aisles. It could be that in the development of out of town, one-stop-
shopping, consumers are missing the variety of shopping formats offered by the
High Street. Food retailers have only partially addressed this by incorporating
bakeries, fishmongers, butcheries and delicatessens mainly around the perimeter
of the supermarkets. Contrastingly, the majority of grocery, frozen food and dry
goods remain in a grid layout. This research suggests that many consumers may
find the current emphasis on the grid layout inconvenient and monotonous.
Furthermore, the tendency to mix in high demand items to pull consumers through
the stores may have been successful when the average shopping visit could cover
the complete shopping area but in the larger format stores this is not as
successful, resulting in time poor shoppers “cherry picking”, leaving "black holes"
[21].

The boutique format was rated almost equally with fixturing, suggesting that
shoppers do seek alternative modes of layout and format. Although respondents
were clearly aware of the shops within shops concept, they were not convinced for
a complete change over to this format in their regular store. Perhaps the solution
lies in a compromise between shops within shops combined with areas of different
fixtures / flooring to break the monotony of the grid pattern. Additionally,
superstores could develop the "zoning" concept originally suggested by Floch [22]
for hypermarkets, whereby those shoppers seeking a rapid shopping visit were
able to locate items in a smaller area without the frustration of shopping the
complete store. The free flow layout has not been fully exploited for supermarket
116
use. This format was not examined in this study and requires greater depth to
investigate its relative disadvantages. Many consumers may also seek more
reassurance from different merchandise locations for the newer categories
associated with higher perceived risk. This study suggests shopping experiences
in a category may influence display preferences. Stores therefore need to monitor
display preferences as they experiment with their merchandise mixes.

One avenue of future research is to examine why preferences for different

display formats appear to be somewhat store specific. One proposition is that
store-specific display preferences (Table 10) might be reflected in variations
between ranges of merchandise within each category. Another factor would be
variations in store profile motives and perceptions. Future research is also
recommended in micromarketing to individual branch level to examine patronage
preferences to reinforce or enhance store positioning. It is suggested that once
optimum store layouts are determined according to size, location of store and
typical length of visit by shopper typology, more logical locations and adjacencies
of both food and non-food categories could be examined at the product group level
for further refinement.

117
References

1. Schlossberg, H. (1992) "Tomorrow's retailing technologies on display today

at smart store", Marketing News, Vol. 26, No. 2, p. 2.
2. Tauber, E. M., (1972) "Why do people shop?", Journal of Marketing, Vol.
36, pp. 47-48.
3. Wilkinson, J.B., Mason, J.B., and Parksoy, C.H., (1982), “Assessing the
impact of short term supermarket strategy variables,” Journal of Marketing
Research, Vol. 11, pp. 72-86.
4. Greenley, G.E and Shipley, D., (1989), "Dimensions of product planning in
retail marketing", Irish Marketing Review, Vol. 4, No. 1, pp. 53-62.
5. Cook, D. and Walters, D. (1991). Retail Marketing, Theory and Practice,
Prentice Hall, Hemel Hempstead.
6. Buck, S., (1994), "Big names get bruised by price fighters", The Times, 31
August, p.18.
7. Duke, R., (1991), "Post saturation competition in UK grocery retailing",
Journal of Marketing Management, Vol. 7, pp. 63-75.
8. Lucas, G. H., Bush, R. P., Gresham, L.G., (1994), Retailing, Houghton
Mifflin, Boston, MA.
9. McGoldrick, P.J., (1990), Retail Marketing, McGraw-Hill, Maidenhead,
Berkshire.
10. Martin, P., (1990), “Trolley fodder”. The Sunday Times, 4 November.
11. Buttle, F., (1984), "Merchandising", European Journal of Marketing, Vol. 18,
Nos. 6/7, pp.104-23.
12. Curhan, R. C., (1974),"The effects of merchandising and temporary
promotional activities on the sales of fresh fruits and vegetables in
supermarkets", Journal of Marketing Research, Vol.11, pp. 286-94.
13. Mendelson, S., (1993), "In aisle magazine sales", Supermarket Business,
Vol. 48, No.10, p 850.
14. Phillips, H. and Bradshaw, R., (1993) "How customers actually shop:
customer interaction at the point of sale", Journal of the Market Research
Society, Vol. 35, No. 1, pp. 51-62.
15. Schiffman, L. G. and Kanuk, L.L (1994) Consumer Behavior, (5th Ed.),
Prentice-Hall, Englewood Cliffs, NJ.
118
16. Lewin, K., (1951) Field Theory in Social Science, Harper, New York.
17. Bitner, M. J. and Obermiller, C., (1985), “The elaboration likelihood model:
limitations and extensions in marketing", Advances in Consumer Research,
Vol 12, 1985, pp. 420-25.
18. Petty, R.E., and Cacioppo, J.T. (1986), “The elaboration likelihood model of
persuasion”, in Berkowitz, L. (Ed.), Advances in Experimental Social
Psychology, Vol.19, Academic Press, Harcourt Brace Jovanovich, San
Diego, CA, pp. 123-205.
19. McHugh, M. and Greenan, K., (1993), "Food shopping: the leisurely way
forward?" Journal of Strategic Change, Vol. 2, pp. 3-12.
20. Davies, G. and Bell, J., (1991) "The grocery shopper - is he different?",
International Journal of Retail and Distribution Management, Vol.19, No.1,
pp 25-28.
21. Hawkes, N., (1995), "The shopper always turns right", The Times, 18
February, p.1.
22. Floch, J. M., (1988), "The contribution of structural semiotics to the design
of a hypermarket", International Journal of Research in Marketing, Vol. 4,
pp. 233-52.

119
 The Dimensions of Assortment: A Proposed Hierarchy of
Assortment Decision Making
Cathy Hart
Senior Lecturer in Retailing and Operations Management
and
Mohammed Rafiq
Senior Lecturer in Retailing and Marketing

International Review of Retail, Distribution and Consumer Research,

Vol. 16 No 3, (July 2006), 333 - 351

Abstract
The merchandise assortment plays a fundamental role in retailing; it conveys the
positioning of the retailer, and is a key determinant of image and patronage.
However, existing literature covering a 50 year period is somewhat fragmented. A
key problem with the literature is that it does not distinguish between different
levels and dimensions of assortment causing potential ambiguity in their
interpretation in the modern retailing context. The lack of standardised terminology
could potentially inhibit future assortment research in, for example, assortment
measurement. In reviewing the literature, the critical dimensions of assortment are
clarified with conceptual definitions and a hierarchical classification of assortment
being proposed. Further, the assortment research is examined, identifying an
imbalance between operational and strategic assortment research. As retail
assortments expand and saturate, the need for assortment measurement research
is argued with associated implications for retail management.

Keywords: Assortment, merchandising, dimensions, definitions, framework.

120
Paper 4. The Dimensions of Assortment: A Proposed Hierarchy of
Assortment Decision Making

Introduction

A key function of retailers is to provide “an assortment of products and services”

for consumers (Levy and Weitz, 2004 p.7). Thus, the product assortment plays an
essential role in retailing strategy (Simonson, 1999, p.366), contributes a major
dimension of store image, (Berry 1969; Lindquist 1974; Mazursky and Jacoby
1986) and is an “important component of the retail mix” (Koelemeijer and
Oppewal, 1999 p. 341). Despite this status, the literature focusing on the
assortment is fragmented. Indeed, over a period of nearly fifty years, assortments
have evolved and changed shape, involving new management processes, yet the
same definitions are still in use causing potential ambiguity in their interpretation in
the modern retailing context. There is a need for more precise terminology,
particularly recognising newer developments such as category management. The
adoption of standardised terminology is also important in order to assist progress
in future assortment research; with no general agreement in the use of existing
terms and dimensions, it is difficult for researchers to successfully develop the
means for measuring and controlling assortments. In contrast, considerable
attention has been assigned to describing and analysing the operational
processes and practice of managing assortments. Furthermore, given the modern
propensity for retailer assortment expansion, overlap and saturation, it is also
surprising that little attention is given to the strategic issue of assortment width and
consistency. Limited discussion exists into “what goods are made available and
how retailers might assort them in innovative ways” (Savitt, 1984), with little
guidance offered for retail managers in the role played by the assortment to
achieve a consistent image (Davies and Ward, 2002, p.153).

This paper seeks to address these shortcomings in several ways. The first
is to review the assortment literature with a view to examining assortment
composition and roles over the previous 50 years. This will enable us to (i) clarify
the levels of assortment and (ii) identify the value of consistency as an assortment
121
dimension. From this we propose updated definitions for assortment and its
dimensions. A third objective of the paper is to outline a hierarchy of assortment
levels to facilitate improved management decision-making. Finally we also
evaluate previous assortment studies to identify where assortment research is
lacking and the directions for future research into measuring and controlling
assortments.

Assortment Definitions and Roles

Assortment definitions

Within the retail and marketing literature, academic interpretations and definitions
of assortment are diverse. An early marketing definition views the “selling
assortment” as the “total number of items which can be sold by a firm in given
transactions” (Balderston, 1956, p. 175). Furthermore, Hollander (1966)
interpreted assortment at the aggregate retail level of “number of lines” on offer.
Simonson (1999, p. 347) expands the description to include “the total set of items
offered by a retailer, reflecting both the breadth and depth of offered product lines”.
The economics view interprets the assortment as a ‘set’ of market goods that may
include substitute, complement and independent items of consumption
(Betancourt and Gautschi, 1990). Thus, according to one stream of literature,
assortment may be interpreted holistically and defined as the total merchandise
offer or “store assortment”.

A second stream of literature defines assortment as the selection (Berman

and Evans, 2001, p. 466), “range of merchandise offered to customers” (Baron,
Davies and Swindley, 1991) or “range of style, size and colour choices offered at a
particular time” (Lee and Kunz, 2001). Thus, the term assortment is used
synonymously with “choice” (Ghosh, 1990, p. 10) or “depth” (Samli, 1998) of
selection in a product group (Ghosh, 1990 p. 10, Newman and Cullen, 2002 p26).
Latterly, product groups have been organised into broader classifications or
categories that represent “an assortment of items (SKUs) the customer sees as
reasonable substitutes for each other” (Levy and Weitz 2004, p. 365). The
category grouping thus introduces a second layer of assortment within the store;
termed both “aggregate category assortments” and “SKU assortments” (Cadeaux,
122
1999). The category level definition of assortment is also expanded to include the
breadth in a category (Dhar, Hoch and Kumar, 2001, p. 171). A subsequent level,
termed the “brand assortment” suggests a subset of assortment whereby the mix
of brands varies independently of the number of categories (Savitt, 1984, p. 51).
In offering a selection of categories, retailers thus engage in multiple assortments.
Confusingly, the singular and plural terms of assortment may be used
interchangeably, referring to either the overall store assortment or individual
assortments of merchandise groups e.g. “category assortment”.

Further examination reveals that retailers do not exclusively compile

assortments. Manufacturers and wholesalers provide an assortment of products
and services (Handelsman and Munson 1985 p108). Thus, the retailer assortment
may be viewed as another layer of assortment in the “Macromarketing System”
(Cadeaux 1992 p 30), with potential disparity emerging between the assortments
offered by retailers and manufacturers, and the assortments desired by consumers
(Cadeaux, 1997). Consumers similarly select an assortment of items but for
personal consumption (Lee and Steckel 1999), thus questioning the traditional
marketing focus on consumer purchase of a single brand (Wind (1977). Indeed,
the earliest definition of the assortment concept emerges as “the collection, which
has been assembled by consumers in the hope and expectation of being prepared
to meet future contingencies” (Alderson 1965). Potentially, therefore, a multi-tiered
system of product assortments may exist across the consumption system.

Assortment roles

A search of the literature reveals that the product assortment assumes a number
of roles in retailing. The economics view holds that the assortment provides a link
between supply and demand; “Assortments are collections of goods and services
that allow consumers to fulfil a wide variety of needs at one point in time through
one-stop shopping” (Koelemeijer and Oppewal, 1999, p.320). This interpretation
implies that the essence of place / time utility is captured and adequately satisfied
through the one-stop shop. Whereas in reality, each shopping experience differs
according to consumers changing needs. Retailers therefore need to match
“changing arrays of products with changing demand for assortments from

123
households” (Cadeaux, 1997, p.197). This continuous matching or gradual
repositioning of assortments with the changing retail environment (Corstjens and
Doyle, 1989) elevates the product assortment to arguably one of the more
“dynamic dimensions of retailing” (Brown, 1990, p.146) given to shifts in breadth,
depth and consistency. Attempts have been made to link these shifts or changing
patterns of assortment with retail evolution. In extending the wheel of retailing
theory, Savitt (1984 p10) hypothesised that as retailers trade up, some of the
assortment may be rationalised. Similarly, the Retail Accordion theory
hypothesizes alternation between wide assortment and narrow line specialist
retailers (Hollander, 1966). It is therefore unsurprising that assortment emerges,
alongside another mix variable; price, as a descriptor or classifier of store formats.
Based on the width or depth of their assortment, retailers may be described as
‘limited line’, ‘speciality’, ‘category killers’, ‘variety stores’ or ‘mass merchandisers’
(Lewison 1997; Levy and Weitz 2004; Gist 1968; Hollander 1966). Originally, a
clear distinction was drawn between retailers offering either “variety” (wide
assortment) or “assortment” (deep assortment) to avoid “semantic ambiguity”
(Gist, 1968). But distinctions became blurred as retailers combined assortment
strategies to meet competition, offering both variety and depth for example, IKEA
offer a wide variety of home furnishings and accessories, with considerable depth
of variants in the majority of product categories when compared with the specialist
furniture retailer, MFI.

Strategically, the assortment decision can be interpreted as a key element

of the marketing mix in attracting consumers into the store (Kahn, 1999). The
“product” variable translates into the total assortment on offer (Onkvisit and Shaw,
1981) and in certain markets, is deemed a more important mix variable than price
(Stassen, Mittelstaedt and Mittelstaedt, 1999). While the retail mix includes a
bundle of goods and services (Lazar and Kelley, 1961), it is the (product)
assortment that tangibalizes the retail strategy and positioning. “The assortment
profile identifies, for the customer and the retail buyer, the positioning intent of the
organization” (Walters and Hanrahan, 2000 p.175). Assortment thus becomes a
strategic positioning tool to attract and retain core customers (Grewal et al. 1999).
For example, the department stores, in response to post-war competition from
discount general merchandise retailers, strategically shifted their product
124
assortments “to those for which they held their strongest differential advantage”
(Bucklin, 1983, p.157) i.e. high quality fashion oriented goods.

Assortment plays a key role in store image. In summarizing the literature

on image, Lindquist (1974) proposed that there were a total of nine different
attributes that contributed to the image perceptions consumers created about a
store. Within these, merchandise assortment / selection received most scholarly
mentions as an image attribute. Indeed, a number of empirical studies have found
that the retailer’s assortment along with other key attributes (e.g. location, price,
promotion, personnel and services offered), is an important factor in store choice
(Kelly and Stephenson 1967; Kunkel and Berry 1968; Buttle 1985; Mazursky and
Jacoby 1986). Moreover, assortment is seen as a key determinant of retail
patronage (Arnold, Oum and Tigert, 1983).

Within store, consumer decisions and choice processes are affected by the
set of product variants offered in the assortment (van Ryzin and Mahajan 1999;
Smith and Agrawal 2000), the current inventory levels of the variants (Mahajan
and van Ryzin, 2001 p.334) and space allocated to each variant (Borin, Farris and
Freeland, 1994). Thus, space allocation and inventory are intrinsically linked with
product assortment decisions (Urban, 1998). This view implies a unidirectional
process whereby the assortment influences consumer behaviour. Similarly, it has
been argued that retailers and marketing texts treat the role of assortment too
narrowly as simply a function in satisfying consumer wants. Indeed, it is
contended that assortments can shape consumer preferences and influence their
future purchasing( Simonson, 1999).

Levels and Dimensions of Assortment

The literature highlights that assortment definitions have one common factor.
Whether assortment describes the whole merchandise offer (Hollander 1966;
Simonson 1999; Walters and White 1987), or simply the depth of the offer
(Lewison 1997; Levy and Weitz 2004), it is essentially concerned with the number
of individual product units. Various ‘groupings’ of products are then applied in
order to manage the scale of the store assortment. For example, product class
125
describes the broad merchandise classification such as health and beauty aids
(Cardozo 1974-5). Certain fashion and mixed merchandise retail stores feature
departments, which are defined here as “a section within a store that carries
groups of merchandise the consumer views as complementary” e.g. The active
sports department in the John Lewis department store which features both sports
equipment and men’s / ladies active wear. Furthermore, the trend for
merchandise scrambling, particularly within the grocery retail sector (for example
Tesco’s expansion into non-food categories such as men’s and ladies apparel)
coupled with the introduction of category management, has facilitated the
development of the category level within a retailer’s merchandise offer. Category
is defined here as “a distinct, manageable group of products that consumers
perceive as related and /or substitutable” (Gruen and Shah, 2000) for example,
breakfast products. A key distinction appears to be in the type of retail business;
with grocery typically adopting the category approach, whereas larger non-food
retailers for example Marks and Spencer or House of Fraser tend to group first by
department, then category. Furthermore, substitutability within the category
suggests a generic; more stable group classification, whereas departments are
likely to experience greater change due to the complementary nature of the
products. For example, the Boots baby department originally carried large items
such as prams and cots before rationalising the offer to concentrate on the faster
moving, more compact, profitable lines of baby and nursery products. Thus within
each category are a number of product lines, a line being defined here as “a group
of related products” (Levy and Weitz, 2004) for example televisions, within the
category of home entertainment at Debenhams.

In clarifying assortment dimensions, an early analysis classifies assortment

into (1) commodity group (e.g. shoes), (2) the number of quality levels in each item
type, (3) the number of sizes, colours and variants within each quality level and (4)
degree of originality, style content (Balderston, 1956). Subsequent literature
expands the first and third classifications into the concept of dimensions,
predominantly referring to “width” and “depth” (Lee and Kunz, 2001). Table 11
summarizes the most frequently cited descriptors of assortment from a review of
retailing texts and assortment articles. From Table 11 it is evident that the
additional dimensions of “breadth” and “consistency” also feature. While the
126
former two dimensions describe the size of the assortment (Handelsman and
Munson, 1985), combined with the latter two dimensions they provide retailers with
a framework for assortment strategy decisions. However, it is noticeable that
some ambiguity exists in the use of width, breadth and depth, depending on
whether assortment is interpreted at the store or category level. Indeed, while only
two of the three terms are often cited, where all three of width, breadth and depth
are used, these refer to different levels of assortment (Samli 1989; Ghosh 1990;
Risch 1991; Levy and Weitz 2004).

In Table 11, assortment depth draws most agreement as the number of

brands (Samli 1998; Kahn 1999) or SKUs (Dhar, Hoch and Kumar, 2001) offered
within a category. Depth is further defined as, “The way of measuring the quantity
of each item available” (Rosenberg, 1995, p10) and “the number of units of each
SKU the store must hold to meet expected demand” (Ghosh, 1990 p296). It is
generally accepted that extent of assortment depth dictates the degree of
specializm offered by the retailer. A more precise interpretation of depth is “the
extent to which items in a retail assortment are net substitutes from a household
viewpoint” suggesting a relatedness of items (Betancourt and Gautschi, 1990, p.
150). This view is reinforced with depth interpreted as the variants within a
product category, and variants as substitutes for each other in consumption
(Kristensen, 1983).

127
Table 11 Dimensions of assortment in retail literature

Authors1 Level2 Width Breadth Depth Consistency

Hollander, 1966 S  

Gist, 1968 S  
C/D 
Kristenson, 1983 S  

Savitt, 1984 S   

Handelsman and Munson, S  

1985
3
Walters & White, 1987 S 
C/D 
Samli, 1989 S  
C/D  
Betancourt and Gautschi, S  
1990
3
Ghosh, 1990 S 
C/D  
3
Risch, 1991 S 
C/D  
3
Mason, Mayer, Ezell, 1994 S 
C/D 
Rosenberg 1995 S   

Messinger & Narasimhan S 

1997
Kahn, 1999 S 
C/D 
Simonson, 1999 S  

Cadeaux, 1999
4
C/D 
Berman & Evans, 2001 S  
C/D 
Dhar, Hoch & Kumar, 2001
C/D  
McGoldrick, 2002 S 
C/D  
Krishnan, Koelemeijer, Rao,
2003 C/D   
Ratchford, 2003 S 
3
Levy and Weitz, 2004 S  
C/D 
Key: C/D = Category/ department assortment; S = Store assortment; V = Variety
Notes:
1. While an extensive range of texts and articles were consulted, a much smaller set provided a definition or
interpretation. Where descriptors were derived, only the original sources have been cited.
2. Dimensions are cited at different levels of consideration e.g. store assortment, category assortment and
brand assortment
3. These authors use the term variety to describe “the number of distinct product classes carried by the store”.
4. Cadeaux (1999) also uses width to describe the brand assortment

128
 While commonly accepted as one of the original dimensions of the
assortment, it is observed in Table 11 that “width” is cited less frequently than
“breadth” in the literature. This is often due to both terms being used
interchangeably (Ghosh, 1990) with little evidence of the terms being used
concurrently. In attempting to differentiate between the two terms, width is used to
describe the number or choice of distinct product classes or categories carried by
the store (Gist 1968; Savitt 1984) and “the extent to which retailers carry
generically different types of goods” (Gist, 1968, p. 41). Width is often used
synonymously with variety, for example Risch (1991) defines it as “the number of
different classifications carried by the store”. Thus, the selection of different
categories or departments denotes the variety of the store assortment (Davies and
Brooks, 1989). As a term, “variety” is also used to describe the total number of
choices offered by the store and hence its interpretation depends on the context in
which it is used whether store or category level.

Table 12 provides a comparison with the terms used to describe the

product mix or assortment in the marketing literature, which may be attributed as a
key source of the assortment terminology in retailing. For example, authors have
most consistently used the term depth as the number of variants in a product line.
As in Table 11, width and breadth are used interchangeably, with the additional
dimension “length” being substituted for breadth (Boone and Kurtz 1999; Kotler
and Armstrong 2003).

129
Table 12 Dimensions of Product Mix / Assortment used in marketing
literature
Length/ Depth
Authors
0B Width
1B
Breadth (# variants of Consistency
2B

(# products/ each product

(# lines) brands within within a line)
lines)
Assael, 1993.  2   (consistency not defined)
(Breadth)
Boone and   (Length)  __
Kurtz, 1999.
Bradley, 1995  __ 3  (how closely product lines
(Breadth) are in customer use,
distribution, technology and
production)
Brassington    __
and Pettitt,
2003.
Dibb, Simkin,  __  __
Pride and
Ferrell,1991..
Evans and  __   (defined as relationship
Berman among product lines in terms of
(1995) their sharing a common end-
use, distribution outlets,
consumer group(s) and price
range, p259-260)
Kinnear, __  (Breadth)  (degree to which products are
Bernhardt, related to each other (p.291)
and Krentler,
1995
Kotler and   (Length)   (how closely related product
Armstrong, lines are in end use, production
2003 requirements, distribution
channels, or some other way
..p.290)
Schoell and 1 __ 3 __
Gulitinan,
1988.
Stanton, Etzel __  (Breadth)  __
and Walker,
1992

Notes:
1. Width and breadth are used interchangeably.
2. Lines also have breadth and depth (p.408-410).
3. Depth is defined as the number of items within each line

The dimension “breadth” has most often been applied inconsistently,

suggesting a lack of standardisation in assortment terminology. Handelsman and
Munson (1985, p.108) define breadth as “the number of different product
categories offered by the retailer” and are supported by Kahn (1999) who
discusses “breadth of assortment across the store (narrow or wide)”. Conversely,

130
breadth is described as “choice offered within each category” (Ghosh, 1990); “the
number of brands offered in a category” (Dhar, Hoch and Kumar, 2001, p. 171),
and the “number of units in each brand and style within each and every generic
class of products”, echoing a previous definition of depth (Samli, 1998, p. 296).
Furthermore, an empirical measure of breadth is proposed as “(average) items per
store”, (Messinger & Narasimhan 1997; Ratchford 2003) contrasting an earlier
interpretation of store assortment. Other interpretations of breadth are proposed
from the demand viewpoint; “the extent to which items in a retail assortment are
net independents”, (Betancourt and Gautschi, 1990) and “breadth of assortment
implies that product categories satisfy separate, fairly basic needs and are
complementary in the purchasing situation” (Kristensen, 1983). We suggest that
greater semantic clarity might be achieved by applying definitions that differentiate
between the store and / or category levels of assortment (see next section for
further elaboration).

The dimension of consistency has least mention in the literature however, it

is potentially one of the most important assortment dimensions from a retail
strategy viewpoint. An early definition derived from marketing and retail buying
texts refers to consistency as “the degree to which the different types of products
that comprise the merchandise assortment are related” (Gist, 1968, p. 254) thus
interpreting consistency as width at the store assortment level. The Wheel of
Retailing theory suggests that retailers not only rationalise some assortments, but
must also attempt to increase assortment consistency (Savitt 1984). Here greater
consistency is interpreted as increased specialization across product areas
suggesting closer relatedness across categories. The concept of ‘relatedness’ is
applied more broadly to “different components of the merchandise assortment”
having particular emphasis on “how they all relate to one another in the customer’s
perception” (Rosenberg, 1995, p. 10) Similarly, associated terms “cohesion”
(Ghosh, 1990, p. 324) and assortment “compatibility” are interpreted as the “effort
taken to coordinate merchandise selection into a coherent statement to the
customer” (Walters and Hanrahan, 2000). Merchandise co-ordination is
particularly important in fashion retailing where a themed “story” is often used for a
seasonal collection; and also for the customer expecting to assemble a complete
outfit from the assortment offered by the UK fashion retailer Next, for example,
131
who developed the lifestyle merchandising concept in the 1980s. A second
interpretation of consistency is rooted at the category level and describes the
stability of an assortment over time, termed “brand / item continuity” (McGoldrick,
2002). This is more recently defined as the “tacit promise made by a retailer to
carry a given set of brands, sizes, flavours, colours etc. from one period to the
next” i.e. full service stores such as Waitrose food stores. Here retailer
commitment to consistency is associated with higher acquisition cost due to
inability to take advantage of special trade deals (Krishnan, Koelemeijer and Rao,
2003) in comparison with for example, discounters such as Aldi or TK Maxx in the
fashion sector. The importance of assortment continuity can be further illustrated
by Marks and Spencer who were criticised by consumers for not providing
sufficient style continuity in their core range of ladies intimate apparel (Hart and
Dewsnap, 2001)

A Framework for Assortment Decision Making

It appears that the interpretation of assortment dimensions varies according to
which level of assortment is being discussed. Within the literature, the notion of
levels of assortment has evolved over time, representing the expanding retail
assortments, particularly in the grocery and department store sector. When
analysing the variations in definitions, terms, semantics and context of
assortments in the literature, it is possible to construct the composition of the
assortment (Figure 8). In order to elucidate terms for classifying past and guiding
future assortment research, the concept of assortment has been divided into a
macro and micro level respectively. The macro level represents the strategic level
or store assortment decision of mix of departments, merchandise classes or
categories to be carried by the store; for example ladies, men’s and children’s
apparel, soft furnishings, electrical goods, toys, etc. at a department store such as
Debenhams. The definition of width is thus proposed here as: the number of
departments or extent of different product classifications offered by the retailer.

The second, micro level represents the department or category level

(depending on size of the store), whereby we propose that breadth is applied to
the category / department level of assortment and is defined as the number of

132
product lines carried within a category. This way, the terms width and breadth are
distinguishable as referring to separate levels of assortment. Furthermore, depth
is thus defined as the number of variants within each product line; each variant
having individual sizes / colours, (termed stock keeping units or SKUs). This
forms the base of the micro level of the assortment.

Figure 8: A hierarchy of assortment decisions

width
MACRO Store Departments
LEVEL Assortment

MICRO Categories Category

LEVEL Assortment

breadth
Product lines

depth
(Number of SKUs
variants
e.g. brands,
styles etc.)

The above dimensions represent a means of describing and quantifying the

size of a specific assortment, whether at the store or category level. However, the
fourth dimension of consistency is notably absent from the classification. At the
macro level consistency can be defined as a means of expressing and measuring
the relatedness of departments / merchandise classes from the consumer
perspective. For the retailer, this is an important dimension as “the choice of
which merchandise groups (or departments), to feature in an assortment is central
to the strategy decision” (Walters & White, 1987), in particular the positioning of
the store. At the micro level, this additional dimension offers retailers a means of
tracking and controlling the category assortment to ensure that it is congruent with
store positioning, particularly where new product lines are being introduced.
However, methods for measuring this dimension at either the macro or micro
assortment level are significantly absent from the literature. Indeed, the rationale
133
for, and understanding of this ostensibly intangible dimension has not been
addressed in the previous assortment literature.

Assortment Research Focus

The constraint of which assortment of products to carry in a finite space has long
motivated scholars and practitioners to research the assortment selection
‘problem’ (Rajaram, 2001). Over time, many similar strands of assortment related
research reappear and these can be broadly grouped into three particular areas of
attention; space management, variety and consumer related. While these three
areas emerged chronologically, they are not mutually exclusive, with overlaps
occurring.

The majority of the literature is focused firmly at the micro level of

assortment i.e. that concerned with merchandise at the product unit (SKU) level.
Given that the accepted understanding of assortment is inherently based around
“breadth” and “depth”, essentially quantitative dimensions, the bulk of studies are
concerned with the number of individual product units within a retailer’s offer,
modelling units within the assortment, or the factors affecting them, in order to
improve operating performance. Rajaram (2001) observes that assortment
management involves choosing a portfolio of products to maximise profits subject
to budgetary and space constraints. It is therefore not surprising to find that a
majority of the previous research has approached assortment from the space
allocation viewpoint (Kahn, 1999, p. 289). Indeed, as retail selling space is a fixed
resource, assortment decisions centred on “which products to stock and how
much shelf space to allocate to those products” (Borin, Farris and Freeland, 1994).
A stream of empirical research is devoted to the problem of optimising shelf-space
allocation (see for instance, Corstjens and Doyle 1981; Bultez and Naert 1988;
Bultez, Gijsbrechts, Naert and Vanden Abeele 1989; Urban 1998). In addition to
optimisation models, previous space management research has featured
experimental tests and commercial applications (Dreze, Hoch and Purk, 1994). In
the grocery sector in particular, commercial space management models drive
assortment decisions, informed by EPoS data, item and direct product profitability
(DPP), combined with product / shelf dimensions to produce planograms.

134
 Variety forms the second major stream of assortment related research
possibly because assortment selection decisions influence consumer store choice,
inventory investment and allocation of display space (Stassen and Waller, 2002).
The importance of assortment variety is underlined by the fact that it is ranked
third in patronage motivations after location and price (Hoch et al., 1999). Indeed,
it is believed that the larger the assortment the more likely consumers are to find a
product that matches their exact needs (Baumol and Ide, 1956). Generally,
consumers are uncertain about future preferences thus prefer greater variety to
provide flexibility in their in-store decision making (Kahn and Lehmann, 1991).
The effect of broadening assortments includes an increase in complementary
sales to the existing customers; that is higher purchase per customer (Kristensen
1983). The satisfaction of a wider range of needs also attracts more customers
from wider socio-economic groups and from greater distances (Balderston 1956
p179). Conversely, high variety has been found to cause confusion among
inexperienced consumers due to the perceived complexity of the choice of options
available (Huffman and Kahn, 1998).

Variety management requires knowledge of substitutability and

complementarity patterns among items in the assortment (Koelemeijer and
Oppewal, 1999, p. 319). However, whilst complementarity has been recognised
as the essence of merchandising, it has received relatively little attention from
academic researchers (Walters, 1991). The role of substitutes in retail
assortments, on the other hand, has been investigated extensively both
theoretically (Katz 1984; Moorthy 1984; Oren, Smith, and Wilson 1984) and
empirically (Kumar and Leone 1988; Moriarty 1985). Much of this work confirmed
the benefits of variety and supported, for instance, the conventional wisdom of
supermarket managers that “more is better” (Boatwright and Nunes, 2001).
However, the growth of new item introductions in the 1990s, led to a saturation of
some assortments (Dreze, Hoch and Purk, 1994), thus prompting research into
item reduction. Surprisingly, it was found that simplifying choice could increase
sales. Consumers welcomed the “elimination of clutter brought on by the
reduction in redundant items” (Boatwright and Nunes, 2001, p. 50). Dhar, Hoch
and Kumar (2001), for instance, showed that reductions in the breadth and depth
of a category assortment are less noticeable to consumers in staple categories.
135
Furthermore, Van Ketel et al. (2003) identified that there is a point of optimal
variety in an assortment whereby assortment attractiveness is maximised; this
being the rationale behind category management. Thus consumers’ perceived
variety of an assortment has been found to be more important in influencing their
satisfaction and store choice (Broniarczyk et al., 1998). Indeed, it is recently
suggested by Morales et al (2005) that the retailer’s organisation of assortment
(i.e. external layout) and the degree of congruency with the consumers’ internal
schema can lead to increased perceptions of variety.

A third body of assortment-focused research may be grouped under the

general heading of consumer and demand related. Various efforts have been
made to understand consumer perceptions of assortment size (Handelsman and
Munson, 1985) and to model consumer utility from the assortment (Baumol and
Ide, 1956). A stream of research is emerging relating to the role that assortments
play in influencing shopper wants and preferences (Simonson, 1999; Lee and
Steckel 1999; Huffman and Kahn, 1998; Stassen, Mittelstaedt and Mittelstaedt
1999; Lesczye and Timmermans 2001). Simonson, (1999), for instance, provides
a theoretical framework for understanding the effect of assortments on shopper
preferences. Empirical work in the area includes Lee and Steckel, (1999) who
study how consumer preferences influence choice within assortment.

Noticeably, all of the previously discussed research is still firmly embedded

at the item level and thus within the “category assortment”. While the customer’s
purchase decision is made at this assortment level, it is not made in isolation.
Furthermore, as the product variable of the marketing mix, assortment has to be
consistent with the other mix variables of price, promotion and visual
merchandising to achieve a congruent image. Yet research into the linkages
between assortment and other retail mix variables is somewhat fragmented.
Notably this includes McIntyre and Miller (1999); Lal and Matutes (1994); Bawa,
Landwehr, and Krishna (1989) and more recently, Lesczye and Timmermans'
(2001) study which examines the impact of pricing, promotions, service, and
assortment on the choice of shopping strategy adopted by consumers. The
emergence of category management has expanded assortment research onto the
category level. For instance, the study of Basuroy, Mantrala, and Walters (2001)
136
demonstrates how in the case of one retailer the adoption of category
management increased the average unit price of the category and reduced its
sales volume and revenues. Dhar, Hoch, and Kumar (2001) demonstrate the role
played by the category in the store's overall portfolio. And, Campo et al's study
(2000) demonstrates the differential impact of location characteristics on
categories' attractiveness.

Given retailers' increasing propensity to manage assortments by category, it

is surprising that little attention has been given to how space should be allocated
between product categories (or departments). Corstjens and Doyle (1981)
recognised the need to consider groups of products rather than individual products
but with no guidance how this might be achieved. An exception to this is the study
by Rinne, Geurts, and Kelly (1987) who addressed the question of space
allocations to departments in a department store setting by proposing “a linear
programming routine which maximises the total gross profit margin for the store on
a monthly basis” (Rinne, Geurts and Kelly 1987, p.31).

In summarising the assortment research in relation to the redefined macro /

micro levels of assortment, studies may be further divided into an operational or
strategic role. While the literature explores both areas, there is a noticeable lack
of empirical and theoretical support for the strategic role of assortment. The
majority of assortment research being concerned with operational decisions and
the achievement of short-term objectives, such as improving operating results
(McIntyre and Miller 1999; Urban 1999), influencing purchase behaviour
(Broniarczyk, Hoyer and McAlister 1998; Simonson 1999), and measuring
performance (Grewal, Levy, Mehrotra and Sharma, 1999). The operational focus
is also noted from the practitioner’s point of view, as Blackwell and Talarzyk (1983,
p.8) suggest: “implementation of the marketing concept appear(s) to be mostly at
the firm’s operational rather than strategic levels”.

137
Figure 9 Assortment level research matrix

LEVEL

MICRO MACRO
1 3
• Shelf space allocation • Allocation by store
classification
• Visual merchandising
• Space allocated to
• Availability departments
OPERATIONAL
• Complements & substitutes • Store layout / location of
departments
• Price bundling
• Assortment overlap /
• Micro merchandising
Specialism or depth of category competition
assortment offered
FOCUS 2 4
• Specialism or depth of • Number of departments
• category assortment offered
Micro merchandising (width) carried

• Price level • Image / positioning of

individual departments
STRATEGIC • Quality of goods
• Store assortment
• Mix of manufacturers / own consistency
brands
• Diversifications into new
departments

Figure 9 shows an 'Assortment Level Research Matrix', which has been

compiled by combining the two predominant areas of assortment literature focus
with the proposed levels of assortment. The Micro-Operational and Micro-Strategic
quadrants are by far the most widely empirically tested areas of the assortment
literature; they are also the themes, which receive the most attention from a
practitioners’ viewpoint. Clearly the micro (product-unit) focus of the left-hand side
of the matrix, holds the most interest for researchers, as only a handful of papers
acknowledge the existence and importance of the macro-level of assortment (Hart
1999; Hart & Davies 1996; Cadeaux 1997 & 1992; Stewart and Hood 1983;
McGoldrick 1979; Cardozo 1974-75). Indeed, few have focused on assortment
movement except for Hart (1999), Sparks (2001) and Simova et al (2003). The
latter for example, examined the number of stores selling clothing, accessories
and mixed merchandise over a six-year period in the Czech Republic. None of the
previous streams of research including portfolio analysis, or category management
(Dhar, Hoch and Kumar, 2001) address the “complex problem of assortment
creation and modification within the marketing system” (Betancourt and Gautschi,
1990, p.147). Surprisingly, strategic assortment dimensions such as width and the
138
issue of which categories to stock (and their associated impact on store image),
have attracted scant research attention (Hart, 1999). As Cadeaux (1992)
highlights, “analysts have looked at factors that influence the probability of new
product acceptance by retailers, but they conduct that analysis for isolated
products rather than for entire product lines”. This could be due to the fact that
“there is little basis for predicting what new products could be sold successfully
with old assortments”, with decisions still being based on merchants’ intuition
(Balderston, 1956, p. 178). However, the aforementioned wheel of retailing
(McNair, 1958) accordion theories (Hollander, 1966) and retail life cycle
(Davidson, Bates and Bass, 1976) implicitly suggest the strategic use of
assortments in repositioning stores over time.

Discussion and Directions for Future Research

This paper sought to review, understand and clarify the definition and meaning of
assortment. In the process, the literature illustrated some ambiguity via the
inconsistent use and interpretation of assortment, offering limited guidance for the
future use of assortment terminology. Thus, an initial contribution of the paper is
to define assortment, its composition and the dimensions by which it may be
described. In so doing, it has become clear that a fourth dimension, consistency,
has been neglected in the literature. Arguably, the need for consistency relates to
the assortment’s role as a key store image / patronage attribute, with the image
that customers hold of a retail store dictating the scope of assortment consistency.
The consistency dimension thus potentially offers an important indicator of a
retailer’s ability to successfully match their assortments with the target consumer.
To date, little attempt has been made to develop a means of measuring this
dimension. The need to measure assortment consistency is similar to that of
measuring store image. “If it cannot be measured, it cannot be managed” (Samli,
1998, p. 207). If retailers wish to introduce a new department or category,
following trial and error process, their key measures of success will be sales /
profit performance. However, these quantitative measures fail to identify whether,
and to what extent the new department is perceived by consumers to be
consistent with the existing offer.

139
 Another reason attributed to the hitherto failure to define assortment is the
lack of a transferable term of measurement to describe the scope of an assortment
“without being compared to another assortment” (Kunz and Rupe, 1999). The
predominantly quantitative approach to measuring assortment performance
highlights a lack of subtlety in measurement. Retailers generally measure the
scope of merchandise assortment in stock keeping units or SKUs (Ghosh, 1990, p.
325). This somewhat blunt tool fails to distinguish between the different levels or
dimensions in which these items are managed. Furthermore, it ignores the
consumer’s view of groups of related products, or categories. Admittedly, at the
macro level, the dynamic nature of the assortment has so far rendered
measurement difficult beyond aggregate terms of percentage change (Bucklin,
1983; Hart, 1999). However, it is argued that a more sensitive form of assortment
measurement is warranted.

In evaluating and categorising previous assortment research, it is

noticeable that the majority of assortment research exists at the micro, individual
item level of assortment. Furthermore, there is a clear emphasis on operational
rather than strategically focused assortment research. That this macro-level of
assortment in a strategic role has been overlooked within the empirical research is
surprising, considering the role that assortment plays within the retail mix.
Moreover, given assortment’s role in store image, the link between assortment and
image has yet to be conceptualised. There is a lack of empirical investigations
into how assortment decisions are made and can be used positively to affect a
retailer's image and thus positioning. If assortment is viewed as the product
element of the retail mix, there appears to be limited understanding of its
interaction / relationship with other mix elements (or atmospherics etc.) as a
positioning tool. As merchandise scrambling continues apace, with the potential
threat of assortment saturation (Hart, 1999), it is suggested that the macro issue of
store assortment width warrants greater attention. We therefore urge greater
research attention into the fourth quadrant of Figure 9.

Fundamentally, the importance of standardising assortment terminology,

similar to that proposed in this paper, provides the first step towards systematically
assessing and quantifying the dimensions of assortment. Subsequent research
140
may then progress to developing and testing specific measures for each
dimension of assortment. Currently, retailers omit to conduct image research at
the category assortment level. In the future, an ability to measure the consistency
of an assortment over time, to check the congruency of a new addition with an
existing assortment, might better equip retailers to design, control and manage
assortments more effectively. Additionally, more research using the category
management perspective is required in this area, particularly given the retailers’
increasing propensity to manage assortments by category.

Finally, the implications of the Internet for assortment strategy cannot be

ignored. Achieving significant differentiation via assortments is likely to become
extremely difficult, given that there are no physical limits on merchandise offered
for on-line retailers. For instance, for the average book buyer, it is very hard to
find any meaningful differences between the selection of books offered by
[Link] and Barnes & [Link]. However, as previously discussed, high
variety can lead to negative evaluations of assortments due to frustration caused
by informational complexity and the difficulty of evaluating numerous choices.
Limiting the extent of relevant information, however, may reduce the perceived
complexity of a choice set (Huffman and Kahn, 1998). In the Internet context, this
may be achieved by providing an electronic shopping assistant /agent, tailoring or
making buying recommendations based on the customer's buying history or other
information provided by the customer (Alba et al. 1997). The store-based retailers,
on the other hand need to focus on merchandise that has important experiential
attributes and take advantage of their ability to provide information tailored to the
needs of customers. Improving experiential aspects of store visits normally
involves improving the store atmospherics and merchandise presentation but
increasingly the entertainment value of shopping is being recognised. Niketown,
and most shopping malls provide a good illustration of this trend (Alba et al, 1997).

141
References
Alba, J., Lynch, J.C., Weitz, B., Janiszewski, C., Lutz, R., Sawyer, A. and Wood,
S. (1997). "Interactive home shopping consumer, retailer and manufacturer
incentives to participate in electronic marketplaces," Journal of Marketing,
61 (July): 38-53.
Alderson, Wroe (1965). Dynamic Marketing Behaviour: A functionalist theory of
marketing. Homewood, IL: Richard D. Irwin: p. 47.
Arnold, Stephen J., Oum, Tae H. & Tigert, Douglas J. (1983). “Determining
Attributes in Retail Patronage: Seasonal, Temporal, Regional and
International Comparisons,” Journal of Marketing Research, 20 (May): 149-
57.
Assael, Henry (1993). Marketing: Principles & Strategy. (2nd Ed.). London: Dryden
Press.
Balderston, F.E. (1956). “Assortment choice in wholesale and retail marketing,”
Journal of Marketing, 21, 2 (October): 175-183.
Baron, Steve, Davies, Barry, and Swindley, David. (1991). The MacMillan
Dictionary of Retailing. Basingstoke, England: MacMillan Press.
Basuroy , Suman, Mantrala, Murali K. and Walters, Rockney G. (2001). “The
impact of category management on retailer prices and performance: Theory
and evidence,” Journal of Marketing, 65 (4): 16-32.
Baumol, William J. and Edward A. Ide (1956). “Variety in retailing,” Management
Science, 3-4: 63-101.
Bawa, K., Landwehr, J. and Krishna, A. (1989). “Consumer response to retailers’
marketing environments,” Journal of Retailing, 65(4): 471-496.
Berman, Barry and Evans, Joel R. (2001). Retail management. A strategic
approach. (8th Ed.). New Jersey: Prentice Hall.
Berry, L.L. (1969). “The Components of Department Store Image: a theoretical
and empirical analysis,” Journal of Retailing, 45: 3-20.
Betancourt, R. and Gautschi, D. (1990). “Demand complementarities, household
production and retail assortments”, Marketing Science, 9 (2): 146-161.
Blackwell, Roger D. and Talarzyk, W. Wayne (1983). “Lifestyle retailing:
competitive strategies for the 1980s,” Journal of Retailing, 59 (4): 7–27.
Boatwright, P. and Nunes, J.C. (2001). “Reducing assortment: An attribute-
based approach,” Journal of Marketing, 65 (3): 50-63.
142
Boone, Louis E., and Kurtz, David L.(1999). Contemporary Marketing. London:
The Dryden Press.
Borin, N., Farris, P.W. and Freeland, J.R. (1994). “A model for determining retail
product category assortment and shelf space allocation,” Decision
Sciences, 25 (3): 359-384.
Bradley, Frank, (1995). Marketing Management: providing, communicating and
delivering value. London: Prentice Hall.
Brassington, Frances and Pettitt, Stephen (2003). Principles of Marketing. (3rd
Ed.). London: Pitman Publishing.
Broniarczyk, S.M., Hoyer, W.D. and McAlister, L. (1998). “Consumers’ perceptions
of the assortment offered in a grocery category: The impact of item
reduction,” Journal of Marketing Research, 35 (May): 166-176.
Brown, S. (1990). “The wheel of retailing: Past and future,” Journal of Retailing, 66
(2): 143-149.
Bucklin, L.P. (1983). “Patterns of change in retail institutions in the United States
with special attention to the traditional department store,” International
Journal of Physical Distribution and Materials Management, 13 (5/6): 153-
168.
Bultez, A. Naert, P., Gijsbrechts, E. and Abelle, P.V. (1989). “Asymmetric
cannibalism in retail assortments”, Journal of Retailing, 65 (2): 153-192.
Bultez, Alain and Naert, Phillippe (1988). “S.H.A.R.P: Shelf allocation for retailers’
profit,” Marketing Science, 7 (Summer): 211-231.
Buttle, F. (1985). “Measuring food store image using Kelly’s Repertory Grid,”
Service Industries Journal, 5 (1): 79–89.
Cadeaux, J.M. (1992). “Industry product volatility and retailer assortments”,
Journal of Macromarketing, 12 (2, Fall): 28-37.
Cadeaux, J.M. (1997). “A closer look at the interface between the product lines of
manufacturers and the assortments of retailers”, International Journal of
Retail and Distribution Management, 25 (6): 197–203.
Cadeaux, J.M. (1999). “Category size and assortment in US macro supermarkets”,
International Review of Retail, Distribution and Consumer Research, 9 (4):
367–377.
Campo, K., Gijsbrechts, E., Goossens, T, and Verhetsel, A. (2000). “The
impact of location factors on the attractiveness and optimal space shares of
143
 product categories”, International Journal of Research in Marketing, 17 (4):
255-279.
Cardozo, Richard N. (1974-75). “How images vary by product class”, Journal of
Retailing, 50 (4): 85-98.
Corstjens, Marcel and Doyle, Peter (1989). “Evaluating Alternative Positioning
Strategies,” Marketing Science, 8 (2): 170-180.
Corstjens, Marcel aqnd Doyle, Peter (1981). “A model for optimising retail space
allocations,” Journal of Marketing, 27 (July): 822-833.
Davidson, William R., Bates, Alpert D. and Bass, Stephen J. (1976). "The Retail
Life Cycle," Harvard Business Review, 54 (November-December): 89-96.
Davies, Barry and Ward, Philippa (2002). Managing retail consumption.
Chichester, England: John Wiley and Sons Ltd.
Davies, G.J. and Brooks, J.M. (1989). Positioning Strategy in Retailing. London:
Paul Chapman Publishing.
Dhar, S.K, Hoch, S.J & Kumar, N. (2001). “Effective category management
depends on the role of the category,” Journal of Retailing, 77 (2): 165–184.
Dibb, Sally, Simkin, Lyndon, Pride, William M. and Ferrell, O. C. (1991). Marketing:
Concepts and strategies. (European Ed.). London: Houghton Mifflin.
Dreze, Xavier, Hoch, Stephen J. and Purk, Mary E. (1994). “Shelf Management
and Space Elasticity,” Journal of Retailing, 70 (4, Fall): 301–326.
Evans, Joel R. and Berman, Barry (1995). Principles of Marketing. (3rd Ed.).
Engelwood Cliffs, N.J: Prentice Hall International.
Ghosh, Avijit (1990). Retail Management. (2nd Ed.). Orlando, Florida: The Dryden
Press, Harcourt Brace.
Gist, Ronald R. (1968). Retailing: Concepts and decisions. New York: John Wiley
and Sons Inc.
Grewal, D., Levy, M., Mehrotra, A. and Sharma, A. (1999). “Planning
Merchandising Decisions to Account for Regional and Product Assortment
Differences,” Journal of Retailing, 75 (3): 405-424.
Gruen, T. W. and Shah, R.H. (2000) “Determinants and outcomes of plan
objectivity and implementation in category management relationships”,
Journal of Retailing, 76 (4): 483-510.

144
Handelsman, Moshe and Munson, J. Michael (1985). “On integrating consumer
needs for variety with retailer assortment decisions,” Advances in
Consumer Research, 12: 108-112.
Hart, Cathy and Davies, Mark (1996). “The location and merchandising of non-
food in supermarkets,” International Journal of Retail and Distribution
Management, 24 (3): 17–25.
Hart, Cathy and Dewsnap, Belinda (2001) “An exploratory study of the consumer
decision process for intimate apparel”, Journal of Fashion Marketing and
Management, 5 (2): 108-119.
Hart, Cathy. (1999). “The retail accordion and assortment strategies: An
exploratory study,” The International Review of Retail, Distribution and
Consumer Research, 9 (April): 111- 126.
Hoch, S. J., Bradlow, E. T., and Wansink, B. (1999). “The variety of an
assortment,” Marketing Science, 18 (4): 527-546.
Hollander, Stanley C. (1966). “Notes on the retail accordion,” Journal of Retailing,
42: 29-40, 54.
Huffman, C. and Kahn, B. (1998). “Variety for sale: Mass customisation or mass
confusion?” Journal of Retailing, 74 (4): 491-513.
Kahn, B., and Lehman, D. (1991). “Modelling choice among assortments,” Journal
of Retailing, 67 (Fall): 274-299.
Kahn, B.E. (1999). “Introduction to the Special Issue: Assortment Planning,”
Journal of Retailing, 75 (3): 289-293.
Katz, M. (1984). "Firm specific differentiation and competition among multiproduct
firms," Journal of Business, 57 (1): 5149-5160.
Kelly, F.R., and Stephenson, R. (1967). “The Semantic Differential: An information
source for designing retail patronage,” Journal of Marketing, 31 (Oct): 43-
47.
Kinnear, Thomas C., Bernhardt, Kenneth L. and Krentler, Kathleen A. (1995).
Principles of Marketing. (4th Ed.). New York: Harpercollins College Div.
Koelemeijer, K. and Oppewal, H. (1999). “Assessing the effects of assortment and
ambience: A choice experimental approach,” Journal of Retailing, 75 (3):
319-345.
Kotler, Philip (1967). Marketing Management: Analysis, Planning and Control.
Englewood Cliffs, N.J: Prentice-Hall Inc.: pp. 289-290.
145
Kotler, Philip and Armstrong, Gary (2003). Principles of marketing. (10th ed.).
Upper Saddle River, NJ: Pearson Prentice-Hall International.
Krishnan, Trichy, Koelemeijer, Kitty and Rao, Ram (2002). “Consistent assortment
provision and service provision in a retail environment,” Marketing Science,
21 (Winter): 54-73.
Kristenson, L. (1983). “Strategic Planning in Retailing,” European Journal of
Marketing, 17 (2): 43-59.
Kumar, V. and Leone, Robert (1988). "Measuring the effects of retail store
promotions on brand and store substitution," Journal of Marketing
Research, 21 (November): 178-185.
Kunkel, J.H. and Berry, L.L. (1968). “A behavioural conception of retail image,”
Journal of Marketing, 32 (Oct): 21-27.
Kunz, Grace I. and Rupe, Dana (1999). “Volume per stock-keeping unit for
assortment: A merchandise planning tool,” Journal of Fashion marketing
and Management, 3 (2): 118-125.
Lal, Rajiv and Matutes, Carmen (1994). “Retail pricing and advertising
strategies,” The Journal of Business, 67 (3): 345-370.
Lazar, W. and Kelley, E.J. (1961). “The retailing mix: Planning and management,”
Journal of Retailing, 37 (1): 34-41.
Lee, J. & Steckel, J.H. (1999). “Consumer strategies for purchasing assortments
within a single product class,” Journal of Retailing, 75 (3): 387-403.
Lee, Seung-Eun and Kunz, Grace I. (2001). “Assortment diversity in relation to
financial productivity: Contributions towards merchandising theory,” The
Journal of Fashion Marketing and Management, 5 (4): 303-312.
Leszczyc, Peter, Popkowski, T.L. and Timmermans, Harry (2001).
“Experimental choice analysis of shopping strategies,” Journal of Retailing,
77 (4): 493-509.
Levy, Michael, and Weitz, Barton A. (2004). Retailing Management, (5th ed.). New
York: McGraw Hill.
Lewison, D.M. (1997). Retailing, (6th ed.). New Jersey: Prentice Hall.
Lindquist, J.D. (1974). “Meaning of Image,” Journal of Retailing, 50: 29-38.
Mahajan, S. and van Ryzin, G. (2001). “Stocking retail assortments under dynamic
consumer substitution,” Operations Research, 49(3) May-June: 334-351.

146
Mason, J. Barry, Mayer, Morris L. & Ezell, Hazel F. (1994). Retailing, (5th ed.).
Illinois: Irwin.
Mazursky, D. and Jacoby, J. (1986). “Exploring the development of store images,”
Journal of Retailing, 62 (2): 145-165.
McGoldrick, P.J. (1979). “Store image: How departmental images differ in a variety
chain,” Retail and Distribution Management, (Sept / Oct): 21-24.
McGoldrick, Peter (2002) Retail Marketing, (2nd ed.). Berkshire, UK: McGraw-Hill
Inc.
McIntyre, S. and Miller, C. (1999). “The selection and pricing of retail assortments:
An empirical approach,” Journal of Retailing, 75 (3): 295-318.
McNair, M.P. (1958). "Significant trends and developments in the post-war period,”
in Smith, A.B. (Ed.), Competitive Distribution in a Free High Level Economy
and its Implications for the University, University of Pittsburgh Press,
Pittsburgh, PA: pp. 1-25.
Messinger, Paul and Narasimhan, C. (1997). “A model of retail formats based on
consumers’ economising on shopping time,” Marketing Science, 16(1): 1-
23.
Moorthy, Sridhar (1984)."Market segmentation, self-selection, and product line
design," Marketing Science, 3 (Fall): 288-307.
Morales, Andrea, Kahn, Barbara E., McAlister, Leigh and Broniarczyk, Susan M.
(2005) “Perceptions of assortment variety: The effects of congruency
between consumers’ internal and retailers’ external organisation”, Journal of
Retailing, 81 (2): 159-169.
Moriarty, Mark (1985). "Retail promotional effects on intra and interbrand sales
performance," Journal of Retailing, 61 (Fall): 27-47.
Newman, Andrew J. and Cullen, Peter (2002) Retailing: Environment and
Operations, Thomson Learning, London.
Onkvisit, S. and Shaw, J. (1981). “Modifying the retail classification system for
more timely marketing strategies,” Journal of the Academy of Marketing
Science, 9 (4, Fall): 436-453.
Oren, S., Smith,S. and Wilson,R. (1984). "Pricing a product line," Journal of
Business, 57 (1): S73-S100.
process for intimate apparel”, Journal of Fashion Marketing and Management, 5
(2): 108-119.
147
Rajaram, Kumar (2001). “Assortment planning in fashion retailing: Methodology,
application and analysis,” European Journal of Operational Research, 129:
186-208.
Ratchford, Brian, T. (2003). “Has the productivity of retail food stores really
declined?” Journal of Retailing, 79: 171-182.
Rinne, H., Geurts, M., and Kelly, J. P. (1987). “An approach to allocating space to
departments in a retail store,” International Journal of Retailing, 2(2): 27-
241.
Risch, Ernest H. (1991). Retail Merchandising. New York; Oxford: Macmillan.
Rosenberg, Jerry M. (1995). Dictionary of retailing and merchandising. New York:
John Wiley & Sons.
Samli, A. Coskun. (1989). Retail marketing strategy: planning, implementation and
control. New York: Quorum Books.
Samli, A. Coskun. (1998). Strategic marketing for success in retailing. Westport,
Connecticut: Quorum Books.
Savitt, R (1984), “The Wheel of Retailing and retail product management,”
European Journal of Marketing, 18 (6/7): 43-54.
Schoell, William F. and Guiltinan, Joseph P. (1988). Marketing: Contemporary
concepts and practices. (3rd ed.). London: Allyn and Bacon Inc.
Simonson, I. (1999), “The effect of product assortment on buyer preferences,”
Journal of Retailing, 75 (3): 347-370.
Smith, S.A., and Agrawal, N. (2000). “Management of multi-item retail inventory
systems with demand substitution,” Operations Research. 48: 50-64.
Sparks, Leigh (2001) “Cataloguing change: Argos catalogues in 1973 and 1998”,
International Journal of Retail and Distribution Management , 29 (10): 427-
441.
Stanton, William J., Etzel, Michael J. and Walker, Bruce J. (1991). Fundamentals
of Marketing. (9th ed.). London: McGraw-Hill.
Stassen, Robert E. and Waller, Matthew A. (2002). “Logistics and assortment
depth in the retail supply chain: Evidence from grocery categories,” Journal
of Business Logistics, 23 (1): 125-143.
Stassen, Robert E., Mittelstaedt, John D. and Mittelstaedt, Robert A. (1999).
“Assortment overlap: its effect on shopping patterns in a retail market when

148
 the distributions of prices and goods are known,” Journal of Retailing, 75
(3): 371-386 .
Stewart, D. and Hood, N. (1983). “An empirical examination of customer store
image components in three UK retail groups,” European Journal of
Marketing, 17 (4): 50-62.
Urban, T. (1998). “An inventory-theoretic approach to product assortment and
shelf-space allocation,” Journal of Retailing, 74 (1): 15-35.
Van Ketel, Eline, van Bruggen, Gerrit H. and Smidts, Ale (2003). “How assortment
variety affects assortment attractiveness,” Proceedings of the European
marketing Academy Conference, May, Strathclyde University, Glasgow.
Van Ryzin, G. and Mahajan, S. (1999). “On the relationship between inventory
costs and variety benefits in retail assortment,” Management Sciences, 45
(11, November): 1496–1509.
Walters, David and Hanrahan, Jack (2000). Retail Strategy. Basingstoke, England:
MacMillan Press.
Walters, David and White, David (1987). Retail marketing management.
Basingstoke, England: Macmillan Press Ltd.
Walters, Rockney G. (1991). "Assessing the impact of retail price promotions on
product substitution, complementary purchase, and interstore sales
displacement," Journal of Marketing, 55 (2): 17-28.
Wind, Jerry (1977). “Toward a change in the focus of marketing analysis: from a
single brand to an assortment,” Journal of Marketing, 41 (October): p. 12: p.
143.

149
 An Exploratory Study of the Consumer Decision Process for
Intimate Apparel

Cathy Hart
Lecturer in Retailing and Operations Management
and
Belinda Dewsnap
Lecturer in Marketing

Journal of Fashion Marketing and Management,

2001, Vol. 5 (2), 108 - 119.

Abstract
In contrast to outer apparel, academic research has neglected consumer
behaviour for intimate apparel or lingerie. It is argued that within this category the
bra deserves singular research attention. This paper presents the results of
exploratory research designed to explore in-depth the bra consumer decision
process. The key findings indicate a highly-involved consumer who is motivated
by a complex range of interlinked factors, and a consumer who desires to be
brand loyal in order to enjoy a less extensive decision process but who is
prevented from doing so by high levels of perceived risk and ‘obstructive’
marketing. The authors offer directions for future empirical research based on the
consumer behavioural constructs of involvement, perceived risk and the self-
concept. Implications for marketing management are also discussed.

Keywords: Consumer behaviour, intimate apparel, involvement, self-concept.

150
Paper 5. An Exploratory Study of the Consumer Decision Process
for Intimate Apparel

Introduction
As a product category, apparel or clothing is regarded as a complex consumer
product (Mitchell 1999). Apparel has therefore been a popular choice for studies
of consumer behaviour, particularly in relation to perceived risk (e.g., Dowling and
Staelin 1994; Jasper and Ouellette 1994) and involvement (Laurent and Kapferer
1985). In contrast to outer apparel, however, very little is understood about
consumer behaviour for the essentially invisible and highly personal category of
intimate apparel. Within this category the bra deserves singular research
attention. The complex physical nature of the bra together with the unique
physiological make-up of each bra wearer combine to suggest that consumer
behaviour for bras will be different to that for other apparel. In terms of physiology,
women’s breasts are not symmetrical, and no two female shapes are the same.
To compound the physiological problems, lifecycle stages affect breast shape, and
child bearing, menopause and hormones all conspire to change breast mass
(Goldsberry et al. 1996). To meet such variation in physiological size and shape
the bra manufacturer grades up to 60-plus commercial sizes across a wide range
of differently ‘engineered’ bra styles. No other garment, not even the apparently
complex footwear category, presents such an extreme marketing problem.
However, in the scant literature that exists it is argued that buyer behaviour is
influenced by more than the physiological. Examination of this literature reveals a
complex set of physiological, psychographic, functional, psychological,
psychosocial, and economic factors (Richards and Sturman 1977; Laurent and
Kapferer 1985; Ashizawa et al 1990; Mintel 1996; Koff and Benavage 1998; Yalom
1998).

This list of diverse influencing factors supports the assertion that consumer
behaviour for bras will be different to that for all other types of apparel. However,
as little empirical evidence exists to clarify how women approach the process of
buying a bra, manufacturers and retailers acknowledge that greater insights would
significantly aid their ability to provide marketing solutions to a market currently
151
estimated at £574m (Mintel, 1997). The present study therefore provides a vital
link between theory and practice via a qualitative study of bra consumers. The
study was commissioned by an international bra brand and represents a critical
first stage piece of research for their distributive retail partner, a leading UK
retailer. As an exploratory study it is designed to provide insights into the bra
buying process, to identify opportunities for further consumer behaviour research,
and to offer recommendations for marketing management.

In the section that follows, the academic basis of the study is presented
together with the study’s research objectives. The second section discusses
methodological issues relating to data collection and analysis, and section three
presents a consideration of the research’s key findings. Concluding the paper,
section four puts forward suggestions for future research and there is a discussion
of the research’s managerial implications.

Background and Research Objectives

The only extant empirical study to focus exclusively on consumer behaviour for
bras is Richards and Sturman’s (1985) research on how consumer lifestyle can
usefully act as a basis for segmenting the bra market. Correspondingly these
authors identify the five segments as the “conservative consumer”, the
“fashionable consumer”, the “brand conscious consumer”, the “outgoing
consumer” and the “home/price oriented consumer”. However, while this research
reveals important information about how consumer life-values and attitudes affect
the bra decision process, this paper does not suggest that the bra is any different
to any other item of fashion apparel. Examination of the remaining non-empirical
literature on the bra indicates that this garment delivers a unique set of benefits
and challenges to consumer and marketer alike, and it is therefore the contention
of this paper that the bra warrants its own research agenda. The bra as a harness
performs the important function of breast support to a vast range of physiological
shapes, sizes and tissue density/elasticity (Farrell-Beck et al 1998). Whether for
everyday, sporting or special occasion use, a bra affects the perceived comfort,
contour, posture, breast health, and physical performance of the wearer (Ashizawa
et al. 1990; Mintel 1996; Yalom 1998). Fit and comfort are therefore deemed

152
critical (Mintel 1997). It is therefore significant to note the anecdotal evidence that
women’s knowledge of bra fit and function is low, and that reportedly 70% of
women in the UK wear the incorrect size of bra (Farrer 1998). At a psychosocial
level, in spite of their ostensibly invisible nature as underwear, “Bras, by their very
design alter the shape of breasts for fashion” (Singer and Grismaijer, 1995). Thus
the external social identity (Feinberg et al 1992) achieved from the bra is derived
from a self-imposed shape and size. Koff and Benavage (1998) found how “Lower
self-esteem was associated with lower breast size satisfaction…while high public
self-consciousness, social anxiety, and appearance preoccupation were
associated with a mismatch between ideal and perceived size”. As the supported
breast presents a visible contoured cue to the world, it is intrinsically linked with a
woman’s sexuality, body-image and social image. Finally, then, from a
psychological standpoint bra styling may affect the wearer’s self-esteem via the
self-image achieved from the type of bra worn (Ostergaard 1999; Shim and Bickle
1994).

Against this background it is therefore surprising that the subject of bra

buying has received only scant attention. Aside from the Richards and Sturman
(1977) bra-focused study of lifestyle segmentation, a later study by Laurent and
Kapferer (1985) examines the concept of involvement across a whole range of
product categories, including the bra, and concludes that the bra consumer is
highly involved with the bra as a product. Finding only these two papers with a
focus on bra buyer behaviour, a wider literature search at the more general level of
apparel research identified the importance of the self-concept construct for
understanding and explaining the apparel consumer’s behaviour. The implications
for the bra decision process of previous research on both of these constructs -
involvement and self-concept - are discussed below. However, from the point of
view of understanding the total bra buyer decision process, to date no studies
have been undertaken. This research is designed to bridge this gap.

Engel and Blackwell (1982, p.22-23) highlight the benefits of using a model
of consumer behaviour as a “frame of reference” for research: a way of specifying
“the underlying variables, the nature of the relationships between them, and the
manner in which behaviour is shaped and affected”. As previous research
153
(Laurent and Kapferer 1985) established a high-involvement process for the bra, it
is considered appropriate to use the Engel Kollat and Blackwell (EKB) model of
high involvement decision process behaviour (Engel and Blackwell 1982) as the
framework for the study. However, it is the EKB model’s central five-stage
decision process framework (derived from the original Engel et al (1968) model)
that is used to structure this exploratory research agenda. Consideration of any
relevant “input” factors, facets of “information processing”, and “decision process
variables” from the more comprehensive EKB model are then explored within the
context of this central five-stage framework. Involvement and self-concept are
both considered as variables that mediate consumption behaviours (Engel et al
1993; Broderick et al 1998). As such, and in line with the proposed application of
the EKB model (Engel et al 1982), this exploratory study considers the impact of
both of these variables on the respective phases of the bra buying process.

In their study of involvement across a range of product categories, Laurent

and Kapferer (1985) found that bras have a high product category involvement
profile. Thus bras are seen by the consumer to have:

- personal relevance and, based on the probability and consequences of a mis-

purchase, associated high levels of risk;
- high pleasure value; and
- based on the product’s strong link with self-concept/personality (Laaksonen
1994), high symbolic value.

Laurent and Kapferer (1985) incorporated the bra and the dress as the two items
of apparel in their study across a wide range of product categories, and it is
interesting to note the comparative findings for these two products. Both were
found to have a high-involvement profile and both scored high averages on all
facets of the profile. However, while both product categories scored similar high
averages on perceived risk factors, the key difference was the bra’s relatively
lower scores on “pleasure value” and “sign (symbolic) value”. From the point of
view of understanding the bra consumer, the Laurent and Kapferer study does not
explain the respective involvement scores. The findings indicate neither why the
bra is seen as a high-risk purchase, nor why when compared with the dress it is
154
perceived to offer relatively lower pleasure value and fewer clues to the
consumer’s self-image.

Self-concept represents the “symbolic” component of the Laurent and

Kapferer’s (1985) involvement profile. Self-concept is relevant to the study of
consumer behaviour because self-perceptions and the image an individual has of
himself/herself directly motivate consumption behaviour (Grubb and Stern 1971;
Malhotra, 1988; Heath and Scott 1998). Laurent and Kapferer (1985), for
example, found how the product category’s “symbolic value” predicted the extent
to which consumers would actively seek information about the product. Although
no research to date has focused exclusively on self-concept in relation to the bra,
the concept has been used as an explanatory and/or predictive variable in
previous clothing research (e.g., Ericksen and Sirgy 1992; Feinberg et al 1992;
Goldsmith et al. 1996). Findings from these studies indicate that as people seek
to reflect a certain self-image through the clothes they buy and wear, so the
clothing they wear is closely linked to their self-concept (Kaiser 1990). However,
these efforts to understand and explain the self-concept in relation to clothing have
focused on outerwear, a category that is characterised by its “visibility” and “public
display” (Goldsmith et al. 1996; Feinberg et al 1992). This information can usefully
be supplemented by the study of the self-concept in relation to the essentially
invisible clothing category of underwear/intimate apparel.

To summarise, the aim of this research is to investigate consumer

behaviour for bras using the Engel et al (1968, 1982) five-stage decision process
model. More specifically, the objective is to explore consumer behaviour for bras
on the basis of both the decision process and the inputs to, and the outcomes
from, that process. Explicating the bra’s high involvement profile (Laurent and
Kapferer, 1985) constitutes the second key objective. Accordingly the aim is to
explore not only how the facets of the construct appear relevant to the bra buyer
decision process but also the reasons and background underlying the bra
involvement profile. Finally, in exploring the bra decision process the study seeks
to identify how, and which component(s) of the self-concept, appear to impact that
process. For the academic, insights gained from this exploratory study will guide
future empirical investigation and from a practitioner perspective, the information
155
and intelligence so gained will support marketing managers in their endeavours to
more accurately segment the bra market, to manage a more effective marketing
mix, and to better predict future bra buying behaviour.

Methodology
The lack of published research in the area of bra buyer behaviour and the
corresponding lack of detailed knowledge required about the bra consumer, her
motivations, values, attitudes, behaviour and bra usage prompted the
development of an exploratory research design for this present study (Churchill
1995). Furthermore, an exploratory approach supports the development of rich
preliminary insights of the broader issues and influences concerning bra
purchasing (Miles and Huberman 1994, p.10). Goldman (1962) has suggested
that the social interaction experienced within a group setting further supports the
delivery of rich research insights. In addition, and of particular significance, it is
suggested that the focus group allows a “spontaneity of discussion and the
disclosure of attitudes, opinions and information on present or prospective buying
and use behavior” (Green, Tull and Albaum 1988). Focus groups were therefore
selected as the appropriate method of data collection. However, it is important to
acknowledge how the increased social interaction of focus groups also implies
“that informant reaction may be influenced and modified” by the presence of peer
members (DeLorme and Reid 1999). Considering the sensitive and personal
nature of the research this may suggest that some respondents in a group
situation may be reluctant to divulge their attitudes and behaviour towards intimate
apparel purchasing. Thus to elicit their underlying motivations, a more discreet
research approach is required. Multiple methods of data collection were combined
into a three-stage research design. The use of combined methods has been
found to deliver synergistic outcomes (DeLorme and Reid 1999), including both
richer data (Fontana and Frey 1994) and triangulation of data (Denzin 1978).
Therefore their use also supports the key objective of gaining an holistic and in-
depth view of bra buyer behaviour.

The three stages utilised, respectively, projective techniques, open

discussion and a blind test experiment. The first part of the session, a word

156
association exercise, was designed to reveal the respondents’ individual
perceptions of the leading bra brands. To counter the risk of ‘peer contamination’
or social influence individual perceptions were captured on paper prior to the
general group discussion. Respondents described eight different bra brands using
a pre-tested list of forty positively and negatively affective descriptors. This
“controlled response” process (Chisnall 1992, p.162) lasted approximately ten
minutes. The second part of the instrument involved a semi-structured open
discussion designed to explore in-depth the bra decision-making process, and to
elicit the relevance of any facets relating to the involvement and self-concept
constructs. Engel et al’s (1968, 1982) basic five-stage buyer decision process
model outlined above provided a structure for the discussion. Previous research
has identified comfort and fit as key criteria in selecting a bra (Mintel 1997).
Exactly how consumers evaluate specific bras across these dimensions is not
known. The third part of the research instrument therefore incorporated a blind
test experiment using actual garments. Seven designs of medium control bra
brands were selected to represent a broad range of current bra styles.
Correspondingly, the twin aims of this stage were to prompt individual feedback on
product evaluation and design preferences, and to stimulate open discussion.
Central to the latter stage was the use of the third person technique (Haire 1950),
used here to explore the congruence of each bra/ brand with respondent self-
image.

The focus groups were selected purposively “to include persons who have a
common background or similar buying or use experience that relates to the
problem to be researched” (Green, Tull and Albaum 1988). The sampling criteria
were limited to four key variables: age, bust size, socio-economic group and
geographic location. To facilitate future comparisons across these variables a
stratified sample was designed to include six focus groups, and with the use of a
recruiting questionnaire, Market Research Society-trained interviewers recruited
forty-eight respondents to participate. The study targeted the middle range of 35-
64 years, a group that would have established their bra buying experience,
attitudes and beliefs. Representing approximately 37 per cent of the UK female
population (Mintel 1996), this age range would also include women in the “family
lifestage”, who had physically achieved “breast maturity” (Singer and Grismaijer
157
1995). To ensure representative distribution across age segments, the age range
was subdivided into three groups: 35-44, 45-54 and 55-64 years with eight
respondents recruited to each group. To reflect the majority distribution of bra
sizes in the population, respondents were recruited from the ‘medium control’
segment that covers “average” to “fuller busted” sizes. Finally, in order to reduce
the potential for sample bias or “sloppy sampling” (Sudman and Blair 1999), the
focus groups were recruited and conducted in three separate locations within the
Midlands and the South East of England.

An independent Market Research Society moderator, selected for

compatibility with the respondents, directed the focus groups. The eight focus
groups lasted on average ninety minutes. All were audio-taped and four were also
video-taped. The tapes were then transcribed and content analysed to determine
common themes and possible interrelationships.

Results and Discussion

The findings of this exploratory study are structured in line with the five buying
decision process stages of the Engel et al (1968, 1982) framework.

Problem Recognition

Existing literature, summarised above, contends that the bra buyer is influenced by
a complex set of physiological, functional and psychological motivating factors. In
terms of problem recognition, the majority of the respondents described how they
are initially prompted into the bra buying process by situational factors. These
included the garment’s loss of functional support through deterioration of elasticity;
the decline in a bra’s aesthetic appearance; the seasonal factor of holidays
prompting product replacement, and a change in the wearer’s own physiology
driving the need for a different bra size. For some, the prompt to purchase is
linked to the hedonic value (Laurent and Kapferer 1985) of “pretty” bras; in other
words, the way in which “pretty” bras are associated with "pleasing someone else",
and the wearing of them seen as a treat or reward: "It makes you feel special
really, spoiling yourself".

158
Information Search

Consistent with the behavioural implications of the high involvement profile for
bras first uncovered by Laurent and Kapferer (1985), the information search for
bras was described in this present study as both critical and extensive.
Considering the consumer’s wide experience of the product and the frequency
with which bras are replaced in her wardrobe, it is surprising that women find it
necessary to undertake an extensive search for each new bra purchase. This may
be attributed to the reported lack of style/size continuity driving the need for
consumers to restart the search process for each new bra purchase. Indeed, the
actual process of shopping for a bra was felt to be a positive, enjoyable experience
for only a third of the respondents, compared with the remainder who found the
experience "a nightmare", "frustrating", "boring", "a necessary evil", "traumatic".
The strongest reason for not enjoying the experience was the lack of attractive
bras in larger sizes, lack of choice in general, the difficulties encountered in finding
bras that offer fit, comfort and the right body shape through clothes. Further
analysis revealed that the experience was considered enjoyable when certain
conditions held. Key among this list of contingent factors were:
- when the respondent found a suitable bra and could return to buy a similar
one or repeat purchase;
- when they had "lots of money to spend";
- if they were a standard size, thus accessing more choice;
- if shopping for a bra to complement a special outfit or occasion.

The high involvement profile for bras (Laurent and Kapferer 1985) also
reflects high levels of risk. These authors found that the extensiveness of the
decision process is influenced above all by the consumer's perceived importance
of the product and the negative consequences of a mis-purchase, and to a lesser
degree by the perceived probability of making a mis-purchase. The risks as
described by the respondents in this study are allied to the difficulties that
consumers experience in securing a well-fitting and physically comfortable bra. A
mis-purchase is only fully confirmed once the garment has been worn and
washed, and the consequences of a mis-purchase are that a potentially expensive
product may then prove unsuitable and be left to languish ‘at the back of a drawer’.
In the Laurent and Kapferer (1985) study the second strongest predictor of an
159
extensive decision process was found to be the perceived differences among
alternatives. Although the respondents in this study did not refer to a number of
highly differentiated brands, they did cite that an additional factor underlining the
risk associated with bra purchase was the lack of consistency or variation in sizes
across different brands and even within the same style from one manufacturer. To
further compound the problem, when the consumer has found a bra that meets her
requirements there is the significant risk that the manufacturer will discontinue the
particular style.

As a variation of high-involvement decision process behaviour, Engel and

Blackwell (1982) introduce the idea of "routine decision making based on high
involvement" (p.33). When satisfaction from product use reinforces future buying
intentions, the decision process will be much simpler. In this situation, problem
recognition is followed by an internal search that immediately reveals a well-
formed purchase intention. This intent then leads directly to a purchase outcome.
Under these conditions high brand loyalty develops and is further reinforced by
continued satisfaction. However, the respondents in this study reported that the
discontinuation of styles and changes in bra brand fit and/or quality are major
obstacles to brand loyalty.

The respondents' express desire was for a far less extensive decision
process, particularly for those bras that they want to repeat purchase. Bra
consumers want to be brand loyal. Yet as a consequence of the perceived risk
factors the consumer admits a tendency to style- and store- rather than brand-
loyalty. They will patronise those stores that offer depth of range, size availability
and specialist advice. With respect to this last factor - specialist advice - an
important theme to emerge from nearly half of the respondents was the need to be
correctly measured and fitted for a bra. As many respondents admitted to not
knowing their correct size, they relied on “a little bit of trial and error” and were
therefore unsuccessful in self-selecting appropriate styles. One-fifth of
respondents reported disillusionment caused by being incorrectly measured in the
past, with a noted difference between sizes recommended by different retailers.
Once again this increases the perceived risk associated with the probability of a
mis-purchase. The perceived variation in fitting and measuring services,
160
combined with the inherent size variation between manufacturers and even within
the same brand, present substantial obstacles to the bra buyer. Seeking to
overcome such barriers to successful purchase, over a third of respondents
underlined the importance of the type of in-store advice associated with the
specialist lingerie retailer as opposed to generalist department store.

It is significant that the majority of women preferred to shop alone for bras,
unlike the preferred social interaction associated with fashion apparel, where “the
customers comparison to referent others appears to be of paramount importance”
(Dodd et al 1998). This finding supports the highly personal nature of the bra, and
reinforces the need to examine the consumer behavioural differences between
outer- and intimate- apparel.

Pre-purchase Evaluation

In terms of the evaluation stage of the buyer decision process, the results indicate
a two-stage process, with a third and final evaluation taking place post-purchase.
Stage one involves the consumer’s appraisal of the bra’s appearance. It is notable
that the respondents could determine the finished shape defined by a bra from a
number of physical styling cues. They could evaluate an evoked set of bras for
potential fit and comfort without trying on the garment. This status had been
achieved as a result of past buyer behaviour experience and, as an input variable
to the consumer evaluation process, represents a particularly important
consideration in a self-service retail environment. The decision rules adopted by
bra buyers are based on product attributes comprising the cut/shape of the bra,
whether under-wired or non-wired, type of fabric, seam positioning and
components. These attributes combine to form an holistic impression of the final
breast shape and contour visible through outer clothing. The fact that the
consumer is motivated to consider her social image is clear evidence of a link
between the bra and self-image, and underlines the importance of self-concept as
a factor that impacts the consumer decision process for bras. For this first stage
visual-evaluation, it is also important to observe that the consumer has a well-
developed understanding of bra brand images and positioning. In analysing the
word association exercise it became apparent that the respondents also held

161
strong attitudes towards the main bra brands. As a result, bra buyers were very
clear about the positioning of competing brands in terms of image, perceived price,
quality and probability of good-fit.

Ericksen and Sirgy (1992) concluded that self-congruity (the extent to which
a product image matches a consumer’s actual self-image) and ideal-congruity (the
extent to which the product image matches the ideal self-image) significantly
predict the outer apparel preference of women. For the purpose of this present
study on intimate apparel, a category that spans the utilitarian and the more
‘frivolous’/special occasion, it is significant to note how this previous research
found ‘ideal’ self-image somewhat more important than ‘actual’ self-image for the
“feminine” (as opposed to business-like) outerwear costume. For the purposes of
the present study, it is the number of bra types or bra wardrobes adopted by a
wearer that represents perhaps the strongest indication of the self-concept as a
mediating factor in the bra decision process. Even where there was a preference
for only one “universal” bra type, the respondent claimed that older, more
comfortable bras were worn in the day whilst new ones were reserved for evening.
There appears to be a clear distinction between older bras feeling more
comfortable and thus suitable for home, while socially - and therefore possibly
related more to either social- or ideal-self concepts than to actual-self concept -
these were perceived to provide neither sufficient support nor the most appropriate
external shape: "You wear your clothes for comfort every day. Whereas when you
go out at night you tend to wear something that’s perhaps not as comfortable; like,
we all squeeze ourselves into high heel shoes because they look good”. The fact
that women will suffer the pain of an uncomfortable bra in order to preserve a
particular contour and shape through clothing signifies that bras are indeed related
to self-image or self-concept.

In addition to affecting product/brand choice, however, self-concept may

influence how consumers view advertising, store image and salespeople, and the
ways they interact with these stimuli (Onkvisit and Shaw 1987). An understanding
of the self-concept’s influence on behaviour can therefore enable marketers to
develop effective marketing mix programmes; an aid to ensuring congruity
between the relevant elements of the marketing mix and the kind of self-concept
162
desired by the consumer (Onkvisit and Shaw 1987). In terms of store choice for
bras, the respondents in the present study appear to seek congruence between
their own profile/image and that of in-store sales staff. This finding is also
consistent with store image theory (Martineau 1958) where sales staff are seen as
a key part of the service proposition. Store choice also appears to be affected by
the presence and associated image of co-consumers. Older respondents reported
that the close proximity of women from younger, slimmer segments could be
obstructive to their bra purchasing.

An assessment of how well a bra fits takes place at the second stage of the
evaluation process. This involves the trying on of bras from a smaller evoked set
and takes place either within in-store fitting rooms or, due to time constraints and
the desire for privacy, at home. If this process is undertaken in store, sales staff
can exert a credible influence on the consumer and her decision process (Baron et
al. 1995), particularly given the reported difficulties that bra buyers encounter in
both establishing their correct size and finding suitable styles and sizes.

Purchase and Post-purchase Behaviour

On average and depending on the quality of the bras and how long they lasted,
the respondents in this study claimed to buy a bra three to four times a year.
Further, to counter the risk of obsolescence consumers tend to buy in bulk when
the product is available. Availability of the chosen bra style is one of the key
factors influencing purchase. Given the importance of comfort and fit, substitution
is rarely an option in bra purchasing. Owing to the previously mentioned problems
of limited availability of styles and sizes and manufacturer-generated product
obsolescence, the majority of the women categorised themselves as convenience
shoppers: “If I see a nice one I think, ‘there’s my size, I’ll buy it’, not because I
really need it. It saves me when I do need one having to go and hunt for one”. In
this way purchase is prompted by the marketing stimulus of an in-stock situation,
and underlines how practitioners can stimulate purchase just by having an item
available.

163
 In terms of number of bra purchases, the majority of the women bought at
least two bras at a time, though it is notable that some respondents claimed they
bought three or more at once. An alternative approach was to buy one bra, try it
on at home, wear it and wash it, then return to buy another exactly the same or in
a variety of colours. This propensity to repeat purchase and to possess multiple
copies of the same garment is unique to the bra, and it further underlines the way
in which the consumer is motivated first and foremost by functionality and
physiology. The strategy to return to repeat purchase is preferred in cases where
disposable income limited but, importantly, is due to the strong belief that a bra
cannot be judged for comfort and fit until it has been worn and washed. This
further underlines the perceived risk associated with bra buying. The
consequences of mis-purchase relate not only to the key objective to secure a
well-fitting and comfortable bra that provides the desired contour and supports the
target self-concept, but also to the need to avoid financial loss. After washing, the
bra cannot generally be returned for reasons of poor fit and will be rendered
useless. In order to reduce these high levels of consumer perceived risk, it is
important for retail sales personnel to be cognisant of this third post-purchase
stage of product evaluation, and their company’s selling and warranty policies
designed, where possible, to give assurances of quality and problem rectification.

Conclusions, Future Research and Managerial Implications

The insights gained from this in-depth investigation support the hypothesis that
consumer behaviour research should not treat the bra as just another item of
apparel. The bra buyer is motivated by a complex array of interlinked factors –
functional, physiological, psychological, psychosocial and economic. The key
contribution the research makes to existing findings is in explicating the individual
facets of Laurent and Kapferer’s (1985) high involvement profile and in revealing
the mechanics of the extensive bra decision process. In terms of the risk facet for
example, this study represents a first step in conceptualising risk based on the bra
consumer's key motivations. Future research could use this information to
produce better measures of the different types of risk, thus facilitating more
effective segmentation based on risk perception analysis and consumer risk-
reducing strategy (Mitchell 1999).

164
 High levels of consumer involvement and an associated extended problem
solving process support the usefulness of the EKB (Engel et al 1968, 1982) high
involvement decision process model for investigating consumer bra behaviour.
However, manufacturers’ and retailers’ efforts to market the bra as a fashion item
deny the consumer her desire for more “routine” decision-making (Engel and
Blackwell, 1982, p.33) based on brand loyalty. It would appear that conventional
theory for high involvement goods does not apply, and that the differentiating
factor appears to be the frequency with which this high-involvement product is re-
purchased. For this type of product, manufacturers and retailers cannot afford to
subject the consumer to risk each time they approach the decision process. In this
situation, too much innovation will be counterproductive to customer retention.
Successful new product strategy must therefore be based on a sound knowledge
of consumer perceived risk. Whether this apparent link between loyalty and
obsolescence-driven risk holds for all types of bra, and also for other products
categories, is a question that future empirical investigation could usefully address.

The difficulties consumers encounter in finding style and size availability

and consistency, coupled with their express wish to be brand loyal, underlines the
need for the manufacturers to work together and with their retail partners to
achieve industry-wide size-standardisation, manufacturing consistency, better
forecasting and more reliable in-store measurement systems. Also, in spite of
years of experience of bra buying the consumer still needs guidance through the
bra decision process. To support successful differentiation therefore, it is
recommended that the retailer offers highly-trained bra fitters and sales specialists.
It is the authors’ contention that if the buying process were facilitated and made
more pleasurable, the customer would shop for bras more frequently. This could
significantly aid market growth. More regular bra buying would also give the
manufacturer and/or retailer the opportunity to establish a closer relationship with
the consumer, and correspondingly, to alert consumers to any planned product
obsolescence.

Laurent and Kapferer (1985, p. 45) found that the bra as a product class
has significant pleasure value. By contrast, and on the basis of the problems
associated with purchase, the bra consumer decision process is seen as anything
165
but pleasurable, although it is acknowledged that this research did not discriminate
between the utilitarian and the special occasion or more frivolous ends of the style
continuum. In developing the involvement profile Laurent and Kapferer (1985)
found no association between the perceived risk facet and the facet relating to
pleasure. This present study uncovered no substantial evidence relating to the
pleasure value of the bra. However, in order to re-validate Laurent and Kapferer's
(1985) involvement profile, it would be useful to re-run the original correlation
analysis but on the basis of more refined measures of the risk and "hedonic"
facets. Finally, future research should seek to explore the proposed link between
a pleasurable bra shopping experience and the consumer’s propensity to re-
engage in that process.

The preliminary findings here on how the buyer’s decision is affected by

self-image also prompt the need for future research. As an exploratory study into
the consumer decision process for bras in general, the differences between
purchasing a bra for ‘everyday wear’ and one for ‘special occasions’ have been
largely overlooked in this study. The former is more of a utilitarian type purchase
and is therefore perhaps more consistent with actual self-concept. Arguably the
‘special occasion’ bra contains the more hedonic, pleasurable, fantasy aspects
(Ostergaard 1999) of bra purchasing and may therefore be related more to ideal
self-concept. An investigation into which particular component of self-concept
(actual or ideal) motivates the bra purchaser, combined with a study of the co-
variant effects of consumer demographics and perceived bra brand image, would
offer valuable insights into the construct of self-concept in relation to bra buyer
behaviour. Any insights gained would be of immense benefit to marketing
strategy and implementation: an aid, for example, to segmenting the bra market
based on self-concept and beyond this, to more effectively designing products and
associated promotional appeals.

This is an exploratory study conducted with a sample of 48 respondents.

The next step is therefore to validate the findings using a larger scale survey.
Also, in order to confirm whether all sub-groups reflect the same bra motivations
and behaviours, future research should replicate the present study with younger
and older groups and with the lowest and highest ends of the size scale.
166
References

Ashizawa, Kumi, Sugane, Akimi and Gunji, Toshihiro (1990), “Breast form
changes resulting from a certain brassiere”, Journal of Human Ergology,
Vol.19, No. 1, pp. 53-62.
Baron, Steven, Harris, Kim and Davies, Barry J. (1995), “Oral participation in retail
service delivery: a comparison of roles of contact personnel and
customers”, European Journal of Marketing, Vol. 30 No. 9, pp. 75-90.
Broderick, Amanda J., Greenley, Gordon, Mueller, Renee D., 1998. Utilising
Consumer Involvement for International Decision-Making in the Food Retail
Market. European Marketing Academy Conference, 27th EMAC
Conference, Stockholm, Track 2, pp. 481-500.
Chisnall, Peter M. (1992), Marketing Research 4th Edition, London, McGraw-Hill.
Churchill, Gilbert.A., Jr. (1995), Marketing research: Methodological foundations,
Fort Worth, The Dryden Press.
Dodd, Christopher A., Clarke, Ian, and Kirkup, Malcolm H. (1998), “Camera
observations of customer behaviours in fashion retailing: methodological
propositions”, International Journal of Retail and Distribution Management,
Vol. 26, No.8, pp. 311-317.
DeLorme, Denise E. and Reid, Leonard N. (1999), “Moviegoers’ experiences and
interpretations of brands in films revisited”, Journal of Advertising, Vol. 28,
2, pp. 71-95.
Denzin, Norman K. (1978), The Research Act: A Theoretical Introduction to
Sociological Methods, Chicago: Aldine.
Dowling, Grahame R. and Staelin, Richard (1994), “A model of perceived risk and
intended risk-handling activity”, Journal of Consumer Research, Vol.21,
No.1, pp. 119-135.
Engel, James F., Kollat David J. and Blackwell, Roger D. (1968) Consumer
Behaviour 4th Edition, New York, Holt, Rinehart and Winston.
Engel, James F. and Blackwell, Roger D. (1982), Consumer Behaviour 4th Edition,
New York, The Dryden Press.
Engel, James F., Blackwell, Roger D. and Miniard, Paul W. (1993), Consumer
Behaviour 7th Edition, Fort Worth, Texas, Dryden Press.

167
Ericksen, Mary K. and Sirgy, Joseph M. (1992), “Employed Females’ Clothing
Preferences, Self-Image Congruence, and Career Anchorage”, Journal of
Applied Social Psychology, Vol. 22 (5), pp. 408-422.
Feinberg, Richard A., Mataro, Lisa and Burroughs, W. Jeffrey (1992), “Clothing
and Social Identity”, Clothing and Textiles Research Journal, Vol.11 No. 1,
pp. 18-23.
Fontana, Andrea and H. Frey, James H. (1994), “Interviewing the Art of Science”.
In: The Handbook of Qualitative Research, N.K Denzin and Y.S. Lincoln
(Eds.), Newbury Park, CA Sage, pp. 361-376.
Goldman, A.E. (1962), “The Group Depth Interview”, Journal of Marketing 26, pp.
61-68.
Goldsberry, Ellen, Shim, Soyeon, and Reich, Naomi (1996), "Women 55 Years
and Older: Part II Overall Satisfaction and Dissatisfaction with the Fit of
Ready-to-Wear", Clothing and Textiles Research Journal Vol. 14 No.2,
pp.121-132.
Goldsmith, Ronald E., Flynn, Leisa Reinecke, and Moore, Mary-Ann (1996), “The
Self Concept of Fashion Leaders”, Clothing and Textiles Research Journal,
Vol. 14, No.4, pp. 242-248.
Green, Paul E., Tull, Donald S. and Albaum, Gerald (1988), Research and
Marketing Decisions, 5th Edition. Prentice-Hall, New Jersey.
Grubb, Edward L. and Stern, Bruce L. (1971), “Self-Concept and Significant
Others”. Journal of Marketing Research, Vol. 8 (August), pp. 382-385.
Haire, Mason (1950), “Projective Techniques in Marketing Research”, Journal of
Marketing, Vol.14, pp. 649-56.
Heath, Adam P. and Scott, Don (1998), “The self-concept and image congruence
hypothesis”, European Journal of Marketing, Vol. 32 No. 11/12, pp. 1110-
23.
Kaiser, Susan B. (1990), The Social Psychology of Clothing, 2nd Edition. Macmillan
Publishing Company, New York.
Koff, Elissa and Benavage, Amy (1998), “Breast Size Perception and Satisfaction,
Body Image, and Psychological Functioning in Caucasian and Asian
American College Women”, Sex Roles, Vol. 38, No. 7/8, pp. 655-673.

168
Jasper, Cynthia R. and Ouellette, Sara J. (1994), “Consumers’ perception of risk
and the purchase of apparel from catalogs”, Journal of Direct Marketing,
Vol.8, No.2, pp. 23-37.
Laaksonen, Pirjo (1994), Consumer Involvement, London, Routledge.
Laurent, Gilles and Kapferer, Jean-Noel (1985), “Measuring Consumer
Involvement Profiles”, Journal of Marketing Research, Vol. 22, pp. 41-53.
Malhotra, Naresh K. (1988), “Self-Concept and Product Choice: An Integrated
Perspective”, Journal of Economic Psychology, Vol. 9, pp. 1-28.
Martineau, P. (1958), “The personality of the retail store”, Harvard Business
Review, Vol. 36 No.1, pp. 47-55.
Miles, Matthew B. and Huberman, Michael A. (1994), Qualitative Data Analysis,
2nd Edition, Sage Publications, Thousand Oaks, CA.
Mintel Marketing Intelligence (1996), Lingerie (August). Mintel International Group
Limited.
Mintel Marketing Intelligence (1997), Bras and Pants (November). Mintel
International Group Limited.
Mitchell, A.A, 1979. Involvement: A Potentially Important Mediator of Consumer
Behaviour. In: Wilkie, W.L. (Ed.), Advances in Consumer Research 6, pp.
191-196.
Mitchell, Vincent-Wayne (1999), “Consumer perceived risk: conceptualisations and
models”, European Journal of Marketing, Vol. 33, No. 1/2, pp.163-195.
Mitchell, Vincent-Wayne and Hogg, Margaret K. (1997), “Exploring the
Risk/Motivation Relationship”, Proceedings, UK Academy of Marketing
Conference, Vol. 1, July, Manchester, pp. 48-57.
Onkvisit, Sak and Shaw, John (1987), “Self-Concept and Image Congruence:
Some Research and Managerial Implications”, Journal of Consumer
Marketing, Vol. 4 (1), pp. 13-23.
Ostergaard, Per (1999), “The feeling of being a real woman: Lingerie ads
interpreted by Danish Females”, Advances in Consumer Research, Vol. 26,
pp. 675-677.
Richards, Elizabeth A. and Sturman, Stephen S. (1977), "Life-Style Segmentation
in Apparel Marketing", Journal of Marketing, Vol. 41 No.4, pp. 89-91.
Shim, Soyeon and Bickle, Marianne C. (1994), "Benefit Segments of the Female
Apparel Market: Psychographics, Shopping Orientations, and
169
 Demographics", Clothing and Textiles Research Journal, Vol. 12, No. 2, pp.
1-12.
Singer, Sydney R. and Grismaijer, Soma (1995), Dressed to Kill: the Link
Between Breast Cancer and Bras. Avery Publishing Group, New York.
Sudman, Seymour and Blair, Edward (1999), “Sampling in the twenty-first
century”, Academy of Marketing Science. Vol. 27, 2, pp. 269-277.
Yalom, Marilyn (1998), A History of the Breast. Pandora, London.

170
 Category Management: A New Approach for Fashion
Marketing?

Belinda Dewsnap
and
Cathy Hart

European Journal of Marketing,

2004, Vol. 38 (7), 809 - 834.

Abstract
As a supply chain management initiative category management has to date been
the almost exclusive preserve of the grocery sector and within that sector, limited
to food categories. This paper proposes that the fashion industry might usefully
follow the grocery industry's lead and implement category management. A
comprehensive review of the literature on category management highlights the
opportunity for fashion marketing to consider the potential of category
management, and the specific research gaps. In operationalising the subsequent
research objectives, the paper reports the results of exploratory, in-depth
consumer research for a particular category of intimate apparel. The managerial
implications of these findings are then discussed in the context of the established
eight-step category management process. The overall tentative conclusion of this
study is that as a consumer-oriented joint planning tool, category management
offers retailer-supplier partnerships in the fashion industry an important adjunct to
the industry’s quick response methods. The paper closes with an agenda for
future research.

Keywords: category management, fashion marketing, quick response.

171
Paper 6. Category Management: A New Approach for Fashion
Marketing?

Introduction
Logistics-based Quick Response (QR) partnerships were initiated by the fashion
industry in the mid 1980s to maximise logistics efficiencies and costs (e.g.,
Giunipero et al., 2001). Almost a decade later the QR model was adapted by the
USA’s grocery sector as Efficient Consumer Response (ECR) (Fiorito et al., 1995;
Kotzab, 1999), and in European grocery markets, category management has been
adopted as a “cornerstone” and development of these original ECR initiatives
(Gruen and Shah, 2000; Kotzab, 1999). Supplier-retailer partnerships are central
to both QR and category management concepts. However, while the efficiency
concepts of QR and the USA’s ECR practices rely principally on retail sales data,
category management uses in-depth information on the consumer to support joint
marketing decisions for the category (Johnson, 1999). Furthermore, category
management ignores traditional product groupings (e.g., breakfast cereals) and
focuses instead on understanding the way consumers shop related products in a
category (e.g., all products constituting breakfast) (IGD, 2002; Kotzab, 1999). In
essence, category management seeks to complement “supply management”
improvements (i.e., improvements in sourcing, purchasing and supply) with
“demand management” activities (i.e., managing assortments and the introduction
of new products) (Gruen and Shah, 2000). The demand activities themselves are
informed by “marketing information” (Hines, 2001, p.23) and are focused on
meeting the needs of an increasingly demanding and highly complex consumer
(IGD, 1999; Johnson, 1999). Although the supply chain efficiencies and cost
savings of QR enhance the retailer-supplier partnership’s business performance, it
is considered that once these types of cost and efficiency improvements have
been made, they merely become “industry norms” (e.g., Johnson, 1999, p.260).
By contrast, category management’s focus on leveraging consumer understanding
to develop category-based consumer “solutions” (Wellman, 1997) offers an
opportunity for creating sustainable competitive advantage (Johnson, 1999; Joint
Industry Report, 1995). Delivering customer value in this way is totally in line with
contemporary views of supply chain management (Hines, 2001).
172
 Category management has enjoyed rapidly growing acceptance and
implementation by retailers and suppliers in the grocery sector around the world
(Basuroy et al., 2001; Dussart, 1998; IGD, 2002). On the basis of its reported
positive outcomes, it is not difficult to understand its high adoption rates. Results
from a recent survey by the UK’s Institute of Grocery Distribution (IGD, 2002)
show how the majority of category management practitioners believe that category
management improves turnover, profitability, market share, inventory levels,
trading relationships, and consumer understanding. Academic empirical research
also reports the positive effects of implementing category management. It is found
to positively affect retailer prices and profitability (Basuroy et al., 2001), and the
implementation of category plans is perceived by suppliers to positively affect
category performance (Dhar et al., 2001). Diffusion of category management to
other sectors is now underway (Basuroy et al., 2001; Dussart, 1998; IGD, 2002;
Kotzab, 1999). However, there is no evidence in the literature or from industry
intelligence that the fashion sector has embraced category management. Indeed
with regard to the latter, discussions with key informants working in the fashion
departments of two of the UK’s leading food retailers and in menswear for one of
the UK’s leading department store chains indicate that as yet no attempt has been
made to adopt category management.

The purpose of this study is therefore to explore how category management

might be applied to fashion marketing. Specifically, the paper reports the results
of an exploratory study that seeks to clarify the role of consumer behaviour
research in category management. This is achieved by identifying how the
collected consumer intelligence can inform a supplier-retailer’s joint category
management planning process.

The remainder of this paper is structured as follows. The next section

presents a comprehensive review of the extant literature on category management
and from this, identifies research opportunities in this area. This section also
outlines the grocery industry’s established category management process.
Building on this review, the second section discusses the data collection and
analysis methods used in the present study. This is followed by a presentation of
the study’s findings. Next, the managerial implications of the findings are
173
discussed via their application to the strategy and tactics planning phases of the
category management process. The paper concludes with an agenda for further
research in the area.

Literature Review

Introduction
The “consumer-centric” thrust of category management (IGD, 2002; Mathews,
1996; Pearce, 1996) is embraced in the grocery industry’s agreed definition of
category management as “the strategic management of product groups through
trade partnerships, which aims to maximise sales and profits by satisfying
consumer needs” (IGD, 1999, Appendix). The grocery industry has developed an
eight-stage category process to guide suppliers and retailers in their category
management efforts, from analysis for the category (“Category Assessment”)
through to the subsequent joint strategy and tactics planning stages (see Figure
10).
Figure 10 The Category Management Process

Category
definition

Category
role

Category
assessment

Category Category
review scorecard

Category
strategies

Category
tactics

Category
implementation

Source: Basuroy et al., 2001, p.17; based on Joint Industry Report on Efficient Consumer
Response (1995).

174
By means of this process “the two sets of knowledge and skills are combined to
tailor the category offer to the retail outlet in question and its customers” (IGD,
2002) and a “category plan” (Gruen and Shah, 2000) is jointly agreed.

Table 13 summarises the extant literature on category management and highlights

the key points of evidence, commentary and criticism contained within this
literature. This literature on category management is categorised into articles
focused on: empirical research; case study-based descriptions; commentaries;
and literature-review based descriptions. Category management has received
much press, particularly in practitioner publications (see Table 13, Group 3), on
specialist industry websites (e.g., [Link]) and at industry conferences.
As Table 13 shows, there is no shortage of descriptive articles, practice guidelines
and commentaries on category management, all of it geared to the grocery
industry. However, the empirical evidence that has been collected is somewhat
limited, and again, all of this is in the grocery sector (e.g., Basuroy et al., 2001;
Dhar et al, 2001; Gruen and Shah, 2000; Zenor, 1994). The purpose of this
literature review is therefore twofold, and the rest of the section is organised under
two headings to reflect these twin aims. The first objective is to further develop the
argument advanced in the Introduction that fashion marketers should consider the
adoption of category management as an extension to their current QR practices,
and the second is to highlight research opportunities in the area of category
management.

Category Management as the next generation QR

As argued above, category management presents a platform for sustainable

competitive advantage based on the ability to market the right products in the right
way to meet “the fast changing needs of a highly complex consumer” (Johnson,
1999, p.260). This relies on retailers sharing transaction data and on suppliers
sharing the in-depth knowledge of the consumer that might explain such
transaction data. It is suggested in the QR literature that the most ‘developed’
implementation of quick response – Stage 3 QR implementation (QRI) - actually
includes category management (Giunipero et al., 2001, p.362; Birtwistle et al.,
2003). However, the three items used to measure the stage 3 QRI construct

175
(Giunipero et al., 2001, p.371) - vendors receive sales data, QR team meets with
vendors, vendors receive stock-out data - do not reflect the consumer centric
thrust of the grocery industry’s conceptualisation of category management.

176
Table 13 Category Management (CM) literature: Theoretical and empirical evidence

Year Author(s) Type of Study Respondents/Sample/ Main Focus/Evidence/CM Benefits/Drawbacks

Publication

Group 1: Empirical Research

1994 Zenor Empirical: Scanner data applied Substantial profit benefit to manufacturer from
A simulation to a pricing model coordinated CM pricing structure, but depends on level of
competition between brands. Impact on retailer and
competitors depends on market structure factors.
1999 IGD Empirical: 120 respondents from Survey results confirm the benefits of CM.
practitioner survey supplier and retailer Underlines the cooperative nature of the process.
organisations Promotes cross-functional working.
2000 Gruen and Shah Empirical: Stage: 128 from Brand management/sales conflict affects performance
Stage 1 qualitative supplier attendees at outcomes.
and Stage 2 testing 1999 ECR conference
of conceptual and from list of
framework category managers
from Food Marketing
Institute (FMI) list
2001 Basuroy et al. Empirical: Aggregate store-level For retailer: Increase in retail prices and category
Intervention time scanner data from 2 profitability where CM implemented and where interbrand
series analysis of retailers – one competition is high and consumer store switching is low.
sales data deploying CM, the For manufacturer: Lower prices and margins. This
other not signals the need for the trading partners’ co-development
of plans if CM is to create a win-win situation.
For consumer: increase in retail prices calls into question
the consumer benefits of CM.

177
2001 Dhar et al. Empirical: Account data for 106 The critical drivers of category performance (e.g.,
Regression major US grocery retail promotion, price, etc.) vary depending upon the role that
analysis chains in the 50 the category plays in the portfolio of the retailer and the
Nielsen SCANTRACK consumer (e.g., staple, niche, variety enhancer, etc).
markets Suggests future research should investigate the benefits
of CM to manufacturers.
2002 IGD Empirical: 64 respondents from Survey confirms the benefits of CM and its importance as
practitioner survey supplier and retailer a core business strategy for the next decade.
organisations Implementation by retailers is the most critical element
that is over-looked.
1998 Broniarczyk et al. Empirical: Study 1: 212 students Number of items stocked might be reduced without
Experiments on 6 shopping trips consequent negative affect on assortment perceptions
looking at the Study 2: 229 and store choice, so long as only low-preference items
impact of item respondents participate are eliminated and category space is held constant.
reduction in a in test store
category

Group 2: Descriptive/Case Study

1996 Pearce Case study Manufacturer Shopper research proved to be the ECR “relationship
enabler”.
1998 Johnson and Descriptive/Mini The clients for the Cites benefits of CM to retailers and suppliers (p.35),
Pinnington case studies to commissioned and the benefits to both of operating a trusting
demonstrate market market research trading relationship.
research forming the mini Concludes with proposal of how CM-related, ad hoc
techniques for CM case studies were market research gives the supplier something of
suppliers to the retail value to offer the trading relationship.
sector (UK and
Greece)
1998 Qureshi and Baker Mini case study A range of in-store Reports how the supplier gained invaluable in-store
format to report market research consumer insights for the category from the market
results of market tools employed with research; this would complement their knowledge of
178
 research study 1000+ shoppers the consumer as a “user” of the products.
1999 Johnson Descriptive/Mini Examples from How ad hoc market research can help CM partners
case studies to various international acquire an in-depth understanding of the consumer’s
demonstrate market markets shopping motivations/ behaviour.
research Suppliers benefit from CM when they have agreed
techniques for CM with the retailer a “purchase marketing strategy” for
the category.

Group 3: Commentary

1993 Harris and Commentary Practitioner Defines CM, and identifies category expertise as a
McPartland publication “shared resource”.
1994 Harlow Commentary Academic- Defines CM, describes CM as an evolved form of
practitioner journal trade marketing, outlines CM principles and the
consequent challenges to companies seeking to
implement CM.
1995 Richards Commentary Practitioner Argues the potential for supplier lock-out when
publication competitors secure category captainships with
retailers.
1996 Mathews Commentary Practitioner Reports the results of a survey of retailers and
publication manufacturers. Noted a maturing of the CM process
and use of “consumer insight” to find and leverage
common ground in the trading relationship.
1997 Freedman et al. Commentary Consultancy The authors are sceptical that supplier and retailer
publication objectives can ever be aligned, but at this stage (i.e.,
1997), “full-blown” CM remains largely untested. US
manufacturers underestimate the cost of developing
joint category plans; European manufacturers admit
to scepticism, with retailer-supplier trust cited as an
issue.
1997 George et al. Commentary Consultancy The authors explain the need for functional
publication “integrators” and describe how, in response to
179
 increasing consumer
and trade customer demands, companies have
reorganised their marketing organisations.
1997 Wellman Commentary Practitioner CM “baggage” (i.e., the cumbersome ‘best practices’
publication templates) makes it unproductive and a barrier to
focusing on the consumer. Advocates CM as a
“cross-merchandising”/cross-category business
model rather than an “organizational business
model”.
1999 di-Nome Commentary Practitioner Differentiation in CM relies on a greater focus on key
publication consumer insights and on category strategy.
Retailer-supplier cooperation and trust are essential.
1999 Dewsnap Commentary Academic- Category management is proposed as a “vehicle for
practitioner journal integration” between sales and marketing.
2001 Gregory Commentary Practitioner Reports the disenchantment with unwieldy CM
publication processes that deliver undifferentiated plans by store.
Advocates more shopper insight as the way forward.
2001 Mitchell Commentary Practitioner CM described as a severe disappointment with
publication cumbersome manuals and templates.
Categories still not being defined by consumer
understanding.
2001 Store and Benady Commentary Practitioner Criticises one UK retailer’s CM attempts as
publication insufficiently consumer-centric and insufficiently
radical.

Group 4: Descriptive/Literature Review

1997 Hutchins Commentary Academic journal Develops a research agenda for academics “with an
interest in supply chain management issues”.
1998 Dussart Review Academic journal Major pitfall: the consumer does not always benefit
from CM. Consumer understanding should be
180
 centre-stage.
1999 Kotzab Descriptive Review based on Distinguishes between textile industry’s Quick
analysis of Response, and grocery industry’s Efficient Consumer
secondary statistical Response (ECR) USA and CM-inclusive ECR
data, literature Europe practices. Discusses ECR background and
review and expert approaches, and how ECR in USA developed from
interviews QR practice in the textile industry. ECR presented a
“distribution model” focused not only on
“replenishment” but also “efficient” promotion, new
product introduction and assortment. Reports
savings offered by the initiatives.

Group 5: Manuals/Guides

1992 Nielsen ‘Category Management: Positioning Your

Organization to Win’
1995 Joint Industry ‘Category Management Report: Enhancing
Report on ECR Consumer Value in the Grocery Industry’
1998 ‘Category Guinness GB Take Home Trade
Management: The
Guinness Guide’
2000 ECR Europe ‘The Essential Guide to Day-to-Day CM’

181
 Notwithstanding this conceptual discrepancy, research in the fashion
industry reveals that implementation of QR to Stage 3 is woefully low anyway. A
study by the consultants, Kurt Salmon Associates (KSA, 1997) found that only
10% of the respondents had implemented this third stage of QR, and a more
recent UK study found that despite the intervening years, the figure had only risen
to 20 per cent (Birtwistle et al., 2003). The impact of QR systems and
technologies has been most noticeable on lead-time, inventory reduction and
reduced stock-outs (Ko and Kincade, 1997). However, it has been argued that the
consequent profit benefits have largely been to the benefit of the retailer, with the
suppliers taking on most of the “cost burden” (Hunter and Valentino, 1995;
Whiteoak, 1999). Against this background, the notion of retailer-supplier
partnerships, and all of the associations of trust, shared information and shared
benefits that the word “partnership” conveys, would appear an unrealistic ideal. As
Birtwistle et al. (2003) report, “There is still a high level of distrust between supply
chain members, which denies one of the fundamental requirements of QR, namely
that information should be shared to the benefit of all.” Such a lack of trust could
explain the reported low levels of full-scale QR implementation in the textile and
apparel industries (Birtwistle et al., 2003; Lowson, 1998; KSA, 1997), and it is this
same lack of trust that could explain why category management, which relies so
heavily on partnership ideals, has not yet been adopted. This was underlined as
follows in a recent research interview with a senior menswear buyer: “We are
sharing more information but neither party trusts each other.” Arguably, in order to
engage in trading partnerships, retailers and suppliers both have to perceive
and/or experience win-win benefits. It is the authors’ contention that the mutual
benefits offered by category management partnerships would encourage
increased openness between retailers and suppliers.

However, the purpose of this study is not to call into question the value of
QR to the fashion industry. Indeed, the aspect of category management that
focuses on logistics-based “optimal replenishment” concept (IGD, 2002) reflects its
QR origins. Rather, this study investigates how the grocery industry’s established
category management concept could act as a useful adjunct to fashion
marketing’s QR. This view is predicated on the proposition that efficient
replenishment does not guarantee that suppliers and retailers in the fashion
182
industry have a priori effectively marketed the right products. Conversely,
category management would enable them to introduce the ‘right products’ to the
marketplace, to merchandise and promote them in the right way, and then to
enable their efficient replenishment. In its development of ECR, the grocery
industry acknowledged that “supply management” and category management was
a powerful combination; specifically, that “to reap the full benefits of ECR” category
management should be linked to “supply management” (IGD, 1999, p.13). This
paper suggests that the fashion industry should consider doing the same.

Research Opportunities: The importance of consumer intelligence and joint

category planning

Johnson (1999) asserts that while electronic point of sale (EPOS) systems provide
volume sales data – that is, the “what” of category sales – category management
depends on understanding the “why” and “how” of consumer behaviour. These
are what the IGD (2002) term “consumer insights”. However, most reports on
category management’s shortcomings highlight the grocery industry’s failure to
fully deploy the requisite in-depth understanding of the consumer (e.g., Dussart,
1998; Gregory, 2001; Mitchell, 2001; Store and Benady, 2001; Wellman, 1997)
and its related failure to “inspire” the consumer with “improved offerings” (Gruen,
2002). A real understanding of the consumer for a particular category would – it is
argued – enable supplier and retailers to adopt radical shifts in the marketing of
the category. It would enable them to move beyond “improved efficiency to
improved consumer experience”, and beyond “single product-based categories to
consumer solutions” (Gruen, 2002, p.18). Indeed, research by the IGD indicating
that category management has a healthy future in the grocery sector (foods and
non-foods) could be an acknowledgement that there is much scope to improve
existing category management techniques.

The collection and analysis of consumer “insights” is a critical part of the

Stage 3, “Category Assessment” phase of the category planning process (Figure
10) 3. Johnson (1999) and Qureshi and Baker (1998) argue a strong case for how
2F

3
The other areas for “assessment” are market, retailer and supplier (IGD, 2002).

183
a whole range of market research techniques are vital for the collection of the
point-of-sale consumer insights and report on market research projects
undertaken in the grocery sector to gather insights into the in-store shopping
process. To date, however, empirical research has not addressed how consumer
insight collected for category management as part of the “category assessment”
stage can be applied to the subsequent planning stages of the category
management process (the “scorecard”, “strategy” and “tactics” stages in Figure
10). Nor has previous research dealt with how category management can be
applied to any non-food, including fashion. This study therefore seeks to fill these
research gaps.

Before category management, neither retailer nor supplier possessed in-

depth knowledge of the consumer's in-store behaviour. Consumer insight is vital
to achieve the aims of category management, but although retailers have access
to EPOS data, few have the resources to conduct and/or analyse consumer
research at the category level (Johnson, 1999). Potentially the supplier can fill this
gap by providing this resource. Although category management is primarily a
retail strategy, in their bid to play the “category captain” role it is the manufacturers
that have taken the initiative to develop the expertise and the category insight
required to support the process (Gruen and Shah, 2000). The key to successful
category management is when these two complementary sides of the trading
relationship work together collaboratively to understand the consumer and to
interpret and fulfil their category needs via the joint category planning process
(Gruen and Shah, 2000; Harris and McPartland, 1993; IGD, 2002). As Gruen and
Shah (2000) found, retailer-supplier continuous involvement in joint category
planning positively affects: “retailer system trust” - conceptualised as the retailer’s
confidence in the potential benefits of category management plans and in the
supplier’s capabilities/delivered information; and the “objectivity” of the category
plan. Correspondingly, the study also found that the “objectivity” of the category
plan, measured in terms of the plan’s freedom from supplier bias and the extent to
which it delivers the consumer perceived best mix of brands and sizes, had a
positive effect on category performance. However, “objectivity” describes the
nature of the plan rather than its contents/constituents. Therefore, this present
study seeks to understand how the consumer intelligence gathered as part of the
184
category assessment process contributes, not to the ‘nature’ of the plan but to its
actual ‘contents’; that is, how the intelligence contributes to the supplier-retailer
strategy and tactics planning stages of the category management process. This
objective reinforces how the application of consumer intelligence extends the
boundaries of empirical category management study.

Methodology
This section details the methodology employed to achieve the research objectives
outlined above.

Preliminary discussions with associates in the fashion industry identified the

potential for a collaborative investigation with a manufacturer in the intimate
apparel sector. The manufacturer, an international brand and its retail customer, a
major UK based apparel retailer operated within the bra category, which
represented all bra styles from fashionable/glamour ranges to the more staple,
functional basic garments. The particular case scenario indicated that retailer and
supplier both lacked fundamental knowledge of the consumer. Retailer category
performance implied declining consumer interest and as such, the retailer had
recently implemented assortment reduction within the category. Conversely, the
supplier reported steady sales performance results from other retailers for the
same category. It also presented a situation of low retailer-supplier collaboration
with no evidence of joint decision-making. Acknowledging the potential benefits of
the research, the manufacturer agreed to sponsor the study.

Research design

Exploratory research is appropriate when little is known of a phenomenon

(Churchill, 1995, p.149). Previous category management studies in the grocery
sector (Johnson and Pinnington 1998; Johnson, 1999; Qureshi and Baker, 1998)
have argued the case for developing quite advanced market research “tools” for
understanding how the consumer shops the various categories in the grocery
sector; for example, recording point of purchase activity, exit interviews and so on.
However, the subject of category management in the fashion sector is as yet
unexplored. Furthermore, the research context here is a situation where the
185
intimate apparel industry overall appears to lack fundamental knowledge of the bra
consumer. Therefore, while sophisticated in-store techniques may be wholly
applicable for a sector that is further advanced in category management, they may
be less appropriate for gathering the required preliminary insights into consumer
behaviour for the bra category. The research therefore pursued an in-depth,
qualitative approach, the objective of which was to gain greater “insight to
consumer beliefs, attitudes and preferences” towards the category (IGD, 2002, p
164).

The structure for the study was provided by a category management

research “framework”, adapted from Johnson (1999). The objective of this
framework was to understand customer motivations and the impact of point of
purchase marketing mix variables. It comprised three key areas of focus:

(i) The pre-purchase environment: What influences the consumer’s brand

choice before entering the retail store; for example, product and packaging design,
advertising, or promotions, brand image / positioning? Why was a specific retailer
chosen?

(ii) The consumer purchase process: What are the consumer’s need states
driving the purchase? Was the purchase planned or impulse and why? Where is
the purchase decision made? How is the category shopped?

(iii) The in-store purchase environment: What are the in-store dynamics -
that is, the in-store variables that influence the consumer purchase process – for
example, availability, assortment variety, visual merchandising, category
location/layout, shelf allocation, service support, in-store promotions?

Data collection methods

Focus groups provide a versatile medium for generating rich research insights
(Fern, 1982; Goldman, 1962). Furthermore, in category management research,
focus groups have generated greater depth to support retail sales and panel data.
They have also been adopted to facilitate the category definition process, to

186
understand category purchase modes (Johnson, 1999), and category usage, and
to identify new product opportunities (IGD, 2002).

A combination of four data collection methods was used, each selected to

answer the research questions within the three areas of the adapted Johnson
framework. This is represented diagrammatically in Figure 11.

To stimulate discussion of the pre-purchase environment, and specifically to

reveal consumer’s perceptions of the leading brands in the category, the research
deployed a word association exercise. This was designed to determine the extent
to which brands were associated with established criteria of fit, comfort, style,
quality and price (Mintel, 1997). Eight different bra brands were selected to
represent the category including, respectively, Berlei, BHS, Warner’s, Marks &
Spencer (M&S), Triumph, Littlewoods, Gossard and Playtex. To counter the risk of
peer contamination, individual perceptions were captured on paper prior to general
group discussion. Each respondent was requested to insert against each bra
brand any number of words or phrases that could be used to describe that brand.
A pre-test revealed that many respondents had difficulty in generating their own
words. Thus, from interviewing a number of bra wearers a list of 40 positive and
negative descriptors was compiled to which respondents could refer if required.
This controlled response process (Chisnall, 1992, p. 162) lasted approximately 10
minutes.

187
 Figure 11 Data collection methods

Word
Association Exercise

Semi-Structured Open
Discussion

“Blind” Experiment

Projective Techniques /
Open Discussion

The second part involved a semi-structured open discussion designed to explore

the consumer purchase process and the in-store purchase environment. This
included questions on usage and bra buying history and also probing on
preferences with respect to in-store variables; for example, merchandising/display
and in-store selling activity.

Third, the research used a ‘blind’ experiment to further explore the pre-
purchase environment and brand perceptions, and to understand the consumer’s
design preferences. For this exercise, each respondent evaluated on paper a
diverse selection of seven unbranded garments representing the category. The
evaluation criteria included product design, perceived price, brand and function.
Respondents were also asked to indicate the likelihood of their purchasing each
individual garment.

The final method of data collection was another open, semi-structured

group discussion. Due to the potentially sensitive nature of intimate apparel, this

188
part of the study explored bra buyer behaviour from the third person perspective.
This involved the use of projective techniques with the unbranded garments used
as prompts (Haire, 1950).

These combined methods of data collection offered the opportunity for

synergistic outcomes (Delorme and Reid, 1999) and for triangulation of the data
(Denzin, 1978). In total, each four-stage focus group lasted on average ninety
minutes. A neutral environment was chosen in which to host the focus groups;
avoiding the stimulation and distraction within one specific retail setting broadened
the discussion to cover the retail environment in general for the category. This
also further supported the exploratory nature of the study.

Sample

A stratified sample was designed to include six focus groups across a range of
ages and bra sizes. Using a recruiting questionnaire, Market Research Society-
trained interviewers screened and recruited forty-eight respondents to participate
in the study. In order to reduce the potential for sample bias (Sudman and Blair,
1999), the focus groups were recruited and conducted in three separate locations
within the Midlands and the South East of England. An independent Market
Research Society moderator, selected for compatibility with the respondents,
directed the focus groups.

Data management

Both the word association exercise and blind experiment were completed on
paper. For the former, individual comments provided against each brand name
were then analysed and scored either positively or negatively for six criteria of fit,
comfort, attractiveness, price, quality and a ‘no response’ that indicated low brand
awareness. For example, comments such as “dull” but “good value” would score a
negative point for attractiveness and a positive point for price, whereas
“glamorous” and “pricey” would score the opposite. The number of positive /
negative frequencies scored for each brand were then calculated. In respect of
the blind experiment, the mean scores for each garment style were calculated
against the evaluation criteria described above.
189
 Each focus group was tape recorded during the open group discussion
sections. The tapes were then transcribed. All of the data was then analysed
following the three key activities of data reduction, data display and conclusion
drawing (Miles and Huberman, 1994). Following the strategy advised by Miles
and Huberman (1994) a combination of within- and cross-case analysis was
undertaken. The within-case analysis generated insights into why certain
behaviours, phenomena and situations existed for a particular focus group. Cross-
case comparisons were then conducted in order to provide insights into how
issues relating to consumer behaviour for the category varied from one focus
group to another.

The combination of data collection methods outlined above generated a

considerable amount of data. For conciseness, the results reported below
represent a synthesis of these data.

Findings
A key objective of this research was to explore how the fashion industry might use
“consumer insights” to inform supplier-retailer category management relationships,
and in so doing, to ascertain whether the industry might usefully deploy category
management. Discussion on how the industry might apply the collected insight to
specific steps of the category management process is covered in the next section.
This current section presents the insights, structured in line with the category
research framework outlined above.

The pre-purchase environment

In analysing the word association exercise it was apparent that the respondents
held strong attitudes towards the eight listed brands. The findings illustrated in
Figure 12 visually depict the respondents’ positive and negative perceptions of
each brand. (For each brand the score reflects the difference between the
respective positive and negative responses.) These indicate that from past
experience, in terms of market positioning, shoppers had formed distinct profiles of
competing brands. From the mere mention of a brand name, respondents were
immediately able to describe the brand’s relative positioning. Not surprisingly, for
190
the market leader, M&S, the balance of responses was positive for all criteria. In
addition, half of the respondents made an unprompted association between M&S
and “quality”. No other brand achieved this same consistent, positive ratings
balance for all criteria: excellent scores on one criteria were always
counterbalanced by significant negative scores on another. For example, Gossard
achieved the highest score for attractiveness, but scored one of the lowest for
perceived fit. Similarly, the Playtex brand achieved the highest mention for fit, but
was negatively rated for attractiveness. Further, economy store brands BHS and
Littlewoods, while scoring positively for price, were negatively scored for fit,
attractiveness and quality. Warner’s scored the highest number of “no response”,
which indicated a very low brand awareness.

The general group discussion generated additional insights regarding the

pre-purchase environment. Surprisingly, given the high levels of brand awareness
demonstrated above, respondents admitted to low brand loyalty - both for own
label and manufacturer brands. Discontinuation of styles was a key complaint: “I
liked the one I had. I buy one and I like it, and then I find that they’ve stopped
doing it. Or they change the range". "The big mistake is that they don’t keep the
range for long enough. They turn it over too quickly". “Every season, change; out
goes the one that you’re familiar with”. The lack of size consistency both inter-
and intra- brand also inhibited brand loyalty: "You can buy one bra, even two bras
from the same manufacturer, and although it says it’s the same size on the packet,
they don’t fit the same". Overall, difficulties encountered finding a well-fitting
garment were seen as major obstacles to brand loyalty. Conversely, style loyalty
featured strongly; once respondents had found a style that they were happy with,
they would continue buying it until it was no longer available: "I’ve got about four
or five bras all the same style. I don’t know what I’m going to do if they stop doing
it to be honest". "It's a problem when manufacturers change a design and they’re
no longer available". Furthermore, with respect to choice of store the consumer
admitted to a tendency to store loyalty. The consumer will patronise those stores
that they have found offer depth of range, size availability and specialist fitting
advice, and unlike other apparel categories respondents expressed a clear
preference for independent, specialist retail stores.

191
Figure 12 Consumer Brand Perceptions: The difference between positive and negative responses

Positive - Negative Responses

192
The blind test experiment illustrated that the respondents had developed very
clear design preferences. Through trial and error they had developed a set of
decision rules for achieving fit and comfort. The first decision was often between
non-wired versus underwired bras, whereby non-wired provided greater comfort
and underwired offered greater support in more attractive designs. Fibre content
was a strong consideration with respondents generally preferring natural fibres.
Individual design elements prompted instant reaction. For example, the position of
seams had been found to cause irritation. Likewise, depth of underband and the
width and position of straps conveyed the degree of support. Collectively, the
combination of salient features, fabric design and shape experienced in the past
provided a quick formula for assessing the potential suitability of a new purchase.
Such was the extent of their accumulated knowledge and experience of a
garment’s styling that most respondents could quickly evaluate the suitability of a
garment without the need for trying-on. This was an important consideration given
the perception of queues and customer service in chain stores: “I do find with X
that it’s hard to try them on. I know they have changing rooms but there are long
queues”. With respect to quality, respondents had clear expectations of fabric
performance, and due to the supporting nature of the garment could detect subtle
changes in function and durability: "I’ve bought a few from there (chain store) over
the last year or so and they’ve not been as comfortable for whatever reason. I’ve
started going elsewhere". "I felt that they’ve changed manufacturer. The quality of
another favourite one went down and that’s when the elastic went and it shouldn’t
have."

The consumer purchase process

The majority of the respondents described how they are initially prompted into the
bra purchase process by a variety of “need states” (Johnson, 1999). These
include the garment’s loss of functional support through deterioration of elasticity,
the decline in the garment’s aesthetic appearance, a change in the wearer’s own
physiology driving the need to buy a different bra size or the seasonal factor of
holidays prompting product replacement; "Holidays spur me on, so if I have one
holiday a year then I’d buy annually ...if I’m packing things I look and I think ‘that
looks awful’. Or that doesn’t go with certain items I’d be wearing". For some, the

193
prompt to purchase is linked to the hedonic value (Laurent and Kapferer, 1985) of
“pretty” bras; in the case of the respondents in this study, the hedonism stemmed
from the way in which “pretty” bras are associated with "pleasing someone else",
and further, how the wearing of them is seen as a treat or reward: “It makes you
feel special, really spoiling yourself".

In terms of the number of wear-occasions adopted by the bra consumer, the

respondents in the study presented a range of consumer typologies, from the
preference for only one bra type - "I’ve got four or five now that I wear on the go ...
they go universal if you like" - to three-style bra wardrobe: "three (bra wardrobes)
in my case. One sports, one ordinary, one special occasion...". The most popular
preference was for two different types of bra, one type for functional everyday use
and another purely for evenings: "I just have a medium support, fairly plain bra for
daytime and then I’ve got fancier stuff for the evening”. Despite the invisible
nature of the garment, there was a clear distinction between home and social use:
"I have bras for work and bras for round the house". Comfort was an important
factor and some women relegated older garments for daytime use: “…old ones
that I wear in the day and nice ones for evening". "I wear the old ones for
gardening - It’s a comfort one". However, discomfort was suffered for social
occasions: “When you go out at night you tend to wear something that’s not as
comfortable - like we all squeeze ourselves into high heel shoes because they look
good."

Consistent with the behavioural implications of the high involvement profile

for bras first uncovered by Laurent and Kapferer (1985), the information search for
bras was described in this present study as both critical and extensive. In light of
the consumer’s wide experience of the product, combined with the frequency with
which bras are replaced in her wardrobe, it is surprising that women find it
necessary to undertake an extensive search for each new bra purchase. This may
be attributed to the reported over-frequent discontinuation of styles and the lack of
size-consistency offered by the competing manufacturers. Together these
increase the consumer’s perceived risk and so drive the need for her to restart the
search process for each new bra purchase. Significantly, only a third of the
respondents considered the actual process of shopping for a bra to be a positive,

194
enjoyable experience. The majority found the experience "a nightmare",
"frustrating", "boring", "a necessary evil", "traumatic". The main reasons for not
enjoying the experience was the lack of attractive bras in larger sizes (“Bras for the
more fuller lady are very boring, old fashioned”), lack of choice in general, difficulty
in finding garments in stock, and the difficulties encountered in finding bras that
offer fit, comfort and the right body shape through clothes.

On average and depending on the quality of the bras and how long they
lasted, the respondents in this study claimed to buy a bra three to four times a
year. Further, to counter the risk of obsolescence consumers tend to buy in bulk
when the product is available. Availability of the preferred bra style is one of the
key factors influencing purchase. Given the importance of comfort and fit,
substitution is rarely an option in bra purchasing. Owing to the problems of limited
availability of styles and sizes and product obsolescence, the majority of the
women categorised themselves as convenience shoppers, buying a bra, not
because they need one, but because one is available for purchase: "If I see a nice
one I think ‘there’s my size, I’ll buy it’, not because I really need it. It saves me,
when I do need one, having to go and hunt for one". In this way purchase is
prompted by the marketing stimulus of an in-stock situation, and underlines how
practitioners can stimulate purchase just by having an item available.

In terms of the number of bra purchases, the majority of the women bought
at least two bras at a time, though it is notable that some respondents claimed
they bought three or more at once. An alternative approach was to buy one bra,
try it on at home, wear it and wash it, then return to buy another exactly the same
or in a variety of colours. This propensity to repeat purchase and to possess
multiple copies of the same garment is arguably quite unique to the bra, and it
further underlines the way in which the consumer is motivated first and foremost
by functionality and physiology, and then by the fear that the manufacturer and/or
retailer may discontinue their preferred style.

It is significant that the majority of women prefer to shop alone for bras.
This contrasts sharply with the preferred social interaction associated with other
fashion categories, where “the customers comparison to referent others appears

195
to be of paramount importance” (Dodd et al., 1998) and underlines the personal,
intimate nature of the category.

The in-store purchase environment

The issue of store loyalty founded on depth of range, size availability and
specialist advice has been highlighted above. Overall, the chain stores were seen
as offering less choice: "It’s a shame because there used to be such a big range
and there isn’t. All you see is 45 of the same one". By contrast, the specialist
lingerie stores were perceived to offer a greater variety of styles and depth of
assortment to cater for a wider market: "I take my two daughters to get their bras
there (specialist store) as well because they have special sizes and will order
anything for you". With respect to the issue of specialist advice an important
theme to emerge from nearly half of the respondents was the need to be correctly
measured and fitted for a bra: "If I find a shop that I like and I tend to keep going
back there. I go to Xs now because they measure me properly". As many
respondents admitted to not knowing their correct size, they relied in part on trial
and error and were therefore unsuccessful in self-selecting appropriately sized
garments. One-fifth of respondents reported disillusionment caused by being
incorrectly measured in the past, with a noted difference between sizes
recommended by different retailers. Once again this increases the perceived risk
associated with the probability of a mis-purchase. The perceived variation in fitting
and measuring services, combined with the inherent size variation between
manufacturers and even within the same brand, present substantial obstacles to
the bra buyer. Seeking to overcome such barriers to successful purchase, over a
third of respondents underlined the importance of the type of in-store advice
associated with the specialist lingerie retailer as opposed to a generalist chain
store. "I prefer a specialist shop - because there’s usually someone there who
specialises…knows what they're talking about".

Surprisingly, over half of the respondents considered price to be less

important than fit, comfort, value and quality: "If I see a nice one I don’t even look
at the price. If I think it’s going to fit, it’s going to serve its purpose and I like it, I’ll
try it on and then look at the price". "I think you get what you pay for - I used to
buy cheap bras but they don’t fit properly". Some women believed the function
196
and shape achieved by the bra to be most critical: "If it made the best of what I’d
got it would be worth any amount of money". Furthermore, women could also be
tempted to pay more for a special occasion or special purpose bra: "If I've seen
something I’ve liked I’ve pushed the boat out because it was for a special
occasion." "I do quite a lot of sport so you need a good fitting sports bra, and
you’re always prepared to pay a lot for that".

On the subject of category merchandising the respondents were equally

split in their views towards different types of display. Half preferred to see bras
displayed on hangers, mainly so that they could be more easily viewed and
handled; the other half – on the basis of being able to more easily select the
required size - expressed a preference for bras to be merchandised in boxes
bearing a photograph of the garment, with just one unboxed item of each style on
display at the fixture: "If I go to X and get one, they’re in boxes with a photograph
of somebody wearing it, I know what they look like and I haven’t got to rake
through - I just grab one in my size and go". Regarding category adjacency -
respondents outlined their preference for intimate apparel departments to be
separated from both general apparel and more importantly, from menswear: “I
wouldn’t want to be looking at my bras when the men are walking through as a
short cut to somewhere else.”

Discussion and Managerial Implications

As outlined in the Methodology, the context for this present study was a situation
where in the absence of retailer-supplier collaborative working, the retailer was
making product range decisions that sub-optimised its sales. The supplier was
also suffering suppressed sales for this particular category compared with its other
accounts. Perhaps worst of all, the consumer was receiving much frustration in
thwarted attempts to seek a solution to her consumer needs. Category
management provides a focused, in-depth collaborative retailer-supplier approach
to: defining the category and agreeing its role; devising category strategy; and
formulating tactics for implementing the strategy. As a preliminary investigation,
the objective was not to collect data specifically for the purpose of working through
the entire category management process (Figure 10). Rather, it was to consider
how – in theory - consumer research findings could inform a category
197
management process for a product category in the fashion sector. Accordingly,
the results presented above are now discussed in the context of the relevant
stages of the category management process.

Category definition

The bra market spans a range of garments, from styles that could be described as
utilitarian/functional to the more glamour/special occasion. The research indicated
that consumers seek a particular garment based on usage occasion and
corresponding base need; for example, comfort for everyday use or glamour for
evening wear. These results represent a marked contrast with the retailer’s more
simplistic and potentially misleading, product-based view of bra categories, which
are based – not on consumer needs or motivations – but on bust size and/or bra
construction (e.g., wired versus non-wired). There is a strong case for suggesting
that the retailer’s narrow definition of bra categories is driving their under-
performance; correspondingly, better knowledge of the consumer’s usage
behaviour and motivations might facilitate better category definition. For example,
further consumer insights about the range of bra wardrobes could inform the
definition of a greater number of bra categories or, more creatively, the intelligence
could be combined with that from other categories to create new categories. For
instance, there may be scope for the creation of a completely new category based
on “special occasions” in which a range of traditional product groupings
(underwear, outerwear, accessories) could be marketed together. It is quite
feasible to propose that the category management process could lead to a similar
re-definition of other fashion categories.

Category role and category strategy

Category management provides a framework for retail and supply partners to

jointly agree the role/business aims of the category and to plan and develop
category strategy (see Figure 10). The results of this study provide initial evidence
of how shared consumer insight can highlight opportunities for category growth
and therefore, the opportunity for jointly developing long-term category strategy for
the categories identified. For instance, in the case of the present research the
identification above of a greater number of wear-occasion related bra categories,
198
would offer the opportunity to significantly grow the market. The identification of
the different categories could also help the retailer understand the ‘role’ of the
particular category to their business. For example, the agreed role of the
glamour/special occasion bra category as a “variety enhancer” or
“occasional/seasonal” category (Dhar et al., 2001; Hoch, 2002) could be to
generate excitement in-store. Conversely, the ‘everyday’/functional bra category
could be designated a “destination” or “staples” category (Dhar et al., 2001; Hoch,
2002), the strategy of which would be to guarantee a level of in-store traffic for
driving revenue in other categories. These same principles would apply equally to
other categories of apparel. Further, this consumer research identified the
importance of in-store bra fitting/advice services. If customer service is central to
a retailer’s positioning, the subsequent strategy for at least one of their bra
categories could be for that category to enhance the store image through this
service provision.

Category tactics

In terms of implementing the strategy developed, the category management

framework highlights a set of “improvement concepts” (IGD, 1999, 2002). These
equate to a detailed operationalisation of the ‘Category Tactics’ Stage 6 (Basuroy
et al., 2001) of the category management process (Figure 10) and are discussed
below in the context of the study's findings.

Optimal assortment

Optimal assortment or range is concerned with maximising profitability through the

effective use of store space and stock. As such, it is concerned with offering a
range or mix of products and individual stock keeping units (SKUs) that balance
inventory costs and consumer needs. Based on the high number of SKUs
required to cater for the different shapes and sizes of women, bra categories pose
a particular challenge to managing assortment. Typically up to 60 different sizes
are stocked in a range of colours across a variety of styles.

The most relevant insight for assortment decisions provided by this

research is the identification of usage-related bra categories (see above).

199
Broniarczyk et al. (1998) found that so long as only “low-preference” items are
eliminated from a retailer’s assortment and that category space is held constant,
retailers might be able to significantly reduce the number of items without
negatively affecting assortment perceptions and store choice. The findings of this
present study indicate that by reducing the variety and number of SKUs within the
more functional bra category, the consumer perceived the retailer to be less
effective in meeting their needs, with the obvious net result of a loss of consumer
loyalty.

Optimal new products

Hines (2001) identifies the timing of partner involvement in the product

development process as an important dimension in retailer-supplier partnerships.
Category management’s optimal new products “improvement concept” (IGD,
1999) deals with the development of new products to target specific
consumer/shopper needs. It assumes that to efficiently manage the roll-out of new
products and the phasing of the old, the retailer draws on the supplier’s intimate
and up-to-date knowledge of the consumer.

The UK’s Department of Trade & Industry analysis of the retail sector (DTI,
2003, p.51) reports how market growth is being driven by the shortening of fashion
cycles; for example, the movement away from “four seasons” to more frequent
product range changes. Yet, as the present research demonstrates, in order to
reflect consumer needs, new product development strategy for bras has to
recognise the function-glamour dichotomy. The fashion industry has developed
QR processes that respond quickly to retail sales data: replenishing the successful
styles and delisting the un-successful. However, these may be too responsive for
some apparel categories, allowing insufficient time for consumers to evaluate the
garment before they wish to return to the store to repeat purchase. With respect
to the bra category, the consumer may not expect to repeat purchase ‘special
occasion’ bra ranges (for example, garments for social evening wear in fashion
colourways), but she does expect to be allowed enough time to evaluate new,
‘functional’ bra ranges (for example, bras for everyday use). Across all fashion
categories, for this type of product manufacturers and retailers cannot afford to
subject the consumer to risk each time they approach the decision process. For
200
the more functional bra category too much innovation may be counterproductive to
customer retention. Successful new product strategy and the associated decision
on the timing of range additions/deletions must therefore be based on a sound
knowledge of consumer perceived risk for the different subcategories that
constitute a retail category.

Consumers in the present study described how the lack of consistency

across manufacturer sizes presents a distinct obstacle to purchase. In terms of
product development, this highlights a clear opportunity for retailers and
manufacturers to work together to achieve industry-wide standardisation in sizes,
perhaps via new technology in body scanning (DTI, 2003). Conceivably, this
partnership approach on garment sizing would apply to other apparel categories.

Optimal promotion

Traditionally, the level of above-the-line promotional activity for the total intimate
apparel sector has been low (Mintel, 1997). This study showed that through the
shopping and wear experience the consumer had developed an acute sense of
the competing brands’ market positioning. This included both strong positive and
negative brand perceptions. This presents a clear opportunity for the industry to
invest in communications to support the manufacturer’s desired positioning.

Within the literature the ‘optimal promotion’ concept deals exclusively with
sales promotions designed to stimulate sales and to create in-store excitement
(e.g., IGD 1999; Dhar et al., 2001). The consumer’s perceived risk of
obsolescence, combined with her experience of product non-availability, drive the
consumer to bulk-purchase. The study findings indicate, then, that retailers do not
have a problem with encouraging customers to buy, so long as they have the
items available for purchase. Accordingly, the scope of the ‘optimal promotion’
concept has been widened to consider the research’s implications for “in-store
promotion” and “communications” overall (Johnson, 1999, p.263).

As outlined above, many consumers perceive the bra purchase as high-risk.

Consumers therefore value in-store measurement and fitting services. However,
these services had proved unreliable in some cases with a fifth of respondents
201
reporting having been incorrectly measured in the past. These findings underline
the importance of in-store personal selling and highlight the potential for retail and
supply partners to work together to improve the specialist fitting skills of their sales
personnel.

With regard to repeat purchases, and in cases where disposable income is

limited, consumers outlined their strategy to return to repeat purchase rather than
buying multiple copies upfront. This is due to their strong belief that a bra cannot
be judged for comfort and fit until it has been worn and washed, and represents an
additional factor for the consumer to see the category as high-risk. With regard to
communication strategy, this is a cue for the retailer, with support from their
supplier partner, to re-design their selling and warranty policies to include the
promise of reimbursement in cases of poor post-wash fitting.

Optimal replenishment

It is noted earlier that the “efficient replenishment” strategy from ECR bears a
strong resemblance to the quick response concept, and that the three
demand/consumer-oriented aspects of ECR (- the category management "tactics"
discussed previously) offer a complement to quick response. Theoretically, then,
category management does not include consideration of replenishment issues.
However, the IGD’s (IGD, 1999) inclusion of the “optimal replenishment” factor in
their description of category management could signify that these demand-based
and replenishment-based concepts are indeed complements of each other. For
this reason, the logistics-based findings from the research are included here.

Product non-availability through obsolescence or stock-outs was seen as a

significant obstacle to purchase. Fearing non-availability, the bra consumer bulk
buys items when they are available. This signals the need for the retailer-supplier
partnership to improve their logistics processes to include improved forecasting
competence and more appropriate inventory policies. However, the key criterion
from a category management perspective is that the corresponding QR
replenishment strategies reflect the identified category differences, with the more
functional/basic SKUs attracting much higher expectations of stock availability.
Supporting Fisher's (1999) view that innovative and functional products demand a
202
different type of logistics "response", consumers value availability on more
functional styles and expect them to be repeatable. They do not expect these
products to be treated as high-fashion items that are introduced and then quickly
deleted from the retailer’s assortment. Arguably this finding would apply to any
apparel category that incorporates basic and fashion merchandise.

Conclusions and Research Next Steps

The empirical literature on category management (Basuroy et al., 2001; Dhar et
al., Gruen and Shah, 2000) focuses exclusively on food categories within the
grocery industry and indicates that in this sector category management still faces
significant development. Further, although this literature reports on the gathering
of consumer insight for particular categories, it does not seek to apply these to the
category management process (e.g., Basuroy et al., 2001). The value of this
study has therefore been to demonstrate (1) how category management might
offer significant benefits to fashion marketing and (2) how consumer insights for
the fashion industry can be applied to the category management process. As the
name suggests, the category management process focuses on a particular
‘category’, in this case the bra category. However, as outlined in the previous
discussion of results, there is good reason to suggest that the model is
generalisable to other fashion sectors.

The present paper argues that for the fashion industry category
management would be complementary to quick response (QR) strategies. As
such, the real value of any insight gathered for category management is in
combining it with the retailer’s sales data used in QR. Additional value comes
from demonstrating how the combination of retailer-supplier information aids joint
category planning efforts and, overall, supports the partnership approach which to
date is largely missing from the fashion industry’s attempts at QR. For example, in
this particular study, without the consumer insight the retailer assumed that falling
sales were due to a declining category. Conversely, with the insight to hand, both
parties in the trading relationship learn that declining sales are the result of not
meeting the customer’s category needs. They both gain invaluable intelligence to
use in their efforts to jointly agree the strategy and tactics for the subsequently
identified categories.
203
 The survey findings in the IGD 2002 report indicate that 96% of the study’s
respondents were confident that category management would be a “core business
strategy” for the next decade. The authors of this report indicate that category
management could be usefully transferred to other non-food sectors: “Category
management evolution is not complete…Non food retailers and providers
throughout the supply chain desire the success gained in the grocery business”
(IGD, 2002, p.220). The inference from this report is that category management
has still to be adopted in non-foods, and as outlined earlier, although fashion is
now a key part of some grocery retailers’ assortment, they have yet to adopt
category management for their fashion ranges. The category management
implementation issues that food suppliers have experienced at store level and the
cumbersome category management process manuals (see Table 13: e.g.,
Gregory, 2002; IGD, 2002) could explain why the grocery sector has not moved on
to implement category management in their non-food sectors. In addition, the
interviews with key informants highlighted how the fashion and grocery sides of
the retailer’s business operate autonomously, with limited sharing of best practice.

Several opportunities for future research spring from the results of this
study. In their investigation into the antecedents and consequences of category
planning and implementation, Gruen and Shah (2000) found that “category plan
objectivity” positively affects “category performance”. This present research has
shown how consumer insight can be used to create the category plan itself, and
arguably the better the contents of the plan (which could be termed plan “quality”),
the better the category performance. This offers the opportunity for future
research to test the Gruen and Shah (2000) framework, expanded to include this
category plan “quality” construct.

As an exploratory study on one product category, the objective of this

research was to provide preliminary insights. The research design therefore limits
the researchers to drawing anything other than tentative conclusions about how
category management could benefit the fashion industry overall. In addition, in
terms of obstacles to category management’s transferability to fashion, it is
conceivable that a demand- and consumer insight-led category management
process that works for grocery products with more stable demand patterns is

204
inappropriate for more supply-led, high fashion ranges. Therefore, in order to
enable more definitive conclusions, a key research next step should be to
investigate its applicability in all fashion sectors and its relative applicability in
categories that span all degrees of exposure to the “fashion cycle” (Ko and
Kincade, 2000). Following Ko and Kincade these could be represented by four
categories as follows: highly fashionable; fashionable; fashionable and basic; and
basic. Further, category management relies on detailed information on the
consumer's in-store behaviour. Therefore, as a study of consumer behaviour
overall, the corresponding lack of specific in-store insights collected in this
research represents a limiting factor. Any future research on category
management in the fashion industry should therefore employ the wide range of in-
store market research techniques described by Johnson (1999) and Qureshi and
Baker (1998).

This exploratory research presents early evidence that category

management could offer distinct opportunities for retailers and suppliers in the
fashion industry to work together to generate sustainable competitive advantage.
However, the concept has not been without its critics (e.g., di-Nome, 1999;
Gregory, 2001; Mitchell, 2001). Most of the criticism surrounds the cumbersome
nature of the manuals and templates supporting the eight-stage process (Figure
10), the over-complexity of the data produced, and its consequent inability to
highlight the key, consumer insights that are so central to the process.
Acknowledging these barriers to implementation, the grocery sector has
streamlined the category management process (ECR Europe, 2000). In
implementing category management, the fashion industry can therefore learn from
the experience of other sectors. For example, the industry would be advised to be
more flexible and pragmatic in applying what would seem like an inflexible process
(IGD, 2002). Indeed as the IGD report points out, “non food retailers…will take
learnings from existing practitioners in order to create bespoke processes to suit
their markets” (IGD, 2002, p.220). This suggests an additional and final avenue
for future research. Studies designed to apply the process to different sectors of
apparel would identify: the relevance of the process to the fashion industry; the
industry's scope for flexibility in applying the process; and how the process can
best be applied within the sector.

205
Acknowledgements

The authors are grateful to the editor and the two anonymous reviewers for their
very valuable comments and suggestions.

206
References

Basuroy, S. Mantrala, M.K. and Walters, R. G. (2001), “The impact of category

management on retailer prices and performance: Theory and evidence”,
Journal of Marketing, vol. 65 (October) pp.16-32.
Birtwistle, G., Siddiqui, N. and Fiorito, S.S. (2003), “Quick Response: Perceptions
of UK fashion retailers”, International Journal of Retail and Distribution
Management, vol. 31, no. 2, pp. 118-128.
Broniarczyk, S.M., Hoyer, W.D. and McAlister, L. (1998), “Consumers’ perceptions
of the assortment offered in a grocery category: the impact of item
reduction,” Journal of Marketing Research, vol. 35, May, pp. 166-176.
Category Management: The Guinness Guide (1998). Guinness Great Britain Take
Home Trade. (Note: This is a proprietary Guinness publication).
Chisnall, P. M. (1992), Marketing Research, 4th Edition, McGraw-Hill, London.
Churchill, G. A., Jr. (1995), Marketing Research: Methodological Foundations, The
Dryden Press, Fort Worth.
DeLorme, D.E. and Reid, L.N. (1999), “Moviegoers’ experiences and
interpretations of brands in films revisited”, Journal of Advertising, vol. 28,
no. 2, pp. 71-95.
Denzin, N.K. (1978), The Research Act: A Theoretical Introduction to Sociological
Methods, Aldine, Chicago.
Dewsnap, B. and Jobber, D. (1999), “Category Management: A Vehicle for
Integration between Sales and Marketing”, The Journal of Brand
Management, vol. 6, pp. 380-392.
Dhar, S.K., Hoch and S.J., Kumar, N. (2001), “Effective category management
depends on the role of the category”, Journal of Retailing, vol. 77, pp.165-
184.
di-Nome, T, (1999), “Category micro-management?”, Supermarket Business, Best
of Class Supplement , pp. 3-6.
Dodd, C.A., Clarke, I. and Kirkup, M.H. (1998), “Camera observations of customer
behaviours in fashion retailing: Methodological propositions”, International
Journal of Retail and Distribution Management, vol. 26 no. 8, pp. 311-317.
DTI (2003), “Competitive Analysis of the Retail Sector in the UK”, Report
submitted to The Department of Trade and Industry, Tender CGS, 1239.
207
Dussart, C. (1998), “Category Management: strengths, limits and developments”,
European Management Journal, vol. 16, no. 1, February, pp. 50-62.
ECR Europe (2000), The Essential Guide to Day-to-Day CM.
Fern, E.F. (1982), “The use of focus groups for idea generation: the effects of
group size, acquaintanceship, and moderator on response quantity and
quality,” Journal of Marketing Research, vol. 9 (February), pp. 1-13.
Fiorito, S.S., May, E.G. and Straughn, K. (1995), “Quick Response in retailing:
Components and implementation”, International Journal of Retail and
Distribution Management, vol. 23 no. 5, pp. 12-21.
Fisher, M.L (1997), “What is the right supply chain for your product?” Harvard
Business Review, March-April, pp 106-116.
Freedman, P.M., Reyner, M. and Tochtermann, T (1997), “European category
management: Look before you leap”, The McKinsey Quarterly, no. 1,
pp.156-164
George, M., Freeling, A. and Court, D. (1997), “Reinventing the marketing
organization”, McKinsey Quarterly, vol. 4, pp. 43-62.
Giunipero, L.C., Fiorito, S.S., Pearcy, D.H. and Dandeo, L. (2001), “The impact of
Vendor incentives on Quick Response”, The International Review of Retail,
Distribution and Consumer Research, vol. 11 no. 4, pp. 359-376.
Goldman, A.E. (1962), “The Group Depth Interview”, Journal of Marketing, 26.
Green, P.E., Tull, D. S. and Albaum, G. (1988), Research and Marketing
Decisions, 5th Edition, Prentice-Hall, New Jersey.
Gregory, H. (2001), “Hard road great ideal”, The Grocer, November 3, pp. 40-41.
Gruen, T.W. and Shah, R.H. (2000), “Determinants and outcomes of plan
objectivity and implementation in category management relationships”,
Journal of Retailing, vol. 76 no. 4, pp. 483-510.
Gruen. T.W. (2002), “The evolution of category management”, ECR Journal, vol. 2
no. 1 (Spring), pp. 17 ibid.
Haire, M. (1950), “Projective Techniques in Marketing Research”, Journal of
Marketing, vol. 14, pp. 649-56.
Harlow, P. (1994), “Category Management: A new era in FMCG buyer-supplier
relationships”, Journal of Brand Management, vol. 2 no. 5, pp. 289-295.
Harris, B. and McPartland, M. (1993), “Category management defined: What it is
and why it works”, Progressive Grocer, vol. 72 issue 9, pp. 5-11.

208
Hines, T. (2001), “From analogue to digital supply chains: implications for fashion
marketing”, in Hines. T. and Bruce, M. (Eds), Fashion Marketing:
Contemporary Issues, Butterworth Heinemann, Oxford, pp. 1-47.
Hoch, S.J. (2002), “How effective is category management?”, ECR Journal, vol. 2
no.1 (Spring), pp. 26 ibid.
Hunter, N.A. and Valentino, P. (1995), “Quick Response – ten years later,
International Journal of Clothing and Science and Technology, vol. 7 no. 4,
pp. 30-40.
Hutchins, R. (1997), “Category management in the food industry: a research
agenda”, British Food Journal, vol. 99 no. 5, pp. 177-180.
IGD [Institute of Grocery Distribution] (1999), Report: “Category Management in
Action”.
IGD [Institute of Grocery Distribution] (2002), Report: “Category Management –
Which Way Now? Debate, evolution and future destination.”
Johnson, M. (1999), “From understanding consumer behaviour to testing category
strategies”, Journal of the Market Research Society, vol. 41 no. 3, July, pp.
259-288.
Johnson, M. and Pinnington, D. (1998), “Supporting the Category Management
challenge: How research can contribute”, Journal of Market Research
Society, vol. 40 no. 1, pp. 33-53.
Joint Industry Report on Efficient Consumer Response (1995), “Category
Management Report: Enhancing Consumer Value in the Grocery Industry”,
Category Management Subcommittee, ECR Best Practices Operating
Committee and The Partnering Group Inc.
Ko, E, and Kincade, D. (1997), “The impact of quick response technologies on
retailer store attributes”, International Journal of Retail & Distribution
Management, vol. 25 no. 2, pp. 90-98.
Ko, E, and Kincade, D. (2000), “Impact of business type upon the adoption of
quick response technologies – The apparel industry experience”,
International Journal of Operations & Production Management, vol. 20 no.
9, pp. 1093-1111.
Kotzab, H. (1999) “Improving supply chain performance by efficient consumer
response? A critical comparison of existing ECR approaches”, Journal of
Business and Industrial Marketing, vol. 14, no 5/6, pp. 364-377.

209
Laurent, G. and Kapferer, J. (1985), “Measuring consumer involvement profiles”,
Journal of Marketing Research, vol. 22, pp. 41-53.
Lowson, B., (1998), “Quick Response for SMEs: A Feasibility Study”, Textile
Institute, UK.
Mathews, R. (1996), “Category management grows up”, Progressive Grocer,
November, pp. 79-81.
Mintel Marketing Intelligence (1997), Bras and Pants (November). Mintel
International Group Limited.
Mitchell, A. (2001), “Category management was too expensive and remote from
consumers”, The Grocer, February 1, p.31.
Nielsen, A.C. (1992), Category Management: Positioning Your Organization to
Win. Chicago: NTC Business Books.
Pearce, A.M (1996), “Insights from industry: Efficient consumer response:
managing the supply chain for ‘ultimate’ consumer satisfaction”, Supply
Chain Management, vol. 1 no. 2, pp.11-14.
Qureshi, B. and Baker, J. (1998), “Category management and effective consumer
response: The role of market research”, Marketing and Research Today,
February, pp. 23-31.
Richards, A. (1995), “Supermarket Superpowers”, Marketing, (August 10), pp. 18-
20.
Storey, J. and Benady, D. (2001), “The Final Curtain?” Marketing Week, (1
February), pp. 28-31.
Sudman, S. and Blair, E. (1999), “Sampling in the twenty-first century”, Academy
of Marketing Science, vol. 27 no. 2, pp. 269-277.
Wellman, D. (1997), “A categorical rejection,” Supermarket Business, vol. 52 part
9, pp.69-73.
Whiteoak, P. (1999), “The realities of quick response in the grocery sector. A
supplier viewpoint,” International Journal of Physical Distribution and
Logistics Management, vol. 29 no. 7/8, pp. 508-520.
Zenor, M. J. (1994), “The profit benefits of category management”, Journal of
Marketing Research, vol. 31 (May), pp.202-213.

210
Enjoyment of the Shopping Experience: Impact on Customers'
Repatronage Intentions and Gender Influence

Cathy Hart, Andrew M. Farrell, Grazyna Stachow, Gary Reed and

John W. Cadogan

The Service Industries Journal,

2007, Vol. 27(5), 583-604

Abstract
In this paper the authors investigate enjoyment of the shopping experience, its
influence on consumers’ intention to repatronise a regional shopping centre and
the effect of gender differences on shopping enjoyment. Four dimensions of
shopping enjoyment are proposed and a 16-item measure is developed to
measure 536 consumer perceptions of the shopping experience across five
counties in the United Kingdom. Findings indicate that shopping experience
enjoyment has a significant positive influence upon customers’ repatronage
intentions. Furthermore, men are found to have a stronger relationship of
enjoyment with repatronage than women. The implications of these results are
discussed, together with managerial implications, study limitations, and future
research directions.

Key words: Regional Shopping Centres, Enjoyment, Repatronage Intentions,

Gender Influences

211
Paper 7. Enjoyment of the Shopping Experience: Impact on
Customers' Repatronage Intentions and Gender Influence

Introduction
One of the key decisions faced by customers is where to shop in terms of
shopping location and store choice [Sit, Merrilees and Birch, 2003]. As a result,
retailer concern with patronage and repatronage is now an important topic of
marketing research [Machleit, Meyer and Eroglu, 2005]. While considerable
research has focused on patronage of retail stores, less is known about consumer
intentions towards patronage of a specific shopping centre and factors that may
influence repatronage. Previous research has demonstrated that store image has
a significant role to play in consumers’ patronage decisions [Oppewal and
Timmermans, 1997]; thus it may be assumed that the image of a shopping centre
may also impact on consumers’ decisions of where to shop. However, while
considerable work has been done on the role of store and even shopping mall
image on consumer patronage [Finn and Louviere, 1996; Wakefield and Baker,
1998; Sit, Merrilees and Birch, 2003] there is, to date, very little research
examining the image of the regional shopping centre, as a more holistic entity, in
determining consumers’ patronage decisions. This is important because
understanding more about what attracts consumers to a specific shopping centre,
followed by effective management of consumer evaluation, should lead to
increased repatronage intentions.

Consumers have many different choices of shopping locations, however.

Regional shopping centres in particular play a key role in the future of the modern
economy [Finn and Louviere, 1996; Phillips and Swaffin-Smith, 2004]. Since
shopping has been identified as the primary purpose for the majority of visits to
town and city centres [Warnaby and Davies, 1997] having the correct retail offer
and facilities to entice visitors into a regional centre results in increased spending,
greater employment opportunities and hence local economic regeneration.
Furthermore, regional centres have been struggling to attract and keep local
customers that may otherwise ‘outshop’ or patronise competing shopping areas

212
[Kirkup and Rafiq, 1999; Whyatt, 2004; Williams, Hubbard, Clark and Berkeley,
2001].

Regional shopping centres represent more than a shopping mall in this

instance. For the purposes of this paper, they include shopping areas with a wide
range of shopping facilities within larger geographical regions, such as town or city
centres. They contain a mixture of traditional high streets, commercial shopping
malls and smaller shopping precincts, and incorporate features which are either
privately owned (merchandise, buildings, signs) or publicly owned (roads, car
parks and amenities) [Bell, 1999]. Perhaps a major difference is that whilst specific
retailers and shopping malls have a long history of co-ordinating their advertising
and promotion efforts, regional shopping centres have only recently begun to
market themselves in terms of image creation and maintenance [Kirkup and Rafiq,
1999; Kupke, 2004; Warnaby and Medway, 2004]. The majority of research in this
field has concentrated on shoppers’ responses at the individual store or, at most,
the shopping mall level; this study focuses on a relatively untouched area of
research, in terms of exploring the impression or holistic image of a shopping
centre as a whole and the influence it can have upon customers’ future shopping
decisions [Bell, 1999; Warnaby and Medway, 2004]. With this in mind, this article
examines how the enjoyment of the shopping experience, as well as the
attractiveness of the shopping centre itself, can influence customers’ behavioural
responses and whether or not they intend to repatronise that shopping centre.
Furthermore, the literature also implies that shopping enjoyment may be a function
of gender. For example Otnes and McGrath [2001] recognise gender differences
in shopping behaviour, involvement in the shopping process and attitudes towards
shopping as a social activity. This is consistent with the view that women behave
differently to men [Stern, 1999] and often display differing approaches to
socialisation [Severiens and Ten Dam, 1998]. Thus there are theoretical reasons
to expect that women may experience shopping enjoyment differently to men.
However, empirical research into the role that gender may play in determining
shopping enjoyment is sparse, and efforts on this front are warranted.

Therefore, the objectives of this research are threefold. Firstly, the research
seeks to generate an understanding of the broad factors that contribute to

213
consumer image perceptions of shopping centres and their associated enjoyment
of the shopping experience. Secondly, the research aims to discover the impact
that enjoyment of shopping in a specific shopping centre has upon customers’
repatronage intentions. Thirdly, the research seeks to examine whether there are
differences in the relationship between enjoyment of shopping and repatronage
intentions based upon whether respondents are male or female.

The paper is organised as follows. Firstly, the existing literature is discussed

in order to develop the theoretical background and the conceptual framework to
the study. Then, based on factors involved in the creation of shopping location
image, enjoyment and repatronage, hypotheses are generated and proposed. The
section following this explains how gender could impact upon enjoyment of the
shopping experience and its relationship to repatronage intentions. The following
section describes the methodology used in the study, including the sample,
measures and scales generated. The following section contains the subsequent
analysis and results, while the final section discusses the findings, puts forward
implications for managers of regional shopping centres and suggests some
directions for future researchers to consider.

Theoretical Background
This section discusses the background literature relating to image, enjoyment,
repatronage and gender. It is established that the concept of image is intrinsically
linked to retail patronage [e.g. Darden and Babin, 1994] and store loyalty
[Mazursky and Jacoby, 1986]. The literature on image therefore provides an
important starting point to understanding the potential linkages with behavioural
concepts of enjoyment and repatronage of shopping centres.

Image

The literature pertaining to image is mainly focused on the retail store and its
ability to project a distinctive identity or ‘personality’ in comparison with its
competitors [Martineau, 1958]. In adapting the store image literature, the image of
a regional shopping centre is similarly hypothesised to create a distinct identity,
which distinguishes it from other, competing centres. The image of a shopping

214
location has been described as a multidimensional concept consisting of features
which are broadly summarised into two categories: firstly tangible or functional
attributes which relate to its physical features; and secondly the more intangible
attributes which represent its atmospheric qualities [McGoldrick, 2002]. The image
of a regional shopping centre is thus considered to be an holistic entity created
from elements such as retail mix, infrastructure and atmosphere [Leo and Philippe,
2002] which can be managed to create a shopping destination which is attractive
to potential customers [Warnaby and Medway, 2004]. A strong image of a
shopping location creates competitive advantage [Sit, Merrilees and Birch, 2003],
being associated with greater willingness to purchase, longer time spent shopping
and higher sales turnover; in effect more positive patronage behaviour [Bitner,
1992; Dennis, Newman and Marsland, 2005; Oppewal and Timmermans, 1997].
Specific image dimensions can influence emotional responses and patronage
decisions in particular shopping situations [Finn and Louviere, 1996; Tai and Fung,
1997]. This study aims to identify the dimensions of regional shopping centre
image which are believed to affect customer behaviour by evoking a range of
emotional responses and to examine how these dimensions influence purchasing
behaviour and patronage decisions.

Enjoyment

Shopping has been referred to as a fun, pleasurable activity that leads to feelings
of ‘joy’ [Jin and Sternquist, 2004]. Positive image has been shown to create higher
levels of pleasurable feelings in customers, and reflects their enjoyment of
spending time in the area [Bell, 1999]. Some attention in the retail literature has
focussed on the notion of excitement [Wakefield and Baker, 1998]. However,
whilst excitement corresponds to an emotional state that elicits high levels of
pleasure and arousal, enjoyment tends to represent a less emotionally intensive
evaluation, which is similar to liking. Of the two emotional states, logic would
dictate that customers would find themselves in a state of enjoyment more often
than a state of excitement, especially if they are regular shoppers. Bell [1999]
supports the notion that affect, or liking, is a more stable concept capturing
consumers’ perceptions of a shopping centre. Enjoyment itself has been reported
as a motivation for ‘bricks’ shopping, as opposed to shopping on the Internet
[Dennis, Newman and Marsland, 2005] so the importance of the construct to
215
modern shopping centres is clear. Therefore, research examining the notion of
enjoyment appears to be more theoretically valid than excitement for research into
general shopping areas where customers are likely to shop more frequently.

Repatronage

Whilst the work examining retail patronage at the store level or shopping mall level
has long attracted the attention of researchers [e.g. Burns and Warren, 1995;
Spies, Hesse and Loesch, 1997; Grewal, Baker, Levy and Voss, 2003], work
investigating these effects at the geographical shopping centre level remains
sporadic. The quality of the shopping experience is generally thought to have a
positive effect on customers’ behavioural intentions [Tai and Fung, 1997;
Wakefield and Baker, 1998], with enjoyment of shopping as a pleasurable
experience associated with raised patronage [Mano, 1999]. Further, it has been
hypothesised that higher levels of excitement lead to greater repatronage
intentions [Wakefield and Baker, 1998]. For the purposes of this study, enjoyment
is thought to be a more relevant construct to delineate than that of excitement;
nevertheless a similar relationship is hypothesised in the current study between
the notion of enjoyment of the shopping experience and repatronage intentions.

Gender

The role that gender plays in consumer behaviour research is certainly worthy of
researchers’ attention [Palan, 2002; Stern, 1988]. Indeed, marketers have been
particularly interested in the roles that gender plays as it can have a significant
impact upon consumers’ behaviour [Stern, 1988; 1999]. Much of the research in
consumer behaviour has focussed upon constructs such as biological gender (i.e.,
male or female) [Palan, 2002], gender roles [Lavin, 1993], or gender identity
[Fischer and Arnold, 1994]. Previous work has identified differences in shopping-
related attitudes based upon biological gender [Palan, 2002] and gender roles
[Lavin, 1993], although distinct gender roles appear to have become somewhat
blurred as evidenced by more recent work [Otnes and McGrath, 2001]. For the
purposes of this paper, gender is used to refer to biological gender and, to a lesser
extent, gender roles rather than other gender-related concepts such as gender
identity [Fischer and Arnold, 1994; Palan, 2002].

216
Conceptual Development
Since the early conceptualisation of store image by Martineau [1958], various
researchers have sought to decompose image into a set of dimensions that are
most valued by consumers and thus may be controlled and measured. In
summarising the store image literature, Lindquist [1974] concluded that nine
different dimensions contributed to consumers’ perceptions of store image; namely
merchandise, service, clientele, physical facilities, convenience, promotion, store
atmosphere, institutional factors and post-transactional satisfaction. While many of
these aspects of store image are also applicable to the broader regional shopping
centre, the regional centre as a whole also contains public spaces which are
beyond the scope of individual retailers to directly manage or influence, such as
architectural design, street layouts, tenant mix, parking, refreshment and toilet
facilities [Bell, 1999; Oppewal and Timmermans, 1999]. These studies (ibid) have
shown that such aspects of the holistic retail area will also influence customer
perceptions and hence patronage intentions.

Consequently, for the purposes of this study, we focus on those factors that
are specific to the shopping centre as a whole, rather than the micro-management
level issue pertinent to individual stores. Specifically, building on attributes of
shopping centre image commonly discussed in the literature, we identify four
dimensions, which we believe are central to the notion of the shopping enjoyment
experience. These attributes are selected for their relevance in defining a
particular shopping centre as an holistic entity, rather than relating to individual
stores.

First accessibility refers to the convenience and logistic aspects of store

location, travel, parking and pedestrian areas. This functional dimension of
shopping centre image has received attention in the literature, and is considered
important on a number of fronts [Sit, Merrilees and Birch, 2003; Dennis, Newman
and Marsland, 2005]. However, while some early research has considered travel
distances in relation to shopping centre patronage, Bell [1999] in particular notes
that these considerations have until now failed to capture emotional responses
which lead to consumers’ image perceptions.

217
 Second, environment refers to consumers’ responses to the more practical
aspects of shopping centre management. Previous research has identified these
aspects to include cleanliness of the shopping centre [Oppenwal and
Timmermans, 1999], opening hours and security [Lehtonen and Maenpaa, 1997],
which, when combined have been shown to influence patronage decisions [Sit,
Merrilees and Birch, 2003].

Third, atmosphere has received considerable attention in the marketing,

retailing and organisational behaviour literature in the context of consumer
response to the physical environment. In the special case of a shopping centre,
atmosphere relates to perceptions of the visual appeal of the architecture and
window displays, appreciation of store variety, and the comfort elements provided
by cafes and restaurants, all of which are associated with positive evaluations of a
shopping centre and hence likelihood to repatronise [Wakefield and Baker, 1998].

Finally, research into services marketing and retail has determined that
attitudes of service personnel have an important part to play in influencing
customers’ shopping experience and the positive or negative impression formed
by the customer [Bitner, Booms and Tetreault, 1990; McGoldrick, 2002;
Parasuraman, Zeithaml and Berry, 1988]. Indeed, according to Mittal and Lassar
[1996, p. 105] personnel displaying enthusiasm and warmth can lead to a
‘personally rewarding shopping and service experience’.

Therefore, in this study we sought to generate items to measure each of the

four dimensions mentioned above. It is hypothesised that these four dimensions
will reflect a higher order latent construct, namely enjoyment of the shopping
experience. Next, we will discuss in more detail the reasoning behind the
formation of the enjoyment of the shopping experience construct, based upon a
brief review of the retail and shopping literature.

Dimensions of enjoyment of the shopping experience: Accessibility

With increased car ownership, customers are prepared to travel longer distances
to patronise shopping centres where the retail mix is sufficiently attractive, rather
than making shopping centre choices based on the nearest convenient location
218
[Dennis, Marsland and Cockett, 2002; Whyatt, 2004]. This suggests that ease of
travel to and within attractive locations has an important role in shopping decision
making, particularly in regard to the growth of competition from Internet shopping
[Dennis, Newman and Marsland, 2005].

Sit, Merrilees and Birch [2003] separate accessibility in a shopping centre

into ‘macro-accessibility’ which refers to transport links from the home location to
the shopping centre, and ‘micro-accessibility’ which refers to parking facilities and
ease of navigation between stores and other facilities within the centre, arguing
that paying attention to both aspects will make the shopping experience more
enjoyable. This is supported by Leo and Philippe [2002] who indicate that poor
directional indicators for travellers to a shopping location result in dissatisfaction
and disappointment with the shopping expedition. Oppewal and Timmermans
[1999] found that one of the aspects of the micro-accessibility of public space,
which were favoured by customers, was compactness of layout, in terms of
accessing stores and space reserved for pedestrian access.

It seems likely, therefore, that factors that contribute towards a customer’s

perceptions of shopping accessibility, such as transport links and car parking
facilities, as well as store location and ease of movement for pedestrians, should
influence their enjoyment of the shopping experience. Based upon this, we
propose that customers’ enjoyment of the shopping experience will reflect the
accessibility of the shopping region.

Dimensions of enjoyment of the shopping experience: Atmosphere

The effects of atmospherics, or physical design elements, on consumers is

recognised by managers and mentioned in virtually all marketing, retailing, and
organisational behaviour texts [Bitner, 1992; Mazursky and Jacoby, 1986;
McGoldrick and Pieros, 1998; Spies, Hesse and Loesch, 1997], and has been
argued to provide an extremely effective point of competitive differentiation [Bell,
1999; Tai and Fung, 1997]. However, despite this attention, little empirical work
has been completed to address the effects of physical surroundings on customers’
behavioural responses in regional shopping centres [Dennis, Murphy, Marsland,
Cockett and Patel, 2002; Kirkup and Rafiq, 1999].
219
 Atmosphere is created by the interaction between the location’s physical
attributes and customers’ subjective perceptions [Hackett, Foxall and Van Raaij,
1993]. Therefore atmosphere is reflected in customers’ reactions to the physical
environment and is related to purchasing behaviour [Spies, Hesse and Loesch,
1997; Smith and Burns, 1996; Tai and Fung, 1997]. Within the store environment,
customers are stimulated by atmospheric aspects such as music, lighting, space,
colours and aromas [McGoldrick and Pieros, 1998; Wakefield and Baker, 1998]. In
a regional shopping centre, atmosphere is created by customers’ responses to the
broader aesthetic appeal of the centre, such as the visual appearance of the
architecture and the variety of the store range or tenant mix [Bell, 1999; Leo and
Philippe, 2002; Wakefield and Baker, 1998]. The attractiveness of storefronts and
window displays, and the availability of cafes and restaurants, also contribute to
atmospherics in a shopping location [Oppewal and Timmermans, 1999; Sit,
Merrilees and Birch, 2003; Warnaby and Medway, 2004].

According to environmental psychologists, individuals respond to places

with two general reactions: approach and avoid [Mehrabian and Russell, 1974].
Approach behaviours include all positive behaviours that might be directed
towards a place, such as desire to stay or explore [Bitner, 1992]. A study by
Donovan and Rossiter [1982] found that in retail settings perceptions of the
shopping environment influenced positive approach behaviours, namely shopping
enjoyment, returning, and time spent browsing. The better the atmosphere, the
more positive the perceptions of the shopping centre are likely to be [Bitner, 1992].
Therefore, we expect that attitudes towards shopping centre atmosphere, resulting
in a positive influence upon shopping enjoyment, will have a positive impact upon
repatronage intentions. This leads us to propose that customers’ enjoyment of the
shopping experience will be reflected by their perceptions of the atmosphere of the
shopping region.

Dimensions of enjoyment of the shopping experience: Environment

Characteristics of the physical shopping environment also influence perceptions of

that environment [Bitner, 1992], and the retail environment also influences
shopping enjoyment and returning behaviour [Donovan and Rossiter, 1982]. This
study argues that factors in the physical environment which are specific to the
220
centre as a whole will influence customers’ enjoyment of shopping. These factors
are within the responsibility of managers of regional shopping centres, and include
security, cleanliness of the shopping centre, and opening hours which are
convenient to customers [Sit, Merrilees and Birch, 2003; Whyatt, 2004].

These factors have received limited attention in the literature [Sit, Merrilees
and Birch, 2003], yet are important considerations in customers’ image
perceptions of the shopping centre, as distinct from individual stores or even
shopping malls. For example, even where stores and shopping malls create a
feeling of safety, this does not easily transfer to the broader shopping centre area
[Lehtonen and Maenpaa, 1997]. Perceptions of high levels of crime associated
with the regional area in general create negative image connotations for the
shopping centre within that area, which are unrelated to individual stores [Sit,
Merrilees and Birch, 2003]. Keeping the public space areas clean and tidy through
maintenance of the streets, hallways and buildings is an important factor in
customers’ perceptions of the pleasantness of the shopping experience [Oppewal
and Timmermans, 1999]. Again, opening hours are determined in many cases by
local authorities rather than by individual stores [McGoldrick, 2002]. As customers
become increasingly time-pressed, and with the availability of extended shopping
hours, store opening hours, which are convenient for those who cannot shop
during ‘normal’ shopping times, are a factor in encouraging many more customers
to patronise the shopping centre [McGoldrick, 2002; Wakefield and Baker, 1998].
This study argues that these three management aspects of the environment of the
shopping centre, namely safety, cleanliness and opening hours, will be reflected in
customers’ attitudes towards enjoyment of the shopping experience.

Dimensions of enjoyment of the shopping experience: Service Personnel

Previous work examining service encounters in retail settings has found that the
attitudes and behaviours of employees can have a positive influence upon
customers’ emotions, satisfaction, perceptions of quality and hence purchase
intentions [Babin, Babin and Boles, 1999; Baker, Parasuraman, Grewal and Voss,
2002; Dabholkar, Thorpe and Rentz, 1996; Parasuraman, Zeithaml and Berry,
1988]. Similarly, it has been found that delays in service, or poor service overall,
can lead to negative impressions being formed of the service received [Bitner,
221
Booms and Tetreault, 1990]. It can be assumed, therefore, that if employees
display positive attitudes or sufficiently impress the customer, then the customer
will leave with a more favourable impression of the retail encounter, at least with
the individual service encounter or individual store.

At the level of the individual retail store, evaluations of service quality are
known to be linked with repatronage intentions, with more favourable perceptions
of service quality resulting in more favourable behavioural intentions such as
recommendation, preference and greater repatronage [Darian, Tucci and Wiman,
2001; Sivadas and Baker-Prewitt, 2000]. These authors indicate that interaction
between the customer and service personnel is a deciding factor in the patronage
decision. Customer service quality is variously described as consisting of
friendliness, knowledge of goods sold, ability to give advice, respect and
responsiveness without being ‘pushy’ [Darian, Tucci and Wiman, 2001; Leo and
Philippe, 2002]. For the purposes of this study, these last four components are
summarised as ‘helpfulness’.

It should of course be noted that much of the prior work investigating

service encounters has been conducted at the individual store level, rather than
attempting to unravel the mystery of shopping centre impressions, as the current
study seeks to do. Unlike the factors representing shopping environment which are
related to the public aspects of the shopping centre, customer service quality may
be more relevant to individual stores than to the overall regional centre, since
individual retailers may have specific staff training programmes which differ from
one another. Nevertheless, Leo and Philippe [2002] show that knowledgeable staff
that gave good advice contributed to perceptions of the entire shopping centre,
rather than an individual store. Sivadas and Baker-Prewitt [2000] consider that,
unlike customer satisfaction, which represents an evaluation of an individual
service encounter, ‘service quality’ refers to a global measure of service
performance, which can be extended to the shopping centre level. Furthermore,
Darian, Tucci and Wiman [2001] have shown that service quality needs only to
meet a basic minimum threshold level to be seen as acceptable to customers;
reaching this threshold level, rather than delivering outstanding service, is more
likely to occur in more stores across the shopping centre as a whole.

222
 Customer service quality can therefore be considered a dimension of the
holistic shopping centre image in that it represents the ‘augmented product’ that
supports the basic merchandising function of the shopping centre and also adds
value to the total enjoyment of the shopping experience [Sit, Merrilees and Birch,
2003]. Therefore, we propose that customers’ enjoyment of the shopping
experience will be reflected in the extent to which customers perceive that they
have received good service in terms of friendliness and helpfulness of staff. It
seems logical that if a customer has a favourable impression of a service
encounter, it indicates that they have enjoyed the encounter and this will lead to a
more favourable impression of the shopping centre. Furthermore, they are more
likely to return to the shopping centre.

Outcomes of Enjoyment

In inferring that enjoyment should have an important role in determining

repatronage intentions for a particular location, it should also be noted that
individual consumers place different importance on the various image attributes of
a shopping centre [Luomala, 2003]. Therefore, one of the objectives of this
research was to attempt to generate a generic measure of shopping enjoyment
and shopping centre attractiveness rather than to provide a limited-scope
measure. In doing so, the current study builds upon the earlier work of Oppewal
and Timmermans [1999], Sit, Merrilees and Birch [2003], and Wakefield and Baker
[1998]. The longer a consumer spends in a retail environment, the more they are
likely to spend [Donovan, Rossiter, Marcoolyn and Nesdale, 1994; Wakefield and
Baker, 1998]. Work by Donovan and Rossiter [1982] found that enjoyment of a
retail experience results in more time being spent shopping and return patronage.
In retailing studies, satisfaction with the retailing experience has been
hypothesised to increase intentions to revisit the shopping area [Dabholkar,
Thorpe and Rentz, 1996; Parasuraman, Zeithaml and Berry, 1988]. Therefore, we
hypothesise the following:

H1: The greater the customer’s enjoyment of the shopping experience,

the greater the likelihood of his or her returning to the shopping centre in
future.

223
The Influence of Gender

Evolutionary psychology contains a theory known as the ‘Savannah Hypothesis.’

This hypothesis argues that human psychology has been shaped by the need to
evolve and adapt to the savannah environment [Dennis and McCall, 2005]. More
specifically: “selection has favoured preferences, motivations and cognitions to
explore and settle in environments abundant with the resources needed to sustain
life” [Buss, 1998, p. 83]. While it is outside of the scope of this paper to initiate
detailed discussion of the Savannah Hypothesis, authors have recently begun to
posit that this evolutionary adaptability may influence the way that we shop
[Dennis, Newman and Marsland, 2005]. It is thought that explanations of our
shopping behaviour may actually date back to the earliest known roles of women
and men, namely gatherer and hunter respectively. Earlier research seems to
confirm the existence of well-defined gender shopping roles [Fischer and Arnold,
1994] although more recently relaxed gender roles seem to have surfaced [Otnes
and McGrath, 2001]. This relaxation of gender roles is perhaps not that surprising,
given that most early conceptualisations of gender roles are biased [Stern, 1999].
In this paper, however, we adopt the Savannah Hypothesis view that gender
differences in shopping are apparent, as discussed next.

It has been argued that men and women differ in relation to how they
process information, their attitudes towards shopping motivations and
responsibility, how they process environmental cues, and their needs in the selling
encounter [Minahan and Beverland, 2005]. Indeed, recent research has begun to
present typologies of the respective shopping styles of men and women [Allegra,
2002; Bakewell and Mitchell, 2004; Campbell, 1997; Dholakia and Chiang, 2003;
Miller, 1998; Otnes and McGrath, 2001; Underhill, 1999]. However, despite this
work, there remains a need for further research examining differences in shopping
behaviour across gender [Otnes and McGrath, 2001]. What authors have found
thus far is that for most shopping experiences females are generally characterised
by: imagining and envisioning the merchandise in use [Campbell, 1997]; weighing
up of the pros and cons of the purchase [Laroche, Saad, Cleveland and Browne,
2000]; taking pride in their ability to shop [Underhill, 1999]; viewing the shopping
process as a leisure activity [Bakewell and Mitchell, 2004]; spending longer
shopping than men [Campbell, 1997]; visiting more shops than men [Allegra,
224
2002]; shopping more often than men [Dholakia, 1999]; including social interaction
as an important part of the shopping experience [Otnes and McGrath, 2001];
including entertainment in the shopping experience [Haytko and Baker, 2004]; and
shopping to express love for families and social networks [Miller, 1998]. Generally,
women view the process of shopping as a very social experience [Campbell,
1997]. These findings imply that women who enjoy shopping are likely to intend to
repatronise a shopping centre based upon this level of enjoyment.

Men, on the other hand, generally have their shopping style characterised
by the following: incisiveness, decisiveness, determination, and excitement at the
‘moment of the kill’ or purchase [Dennis Newman and Marsland, 2005];
instrumentality and product specificity [Campbell, 1997]; lack of patience and
trying to complete the shopping activity in the shortest possible time [Bakewell and
Mitchell, 2004]; preference for top brands as both symbols of economic power
[Underhill, 1999] and time savers [Bakewell and Mitchell, 2004]; and shopping for
power and achievement, so called ‘shopping to win’ [Otnes and McGrath, 2001].
Here, we notice that men are characterised by a lack of patience and a desire to
finish the shop as soon as possible.

Notably, the differences between genders most appropriate to this study are
the lack of patience on the part of men, and the desire to shop more often on the
part of women. While there is indication that enjoyment of the shopping experience
will result in heightened repatronage intentions for both men and women, we
expect a stronger relationship between these two major constructs for women due
to their more ‘involved’ approach to shopping. For a woman enjoying shopping, the
intention to repatronise a shopping region should be appreciated as an opportunity
to spend more time shopping, visit more shops, and to shop more often. Men,
seeing shopping as a fast, practical solution to a problem, are unlikely to want to
draw out their shopping experience, even if they are enjoying it. This is consistent
with gender role theory that argues women and men adopt different approaches
when the notion of socialisation is considered [Severiens and Ten Dam, 1998].
Hence, while there may be a relationship between enjoyment of the shopping
experience and intentions to repatronise for men, we expect this relationship to be
weaker than that for women. This is simply because the involvement of women in

225
the shopping process appears to be greater [Otnes and McGrath, 2001], so the
existence of a stronger relationship between shopping and outcomes seems more
logical. This leads to the formulation of the following hypothesis, building upon H1
earlier:

H2: The strength of the relationship between the customer’s enjoyment

of the shopping experience and the likelihood of his or her returning to
the shopping centre in future will be greater for women than for men.

Methodology
The specific aim of the research was to investigate the construct of enjoyment of
the shopping experience and its relationship towards repatronage of United
Kingdom shopping districts. To this end, a quantitative approach was adopted,
employing questionnaires submitted to respondents during face-to-face interviews.
Such methods allow for the controlled collection of large amounts of specific
information from respondents [Churchill and Iacobucci, 2004].

Sample

The survey was conducted during daytime shopping hours in December 2004, as
part of a wider study commissioned by the Sector Skills Council for the retail
sector 4. As the primary area of interest was the East Midlands region of the United
3F

Kingdom, customers in 17 regional towns and cities were targeted. Interviewers

were asked only to target those customers who were resident in the area of
interest. The study targeted 550 customers, 50 each in the cities of Derby,
Leicester, Lincoln, Northampton and Nottingham, and 25 each in the towns of
Boston, Buxton, Chesterfield, Corby, Grantham, Hinckley, Kettering,
Loughborough, Mansfield, Newark, Oakham and Wellingborough. Once interviews
were completed, responses from those who lived outside of the East Midlands
were removed, resulting in a total of 536 usable responses.

4
[Link]

226
 A stratified random sample was used. Overall, the sample was 61% female,
39% male. Females between the ages of 16-25 and 36-45 years of age made up
the largest proportion of the total sample (21.1% and 21.5% respectively). Most
males were between 26-35 and 46-55 (14.7% and 16.4% of the total sample
respectively). 36% of respondents had children, and 62.3% were employed. The
sample was predominantly Caucasian (89.7%).

Measurement

Respondents were asked to name the location where they shopped most and to
give their perceptions of their current location as a shopping destination. In order
to generate the scale to measure shopping enjoyment and location attractiveness,
questionnaire items were adapted from previous work by Wakefield and Baker
[1998], Oppewal and Timmermans [1999] and Sit, Merrilees and Birch [2003].

Items were selected for relevance to the shopping region as a whole as

opposed to describing characteristics of individual stores; these items are within
the remit of managers of the shopping region rather than being the responsibility of
individual store owners or managers. For example, Sit, Merrilees and Birchs’
[2003] dimensions of macro-accessibility, micro-accessibility, personal service,
food stops, and security were utilised for the scale, while items more relevant to
individual stores, such as music, product range and colour scheme, were omitted.
Items employed by Oppewal and Timmermans [1999] to assess customer
perceptions of public spaces, window displays, pedestrianised areas, and
maintenance (or cleanliness) were also used. Finally, items appearing in the work
on shopping malls of Wakefield and Baker [1998] were adapted to measure
architecture and design factors, store variety, desire to stay and repatronage
intentions, again omitting items such as aroma and temperature, which refer to
individual stores. In total, 16 items were generated from the literature to measure
the four dimensions of enjoyment of the shopping experience proposed here.

Once items had been generated, they were subjected to peer review from
six academics with backgrounds in retail and/or methodology. At the conclusion of
the peer review process, four protocols were carried out on the questionnaire, in

227
order to assess potential difficulties in questionnaire completion or understanding.
Neither the peer reviews nor the protocols highlighted any significant concerns.

Overall, the measure was designed to reflect the amount of enjoyment

derived from a shopping centre experience, with the latter comprising the following
four dimensions: accessibility of the shopping centre, atmosphere of the shopping
centre, attitudes towards shopping centre environment, and attitudes towards
shopping centre personnel [c.f., Oppewal and Timmermans, 1999; Sit, Merrilees
and Birch, 2003; Wakefield and Baker, 1998].

The endogenous variable was intentions to repatronise the shopping centre.

This outcome was measured by a single item: “It is likely that I will shop in
(location) in the future.” Respondents indicated the extent to which they agreed
with each questionnaire item using a 10-item scale ranging from 1 (very low
agreement) to 10 (very high agreement). In addition to the scale questions,
respondents also supplied demographic data, including age, gender, and ethnicity.

Scale properties

Initially, the 16 items comprising the shopping enjoyment and location

attractiveness scale had their variability assessed. All 16 items displayed
acceptable means and standard deviations given the scale range from one to ten
(means of individual items ranged from 5.24 to 8.15; standard deviations ranged
from 1.649 to 2.634). Following the recommendations of Gerbing and Hamilton
[1996], the 16 items were then subjected to an exploratory factor analysis (to
identify poorly performing items) and subsequent confirmatory factor analysis (for
additional scale purification). All structural equation models were assessed using
LISREL 8 [Jöreskog and Sörbom, 1993] utilising the maximum likelihood
procedure and the covariance matrix as input. Two items were removed during
this analysis due to poor factor loadings and/or cross-loadings on more than one
factor. The items that were removed pertained to crowdedness of the shopping
centre (“it is too crowded”) and the range of fashionable goods sold in stores (“the
stores sell a good range of fashionable items”). The results of the confirmatory
factor analysis for the remaining 14-item measurement model are shown in Table
14.
228
Table 14 Model Fit Measures

Model χ2 (df) p RMSEA NNFI CFI GFI

Measurement 306.80 (86) .000 .073 .937 .949 .922
Restricted 477.36 (206) .000 .074 .933 .924 .884
Unrestricted 459.30 (205) .000 .096 .937 .938 .888
Note: RMSEA = root mean square error of approximation; NNFI = Non-Normed Fit Index; CFI =
Comparative Fit Index; GFI = Goodness-of-Fit Index.

As indicated in Table 14, the measurement model has adequate fit statistics
with regards to all of the fit indices presented. A full list of the items retained
following factor analysis, and the dimensions of enjoyment of the shopping
experience to which they were assigned, can be found in Appendix A. The
properties of the scale for enjoyment of the shopping experience are shown in
Table 15 below.

Table 15 Correlations and Measurement Properties

Mean S.D. 1. 2. 3. 4. 5.
1. Accessibility 7.75 1.43 .70
2. Atmosphere 7.19 1.94 .56 .89
3. Environment 7.54 1.81 .69 .81 .83
4. Personnel 7.75 1.81 .55 .65 .79 .95
5. Repatronage 8.89 1.37 .32 .37 .45 .37 .70 a
Note: item inter-correlations are shown; composite reliability of each dimension is shown on the
a
diagonal in italics; : reliability of the single indicant set at .70

Analysis
Once the properties of the scale had been assessed, a model was run depicting
the influence of enjoyment of the shopping experience upon customers’
repatronage intentions. A conceptual diagram of the structural model is shown in
Figure 13 below.

In order to assess gender differences in the γ parameter between men and

women, two competing models were run via multi-group analysis. Firstly, a

229
restricted model was run in which the γ value was fixed to be identical between
men and women. The fit statistics for this restricted model are shown in Table 14.
Following this, a second model was run in which the γ parameter was estimated
freely. The fit statistics for this unrestricted model are also shown in Table 14. As
can be seen, the fit statistics for the unrestricted model show a substantial
improvement over the restricted model. This confirms that there is a difference in
the strength of the relationship between enjoyment of the shopping experience
and intentions to repatronise the shopping centre across gender. More specifically,
in the unrestricted model, χ2 decreases by 18.06 with an associated decrease of
one degree of freedom, which is significant at p < .05 (to be significant at the 5%
level, the critical value for a Chi-Square change with one degree of freedom is
3.84). Similarly, we see an increase in the Non-Normed Fit Index, the Comparative
Fit Index, and the Goodness-of-Fit Index of the unrestricted model over the
restricted model. Therefore, based upon these results we used the unrestricted
model to inform our hypotheses.

Figure 13 Conceptual Model of Shopping Enjoyment

Accessibility

Atmosphere
Enjoyment γ
of Repatronage
Intentions
Environment Shopping

Personnel

Results and Discussion

The original hypothesis of this study was that the enjoyment of the shopping
experience at a regional shopping centre is positively related to customers’

230
repatronage intentions. It was hypothesised that enjoyment of the shopping
experience is reflected by four dimensions: accessibility, atmosphere, environment
and service. Factor analysis using LISREL 8 confirmed the presence of a four-
dimensional, reflective enjoyment of the shopping experience construct as
evidenced by the fit indices presented in Table 14 being above the recommended
statistical thresholds [Hu and Bentler, 1999].

The results from the analysis show that the relationship (γ parameter)
between enjoyment of the shopping experience and intentions to repatronise the
shopping centre is positive and significant for both males and females in our study,
lending support to H1. This indicates that there is a significant, positive relationship
between enjoyment of the shopping experience (as measured by the scale
developed here) and intentions to repatronise a shopping region.

However, contrary to our expectations, the relationship between the two

focal constructs appears to be stronger for males than for females. More
specifically, when considering the results of the multi-group analysis, the results
show that there is a stronger relationship (γ parameter) between enjoyment of the
shopping experience and intentions to repatronise the shopping centre for males
than for females. For males, γ = .77 (t-value = 7.81) and for females γ = .31 (t-
value = 4.46). Hence, the findings here do not provide support for H2. Indeed, the
standardised parameter estimate for males is almost two-and-a-half times the
magnitude of the standardised parameter estimate for females.

While this may at first appear counterintuitive given the arguments put
forward in support of H2, there may in fact be perfectly logical reasons for this
finding. We assumed in our arguments that men, due to impatience when
shopping, are unlikely to have as strong a relationship between enjoyment of
shopping and repurchase intentions. Predominantly, we hypothesised that this
was as a result of women’s predisposition towards enjoyment in and pride towards
shopping, as well as increased involvement in shopping. We extrapolated out this
superior involvement in the shopping experience to conclude that women should
have a stronger disposition towards repatronage if they enjoyed the shopping
experience.

231
 However, given that men are shrewd and incisive when shopping [Dennis,
Newman and Marsland, 2005] their enjoyment of a shopping experience may be
driven by their desire to fulfil certain of their shopping experience expectations.
Perhaps shopping is a functional activity for men, and enjoyment of shopping for
men is actually driven by their shopping experience allowing them to be decisive,
and to complete their shopping quickly and efficiently. Another reason could be
that the gender role associated with men when shopping could be outdated, or
incorrect [c.f., Stern, 1999].

On the other hand, for women, enjoyment is related to shopping as a leisure

activity. Shopping for pleasure for women includes elements of browsing and
investing time in comparison of available alternatives before making purchase
decisions. To do this, women are more likely to ‘shop around’ [Campbell, 1997].
Dennis, Newman and Marsland [2005] have shown that, for women, enjoyment of
shopping is driven less by location than for men, and this could explain why, if
shopping is to be an enjoyable experience, women like to have a variety of
location options available to them. Hence a weaker relationship of enjoyment with
repatronage may be explained by the fact that for women to enjoy shopping they
prefer a range of locations to consider when selecting a shopping destination.

If women see the availability of choice as adding to their pleasure in

shopping, this would reduce their loyalty to a specific shopping location, without
directly diminishing their enjoyment of the location. This concurs with Sivadas and
Baker-Prewitt [2000] and Miranda, Konya and Havrila [2005], among others, who
have shown that favourable evaluations of the shopping experience do not
necessarily result in loyalty and repatronage, suggesting that other factors are
involved in patronage choices. Due to their propensity to be decisive and saving
time in shopping, men appear to show more loyalty in their shopping choices.
Loyalty to a shopping location can be a mechanism to simplify the shopping
process for men by reducing the confusion caused by too much choice [Bakewell
and Mitchell, 2004]. This could explain the stronger relationship between
enjoyment of the shopping experience and repatronage intentions for men than
women.

232
Conclusions and Managerial Implications
The results of the analysis give clear theoretical and managerial implications.
From a theoretical perspective, the research relates to prior work both regarding
gender differences in shopping and influences on repatronage intentions. From a
gender-related perspective, our results appear to refute the notion of the
Savannah Hypothesis. Based upon this hypothesis, we expected the relationship
between enjoyment and repatronage intentions to be stronger for women than
men, as women enjoying shopping should see repatronage as an opportunity to
extend their shopping experience, whereas men generally seek to keep shopping
time to a minimum. This difference was based upon our expectation that there are
clearly defined gender differences when it comes to shopping [c.f. Fischer and
Arnold, 1994]. However, our results appear to support the notion that for men
enjoying shopping, they indicate a stronger intention to return than do women.

This could raise potential issues for shopping centre management. Based
on previous research, marketing management may be predisposed to ensuring
that women enjoy their shopping experience and may target advertising and
promotional activities more towards women than men. In fact, they may find that
the same level of activity directed towards men could pay greater dividends in
terms of repatronage. Perhaps we are seeing a change in the definition of
shopping roles, and this could account for the unexpected results of our gender-
based repatronage intentions hypothesis. Perhaps the approaches of men and
women to shopping are changing due to the socialisation aspect of the activity
[Severiens and Ten Dam, 1998]. The generation gap between young and mature
male shoppers may be challenging traditional shopping motivations; ‘Younger
males today are more accustomed to shopping and seem to gain more enjoyment
out of it’ [Minahan and Beverland, 2005, p. 174]. Indeed, whilst earlier work
confirmed the existence of well-defined gender shopping roles [Fischer and
Arnold, 1994; Lavin, 1993] more recent work appears to refute these roles and
appears more in favour of relaxed gender roles [Otnes and McGrath, 2001].

A major managerial outcome is the understanding that customers who

enjoy their shopping intend to return to the shopping area in future. This raises a
number of issues relating to repatronage intentions. Our results give indication that

233
customers derive enjoyment from regional shopping experiences from their
assessment of accessibility, atmosphere, environment, and shopping centre
personnel. This result is similar to that provided by Wakefield and Baker [1998]
who found that excitement has a positive influence upon repatronage intentions. It
stands to reason then that if a shopping centre facilitates fast, efficient shopping,
this would appeal to men, who would enjoy shopping in that region, and may
therefore be more likely to return to the location in the future.

This research also has implications for managers of retail stores in regional
shopping centres. There is a clear need for bodies such as local authorities and
town centre managers to coordinate the shopping experience in these regional
centres so that customers’ enjoyment is maximised, in a similar way to how more
discrete shopping centres, such as shopping malls or plazas, are managed [Kirkup
and Rafiq, 1999]. This research provides a useful benchmark that retailers can use
to manage the shopping experience. With increasing participation in shopping by
male consumers who prefer to shop in a fast and efficient way, attention needs to
be paid to making shopping centres easily accessible and attractive to this
segment of shoppers. For example, mature male consumers prefer to shop by
product type rather than brand [Minahan and Beverland, 2005]; therefore town
centre managers could in the future seek to group stores by type in functional
focused ‘hotspots’; also providing clear directions, signage and customer
information / advice points within the shopping centre [Laroche, Saad, Cleveland
and Browne, 2000]. At the same time, since women comprise a higher proportion
of the shopping population, there is a need to promote aspects of the shopping
centre as a relaxing and fun leisure activity to increase female enjoyment of the
shopping location, to retain these customers and increase the likelihood of
repatronage.

Limitations and Future Research Directions

The nature of the scale in this paper indicates specific areas that customers use to
judge their enjoyment of a shopping experience in regional shopping centres.
However, in considering any research, it is important to evaluate the limitations of
the work. First, this study is cross-sectional, measuring respondents’ enjoyment of
the shopping experience and intentions to repatronise at only one point in time. As
234
such, conclusions regarding causal order amongst the variables of interest cannot
be made, and care should be exercised when interpreting the findings.

Second, although this study has confirmed a number of variables relating to

enjoyment of shopping locations, it may be likely that certain of these variables will
be more influential than others in affecting levels of enjoyment and repatronage
decisions. Further research could examine specific factors that customers use to
judge their enjoyment of a shopping experience in regional centres. Further
research would also be valuable in identifying other consumer segments than the
ones identified by gender in this study, and to segment for example according to
respondents' age or travel distance. Third, it should be noted that there are
differences between intentions to repatronise and actual repatronage behaviours
on the part of shoppers. Future work in the area could seek to examine whether
intentions to repatronise a shopping centre translate into actual repatronage
behaviours.

Fourth, we measured a limited number of variables in our work. There may

be additional factors that could contribute towards shopping enjoyment. One
possibility is that the quality and quantity of entertainment offered may influence
enjoyment of the shopping experience. For instance, some shops or shopping
centres provide additional entertainment during festive seasons (such as
Christmas, Thanksgiving, or Easter). In this respect, seasonality aspects could
also affect enjoyment of the shopping experience, and this issue warrants
empirical study.

It is also the case that enjoyment of the shopping experience may impact
upon more than just intentions to repatronise the centre. For instance, it could be
argued that shopping enjoyment may influence the discretionary amount that
shoppers are prepared to spend in a shopping trip; it is established that increased
time spent in the retail environment results in higher spending [Donovan, Rossiter,
Marcoolyn and Nesdale. 1994; Wakefield and Baker, 1998] thus enjoyment may
increase duration of time spent on the shopping visit and spending potential.
Willingness to spend more money or willingness to recommend the shopping
centre to others are other potential outcome variables.

235
 It is also important to consider the relationship between service quality and
enjoyment and future work could seek to explain how these constructs specifically
relate to one another. For example it is likely that certain aspects of shopping
enjoyment may also impact on service quality perceptions. As such, the role that
shopping enjoyment may play in shaping service quality needs to be investigated.
For instance, perceptions of the shopping centre environment could have an
impact on customers’ feelings of assurance regarding the shopping centre as a
whole. Consequently, it could be that effective management of the shopping
environment itself could influence perceptions of service quality (e.g., at the store-
level and more broadly in terms of the centre itself). Furthermore, it could be
interesting to see how the concept of entertainment, as discussed by Sit, Merrilees
and Birch [2003], relates to excitement, enjoyment and quality perceptions of
customers, and how in turn it may influence repatronage.

Therefore, while our study provides important results for managers in

shopping centres, we are encouraged that there is sufficient scope for further
research into this area. Additionally, the research concentrated on aspects of
regional shopping centres, which predominantly relate to the retail provision in
these areas. Regional town and city centres are beginning to focus their marketing
on a whole package of customer attraction which includes retail but also aspects
such as entertainment, leisure facilities, hospitality and tourism. Future research
into customer enjoyment and patronage could investigate how these aspects can
also influence patronage decisions.

Finally, since the study was conducted in a specific region of the United
Kingdom, it would be interesting to examine the stability of the results first
nationally, and then cross-nationally. It may be that customers’ enjoyment of the
shopping experience could vary across geographic regions or international
borders. Indeed, it is for this reason that there is a healthy stream of international
consumer behaviour research. However, taking these limitations into account, we
feel that the current study makes a significant contribution to the retail consumer
behaviour literature and has identified a number of potential areas for future
research into this important area of the literature.

236
References

Allegra (2002) Shopping Centres in the UK, London: Allegra Strategies.

Babin, L.A., B.J. Babin and J.S. Boles (1999) The effects of consumer perceptions
of the sales person, product and dealer on purchase intentions, Journal of
Retailing and Consumer Services, 6(2), pp. 91-97.
Baker, J., A. Parasuraman, D. Grewal and G.B. Voss (2002) The influence of
multiple store environment cues on perceived merchandise value and
patronage intentions, Journal of Marketing, 66(2), pp. 120-142.
Bakewell, C. and V-W. Mitchell (2004) Male Consumer Decision-Making Styles,
International Review of Retail, Distribution and Consumer Research, 14(2),
pp. 223-40.
Bell, S. (1999) Image and Consumer Attraction to Intraurban Retail Areas: An
Environmental Psychology Approach, Journal of Retailing and Consumer
Services, 6(2), pp. 67-78.
Bitner, M.J. (1992) Servicescapes: The Impact of Physical Surroundings on
Customers and Employees, Journal of Marketing, 56(2), pp. 57-71.
Bitner, M.J., B.H. Booms and M.S. Tetreault (1990) The Service Encounter:
Diagnosing Favorable and Unfavorable Incidents, Journal of Marketing,
54(1), pp. 71-84.
Burns, D.J. and H.B. Warren (1995) Need for uniqueness: shopping mall
preference and choice activity, International Journal of Retail and Distribution
Management, 23(12), pp. 4-12.
Buss, D.M. (1998) Evolutionary Psychology: The New Science of The Mind,
Boston: Allyn and Bacon.
Campbell, C. (1997) Shopping, Pleasure and the Sex War, in P. Falk and C.
Campbell (eds.), The Shopping Experience, London: Sage.
Churchill, G.A. and D. Iacobucci (2004) Marketing Research: Methodological
Foundations, Chicago, IL: Dryden Press.
Dabholkar, P.A., D.I. Thorpe and J.O. Rentz (1996) A Measure of Service Quality
for Retail Stores: Scale Development and Validation, Journal of the Academy
of Marketing Science, 24(1), pp. 3-16.

237
Darden, W.R. and B.J. Babin (1994) Exploring the concept of affective quality:
expanding the concept of retail personality, Journal of Business Research,
29(2), pp. 101-109.
Darian, J.C., L.A. Tucci, and A.R. Wiman (2001) Perceived salesperson service
attributes and retail patronage intentions, International Journal of Retail and
Distribution Management, 29(5), pp. 205-213.
Dennis, C., D. Marsland and T. Cockett (2002) Central Place Practice: Shopping
Centre Attractiveness Measures, Hinterland Boundaries and the UK Retail
Hierarchy, Journal of Retailing and Consumer Services, 9(4), pp. 185-99.
Dennis, C. and A. McCall (2005) The Savannah Hypothesis of Shopping, Business
Strategy Review, 16(3), August, pp. 12-16.
Dennis, C., J. Murphy, D. Marsland, T. Cockett and T. Patel (2002) Measuring
Image: Shopping Centre Case Studies, International Journal of Retail,
Distribution and Consumer Research, 12(4), pp. 355-73.
Dennis, C., A. Newman and D. Marsland (2005) Objects of Desire: Consumer
Behaviour in Shopping Centre Choices, Basingstoke: Palgrave MacMillan.
Dholakia, R.R. (1999) Going shopping: key determinants of shopping behaviours
and motivations, International Journal of Retail and Distribution Management,
27(4), pp. 154-165.
Dholakia, R.R. and K-P. Chiang (2003) Shoppers in Cyberspace: Are They From
Venus or Mars and Does it Matter?, Journal of Consumer Psychology,
13(1/2), pp. 171-6.
Donovan, R.J. and J.R. Rossiter (1982) Store Atmosphere: An Environmental
Psychology Approach, Journal of Retailing, 58(1), pp. 34-57.
Donovan, R.J., J.R. Rossiter, G. Marcoolyn and A. Nesdale (1994) Store
atmosphere and purchasing behaviour, Journal of Retailing, 70(3), pp. 283-
94.
Finn, A. and J.J. Louviere (1996) Shopping Center Image, Consideration, and
Choice: Anchor Store Contribution, Journal of Business Research, 35(3), pp.
241-51.
Fischer, E. and S.J. Arnold (1994) Sex, Gender Identity, Gender Role Attitudes,
and Consumer Behavior, Psychology & Marketing, 11(2), pp. 163-82.

238
Gerbing, D.W. and J.G. Hamilton (1996) Viability of Exploratory Factor Analysis as
a Precursor to Confirmatory Factor Analysis, Structural Equation Modeling,
3(1), pp. 62-72.
Grewal, D., J. Baker, M. Levy and G.B. Voss (2003). The effects of wait
expectations and store atmosphere evaluations on patronage intentions in
service-intensive retail stores. Journal of Retailing, 79(4), pp. 259-268.
Hackett, P.M.W., G.R Foxall and W.F. Van Raaij (1993). Consumers in Retail
Environments, in T. Garling and R.G. Colledge (eds.) Behaviour and
Environment: Psychological and Geographical Approaches, Elsevier Science
B.V.
Haytko, D.L. and J. Baker (2004) It’s all at the mall: exploring adolescent girls’
experiences, Journal of Retailing, 80(1), pp. 67-83.
Hu, L-T. and P.M. Bentler (1999) Cutoff Criteria for Fit Indices in Covariance
Structure Analysis: Conventional Criteria Versus New Alternatives, Structural
Equation Modeling: A Multidisciplinary Journal, 6(1), pp. 1-55.
Jin, B. and B. Sternquist (2004) Shopping is Truly a Joy, The Service Industries
Journal, 24(6), pp. 1-18.
Jöreskog, K.G. and D. Sörbom (1993) LISREL 8: Structural Equation Modeling
with the SIMPLIS Command Language, Hillsdale, NJ: Lawrence Erlbaum.
Kirkup, M.H. and M. Rafiq (1999) Marketing Shopping Centres: Challenges in the
UK Context, Journal of Marketing Practice: Applied Marketing Science, 5(5),
pp. 119-33.
Kupke, V. (2004) Identifying the Dimensions to Retail Centre Image, Journal of
Property Investment and Finance, 22(4), pp. 298-306.
Laroche, M., G. Saad, M. Cleveland and E. Browne (2000) Gender differences in
information search strategies for a Christmas gift, Journal of Consumer
Marketing, 17(6), pp. 500-524.
Lavin, M. (1993) Husband-Dominant, Wife-Dominant, Joint: A Shopping Typology
for Baby Boom Couples, Journal of Consumer Research, 10(3), pp. 33-42.
Lehtonen, T-K. and P. Maenpaa (1997) Shopping in the East Centre Mall, in P.
Falk and C. Campbell (eds.), The Shopping Experience, London: Sage.
Leo, P-Y. and J. Philippe (2002) Retail Centres: Location and Customer
Satisfaction, Service Industries Journal, 22(1), pp. 122-146.

239
Lindquist, J., (1974) Meaning of image: a survey of empirical and hypothetical
evidence, Journal of Retailing, 50(4), pp. 29-38.
Luomala, H.T., (2003) Understanding how retail environments are perceived: a
conceptualization and a pilot study, International Review of Retail,
Distribution and Consumer Research, 13(3), pp. 279-300.
Machleit, K.A., T. Meyer and S.A. Eroglu (2005) Evaluating the Nature of Hassles
and Uplifts in the Retail Shopping Context, Journal of Business Research,
58(5), pp. 655-663.
Mano, H. (1999) The Influence of Pre-Existing Negative Affect on Store Purchase
Intentions, Journal of Retailing, 75(2), pp. 149-172.
Martineau, P. (1958) The personality of the retail store, Harvard Business Review,
36(1), pp. 47-55.
Mazursky, D and J. Jacoby (1986) Exploring the Development of Store Images,
Journal of Retailing, 62(2), pp. 145-165.
McGoldrick, P.J and [Link] (1998) Atmospherics, pleasure and arousal: the
influence of response moderators, Journal of Marketing Management, 14(1-
3), 173-197.
McGoldrick, P. (2002) Retail Marketing, McGraw Hill Education, Berkshire.
Mehrabian, A. and J.A. Russell (1974) An Approach to Environmental Psychology,
Cambridge, MA: Massachusetts Institute of Technology (MIT).
Minahan, S. and M. Beverland (2005) Why Women Shop, Milton, Qld:
Wrightbooks.
Miller, D. (1998) A Theory of Shopping, Ithaca, NY: Cornell University.
Miranda M.J., L. Konya and I. Havrila (2005) Shoppers’ satisfaction levels are not
the only key to store loyalty, Marketing Intelligence and Planning, 23(2), pp.
220-232.
Mittal, B. and W.M. Lassar (1996) The Role of Personalization in Service
Encounters, Journal of Retailing, 72(1), 95-109.
Oppewal, H. and H. Timmermans (1997) Retailer Self-Perceived Store Image and
Competitive Position, International Review of Retail, Distribution and
Consumer Research, 7(1), pp. 41-59.
Oppewal, H. and H. Timmermans (1999) Modeling Consumer Perception of Public
Space in Shopping Centers, Environment and Behavior, 31(1), pp. 45-65.

240
Otnes, C. and M.A. McGrath (2001) Perceptions and Realities of Male Shopping
Behavior, Journal of Retailing, 77(1), pp. 111-37.
Palan, K.M. (2002) Gender Identity in Consumer Behavior Research: A Literature
Review and Research Agenda, Academy of Marketing Science Review,
2001(10), pp. 1-26 [Available at: [Link]
[Link]]
Parasuraman, A, V.A. Zeithaml and L.L. Berry (1988) SERVQUAL: A multiple-item
scale for measuring consumer perceptions of service quality, Journal of
Retailing, 64(1), pp. 12-48.
Phillips, M. and C. Swaffin-Smith (2004) Market Towns – Victims of Market
Forces? International Journal of Retail and Distribution Management, 32(11),
pp. 557-68.
Severiens, S. and G. Ten Dam (1998) Gender and Learning: Comparing Two
Theories, Higher Education, 35(3), pp. 329-350.
Sit J., B. Merrilees and D. Birch (2003) Entertainment-Seeking Shopping Centre
Patrons: The Missing Segments, International Journal of Retail and
Distribution Management, 31(2), pp. 80-94.
Sivadas, E. and J.L. Baker-Prewitt (2000) An examination of the relationship
between service quality, customer satisfaction and store loyalty, International
Journal of Retail and Distribution Management, 28(2), pp. 73-82.
Smith, P. and D.J. Burns (1996) Atmospherics and retail environments: the case of
the ‘power aisle’, International Journal of Retail and Distribution Management,
24(1), pp. 7-14.
Spies, K., F. Hesse and K. Loesch (1997) Store atmosphere, mood and
purchasing behaviour, International Journal of Research in Marketing, 14(1),
pp. 1 – 17.
Stern, B.B. (1988) Sex-Role Self-Concept Measures and Marketing: A Research
Note, Psychology & Marketing, 5(1), pp. 85-99.
Stern, B.B. (1999) Gender and Multicultural Issues in Advertising: Stages on the
Research Highway, Journal of Advertising, 28(1), pp. 1-9.
Tai, S.H.C. and A.M.C. Fung (1997) Application of an environmental psychology
model to in-store buying behaviour, International Review of Retail,
Distribution and Consumer Research, 7(4), pp. 311-337.
Underhill, P. (1999) Why We Buy, London: Orion.

241
Wakefield, K.L. and J. Baker (1998) Excitement at the Mall: Determinants and
Effects on Shopping Response, Journal of Retailing, 74(4), pp. 515-39.
Warnaby, G. and [Link] (1997) Commentary: Cities as Service Factories?
Using the Servuction System for Marketing Cities as Shopping Destinations,
International Journal of Retail and Distribution Management, 25(6), pp. 204-
10.
Warnaby, G. and D. Medway (2004) The Role of Place Marketing as a
Competitive Response by Town Centres to Out-of-Town Retail
Developments, International Review of Retail, Distribution and Consumer
Research, 14(4), pp. 457-77.
Whyatt, G. (2004) Town Centre Management: how theory informs a strategic
approach, International Journal of Retail and Distribution Management, 32(7),
pp. 346-353.
Williams, P., P. Hubbard, D. Clark and N. Berkeley (2001) Consumption, exclusion
and emotion: the social geographies of shopping, Social and Cultural
Geography, 2(2), pp. 203-219.

242
 APPENDIX A
Items Measuring Dimensions of Location Attractiveness

Directions: On a scale from 1 to 10, where 1 is the lowest and 10 is the highest
score you can give, please rate the shopping area on the following statements

ITEM(S) DIMENSION

Accessibility

The location of stores is convenient* Location

There is adequate parking *** Parking

Pedestrianised areas make it easy to get to move around ** Pedestrianisation

Travelling to shop here is easy and straightforward *** Travel

Atmosphere
The buildings in this place look attractive * Appearance

The overall atmosphere makes shopping enjoyable ** Atmosphere

There is a good selection of cafés and restaurants * / ** Food Stops

There is an excellent variety of stores * Variety

The window displays are attractive ** Fashion / Style

Environment
The shopping area looks clean and tidy *** Cleanliness

Shopping hours are convenient *** Opening Hours

It feels like a safe place to shop *** Safety

Personnel
The service in the shops is good *** Customer Service

The attitude of staff in the shops is helpful and friendly *** Staff Attitudes

* Item adapted from Wakefield and Baker [1998]

** Item adapted from Oppewal and Timmermans [1999]
*** Item adapted from Sit, Merrilees and Birch [2003]

243
 Conceptualising Town Centre Image and the Customer
Experience

Cathy Hart
Grazyna Stachow
John W. Cadogan

Journal of Marketing Management,

2013, Vol 29 (15 - 16) 1753 - 1781.

Abstract
Compared with store image literature, the town centre is underexplored as a retail
image context. This paper examines ‘what is town centre image’ by conducting an
empirical study of 536 consumers to elicit top of mind images of five town centres.
Following extensive content analysis and rigorous replication, the findings reveal a
broader, holistic image construct than previously assumed in the literature. New
psychological, intangible elements of town centre image are identified confirming the
need to measure image specific to the retail context. Further, drawing on the
customer experience framework, a conceptual model of town centre image is
developed. The model provides the basis for future conceptualisation and
operationalisation of town centre image, combined with an approach for managing
the customer experience in town centres.

Summary statement of contribution

This study demonstrates contribution to retail marketing theory by:

[Link] the first conceptualisation of holistic image in a town centre context,

thereby emphasising the neglected, psychological elements of image.
[Link] the salient dimensions and components driving town centre image,
thus facilitating operationalisation of image.
[Link] a conceptual model of town centre image based on and integrated
with the customer experience; which will enable stakeholders to develop
management strategies using the identified town centre components.

Keywords: town centre, retail image, customer experience, place marketing

244
Paper 8. Conceptualising Town Centre Image and the Customer
Experience

Introduction
Downtown and town centre shopping areas play a crucial role in the economic
health of towns and cities worldwide and need to attract consumer patronage in
order to survive and prosper (Robertson, 1999). However, competition for patronage
has steadily increased, not only from out of town planned shopping centres and
online shopping, but also from neighbouring cities and more distant shopping
venues (Weltevreden & Van Rietbergen, 2007). Periods of economic downturn
further reduce patronage, leading to store closures and unemployment. Since 2009
retailers have closed tens of thousands of stores across the US and Europe, with
vacancies reaching an all time high (Grewal, Levy & Kumar, 2009). As store
vacancies rise, the attractiveness of the shopping area reduces, changing the
consumer’s perception of that centre (Mullis & Kim, 2011; Whysall, 2011).

The consumer’s image of a shopping area is an integral part of the consumer

decision-making process, driving patronage and shopping centre equity (El Hedli &
Chebat, 2009). Few retail stores exist in isolation and the success of each store is
dependent upon the image of the total shopping area (Bell, 1999). A less positive
centre image can result in reduced patronage and outshopping to competing centres
(Miller & Kean, 1997; Mullis & Kim, 2011). Understanding the consumer’s image of a
centre thus becomes highly relevant to place marketers and centre managers, to
differentiate the centre and to protect its long-term economic viability (Hart, Farrell,
Stachow, Reed & Cadogan, 2007; Runyan & Huddleston, 2006).

The study of image in the retail environment is well documented in an

extensive body of literature extending over 50 years (c.f. Stern, Zinkhan, & Jaju,
2001). It is acknowledged that store image influences consumers’ patronage
behaviour, and researchers have sought to conceptualise and measure consumer
perceptions in order to manage and control image (Samli, 1998). Yet despite the
strategic importance of town centres, surprisingly few empirical studies investigate
the consumers’ image of a town or city centre as a retail environment. Further, the

245
conceptualisation of town centre image lags behind other image constructs,
exposing an important research gap. Without clear understanding of the town centre
image construct, successful operationalisation of image is impossible, inhibiting
measurement, and hence hindering development of the ‘place brand’ and effective
management of image for place marketing strategy. It may also be detrimental to
developing the customer experience for town centre shoppers. Customer
experience is increasingly recognized as a key factor in building retailer loyalty and
success (Verhoef et al., 2009), and the relationship between customer experience
and image has previously been noted (Berry, 1969; MacInnis & Price, 1987).
However, customer experience in a wider retail environment is relatively under
researched and therefore understanding the image of a town centre may, in turn,
help strengthen the customer experience in that location.

This paper addresses the research gap by investigating what we mean by town
centre image, utilising a customer experience perspective as a guiding framework.
The study will benefit research, theory and practice in the following ways. First, a
model of town centre image is proposed, facilitating image conceptualisation.
Second, drawing from a study eliciting ‘top of mind’ image characteristics from 534
respondents, we identify new, consumer-generated dimensions and attributes
driving town centre image, providing the basis for future operationalisation of this
image construct. Third, the findings emphasise the neglected psychological,
intangible elements of retail image and reinforce its holistic, gestalt nature. Fourth,
evaluating the findings with extant literature identifies a broader construct than
previously assumed in the literature, confirming the need to measure image specific
to the retail context that is the town centre environment. We conclude by discussing
how the image construct provides the basis for orchestrating ‘clues’ within the
shopping environment to guide customers through the buying process and
translates into practical strategies for managing the customer experience in town
centres (Verhoef et al., 2009).

Literature review
Often termed the ‘High Street’ in the UK, town centres are nevertheless an
international phenomenon. In Europe and elsewhere, the town centre delineates an
area which is ‘central to the town as a whole, which … forms the most important
246
retail area in the town … [and] also serves a wider purpose as a business, cultural
and entertainment focus for the community’ (Guy, 1994, p. xvi). While in North
America the downtown no longer occupies the highest level of the retail hierarchy, it
nevertheless maintains a significant role as a shopping destination for consumers
who choose to shop there (Padilla & Eastlick, 2009; Sneed, Runyan, Swinney &
Lim, 2011).

The context for town centre image emerges from two strands of literature. (1)
Spatial shopping behaviour, including classical geographic theories such as Central
Place Theory (Christaller, 1933) or gravity type models based on the attraction and
mass of a centre versus distance (Huff, 1963; Reilly, 1929); and (2) research into
perceptions of composite measures of town centres, or image (e.g. Bell, 1999;
Nevin & Houston, 1980). Both strands ultimately seek to predict consumer
patronage of retail locations. However, while offering some theoretical and
methodological consistency to patronage research (Bell, 1999), the geographic
approach is criticised for its broad, unrealistic assumptions regarding shopping
behaviour (Timmermans, 1981). Instead, researchers suggest that subjective
impressions in the form of image related dimensions ameliorate the impact of
distance (Bell, 1999; Bucklin, 1967).

Yet despite its importance, the literature on town centre image remains limited
in comparison to store and mall image. Researchers have for over 50 years studied
the store image construct based on Martineau (1958)’s definition of image as both
tangible and intangible dimensions, or “the way in which the store is defined in the
shopper’s mind, partly by its functional qualities and partly by an aura of
psychological attributes” (p. 47). Most influential has been Lindquist (1974)’s meta
analysis which established nine core dimensions of store image (merchandise,
service, clientele, physical facilities, convenience, promotion, atmosphere,
institutional factors and post-transaction satisfaction). Further, the mall image
literature evidences a research stream focused on atmospherics in terms of ambient
odour, music and colours (e.g Michon, Chebat & Turley, 2005) and excitement
(Wakefield & Baker, 1998). Subsequent researchers have explored the
psychological aspects of store image, hypothesizing that image has affective or

247
emotional content (Darden & Babin, 1994; Mazursky & Jacoby, 1986) and
speculated as to its holistic, gestalt nature (Dichter, 1985; Keaveney & Hunt, 1992).

In contrast, relatively few studies specifically research town centre image as a

distinct entity. A number of authors compare consumer perceptions of a town centre
with shopping malls (e.g. Howell & Rogers, 1980; Leo & Philippe, 2002; Nevin &
Houston, 1980). Other studies survey the town centre as a collective of ‘shopping
centres’ in a geographic area without distinguishing between different shopping
formats, such as urban agglomerations or in town malls (e.g. Bell, 1999; Gentry &
Burns, 1977; Hunter, 2006). Consequently, a key shortcoming of the literature is the
lack of empirically established image variables pertinent to the town centre
environment. Due to the scarcity of town centre specific measures, researchers
regularly transfer or adapt store image variables. Indeed, a number of authors (e.g.
Bell, 1999; Gentry & Burns, 1977; Hunter, 2006; Nevin & Houston, 1980) base their
image research on measures adapted from Lindquist’s (1974) synthesis of store
image literature. This store image structure of nine dimensions, containing 32
component attributes, generates considerable influence on town centre studies.
Table 16 includes studies that research image aspects of a downtown or town
centre and lists the component attributes employed. It is noticeable that the majority
of town centre attributes overlap with many of the original store image attributes
identified by Lindquist (1974).

248
Table 16 Town centre attributes in the literature

Ease of internal movement / compactness

Self-image congruency / other customers

Refreshments / comfort areas / facilities

Location / traffic / travel convenience

Place to spend time / take children

Structure and design / layout
Personnel / service quality

Activities/ special events

Advertising / promotions
Range / quality of goods
Range / quality of shops

Buildings / landscaping

Non-retailing functions
Visual appearance

Personal / friendly
Shopping hours
Style / fashion

Temperature
Price / value

Atmosphere

Cleanliness

Reputation
Crowding

Old / new

Lighting
Parking

Safety
Busy
Size
Lindquist 1974 x x x x x x x x x x x x x x x x
Downs 1970 x x x x x x x x x x x x x x
Gentry & Burns 1977 x x x x x x x x x x x x
Hauser & Koppelman 1979 x x x x x x x x x
Nevin & Houston 1980 x x x x x x x x x x x x
Howell & Rogers 1980 x x x x x x x x x x x x
Gautschi 1981 x x x x x x x x x x x x
Timmermans et al 1982 x x x x x x x x x x x x x x x x x x x x x x x
van Raaij 1983 x x x x x x x x x x x x
Wee 1986 x x x x x x x x x x x x x x x x x
Feinberg et al 1989 x x x x x x x x
Hackett & Foxall 1994 x x x x x x x x x x x x x x x
Bell 1999 x x x x x x x x x x x
Oppewal & Timmermans 1999 x x x x x x x
Leo & Philippe 2002 x x x x x x x x x x x x x x x
Kupke 2004 x x x x x x x
Hunter 2006 x x x x x x x x x x x x
Teller & Elms 2010 x x x x x x x x x x x x x

249
 Attributes are defined as ‘the narrowest, most specific constructs’ of image,
and dimensions refer to the ‘most general constructs’ containing groups of attributes
(Hansen & Deutscher, 1977 p. 59). Table 17 lists image dimensions identified in
town centre studies. While some of Lindquist (1974)’s nine store image dimensions
are recognizable in town centre image studies, here the literature diverges. Notably,
the order and number of town centre dimensions vary considerably from three to
nine, with limited consistency between the terminology and meaning of these
dimensions. While assortment appears the most regularly cited image dimension, its
interpretation is unclear in this context; do consumers evaluate individual goods or
stores, or the total number of stores in a shopping centre? (R. M. Downs, 1970).
Similarly, atmosphere features strongly in the image literature. However,
perceptions of atmosphere are likely to vary between a created, controllable
environment such as a store or mall where atmosphere is manipulated through
sounds, smells and décor (Baker, Grewal, & Parasuraman, 1994; Michon et al.,
2005), compared with an evolved environment such as a town centre that is less
controllable. The remaining image attributes used in the town centre studies show
inconsistency across studies. For example, authors less frequently mention size,
crowding and busyness. Hence, the literature demonstrates neither agreement for
the salient dimensions of town centre image, nor any consistent ‘taxonomy’ of
dimensions or attributes with which to measure town centre image (Bell, 1999, p.
68).

Furthermore, studies using store image related measures to assess urban

patronage deliver conflicting results. Authors find that dimensions explaining
patronage in shopping malls differ from those in town centres (Hackett & Foxall,
1994; Leo & Philippe, 2002). Further, Nevin and Houston (1980) and Feinberg,
Sheffler, Meoli and Rummel (1989) find their derived image dimensions are less
successful in predicting patronage of shopping malls than for town centres. Hence,
dimensions that are effective in measuring store or mall image may not consistently
transfer to town centres.

250
Table 17 Dimensions of town centre image identified in the literature
Downs (1970) (1) Service quality, (2) price, (3) structure & design, (4) hours, (5)
internal pedestrian movement, (6) shop range & quality, (7) visual
appearance, (8) traffic conditions, (9) atmosphere*
Gentry and (1) Price, variety, product quality / stores, (2) parking, traffic, (3)
Burns (1978) buildings, customers, (4) hours, advertising
Hauser and (1) Variety, (2) quality & satisfaction, (3) value, (4) parking.
Koppelman
(1979)
Nevin and (1) Assortment, (2) facilities, (3) market posture
Houston (1980)
Howell and (1) Atmosphere, (2) personnel, (3) fashion shopping, (4)
Rogers (1980) advertising, (5) convenience, (6) accessibility, (7) proximity
Gautschi (1981) (1) Transport, (2) assortment, (3) design, (4) prices, (5) hours, (6)
dress, (7) crowds
Timmermans et (1) Size of assortment, (2) accessibility, (3) atmosphere, (4)
al. (1982) physical layout
Van Raaij (1983) (1) General evaluation, (2) environment, (3) efficiency, (4)
accessibility, (5) social
Wee (1986) (1) Assortment, (2) facilitative, (3) maintenance, (4) operational

Feinberg et al. (1) Sales people, (2) merchandise, (3) reputation, (4) atmosphere,
(1989) (5) social experience
Hackett and (1) Service quality, (2) access / facilities, (3) social, (4) choice &
Foxall (1994) variety
Bell (1999) (1) Product & store range & quality, (2) customer service, (3) visual
amenity, (4) convenience, (5) price fairness
Leo and Philippe (1) Retail mix, (2) environment, (3) accessibility, (4) price
(2002)
Kupke (2004) (1) Parking, (2) price, (3) atmosphere, (4) service, (5) food facilities,
(6) range of shops, (7) hours
*Italics denote insignificance

251
 While most town centre studies survey consumer perceptions, surprisingly
few researchers elicit town centre specific attributes direct from the consumer. Yet,
authors argue that image measures based on attributes pre-specified from other
contexts might result in the omission of attributes that are relevant to consumers,
or the inclusion of attributes that are irrelevant to their perceived image (Gautschi,
1981; Zimmer & Golden, 1988), representing the researcher’s assumptions, rather
than the reality of consumers’ individual image perceptions (Keaveney & Hunt,
1992).

The diversity of the town centre environment and the way consumers
experience this, suggests a wider, multifaceted image construct. Town centres are
complex environments comprising extensive retail and non-retail functions including
housing, education, health, employment, leisure and entertainment, and public
services (Warnaby, Bennison, Davies, & Hughes, 2002). They provide a diverse and
interesting shopping experience due to unique features such as historical buildings
and related themed areas (Padilla & Eastlick, 2009; Robertson, 1997; Runyan &
Huddleston, 2006). These features lead to a sense of attachment to the community,
which the downtown represents (Runyan & Huddleston, 2006). Consumers identify
with individuals and social groups when they shop in a town centre, potentially
forming links and reciprocal obligations towards their community as a whole (Miller &
Kean, 1997). Thus, town centres are likely to hold more personal meaning to
consumers through habituation and familiarity than simply performing as a shopping
destination (Relph, 1976). Accordingly, authors suggest town centre image includes
a broader domain of attributes than captured in store image studies, which ignore
the characteristics of the wider public and social environment surrounding the town
or city centre (Oppewal & Timmermans, 1999). Hence, adapting store image
attributes and dimensions may not adequately capture the ‘total impression’ the
town centre makes on consumers’ minds (Dichter, 1985).

A final observation refers to the domain of the image construct. Bell (1999)
suggests that consumers are likely to adopt a holistic approach to town centre
image perceptions. The holistic view of image is presented as a ‘gestalt’ or overall
impression into which the dimensions of a location’s image are integrated to form
individual, subjective perceptions (Hartman & Spiro, 2005; MacInnis & Price, 1987).

252
Alternatively, image is interpreted as an abstraction simplifying complex information
(Keaveney & Hunt, 1992; Mitchell, 2001). As such, a gestalt image may instead
represent ‘a couple of ideas rather than an array of individual, specifically focused
attributes’ which could include evaluations, global perceptions or overall
impressions, as opposed to a summary of its dimensions (Zimmer & Golden, 1986,
p. 287). Despite this, researchers predominantly measure the tangible, functional
dimensions of town centre image, with little attention given to intangible elements
such as consumers’ psychological beliefs and attitudes or the holistic approach to
image (Bell, 1999). Moreover, researchers have not yet established what exactly the
holistic, gestalt notion of town centre image consists of (Stern et al., 2001).

Thus the literature exposes a fundamental gap in the conceptualisation and

operationalisation of town centre image. Existing image measures comprise a
combination of (a) measures that are relevant and used, and (b) measures that are
used but are irrelevant to that image context. However, a third group emerges: (c)
measures which are potentially relevant but not considered to date. With limited
understanding of the latter two groups, it is unclear if existing studies actually
measure what consumers perceive as the image of a town centre. Fundamentally,
these measures may provide a false picture of the town centre for managers and
marketers, undermining the effectiveness of marketing strategy.

Customer experience

In order to understand the measures specific to town centre image, an important

priority for research is to develop an appropriate conceptual underpinning. However,
the diverse nature of town centre image needs to consider the wider context in
which it operates. Customer experience has been identified as an important factor in
the formation and enhancement of retail image (MacInnis & Price, 1987). Berry
(1968, p. 24) states that ‘an image is acquired through experience and thus is
learned’, therefore the customer experience research stream may provide a suitable
conceptual framework for town centre image. There has been increasing academic
attention focussed on the consumer shopping experience and the notion of
managing the customer experience in order to create value for customers (Berry,
Carbone, & Haeckel, 2002). Recent literature defines customer experience as the
consumer’s ‘internal and subjective response’ to any direct / indirect contact with a
253
retail location (Meyer & Schwager, 2007, p. 118). This contact spans experiences in
retail stores and across all channels, and is ‘holistic in nature’ (Verhoef et al., 2009).
From a retail perspective, the customer experience can begin with the journey to the
store (Gilmore & Pine, 2002) and thus the shopping location becomes part of the
experience. Conceptualised in a retail or services setting, the customer experience
entails a personal involvement at different levels, identified as cognitive / physical,
sensory / affective, and social (Gentile, Spiller & Noci, 2007; Schmitt, 1999; Verhoef
et al., 2009). The four levels of customer experience potentially overlap with notions
of image in interesting ways and suggest a theoretical framework within which the
town centre image concept may be understood.

First, cognitive and physical evaluations feature prominently in retail image

studies and are strongly associated with patronage (Darden & Babin, 1994; Gauri,
Trivedi & Grewal, 2008). Consumers experience the physical environment through
cognitive processing of information derived from its tangible characteristics, or
stimuli that activate knowledge structures (MacInnis & Price, 1987; Stern et al.,
2001). Given that image ‘simplifies complex information’ (Mitchell, 2001, p. 167),
these knowledge structures contain elements in the physical environment, such as
the range of stores and goods, accessibility and parking, entertainment and
refreshments, street layouts and architecture, which may be salient to consumers.
Where the consumer perceives these elements more favourably, it is likely to lead to
greater patronage (Puccinelli et al., 2009).

Second, at a sensory and affective level, consumers experience the retail

environment in ways that involve their senses and emotions (Darden & Babin, 1994;
Oxenfeldt, 1974; Schmitt, 1999). The influence of atmosphere on consumers’ image
is evident in store and mall contexts, and research suggests that the interaction of
various atmospheric stimuli will strongly influence patronage behaviour (Baker,
Parasuraman, Grewal & Voss, 2002; Michon et al., 2005; Wakefield & Baker, 1998).
Darden and Babin (1994) show that retail environments evoke emotional responses
such as pleasure, energy and sleepiness, which in turn, influence patronage in
complex ways (Laroche, Teng, Michon, & Chebat, 2005; Machleit & Eroglu, 2000).
In addition, consumers seek entertainment and excitement in a shopping experience
(Sit, Merrilees & Birch, 2003; Wakefield & Baker, 1998). The numbers and density of

254
other shoppers also contribute to feelings of vibrancy and vitality in a town centre
(Oppewal & Timmermans, 1999; Ravenscroft, 2000). Affect influences patronage
through mood, since consumers shop for longer and spend more when they are
feeling positive; through evaluation, in that a more positive emotion results in a more
positive evaluation of a retail experience and are more likely to return; and post-
purchase satisfaction so they are more likely to recommend (Puccinelli et al., 2009).
Hence, consumers with a more positive affective image are likely to demonstrate
increased patronage.

Finally, the social environment and the consumer’s perceived relationship with
the place and other people influence the image formed. Brand image theory
suggests that image has a social aspect which associates a consumer with a
desired group, role or self-image (Park, Jaworski and MacInnis, 1986). Consumer
attitudes and behaviour are often determined by perceptions of typical shoppers, as
compared to consumers’ own personal images (Sirgy et al., 1997; Verhoef et al.,
2009). A consumer who perceives his/her self-image as congruent with that of other
shoppers is believed to view a shopping area more favourably (Sirgy et al., 1997;
Sirgy & Samli, 1985), leading to greater satisfaction with the customer experience
and hence increased patronage (Verhoef et al., 2009). Consumers who enjoy the
social element of the customer experience are likely to shop more frequently in that
location and therefore develop greater familiarity with it (Haytko & Baker, 2004;
Miller & Kean, 1997). Familiarity in part stems from feeling comfortable in the
environment due to more frequent interaction, increased trust and identification with
other people who are perceived as similar (Haytko & Baker, 2004; Line, Runyan,
Costen, Frash & Antun, 2012). In a consumer’s home town location, familiarity may
result in a sense of attachment to the community through social identity and social
capital (Miller, 2001). The social nature of the customer experience leads to the
likelihood the group members will communicate their perceptions of the town centre
to others (Granovetter, 1985; Line et al. 2012). If these are favourable, such positive
word of mouth recommendations will enhance the reputation of the location (Brown,
Barry, Dacin & Gunst, 2005; Dick & Basu, 1994).

Figure 14 synthesises the customer experience literature in a generic model of

retail customer experience. Accordingly, the emerging customer experience

255
literature provides a framework within which to contextualise dimensions of town
centre image. Indeed, the levels of customer experience (cognitive / physical,
sensory / affective and social) provide a starting point for developing a holistic
conceptualisation of image that may play a role in shaping patronage behaviour.

Figure 14 Generic model of retail customer experience

Adapted from Grewal et al., (2009) and Verhoef et al., (2009)

With this in mind, the main objective of this study is to investigate and provide
an answer to the question of ‘what is town centre image’. Specifically, we first
examine the composition of town centre image by eliciting consumer generated
attributes and dimensions, which will facilitate the development of a conceptual
model. Second, an associated objective for this research is to establish which image
attributes are relevant to town centres and which may be redundant to this context.
Third, recognising the importance of managing the customer experience in the retail
environment, we conceptualise town centre image using a framework based on the
customer experience.

Method
To capture consumers’ perceptions of town centre image, a content analysis
methodology was adopted. The methodology was adapted from Zimmer and Golden
(1988) who used unstructured methods to elicit consumers’ unprompted image
responses, avoiding any specification bias related to researcher-imposed
questioning (Zimmer & Golden, 1988). Unstructured methods are more flexible in

256
generating detailed responses, including new, unexpected and ‘non-fitting’
responses, which are potentially the most interesting in extending the
conceptualisation of image (Keaveney & Hunt, 1992). Further, these methods
enable respondents to give ‘aggregate’ or holistic responses, which are better suited
to interpreting the meaning consumers attach to the complexity of an environment
(Stern et al., 2001, p. 220). Hence, this method is particularly conducive to
identifying the holistic content of town centre image.

Sample and data collection

An intercept survey of 536 consumer respondents was conducted in seventeen

towns and cities across the East Midlands region in the UK representing 1.4 million
inhabitants. The overall sample was approximately 61% female and 39% male,
aged between 16 and 66. Females between 16 and 25 and 36 and 45 years of age
made up the largest proportion of the total sample (21.1% and 21.5% respectively).
Most males were between 26 and 35 and 46 and 55 (14.7% and 16.4% of the total
sample respectively). A total of 36% of respondents had children, and 62.3% were
employed. The sample was predominantly Caucasian (89.7%).

To avoid bias towards specific attributes, we asked consumers ‘please give

your overall impression of the following, as places for shopping’ (adapted from
Zimmer & Golden, 1988). The question aimed to capture consumers ‘top of mind’
spontaneous descriptors, which are believed to contain the most salient image
perceptions (Woodside & Trappey, 1996) and was placed immediately at the start of
a separate survey. Five cities of varying size in the region provided cues for the
responses, with the majority of respondents believed to have awareness of two or
more of these. Each respondent could give as much or as little information as
required. A pretest of 10 consumers in one town centre confirmed the clarity of the
wording for all respondents with no ambiguity reported. Trained Market Research
Society (MRS) interviewers administered the survey as part of a commissioned
study into retailing in the region, and intercepted shoppers during normal shopping
hours in central locations.

257
Content analysis

The responses thus generated were analyzed using content analysis to establish
the frequencies with which respondents’ statements could be located within
designated categories, and to assess their relative importance. All responses were
aggregated into a single dataset, since it was not the purpose of the research to
determine the image of each individual city, rather to identify the types of statements
consumers use when perceiving town centre image. Responses were listed as
individual words or phrases containing a single idea, generating 2013 image
statements in total.

The purpose of content analysis is to reduce a large number of words or

phrases to a smaller number of meaningful categories (Weber, 1990). Accordingly,
one researcher initially reduced the total number of responses (2013) by grouping
similar or identical responses (such as ‘choice’ and ‘selection’, ‘hard’ and ‘difficult’)
into themes each containing one individual idea. This process generated 432 distinct
themes relating to consumers’ perceptions of town centre image. To minimise
researcher bias in the analysis and to ensure reliability in the results, independent
coders were trained and an iterative process of analysis and replication was carried
out (Krippendorff, 2004; Weber, 1990). A coding template was prepared from the list
of 29 attributes in Table 16 identified in the town centre literature. Then six coders,
who were research assistants or academics in areas outside of retailing, were
instructed to independently code as many of the 432 themes as possible into the 29
categories in the template. Although individual totals varied, coders allocated
approximately 70 per cent of the 432 themes (representing an average of 1250
statements each) to a category. The remaining statements (representing around
760 statements or 144 themes) were not considered to ‘fit’ any of the categories
currently existing in the literature.

The coders were informed that the ‘non-fitting’ responses were considered
equally of interest since they might provide a more complete explanation of the
image construct (Keaveney & Hunt, 1992, p. 167). The ‘non-fitting’ themes were
discussed with the coders who were asked to suggest additional categories, into
which they believed these non-fitting themes might fit. A number of common
categories emerged which were discussed by the authors and the coders until
258
agreement was reached on their exact labeling. As a result eight additional
categories were created which were labelled as: ‘named shopping locations’,
‘named stores’, ‘character and tradition’, ‘familiarity’, ‘boring / exciting’, ‘comparisons
with other places’, ‘likes and dislikes’, and ‘positive and negative overall comments’.
Furthermore, due to some reported confusion among coders about the ‘non-retailing
functions’ category in Table 16, responses to this category were grouped into two
separate new categories: ‘leisure / services’, and ‘entertainment venues’. In all, ten
new categories were created to capture the ‘non-fitting’ responses which were
added to the template, resulting in 38 categories (see Table 18, column 4 for the
final list of 38 categories). Statement frequencies were totalled for each category.

A second phase of content analysis then followed. Two research assistants,

working independently, re-entered the 432 themes into the expanded template. The
results from these two coders were tested for inter-rater reliability. The most
commonly used measure of inter-rater reliability is the percentage of coder
agreements to the total number of coding decisions (Kassarjian, 1977). However,
this measure may include agreements, which are reached by error or chance;
hence, agreement was assessed as a function of an assumed true level of reliability
(Bitner, Booms, & Mohr, 1994; Perreault & Leigh, 1989; Rust & Cooil, 1994).
Perreault and Leigh (1989)’s formula for inter-rater reliability (I) is represented by the
equation:
I = {[F/N – (1/K)][K/(K – 1)]}0.5

where F is the observed number of agreements between coders, N is the total

number of judgments made, and K is the number of categories into which items are
placed (the values of I falling between 0 and 1 with higher values representing
greater reliability).

Initial results achieved agreement of 0.847 between the two coders.

Discrepancies were discussed between the coders (Bitner, Booms, & Tetreault,
1990; Boivin, 1986) and the discrepant themes were re-assigned into the coding
scheme. This time the level of agreement reached 0.976, with 21 themes still unable
to be agreed. This level of agreement was considered sufficiently high (I > 0.9)
(Perreault & Leigh, 1989), however since all themes were of interest the principal

259
author independently selected the most appropriate category for these themes
(Bitner et al., 1994). In this way, all 432 themes were allocated to categories. The
resulting categorisation revealed that some categories contained large numbers of
responses, whereas others characterised by lower frequencies; two categories
obtained from the literature contained no items at all.

The authors then independently grouped the final 38 categories into a set of
dimensions. All categories were included, regardless of frequency of mention. Using
an iterative process of analysis and discussion, agreement was reached so that
eight dimensions captured all 38 categories. These were agreed as retail
assortment, global impression, atmosphere, accessibility, layout / scale, visual
appearance, non-retail activities and place attachment (see Table 18, column 1).
Next, a substantive validity test was carried out to determine the extent to which
each category is deemed to fit with its expected dimension (Anderson & Gerbing
1991; Holden & Jackson, 1979). The test is particularly advantageous in determining
validity in small samples such as in this study (Anderson & Gerbing, 1991). The test
was conducted by thirteen graduates in areas unconnected to retail, who were
provided with definitions of the 38 categories and the eight dimensions, with
instructions to enter each of the categories into one of the dimensions. The resulting
information was measured using the substantive validity coefficient (Csv), where
Csv = (nc – n0) / N, and nc = the number of coders assigning a category into the
expected dimension; n0 = the highest number of times that the category was
assigned to a dimension other than the expected one; and N = the total number of
coders. The values for Csv range from -1 to 1, where a higher value represents
greater validity and a negative value indicates the category is deemed to belong to
another dimension.

The Csv scores were calculated for each of the 38 categories, and then
aggregated for the eight dimensions. Two categories (‘advertising / promotions’,
‘shopping hours’) totaled zero statements during the allocation process and were
eliminated. Two other categories (‘service quality’ and ‘place to spend time’)
received negative scores without any clear alternative dimension indicated. The
three service quality statements related to stores rather than the centre as a whole.
The literature suggests that ‘a place to spend time’ is a more nebulous construct. As

260
a result, the latter two categories were eliminated. Following their omission, the Csv
scores were re-calculated. Scores across all dimensions were > 0.5 giving
confidence that the eight dimensions reliably represent the construct of interest,
namely town centre image.

Results
In response to the broad research question of ‘what is town centre image?’ the
content analysis results suggest that town centre image can be represented by
seven dimensions, and one dimension, which appears to operate at a higher level,
representing global impressions. Importantly, three of these dimensions are new to
the literature, namely global impressions, non-retail activities and place attachment.
Additionally, the three dimensions of retail assortment, atmosphere and appearance,
while recognised in the literature, exhibit different constituent elements for this
image context. These will be described in turn below.

Referring to Table 18, the most prominent dimension is assortment, based on

the number of respondent statements. However, the characteristics that describe
the assortment dimension in the context of town centre image are qualitatively
different to the literature. The ‘range and quality of shops’ is the dominant category
(n=421) compared with the ‘range and quality of goods’ (n=56) which feature
strongly in the town centre image literature. Interestingly, a new category of ‘named
shopping locations’ emerged, referring to specific shopping streets, in-town
shopping malls, markets or stores in the town centre. A further category ‘named
stores’, also resulted from the content analysis whereby respondents named specific
anchor department stores and individual retailer brands in relation to that town
centre.

Although often included in town centre image studies, comments relating to

‘price / value’ and ‘style / fashion’ were less numerous. These categories may not
refer to the town centre as a whole, but rather to individual stores. Similarly, the
categories of ‘advertising / promotions’, ‘shopping hours’ and ‘service quality’,
regularly included in town centre image studies received very few mentions.
Advertising may be perceived as focused on store promotions, while efforts to
promote the town centre as a whole remain unnoticed by shoppers. Shopping hours,
261
while a previous town centre attribute, are perhaps assumed by consumers,
whereas mall opening hours may vary, being manipulated by management to suit
consumers and location. Thus, these categories appear more applicable to the store
image or mall image than town centres. The category of ‘refreshments, retail related
facilities’, scored a low but positive Csv score which just secured it within the
assortment dimension. Therefore, in town centre image, assortment encompasses
the entire range and quality of retail provision such as stores (multiples, independent
traders), specific groups of retailers (in streets, shopping malls and markets),
together with the goods they sell, and including services and facilities which support
the shopping process.

262
 Table 18 Dimensions of town centre image and experience following Csv
tests
csv score
Total no of dimension statement statements per
Dimension statements % of total Category frequency % of total category
Retail assortment 654 32.49 Range / quality of shops 421 20.91 1.00
Named shopping locations (malls, streets, markets) 89 4.42 0.54
Range / quality of goods 56 2.78 1.00
Price / value 28 1.39 0.38
Refreshments / retail-related facilities (cafes, seats) 22 1.09 0.08
Style / fashion 18 0.89 0.38
Named stores 17 0.84 1.00
Service quality* 3 0.15 -0.15
Advertising / promotions* 0 0.00 0.23
Shopping hours* 0 0.00 0.15
Atmosphere 181 8.99 Busy 78 3.87 0.62
Crowding 32 1.59 0.77
Boring / exciting 30 1.49 0.46
Safety 12 0.60 0.15
Atmosphere 11 0.55 1.00
Personal / friendly 10 0.50 0.31
Senses (temperature, noise) 8 0.40 1.00
Accessibility 179 8.89 Location / travel (convenience, traffic) 144 7.15 1.00
Parking 35 1.74 0.85
Layout/scale 158 7.85 Size 60 2.98 0.69
Structure and design / layout (under cover, pedestrianised) 54 2.68 0.77
Ease of internal movement (compact, easy to get round) 44 2.19 0.54
Appearance 135 6.71 Visual appearance 50 2.48 0.77
Cleanliness 33 1.64 0.54
Character / tradition (historic, olde worlde) 26 1.29 0.08
Old / new (modern, deteriorated) 18 0.89 0.46
Buildings / landscaping 6 0.30 1.00
Lighting 2 0.10 0.77
Non-retail activities 117 5.81 Leisure / services (administrative, tourist attractions, sports) 54 2.68 1.00
Place to spend time/ "day out"* 23 1.14 -0.31
Activities/ special events (Xmas, Ramadan) 20 0.99 0.38
Entertainment venues (bars, clubs, theatre, night life) 20 0.99 0.77
Place attachment 106 5.27 Familiarity (my home town, always shop here) 66 3.28 0.85
Other people 34 1.69 0.62
Reputation 6 0.30 0.23
Global impressions 483 23.99 Descriptors (good, brilliant, appalling) 339 16.84 0.85
Likes / dislikes (I like it, my favourite, disappointing) 100 4.97 0.46
Comparisons (not as good as ..., better than...) 44 2.19 0.62
2013
* problematic categories omitted from final calculations

263
 Atmosphere consistently features in the retail marketing literature (Kotler,
1973) and the results of the content analysis confirm this as a core dimension of
town centre image. The most significant categories within this dimension were the
linked themes of ‘busy’ and ‘crowding’, frequently in terms of ‘too busy’ or
‘overcrowded’. The remaining categories representing atmospheric conditions
include lighting, temperature and noise. The content analysis produced a further
category labelled ‘boring / exciting’, expressed in terms of ‘fun’, interesting’ or
‘nothing happening’, not previously included in town centre studies but evident in the
mall image literature (Sit et al. 2003; Wakefield & Baker 1998). In addition, coders
allocated ‘personal / friendly’ within this dimension, reinforcing the literature; for
example Nevin and Houston (1980) considered atmosphere as ‘friendly’, and R. M.
Downs (1970) included items based on a ‘busy, relaxed and personal atmosphere’.
In sum, atmosphere in a town centre represents how consumers experience and
interpret the sensory and affective aspects of the environment.

Representing aspects of travel and access to the town centre, the accessibility
dimension confirms that consumers include distance and travel costs in their town
centre image perceptions; although distance travelled appeared less relevant than
factors inhibiting access to the shopping area (e.g. traffic congestion and parking).
This dimension features in the town centre image literature (R. M. Downs, 1970;
Gautschi, 1981; Timmermans, van der Heijden and Westerveldt, 1982) and supports
the distance variable in the attractiveness and patronage literature (Bell, 1999; Leo
& Phillipe, 2002). The provision of ‘easy’, ‘plenty’ and ‘inexpensive parking’ was
mentioned by respondents and is regarded an important factor in town centre
patronage (Bell, 1999; Wee, 1986), considering that out of town malls provide easy
access and free parking.

A separate dimension of layout / scale emerged to distinguish internal mobility

from external access to the town centre. Some image studies combine accessibility
with internal movement within the centre in one dimension (R. M. R. M. Downs,
1970; Hackett & Foxall, 1994; Nevin & Houston, 1980). However, the responses,
coders and Csv scores all suggest that consumers evaluate external accessibility as
a separate variable to internal mobility. Layout refers to the ease of physical
movement by consumers within the town centre. Linked to this are the scale, or

264
perceived size and spread of the centre; whether it is compact and easy to negotiate
or spread out. Respondents also referred to the structure and design of the town
centre as encompassing enclosed, indoor or open aspects, and pedestrianised
areas.

The visual appearance of centres features to a lesser extent in the town centre
literature. While the appearance of buildings is more commonly included in town
centre measures, here this dimension includes cleanliness, lighting and age, rather
than the architecture of individual stores. Notably, a new category of ‘character /
tradition’ indicates that consumers respond to aesthetic appeals based on history
and tradition in the town centre. This finding reinforces the notion that historical
architecture, although not always visible at street level, nevertheless plays an
important role in distinguishing and differentiating town centres to consumers
(Sharp, 1968; Warnaby, 2009).

Critically, and of great interest, are the final three dimensions representing
town centre image which emerge from the data: ‘non-retail activities’, ‘place
attachment’ and ‘global impressions’. These dimensions do not feature in town
centre image studies to date. The dimension labelled ‘non-retail activities’ was
allocated to a separate dimension but is complementary to retail provision.
Respondents consider these activities an important part of town centre image,
referring to entertainment venues such as bars, clubs, theatre and nightlife, together
with leisure and sports venues, visitor attractions, administrative functions and
workplaces. These reflect the wider purpose of the centre beyond retailing. The
category ‘a place to spend time’ has been included in the literature (Hackett &
Foxall, 1994; Leo & Philippe, 2002; Nevin & Houston, 1980) and is linked to this
dimension as providing wider entertainment for shoppers. However, this notion
proved less clear in the analysis, and is thus omitted from the final image
dimensions.

The smallest dimension - ‘place attachment’ - captured the sense of personal

attachment or belonging to the place where consumers shop frequently. This
dimension contained comments involving familiarity and frequency of patronage, for
example ‘I always shop here’, ‘it feels comfortable’ and ‘my home town’. When a

265
town centre was less familiar to respondents, they referred to the town or city’s
reputation, demonstrating knowledge gained through secondary sources rather than
first hand. Perceptions of other shoppers also contributed to a sense of attachment
to the town centre.

The most significant new dimension emerged from respondents’ global

impressions of the town centre and comprised nearly one quarter of all respondent
statements. Within this, the largest category comprised simple descriptors
describing the town centre positively or negatively, for example ‘good’, ‘bad’,
‘gorgeous’, or more expressive comments such as ‘excellent’, or ‘appalling’
summarising their perceptions of the town centre. A further category listed
subjective attitudes relating to likes or dislikes; for example ‘enjoyable’,
‘disappointing’ and ‘my favourite’. The third category included comparisons
between the town centre and other known locations, for example ‘looks poorer
than X’, ‘better than my home town’ and ‘needs to catch up with other towns and
cities’. Comments such as ‘same as all towns’ and ‘best shopping centre’ also
suggest that consumers form prototypes or abstract typical town centre examples.
These holistic responses echo the overall impressions identified by Zimmer and
Golden (1988)’s image study findings, and were allocated to a new dimension
created as a result of this analysis labelled ‘global impressions’.

Discussion
The objectives of this study were to identify what is town centre image, and to
develop a conceptual model, which explores town centre image in the wider context
of the customer experience. Building on the generic model of retail customer
experience introduced in Figure 14, the results are discussed in relation to these
objectives in the following section, and describe the conceptual model presented in
Figure 15.

The findings imply a town centre image construct that is ‘a composite of

beliefs held by a consumer on a number of dimensions’ (Houston & Nevin, 1981,
p. 678), in which certain dimensions are functional and tangible, and others
psychological and intangible (Lindquist, 1974; Martineau, 1958; Mazursky &
Jacoby, 1986), with a further dimension operating at a higher level of abstraction.
266
Five dimensions refer to functional elements, representing the features necessary
for the town centre to operate as a retail centre. These include retail assortment,
accessibility, layout / scale, appearance and non-retail activities. Two dimensions
are associated with psychological, intangible elements (Martineau, 1958), namely
atmosphere and place attachment. Furthermore, the town centre image structure
suggests a ‘multidimensional gestalt’ composed of global impressions (Stern et al.,
2001, p. 218; Zimmer & Golden, 1988).

When compared with the literature it is the exceptions and omissions within the
dimension findings that deliver new insight into the consumers’ image of a town
centre. Only two functional dimensions (accessibility and layout / scale) evidence
similar composition to that previously included in the literature, whereas three
previously included dimensions (retail assortment, atmosphere, appearance)
demonstrate a different composition within each dimension. Certain categories
(service quality, advertising / promotions and shopping hours) appear less relevant
to the town centre context and consumers’ aggregation of town centre image.
Importantly, the research findings reveal two new dimensions (non-retail activities,
place attachment), a higher-level global impressions dimension and nine additional
categories previously absent from town centre image studies (named shopping
locations, named stores, descriptors, likes/dislikes, comparisons, boring/exciting,
character/tradition, entertainment venues, familiarity).

A conceptual model of town centre image and experience

In the absence of a previously-developed and validated conceptual framework in

the retail image literature, the Customer Experience Framework proposed by
Verhoef et al. (2009) offers a preliminary framework for explaining the dimensions
and constituents identified in Table 18, which combine into a contextualised model
of town centre image in Figure 15. Figure 15 proposes that consumers’ images of
town centres are formed following similar dimensions to the framework suggested
by the customer experience. Thus, as the consumer experiences the centre, the
cognitive / physical, sensory / affective and social levels integrate in the creation of
a town centre image. In addition, image becomes part of a process whereby
consumers in some cases combine information regarding the town centre into an
image gestalt or global impression (MacInnis & Price, 1987; Stern et al. 2001). The
267
dimensions identified will influence consumers’ town centre image formation in the
following distinct ways (Finn & Louviere, 1996; Pan & Zinkhan, 2006).

Figure 15 Contextualised model of town centre image and experience

# New dimensions
* Containing new characteristics

Cognitive / physical

This element contains five functional dimensions comprising assortment,

accessibility, layout / scale, appearance and non-retail. The strength of the retail
assortment, combined with easy access, internal layout, and perceived size / scale,
suggest that consumers cognitively evaluate the town centre according to whether it
functions as a retail centre, fulfilling the utility of a shopping trip. The strong
emphasis on the store assortment suggests that consumers primarily evaluate the
overall range and quality of shops before individual goods in a town centre.

Furthermore, by naming specific shopping locations (streets or in-town malls),

consumers interpret town centres as not one entity of retail stores but different
268
groups of stores that collectively form an overall image, supporting Houston and
Nevin’s (1981) conclusion that a downtown area is a conglomerate of separate
shopping streets. Moreover, naming specific retail brands in the town centre,
suggests that other images, such as anchor stores, influence the overall town centre
image (Chebat, Sirgy, & St-James, 2006; Finn & Louviere, 1996).

Consumers’ image of town centres is shaped by the physical appearance

represented by the public space in terms of buildings and landmarks, community
artworks and historical associations (Gentile et al., 2007; Runyan & Huddleston,
2006). While the assortment dimension represents increasingly homogenised
stores and high streets (Dawson, 2002), character or tradition potentially
distinguish one town centre from another. Consumers recognise and value visual
attributes, particularly whether the centre is well maintained, clean, modern or
deteriorated. This dimension holds particular relevance given the negative impact
of store closures on the visual landscape. Notably, the non-retail activities, while
outside of retailer control, nevertheless influence the consumer’s image and
experience of that location. Compared to mall research, which includes leisure,
services and entertainment venues, this dimension receives little recognition in
town centre studies to date (Sit et al., 2003; Wakefield & Baker, 1998; Warnaby et
al., 2002).

Sensory / affective

The experiential aspects of image suggest that consumers seek shopping

experiences that provide sensory pleasure and satisfy emotional needs (Holbrook &
Hirschman, 1982; Wakefield & Baker, 1998). The atmosphere or ambience
perceived in the town centre generates emotional affective responses (Turley &
Milliman, 2000), which ‘address sight, touch, taste, and smell so as to arouse
aesthetical pleasure, excitement, satisfaction, sense of beauty’ (Gentile et al., 2007,
p. 398). The new category labelled boring / exciting, demonstrates that town centres
can also provide excitement, variety and stimulation (Park et al., 1986; Wakefield &
Baker, 1998). The linked themes of ‘busy’ and ‘crowding’ indicate that some
consumers have negative reactions to a vibrant atmosphere and avoid certain town
centres. Out of town malls increasingly offer ‘an entire customer experience’ to
attract and keep customers (Haytko & Baker, 2004, p. 68) whereas a town centre
269
experience may be fragmented across a wider, physical location, which may be
harder to manage and control. It is important therefore that centre managers
understand the consumers’ interpretation of atmosphere and its impact for a specific
town centre in order to encourage patronage

Social

The consumer’s relationship and familiarity with the centre should also contribute
to the image formed in a town centre (Haytko & Baker, 2004). In this study, the
social element reflects the sense of connectedness with, or attachment to, the
town centre, particularly for those respondents who referred to the town centre as
being in ‘my home town’ (Runyan & Huddleston, 2006). Store image studies refer
to a ‘sense of belonging’ (Lindquist, 1974, p. 30), or ‘deep attachment’ (Martineau,
1958, p. 54), while in a mall context, place attachment is defined as a ‘bond or link
between people and specific places’ (Allard, Babin, & Chebat, 2009, p. 42). The
respondent perceptions of other shoppers and how they ‘fit in’ imply a sense of
familiarity and trust, suggesting similar processes apply in a town centre. This is
believed to occur through homophily, which is the notion that consumers associate
with others whose images they perceive as similar to their own (Line et al., 2012;
Sirgy et al., 1997). Homophily and feeling comfortable with certain other groups is
linked to social identity (Tajfel, 1978), one of the sources of which is the
neighbourhood or community (Forrest & Kearns, 2001). It also contributes to social
capital through greater involvement with the community (Miller, 2001), which
Forrest and Kearns (2001) argue leads to place attachment.

The more consumers interact with a town centre, and if their attachment
becomes positive, the more likely they are to develop loyalty and recommend the
location to others, enhancing its reputation (Brown et al., 2005; Orth, Limon, & Rose,
2010). Therefore, the social component is also likely to influence patronage
behaviour through perceptions of other shoppers and place attachment. Yet retail
researchers have only recently considered the notion of place attachment, and
currently its role in relation to town centre image is poorly understood (Mullis & Kim,
2011; Sneed et al., 2011). This is surprising since place attachment may be

270
considered more important in town centres, as place and identity are believed inter-
related (Cresswell, 1996).

Global impressions

The global impressions dimension supports the notion that ‘non-fitting’ responses
are potentially most interesting to the holistic image concept (Keaveney & Hunt,
1992, p.167). Within this dimension, global positive / negative responses summarise
consumers’ subjective reactions to that location, these descriptors appearing to
simplify informational stimuli from the retail environment (Mitchell, 2001). The global
impressions dimension identified in this research concurs with Zimmer and Golden
(1988)’s suggestion that global image impressions are formed from ‘a consumer
abstraction process wherein many attributes are considered initially and then a
higher-level, affectively based category is formed’ (p. 273). When processing
informational stimuli, consumers unconsciously form judgments, developing
affective responses in terms of likes and dislikes or subjective descriptions (Stern et
al. 2001; Darden & Babin 1994; Cohen, 1982). Alternatively, consumers may form
images simply as an overall impression on the basis of ‘a couple of ideas’ which
most closely represent the location for them (Zimmer & Golden, 1988, p. 287),
consistent with automatic cognitive processing theory where spontaneous image
responses form with little conscious thought, and thus contain only a few items of
information (Fazio, 1989; Grunert, 1996; Louviere & Gaeth, 1987).

Categorisation and image of other locations

Originally included in the global impressions dimension was the use by consumers
of ‘labels, prototypes and exemplars’, to compare a town centre with other known
good examples of the category (Zimmer & Golden, 1988). Category-based
information processing integrates new information with existing information through
schemas held in the memory and is believed to assist consumers through reducing
effort in image formation (Keaveney & Hunt, 1992). Rather than viewing each
location through a system of separate attributes, consumers will initially avoid effort
through comparing it with other known locations (Keaveney & Hunt, 1992).
Following subsequent discussion, this category appeared to differ from the other two
categories in the global impressions dimension. Positive or negative evaluations and

271
descriptors perform an evaluative role, whereas the comparison category
demonstrates varying levels of awareness of other locations with which to compare
the location in question. Respondents could have varying levels of awareness of
other locations for example due to their backgrounds, mobility and past experiences,
and hence this category was deemed to have a moderating effect on the resulting
image construct rather than acting as a component of the image construct.

Consumer and situation influences

Considering the ‘multiple factors both within and outside the retailer’s control’ that
influence the image of the town centre, our model includes consumer and situation
influences (Verhoef et al., 2009, p. 33). The relationship between the individual
dimensions of town centre image and patronage is likely to be influenced by the
consumer’s purpose of visit, whether shopping or wider purposes. Shopping
motivations may include utilitarian or hedonic / experiential typologies (Holbrook &
Hirschman, 1982) where for example, ‘grazer’ shoppers (Bloch, Ridgway, &
Dawson, 1994) may consider the availability of cafes and restaurants as important
as retail assortment. Personal and socio demographic characteristics and
expectations also shape the image perceived (Haytko & Baker, 2004; Miller & Kean,
1997; Otnes & McGrath, 2001).

Given the nature of the town centre environment, situation factors such as the
type of town centre, proximity to competitive centres, the local infrastructure and
support available such as Business Improvement Districts, will affect maintenance
and the perceived visual appearance. Macro issues such as economic climate and
government policy will influence the viability and perceived vitality of the town centre
(Ravenscroft 2000).

The consumer’s previous and future town centre experiences will

accordingly reinforce or amend the likelihood that this image will influence
patronage over time (Berry, 1969). Further to Verhoef et al. (2009) we confirm
town centre image as involving the customer’s cognitive / physical, sensory /
affective and social image responses to the totality of the town centre experience.
In this respect, town centre experiences encompass every interaction with a town

272
or city centre offering from the marketing, access to, movement within the centre
through to the outcome of that visit.

Conclusions and implications

This study aimed to fill a literature gap by conceptualising town centre image. Our
findings suggest that consumers form images of town centres, which are distinct
from other retail locations such as stores and malls. Previous image
conceptualisations attract criticism for excluding the consumer and failing to capture
the image gestalt (Zimmer & Golden, 1988). By focusing on a customer experience
framework, we uncover a process of consumer image generation (Stern et al.,
2001), and identify a set of ‘dimensions’ that influence town centre image formation,
including new categories and dimensions not hitherto considered by the town centre
image literature.

Various studies hypothesise the holistic image construct (e.g. Mazursky &
Jacoby, 1986; Bell, 1999). With the exception of Zimmer and Golden (1988) in the
store image context, no research attempts to explain the content of the holistic
component of image (Stern et al., 2001). Indeed, few researchers attempt to
measure consumers’ holistic or global impression of image due to the complexity
involved in capturing and analysing ‘an overall picture that is more than the sum of
the parts’ (Marks, 1976, p. 37). By categorising all consumer responses, including
those usually discarded due to their ‘non-fitting’ nature (Chowdhury, Reardon and
Srivastava, 1998; Keaveney & Hunt, 1992) we demonstrate a first attempt to
conceptualise holistic image in a town centre context.

Similar to Zimmer and Golden (1988), we find that functional attribute specific
descriptors only represent part of the consumer’s image. It appears that consumers
may form images of retail locations employing either or both attribute or holistic
approaches and involving more psychological, intangible elements than previously
included. Moreover, town centre image comprises beliefs and attitudes evidenced in
respondents’ use of summary judgments or global impressions of complex retail
environments.

273
 The inability of researchers to agree a consistent set of salient dimensions
driving town centre image may be due to incomplete image measures adapted from
other retail environments. It is clear that further empirical work is warranted in terms
of operationalising image, and to recalibrate the internal composition of dimensions
for the town centre context. Given the strength of the global impression dimension,
one priority is to investigate methods to capture this image component.

Managing image and the customer experience in the town centre

In isolation, retail image has limited value, but gains strategic meaning when
related to patronage or loyalty (P. E. Downs & Haynes, 1984). We provide a
practical application of the image variable through its links with the customer
experience. Increasingly, consumers will expect businesses to use coordinated
strategies to support and add value to the customer experience. As Houston and
Nevin (1981) suggest ‘the image variable offers the most practical focus for
downtown revitalisation efforts’ (p. 677). In town centre terms, this will translate
into a coordinated approach to managing its image beyond simply marketing the
place as a brand.

To retain and increase consumer patronage, marketers will need to understand

the ‘touch points’ of key contacts with the town centre (Frow & Payne, 2007). If
customer experience is the sum of all experiences a customer has with a provider of
goods and services, then town centre management will actively need to manage
that experience, together with their retail partners and other stakeholders such as
local government and retail property developers. Identifying the unique combinations
of positive or negative image drivers by which consumers distinguish individual town
centres can then support proactive customer experience strategies. For example,
the complexity of town centre image suggests that consumers use other images as
‘anchors’ to both simplify and describe the town centre. Favourite brands,
departments, stores and in town shopping streets or malls influence the consumer’s
image of the town centre. These multiple retail images require careful management
to ensure congruity with each other and the town centre image (Ailawadi & Keller,
2004). If a retail store image improves by linking with a high brand image, and the
retail image suffers if linked with a low image brand (Jacoby & Mazursky, 1984),
then introducing incongruent retail brands may have a negative effect on the overall
274
town centre image. To manage the town centre experience effectively, town centre
management will need to identify which combination of brands achieve most
synergy for that location. The character and tradition of a town centre may be valued
more highly than concerns for central parking. Researching consumers’ ‘touch
points’ for that town centre could generate new opportunities for shoppers to
celebrate the local heritage of specific streets or buildings, for instance, by creating
a ‘heritage’ access route from outlying car parks with supporting signage, seating,
and historical information points. This could be reinforced with associated events
organised through local retailers and community groups. Monitoring the customer
experience and image will accordingly help create and communicate the town centre
as a ‘place brand’, enhancing the competitiveness, long-term patronage and viability
of the centre.

Future research needs to consider a number of issues. First, as noted above,

the model and dimensions we have identified require testing. Second, current town
centre image research assumes attribute based processing, ignoring prior town
centre experience and potential category based processing (Keaveney & Hunt,
1992). Our findings suggest that consumers make inferences about a town centre
by comparing it with their experience of similar town centres; it would be useful to
understand the circumstances under which this occurs. Third, the type of shopping
trip or shopping orientations (Darden & Reynolds, 1971) may impact the perceived
image, raising the question of the extent to which enjoyment, or recreational
shopping, influence the town centre experience. It is suggested that image is
determined by the shopping motivation of the consumer over time. Thus a
longitudinal study may deliver greater insights (Osman, 1993). Fourth, we have not
considered additional variables such as size and type of centre. How might the
dimension salience vary for smaller shopping areas such as market towns,
compared with larger city centres? Finally, it would be desirable to explore how
consumers categorise town centres to form a generic town centre schema. Longer
term, this knowledge may also inform the predominantly stakeholder view of town
centre attractiveness rankings.

275
References

Ailawadi, K.L. & Keller, K.L. (2004). Understanding retail branding: conceptual
insights and research priorities. Journal of Retailing, 80(4), 331-342.
Allard, T., Babin, B.J. & Chebat, J. (2009). When income matters: Customers
evaluation of shopping malls’ hedonic and utilitarian orientations. Journal of
Retailing and Consumer Services, 16(1), 40-49.
Anderson, J.C. & Gerbing, D.W. (1991). Predicting the performance of measures
in a confirmatory factor analysis with a pretest assessment of their substantive
validities. Journal of Applied Psychology, 76(5), 732-740.
Baker, J., Grewal, D., & Parasuraman, A. (1994). The influence of store
environment on quality inferences and store image. Journal of the Academy of
Marketing Science, 22(4), 328-339.
Baker, J., Parasuraman, A., Grewal, D., & Voss, G.B. (2002). The influence of
multiple store environment cues on perceived merchandise value and
patronage intentions. Journal of Marketing, 66(2), 120-141.
Bell, S.J. (1999). Image and consumer attraction to intraurban retail areas: An
environmental psychology approach. Journal of Retailing and Consumer
Services, 6(2), 67-78.
Berry, L. L., Carbone, L. P., & Haeckel, S. H. (2002). Managing the total customer
experience. MIT Sloan Management Review, 43(3), 85-89.
Berry, L.L. (1969). The components of department store image: a theoretical and
empirical analysis. Journal of Retailing, 45(1), 3-20.
Bitner, M. J., Booms, B. H., & Tetreault, M. S. (1990). The service encounter:
Diagnosing favorable and unfavorable incidents. Journal of Marketing, 54(1),
71-84.
Bitner, M.J., Booms, B.H., & Mohr, L.A. (1994). Critical service encounters: the
employee's viewpoint. Journal of Marketing, 58(4), 95-106.
Bloch, P. H., Ridgway, N. M., & Dawson, S. A. (1994). The shopping mall as
consumer habitat. Journal of Retailing, 70(1), 23-42.
Boivin, Y. (1986). A free response approach to the measurement of brand
perceptions. International Journal of Research in Marketing, 3(1), 11-18.
Brown, T. J., Barry, T. E., Dacin, P. A., & Gunst, R. F. (2005). Spreading the word:
Investigating antecedents of consumers’ positive word-of-mouth intentions

276
 and behaviors in a retailing context. Journal of the Academy of Marketing
Science, 33(2), 123-138.
Bucklin, L.P. (1967). The concept of mass in intra-urban shopping. Journal of
Marketing, 31(4), 37-42.
Chebat, J., Sirgy, M.J., & St-James, V. (2006). Upscale image transfer from malls
to stores: A self-image congruence explanation. Journal of Business
Research, 59(12), 1288-1296.
Chowdhury, J., Reardon, J., & Srivastava, R. (1998). Alternative modes of
measuring store image: An Empirical assessment of structured versus
unstructured measures. Journal of Marketing Theory and Practice, 6(2), 72–
86.
Christaller, W. (1933). Central places in Southern Germany. Englewood Cliffs, NJ:
Prentice Hall.
Cohen, J.B. (1982). The role of affect in categorization: Toward a reconsideration
of the concept of attitude. Advances in Consumer Research, 9(1), 94-100.
Cresswell, T. (1996). In place/out of place. Minnesota: University of Minnesota
Press.
Darden, W.R. & Reynolds, F.D. (1971). Shopping orientations and shopping usage
rates. Journal of Marketing Research, 8(4), 505-508.
Darden, W.R., & Babin, B.J. (1994). Exploring the concept of affective quality:
expanding the concept of retail personality. Journal of Business Research,
29(2), 101-109.
Dawson, J.A. (2002). The changing High Street. In A. M. Findlay & L. Sparks
(Eds.), Retailing: Critical Concepts (375-391). London: Routledge.
Dichter, E. (1985). What’s in an image? Journal of Consumer Marketing, 2(1), 75-
81.
Dick, A. S., & Basu, K. (1994). Customer loyalty: Toward an integrated conceptual
framework. Journal of the Academy of Marketing Science, 22(2), 99-113.
Downs, P.E. & Haynes, J.B. (1984). Examining retail image before and after a
repositioning strategy. Journal of the Academy of Marketing Science, 12(4), 1-
24.
Downs, R.M. (1970). The cognitive structure of an urban shopping center.
Environment and Behavior, 2(1), 13-39.

277
El Hedhli, K., & Chebat, J. (2009). Developing and validating a psychometric
shopper-based mall equity measure. Journal of Business Research, 62(6),
581-587.
Fazio, R.H. (1989). On the power and functionality of attitudes: The role of attitude
accessibility. In A.R. Pratkanis, S.J. Breckler & A.G. Greenwald (Eds.),
Attitude Structure and Function (153-179), Hillsdale, NJ: Erlbaum.
Feinberg, R.A., Sheffler, B., Meoli, J., & Rummel, A. (1989). There's something
social happening at the mall. Journal of Business and Psychology, 4(1), 49-
63.
Finn, A., & Louviere, J.J. (1996). Shopping center image, consideration, and
choice: Anchor store contribution. Journal of Business Research, 35(3), 241-
251.
Forrest, R., & Kearns, A. (2001). Social cohesion, social capital and the
neighbourhood. Urban Studies, 38(12), 2125-2143.
Frow, P. & Payne, S.A. (2007). Towards the perfect customer experience, Journal
of Brand Management, 15 (2), 89-101.
Gauri, D.K., Trivedi, M., & Grewal, D. (2008). Understanding the determinants of
retail strategy: an empirical analysis. Journal of Retailing, 84(3), 256-267.
Gautschi, D.A. (1981). Specification of patronage models for retail center choice,
Journal of Marketing Research, 18(2), 162-174.
Gentile, C., Spiller, N., & Noci, G. (2007). How to sustain the customer experience:
an overview of experience components that co-create value with the
customer. European Management Journal, 25(5), 395-410.
Gentry, J.W., & Burns, A.C. (1977). How ‘important’ are evaluative criteria in
shopping center patronage? Journal of Retailing, 53(4), 73-86.
Gilmore, J. H., & Pine II, B. J. (2002). Customer experience places: The new
offering frontier. Strategy & Leadership, 30(4), 4-11.
Granovetter, M. (1985). Economic action and social structure: The problem of
embeddedness. American Journal of Sociology, 91(3), 481-510.
Grewal, D., Levy, M., & Kumar, V. (2009). Customer experience management in
retailing: An organizing framework. Journal of Retailing, 85(1), 1-14.
Grunert, K.G. (1996). Automatic and strategic processes in advertising effects.
Journal of Marketing, 60(4), 88-101.

278
Guy, C. (1994). The Retail Development Process: Location, Property, and
Planning. London: Routledge.
Hackett, P.M.W., & Foxall, G.R. (1994). A factor analytic study of consumers'
location specific values: a traditional high street and a modern shopping mall.
Journal of Marketing Management, 10(1-3), 163-178.
Hansen, R.A., & Deutscher, T. (1977). An empirical investigation of attribute
importance in retail store selection. Journal of Retailing, 53(4), 59-72.
Hart, C., Farrell, A.M., Stachow, G., Reed, G., & Cadogan, J.W. (2007). Enjoyment
of the shopping experience: Impact on customers' repatronage intentions and
gender influence. The Service Industries Journal, 27(5), 583-604.
Hartman, K. B., & Spiro, R. L. (2005). Recapturing store image in customer-based
store equity: A construct conceptualization. Journal of Business Research,
58(8), 1112-1120.
Hauser, J.R., & Koppelman, F.S. (1979), Alternative perceptual mapping
techniques: Relative accuracy and usefulness. Journal of Marketing
Research, 16(4), 495-506.
Haytko, D. L., & Baker, J. (2004). It’s all at the mall: Exploring adolescent girls’
experiences. Journal of Retailing, 80(1), 67-83.
Holbrook, M. B., & Hirschman, E. C. (1982). The experiential aspects of
consumption: Consumer fantasies, feelings, and fun. Journal of Consumer
Research, 9(2), 132-140.
Holden, R.R., & Jackson, D.N. (1979). Item subtlety and face validity in personality
assessment. Journal of Consulting and Clinical Psychology, 47(3), 459-468.
Houston, M., & Nevin, J. (1980). Retail shopping area image: Structure and
congruency between downtown areas and shopping centres. Advances in
Consumer Research, 18, 677-681.
Howell, R., & Rogers, J. (1980). Research into Shopping Mall Choice Behavior.
Advances in Consumer Research, 8, 671-681.
Huff, D.L. (1963). A probabilistic analysis of shopping center trade areas. Land
Economics, 39(1), 81-90.
Hunter, G.L. (2006). The role of anticipated emotion, desire, and intention in the
relationship between image and shopping center visits. International Journal
of Retail and Distribution Management, 34(1), 709-721.

279
Jacoby, J., & Mazursky, D. (1984). Linking brand and retailer images: Do the
potential risks outweigh the potential benefits? Journal of Retailing, 60(2),
105-122.
Kassarjian, H.H. (1977). Content analysis in consumer research. Journal of
Consumer Research, 4(1), 8-18.
Keaveney, S.M., & Hunt, K.A. (1992). Conceptualization and operationalization of
retail store image: A case of rival middle-level theories. Journal of the
Academy of Marketing Science, 20(2), 165-175.
Kotler, P. (1973). Atmospherics as a marketing tool. Journal of Retailing, 49(4),
48-64.
Krippendorff, K. (2004). Content Analysis: An Introduction to Its Methodology.
Thousand Oaks: Sage.
Kupke, V. (2004). Identifying the dimensions to retail centre image. Journal of
Property Investment & Finance, 22(4), 298-306.
Laroche, M., Teng, L., Michon, R., & Chebat, J. (2005). Incorporating service
quality into consumer mall shopping decision making: a comparison between
English and French Canadian consumers. Journal of Services Marketing,
19(3), 157-163.
Leo, P., & Philippe, J. (2002). Retail centres: location and customer satisfaction.
The Service Industries Journal, 22(1), 122–146.
Lindquist, J.D. (1974). Meaning of image: a survey of empirical and hypothetical
evidence. Journal of Retailing, 50(4), 29-38.
Line, N. D., Runyan, R. C., Costen, W., Frash, R., & Antun, J. M. (2012). Where
everybody knows your name: Homophily in restaurant atmospherics. Journal
of Hospitality Marketing & Management, 21(1), 1-19.
Louviere, J.J., & Gaeth, G.J. (1987). Decomposing the determinants of retail
facility choice using the method of hierarchical information integration: a
supermarket illustration. Journal of Retailing, 63(1), 25-48.
Machleit, K.A., & Eroglu, S.A. (2000). Describing and measuring emotional
response to shopping experience. Journal of Business Research, 49(2), 101-
111.
MacInnis, D.J., & Price, L.L. (1987). The role of imagery in information processing:
review and extensions. Journal of Consumer Research, 13(4), 473-491.

280
Marks, R.B. (1976). Operationalizing the concept of store image. Journal of
Retailing, 52(3), 37-46.
Martineau, P. (1958). The personality of the retail store. Harvard Business Review,
36(1), 47-55.
Mazursky, D., & Jacoby, J. (1986). Exploring the development of store images.
Journal of Retailing, 62(2), 145-165.
Meyer, C., & Schwager, A. (2007). Understanding customer experience. Harvard
Business Review, 85(2), 116.
Michon, R., Chebat, J.C., & Turley, L. (2005). Mall atmospherics: the interaction
effects of the mall environment on shopping behaviour. Journal of Business
Research, 58(5), 576-583.
Miller, N. J., & Kean, R. C. (1997). Factors contributing to inshopping behavior in
rural trade areas: Implications for local retailers. Journal of Small Business
Management, 35(2), 80-94.
Mitchell, V. (2001). Re-conceptualizing consumer store image processing using
perceived risk. Journal of Business Research, 54(2), 167-172.
Mullis, K., & Kim, M. (2011). Factors determining inshopping in rural US
communities: Consumers' and retailers' perceptions. International Journal of
Retail & Distribution Management, 39(5), 326-345.
Nevin, J.R., & Houston, M.J. (1980). Image as a component of attraction to
intraurban shopping areas. Journal of Retailing, 56(1), 77-93.
Oppewal, H., & Timmermans, H. (1999). Modeling consumer perception of public
space in shopping centers. Environment and Behavior, 31(1), 45-65.
Orth, U.R., Limon, Y., & Rose, G. (2010). Store-evoked affect, personalities, and
consumer emotional attachments to brands. Journal of Business Research,
63, 1202-1208.
Osman, M.Z. (1993). A conceptual model of retail image influences on loyalty
patronage behaviour. International Review of Retail, Distribution and
Consumer Research, 3 (2), 133-148.
Otnes, C., & McGrath, M. A. (2001). Perceptions and realities of male shopping
behavior. Journal of Retailing, 77(1), 111-137.
Oxenfeldt, A.R. (1974). Developing a favorable price-quality image. Journal of
Retailing, 50(4), 8-14.

281
Padilla, C., & Eastlick, M.A. (2009). Exploring urban retailing and CBD
revitalization strategies. International Journal of Retail and Distribution
Management, 37(1), 7-23.
Pan, Y., & Zinkhan, G.M. (2006). Determinants of retail patronage: A meta-
analytical perspective. Journal of Retailing, 82(3), 229-243.
Park, C. W., Jaworski, B. J., & Maclnnis, D. J. (1986). Strategic brand concept-
image management. Journal of Marketing, 50(4), 135-145.
Perreault Jr, W.D., & Leigh, L.E. (1989). Reliability of nominal data based on
qualitative judgments. Journal of Marketing Research, 26(2), 135-148.
Puccinelli, N.M., Goodstein, R.C., Grewal, D., Price, R., Raghubir, P. & Stewart, D.
(2009). Customer experience management in retailing: understanding the
buying process. Journal of Retailing, 85(1), 15-30.
Ravenscroft, N. (2000). The vitality and viability of town centres. Urban Studies,
37(13), 2533-2549.
Reilly, W. J. (1929). Methods for the study of retail relationships. University of
Texas Bulletin, 2944.
Relph, E. (1976). Place and Placelessness. London: Pion.
Robertson, K. A. (1997). Downtown retail revitalization: A review of American
development strategies. Planning Perspectives, 12(4), 383-401.
Robertson, K.A. (1999). Can small-city downtowns remain viable? A national study
of development issues and strategies. Journal of the American Planning
Association, 65(3), 270-271.
Runyan, R.C., & Huddleston, P. (2006). Getting customers downtown: The role of
branding in achieving success for central business districts. Journal of Product
and Brand Management, 15(1), 48-61.
Rust, R.T., & Cooil, B. (1994). Reliability measures for qualitative data: Theory and
implications. Journal of Marketing Research, 31(1), 1-14.
Samli, A.C. (1998). Strategic marketing for success in retailing. Westport, CT:
Quorum Books.
Schmitt, B. (1999). Experiential marketing. Journal of Marketing Management,
15(1), 53-67.
Sharp, T. (1968). Town and townscape. London: John Murray.

282
Sirgy, M.J., & Samli, A.C. (1985). A path analytic model of store loyalty involving
self-concept, store image, geographic loyalty, and socioeconomic status.
Journal of the Academy of Marketing Science, 13(3), 265-291.
Sirgy, M.J., Grewal, D., Mangleburg, T., Park, J., Chon, K., Claiborne, C.B., Johar,
J.S. & Berkman, H. (1997). Assessing the predictive validity of two methods of
measuring self-image congruence. Journal of the Academy of Marketing
Science, 25(3), 229-241.
Sit, J., Merrilees, B., & Birch, D. (2003). Entertainment-seeking shopping centre
patrons: The missing segments. International Journal of Retail and Distribution
Management, 31(2/3), 80-94.
Sneed, C. T., Runyan, R., Swinney, J. L., & Lim, H. J. (2011). Brand, business
mix, sense-of-place: Do they matter downtown? Journal of Place Management
and Development, 4(2), 121-134.
Stern, B., Zinkhan, G.M., & Jaju, A. (2001). Marketing images: Construct definition,
measurement issues, and theory development. Marketing Theory, 1(2), 201-
224.
Tajfel, H. (1978). Differentiation between Social Groups: Studies in the Social
Psychology of Intergroup Relations, Academic Press, London.
Teller, C., & Elms, J. (2010). Urban place marketing and retail agglomeration
customers. Journal of Marketing Management,
DOI:10.1080/0267257X.2010.517710
Timmermans, H.J.P. (1981). Multiattribute shopping models and ridge regression
analysis. Environment and Planning A, 13(1), 43-56.
Timmermans, H.J.P., Van Der Heijden, R., & Westerveld, H. (1982). Cognition of
urban retailing structures: a Dutch case study. Tijdschrift Voor Economische
en Sociale Geografie, 73(1), 2-12.
Turley, L.W., & Milliman, R.E. (2000). Atmospheric effects on shopping behavior:
A review of the experimental evidence. Journal of Business Research, 49(2),
193-211.
Van Raaij, W.F. (1983). Shopping center evaluation and patronage in the city of
Rotterdam. Papers on Economic Psychology, 27. Rotterdam, the Netherlands:
Erasmus University Press.

283
Verhoef, P.C., Lemon, K.N., Parasuraman, A., Roggeveen, A., Tsiros, M., &
Schlesinger, L.A. (2009). Customer experience creation: Determinants,
dynamics and management strategies. Journal of Retailing, 85(1), 31-41.
Wakefield, K.L., & Baker, J. (1998). Excitement at the mall: Determinants and
effects on shopping response. Journal of Retailing, 74(4), 515-539.
Warnaby, G. (2009). Look up! Retailing, historic architecture and city centre
distinctiveness. Cities, 26(5), 287-292.
Warnaby, G., Bennison, D., Davies, B.J., & Hughes, H. (2002). Marketing UK
towns and cities as shopping destinations, Journal of Marketing Management,
18(9-10), 877-904.
Weber, R.P. (1990). Basic content analysis. Newbury Park, CA: Sage.
Wee, C.H. (1986). Shopping area image: its factor analytic structure and
relationships with shopping trips and expenditure behaviour. Advances in
Consumer Research, 13, 48-52.
Weltevreden, J.W.J., & Van Rietbergen, T. (2007). E-shopping versus city centre
shopping: the role of perceived city centre attractiveness. Tijdschrift voor
Economische en Sociale Geografie, 98(1), 68-85.
Whysall, P. (2011). Managing decline in inner city retail centres: From case study
to conceptualization. Local Economy: The Journal of the Local Economy
Policy Unit, 26(1), 3-17.
Woodside, A.G., & Trappey, R.J. (1996). Customer portfolio analysis among
competing retail stores. Journal of Business Research, 35(3), 189-200.
Zimmer, M.R., & Golden, L.L. (1988). Impressions of retail stores: a content
analysis of consumer images. Journal of Retailing, 64(3), 235-236.

284