See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/343831739

Artificial Intelligence in Retail: The AI-Enabled Value Chain

Article in Australasian Marketing Journal (AMJ) · August 2020

DOI: 10.1016/j.ausmj.2020.07.007

CITATIONS READS

0 1,269

4 authors:

Kim Oosthuizen Dr Elsamari Botha

Stellenbosch University Stellenbosch University
1 PUBLICATION 0 CITATIONS 32 PUBLICATIONS 432 CITATIONS

SEE PROFILE SEE PROFILE

Jeandri Robertson Matteo Montecchi

25 PUBLICATIONS 34 CITATIONS
King's College London
16 PUBLICATIONS 87 CITATIONS
SEE PROFILE
SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Value creation practices that affect online interaction on social media: Premium Wines on Facebook View project

Artificial intelligence in retail: The AI-enabled value chain View project

All content following this page was uploaded by Jeandri Robertson on 25 August 2020.

The user has requested enhancement of the downloaded file.

JID: AMJ
ARTICLE IN PRESS [m5G;August 24, 2020;8:27]
Australasian Marketing Journal xxx (xxxx) xxx

Contents lists available at ScienceDirect

Australasian Marketing Journal

journal homepage: www.elsevier.com/locate/ausmj

Artificial intelligence in retail: The AI-enabled value chain

Kim Oosthuizen a, Elsamari Botha b,∗, Jeandri Robertson c, Matteo Montecchi d
a
University of Stellenbosch Business School, Cape Town, South Africa
b
Department of Management, Marketing and Entrepreneurship, University of Canterbury; and University of Stellenbosch Business School, Cape Town, South
Africa
c
School of Management Studies, Marketing Section, University of Cape Town, Cape Town, South Africa
d
King’s Business School, King’s College London, London, United Kingdom

a r t i c l e i n f o

Article history:
Received 7 February 2020
Revised 24 July 2020
Accepted 28 July 2020
Available online xxx

1. Introduction try is then explained, followed by a particular focus on the role

that AI has played in disrupting this industry. We then map var-
Traditional retailers’ business models are facing disruption by ious AI technologies based on Gartner’s (Sicular et al., 2019), to
new entrants who can deliver greater value to customers more each stage in the value chain and show that some AI technology
efficiently. In recent years, authors have argued that the tradi- investments can serve multiple purposes in the value chain. Then,
tional value chain drives inefficiencies (Begley et al., 2018) and we use Christensen’s jobs-to-be-done approach (Christensen, 2003;
that the value chain is shortening as manufacturers, third parties Christensen et al., 2016a; Christensen et al., 2016b) to better under-
and customers are increasingly engaging with customers directly stand the value of AI in the retail industry. The main aim of this
(Reinartz et al., 2019). These inefficiencies combined with the in- article, therefore, is to better understand what an AI-enabled retail
ability to adapt to a changing competitive landscape leaves tradi- value chain would look like.
tional retailers vulnerable to disruption from market entrants. To This article provides two important contributions to the emerg-
remain competitive and survive in an ever-changing and diversified ing literature on AI and its implementation in marketing and
customer market, retailers need to become leaner (Campbell et al., retailing. First, we show how AI technologies can be used
2020), more agile (Goworek, 2014), and innovate their value chain across various retail value chain activities. While several au-
by adopting new technologies (Lee et al., 2018). Of the new thors have addressed the relevance of AI to business in general
technologies that are impacting the retail industry, AI has been (Kietzmann et al., 2018; Paschen et al., 2019a; Poole and Mack-
earmarked as the most transformative (Kietzmann et al., 2018; worth, 2010; Ransbotham et al., 2017; Van Esch et al., 2019), the
Lee et al., 2018; Silva et al., 2019). Yet while there is great ex- strategic role and implementation of AI in retailing organizations
citement about artificial intelligence (AI), it has yet to fully deliver has been subject to limited critical scrutiny (Van Esch, 2019). By
on its promise (Ransbotham et al., 2017) and academics and prac- mapping specific AI technologies against the retail value chain,
titioners are in the early stages of understanding the application we provide retail managers with some guidance regarding which
of AI (Van Esch et al., 2020). This article introduces a conceptual AI technology investments to prioritize, or how current AI invest-
framework to understand the role that AI can play in the retail ments can be leveraged.
value chain by proposing an AI-enabled retail value chain. Second, while authors argue that the retail value chain
As a starting point, we provide a brief overview of the tradi- needs revisiting because of new technologies (Hagel et al., 2016;
tional retail value chain and the activities, stakeholders, and tech- Reinartz et al., 2019), no academic studies, to the knowledge of
nology involved at each stage. The disruption of the retail indus- the authors, have suggested exactly how the retail value chain
should change. Guided by the jobs-to-be-done approach to inno-
vation (Christensen et al., 2016a; Christensen et al., 2016b) we
∗
Corresponding author. identify four key roles for AI solutions in the retail value chain:
E-mail addresses: [email protected] (K. Oosthuizen), knowledge and insight management, inventory management, op-
[email protected] (E. Botha), [email protected] (J. Robert- erations optimization, and customer engagement. This approach
son), [email protected] (M. Montecchi).

https://doi.org/10.1016/j.ausmj.2020.07.007
1441-3582/© 2020 Australian and New Zealand Marketing Academy. Published by Elsevier Ltd. All rights reserved.

Please cite this article as: K. Oosthuizen, E. Botha and J. Robertson et al., Artificial intelligence in retail: The AI-enabled value chain,
Australasian Marketing Journal, https://doi.org/10.1016/j.ausmj.2020.07.007
JID: AMJ
ARTICLE IN PRESS [m5G;August 24, 2020;8:27]

2 K. Oosthuizen, E. Botha and J. Robertson et al. / Australasian Marketing Journal xxx (xxxx) xxx

is customer-centric (Bettencourt and Ulwick, 2008), not Tayloristic pany’s success (DalleMule and Davenport, 2017; Sankaran et al.,
and process-driven and, therefore, better suited to the complex na- 2019), and data-driven decisions are becoming increasingly im-
ture of business amidst new technologies (McChrystal et al., 2015). portant in supply chains (Sankaran et al., 2019). Third, the more
Contrary to the more traditional silo-mentality and linear view of complex and long the value chain is, the more expensive prod-
the value chain, we argue that AI solutions can perform multiple ucts and services become, and the longer it takes to reach cus-
roles simultaneously, thus establishing interconnectivity between tomers. Finally, overly complex value chains leave organizations
the different value chain activities. First, however, the digital dis- vulnerable to digital disruption from smaller, more agile firms that
ruption of the traditional retail value chain is discussed. leverage new technologies to reduce costs and scale up quickly
(Gupta, 2018; Verhoef et al., 2019).
2. Digital disruption of the traditional retail value chain According to Ransbotham et al. (2017) expectations for the
commercial application of AI in business, particularly in retailing,
In his seminal work, Michael Porter (1985) used the term are sky-high. While there are, however, existing analytical tools for
value chain to describe a set of activities performed to design, managers to gauge AI’s influence on the retail and other industries
produce, market, deliver, and support products within businesses (U Paschen et al., 2019a), how to create and leverage AI’s value for
(Hagel et al., 2016). Often referred to as the supply chain (Levy and commercial advantage in the value chain still seems complex to
Weitz, 2009), the value chain is a set of processes that deliver most. To assist in this delineation, the following section explicates
value across primary activities (for example, inbound logistics, op- what exactly AI is and how it is currently applied in retailing.
erations, outbound logistics, marketing and sales, and service),
and secondary activities (including firm infrastructure, human re- 3. Artificial intelligence in the retail value chain
source management, technology development and procurement).
The activities in the traditional value chain move in a sequence 3.1. Artificial intelligence
of linear steps, which facilitates the process from product design,
to the point of consumption (Reinartz et al., 2019). In retailing, Most extant conceptualizations of AI make reference to com-
the value chain encompasses all the stakeholders and processes puter systems with human-like intelligence (Wierenga, 2010),
needed for retailers to deliver an end product or service to a cus- which encompasses these systems’ abilities “to interpret external
tomer (Levy and Weitz, 2009). From the supplier, to the manufac- data correctly, to learn from such data, and to use those learn-
turer, to the retailer - each stakeholder in the value chain con- ings to achieve specific goals and tasks through flexible adaptation”
tributes toward adding value to the customer. Table 1 details the (Kaplan and Haenlein, 2019: 17). In their definition, the Oxford
stages involved in the traditional retail value chain, including the Dictionary includes tasks such as visual perception, speech recog-
objectives and typical activities in each stage (Hagel et al., 2016; nition, decision-making, and the translation between languages,
Rieple and Singh, 2010). The stakeholders and technologies typi- with the Merriam-Webster Dictionary emphasizing that AI imi-
cally involved in each stage have been included. tates intelligent human behavior. Poole and Mackworth (2010: 3)
New digital technologies have, however, disrupted the tradi- offer a slightly nuanced explanation by framing AI as “computa-
tional retail business model by changing marketplaces from brick tional agents that act intelligently”. This definition describes AI as
and mortar only to omnichannel, which significantly alter the agents capable of perceiving an environment in order to take ac-
customer purchase journey (Bolton et al., 2019; Carlsson, 2018; tion, with the goal being to maximize the likelihood of achieving
Van Esch et al., 2019a and b). Customers are more connected success (Paschen et al., 2019a). From a systems perspective, the
than ever (Kietzmann et al., 2011), and the transformation from definition also implies a rational view of AI, where an AI system
store to omnichannel retailing has elevated their service expec- would, given what it knows, act to achieve the best possible out-
tations (Oh and Polidan, 2018). As the rate with which these come (Paschen et al., 2019a).
new technologies enter the market increases (Brynjolfsson and For the purposes of this paper, we adopt the definition by
McAfee, 2016; Gupta, 2018), the accelerated rate of change blurs Poole and Mackworth (2010), as it highlights two conceptual de-
the market boundaries and holds unpredictable consequences for lineations. First, it highlights the different evolutionary stages of
retailers (Day and Schoemaker, 2019). For instance, innovative and AI (Haenlein and Kaplan, 2019). What might have been considered
fast-growing digital entrants like Alibaba and Amazon, have al- intelligent behavior displayed by a machine five years ago, is now
ready adversely affected traditional retailers like Toys ‘R’Us and Ra- hardly noteworthy. Three stages of AI are differentiated: artificial
dioshack, by using their digital resources to not only disrupt the narrow intelligence (applied, below human-level AI, e.g., Siri voice
retail industry but also seemingly unrelated industries, like bank- recognition), artificial generalized intelligence (strong, human-level
ing and global shipping (Verhoef et al., 2019). AI, e.g., Siri developing the ability to autonomously perform tasks
As new digital technologies continue to transform the retail in- like driving a car), and artificial superintelligence (conscious/self-
dustry (Hagberg et al., 2016; Romero and Martínez-Román, 2015; aware, above human-level AI, e.g., Siri developing superhuman
Van Esch et al., 2019), the retail value chain needs to evolve with capabilities to instantaneously solve complex problems). Most of
it (Fiorito et al., 2010). However, majority retailers still employ the the commercial AI technologies available today are classified as
traditional value chain, or variations thereof (like the introduc- “narrow” and almost all of the AI technology to be integrated
tion of multiple channels to serve customer needs), which holds into business in the next ten years will be “narrow” or “applied”
the following four risks. First, while each stage of the retail value (Kelly, 2017; Marr, 2017). For example the use of AI to create ads
chain adds value, it also adds complexity by increasing the num- (Bakpayev et al., 2020). This applies to retailing as well.
ber of stakeholders and their accompanying support structures in- The second defining characteristic highlighted in the Poole and
volved. Complicated value chains inhibit retailers to understand Mackworth (2010) definition is the notion that AI represents
and swiftly respond to customer preferences (Hagel et al. 2016). knowledge, expertise, and intuition to solve problems. AI requires
Second, various stakeholders are likely to use their own platforms tailored knowledge to be built into a “carefully constructed sys-
and software, making it difficult to integrate systems and man- tem” (Kaplan, 1984: 52), where the storage of past knowledge
age data. Legacy systems often inhibit organizations’ agility to re- should reflect experiences that would inform subsequent intelli-
spond to changing customer needs and use the data for compet- gent behavior (Paschen et al., 2019a). In AI systems, these knowl-
itive advantage (Westerman et al., 2014). The ability to manage edge representations could include inputs (structured and unstruc-
mass amounts and different sources of data are critical to a com- tured data); processes (machine learning); or self-generated AI-

Please cite this article as: K. Oosthuizen, E. Botha and J. Robertson et al., Artificial intelligence in retail: The AI-enabled value chain,
Australasian Marketing Journal, https://doi.org/10.1016/j.ausmj.2020.07.007
 JID: AMJ
Australasian Marketing Journal, https://doi.org/10.1016/j.ausmj.2020.07.007
Please cite this article as: K. Oosthuizen, E. Botha and J. Robertson et al., Artificial intelligence in retail: The AI-enabled value chain,

Table 1
Traditional Retail Value Chain

Retail Value Chain Technology Application

Stages Activities Stakeholders Key Driver Decision Outcome Examples

Design Design initiation, Design Retail; Product trends; Customer Which products to design; Pattern Design and CAD system; Fabric quality
concept; Decision making Manufacturer; needs; Bring new product to Product specification; specifications; Productions checking; Online product
process; Technical design Suppliers; market quicker; Market Materials planning and control; searches
Customers research; Product quality Garment evaluation
Sourcing/Procurement Planning and production Retailer; Tailoring assortment for Which products will fulfil Sales budget planning; Assortment planning;

K. Oosthuizen, E. Botha and J. Robertson et al. / Australasian Marketing Journal xxx (xxxx) xxx
control; Purchasing or Manufacturer; customer needs; customer needs? When is Merchandise strategy; Merchandise financial
building inventory Suppliers Understanding customer the product needed in Merchandise financial planning; Size profile
segments; Product selection; store? Buy quantity; Align planning; Assortment optimization; Excel
Price negotiation; Liaising to budget; Which supplier planning;
with suppliers or manufacturer? Supplier/Manufacturer
selection; Order placement

ARTICLE IN PRESS
Manufacturing and Cutting, Sewing, Finishing, Manufacturer; Accurate forecasting; One-time Cutting; Sewing; Finishing Cutting quantity; Job ERP systems; Assembly;
assembly Packing, Acquiring, Storing Supplier; Shipping ordering; Product flow and distribution; scheduling; Assembly line; Work force scheduling;
and preparing raw visibility; Product quality Productions scheduling; Fabric laying; Cutting; Materials management
materials Factory management; Overall handling;
Quality checking Workforce planning
Inventory Managing and distributing Retail; Suppliers; End-to-End inventory Monitoring deliveries; Movement of products Product information
management and products to be sold Logistics; visibility; Consistent and Monitoring inventory; from manufacturer to DC management; Order
distribution Distribution accurate sales and inventory Scheduling DC to store or store sales forecasting; processing; Demand
forecasting; Monitoring deliveries Allocation to stores; planning; Product
product deliveries Forecasting; Determining allocation
replenishment
information; Controlling
inventory levels;
Workforce planning
Store operations and Managing point of sale Retail; Suppliers; Performance management; Maximising sales; Promotions and markdown Pricing management;
Sales and executing purchase Logistics; Price and markdown Minimising markdowns; planning; Reordering; Key Promotional planning;
transition Customer; optimisation; Monitoring stock Managing product life products; Reverse Product lifecycle
Competitors levels cycle; Price point logistics; Repairs, returns management; POS system
management; Customer and maintenance support;
service Traffic management
Fulfilment Delivering products to the Retail; Suppliers; Matching demand to product Managing out of stocks; Product replenishment; Demand prediction,
customer Manufacturer supply Maximising sales; Logistics management; inventory management;
Managing inventory; Movement of products to Ordering items
Reorder negotiation stores; Forecast accuracy;
Inventory placement
optimisation for omni
fulfilment; Workforce
planning

[m5G;August 24, 2020;8:27]

Customer use and Helping customers Retail; Customer; Offering customers Identify high value Satisfied customers; Online platform; POS
support maximise value; Using and Logistics personalised or customized customers and products; Personalised offerings; systems; CRM
maintaining products offerings Supporting customer Product recommendations management
queries
SOURCE: Adapted from Cammett, 2006; Hagel et al., 2016; Lee et al., 2018; Michael, 1985; Reinartz et al., 2019; Rieple & Singh, 2010

3
JID: AMJ
ARTICLE IN PRESS [m5G;August 24, 2020;8:27]

4 K. Oosthuizen, E. Botha and J. Robertson et al. / Australasian Marketing Journal xxx (xxxx) xxx

Table 2
Example applications of Retailers’ AI Technology in the Value Chain

AI Technology (Gartner
Value Chain Stage Retailer Hype Cycle) Detail of Technology Use

Design Adidas Machine learning Using Machine learning Adidas is using a "speed factory" to help customers
design their own personalised shoes and making them in 24 hours and
shipping to customers
Sourcing/Procurement Simons Insight engines Simons implemented analytics for improved insight on projected demand
and inventory optimization recommendations that get the right product to
the right store proactively
Manufacturing Mohammadi Fashion Intelligent applications AI enabled sewing machines that knit sweaters
Sweaters
Inventory management ThredUp Deep learning Scanning close to 100 million unique inbound items via image recognition
and distribution enabling automated visual tagging of products and assigning a unique item
code
Fulfilment The Home Depot Edge AI The Home Depot Inc. is connecting in-store robotics with an intelligent
enterprise approach. It is using drones and robotics to create an efficient
in-store experience that delights customers and provides faster order
fulfillment
Store operations and Sephora Augmented Intelligence The app allows customers to try products virtually via augmented reality.
Sales The tool scans a customer’s face, figures out where their lips and eyes are
then matches colours and suggests products for consumers to buy.
Customer use and Ikea Augmented Intelligence The augmented reality helps the customer virtually "place" true to scale 3D
support furniture in their home.
SOURCE: Adapted from Araujo, 2019; de Leon, 2019; Emont, 2018; Galer, 2018; Ikea, n.d.; RIS, 2020; Sephora, 2017

output (robotics) (Paschen et al., 2019a, Paschen et al., 2019b). As Using the retail value chain presented in Table 1 to better un-
AI continuously advances, our understanding of its various applica- derstand the possible role of AI therein, we identified retailers
tions similarly needs to develop at a breakneck pace. Research into who currently apply AI technologies in their value chain. Table 2
AI and its implications for business has burgeoned with studies illustrates examples of retailers applying various AI technologies
looking at AI in recruitment (Black and Van Esch, 2020; Van Esch throughout the value chain.
et al., 2019, 2020; Van Esch & Black, 2019) and marketing (Bak- Retailers are already beginning to apply AI applications in parts
payev et al., 2020, 2020; Mogaji et al., 2020; Paschen et al., 2020; of the value chain (Bughin et al., 2017). However, both researchers
Xu et al., 2020, 2020) amongst others. Feng et al. (2020) provide and practitioners are only in the early stages of fully understanding
a summary of all the key AI-related studies in marketing. The fol- the application of AI (Van Esch et al., 2020). And as is evidenced in
lowing section looks at the application of AI technologies in retail Table 2, some AI applications were used for more than one value
in particular, in order to better understand their role in the retail chain activity. For example, Augmented Intelligence was used for
value chain. store operations and sales by one retailer, and customer use and
support by another. This prompted the authors to take a broader
view of where each AI technology can be applied in the retail value
chain.
3.2. AI in retail While an understanding of current application of AI in retail
can help identify current gaps in its use, it does not provide in-
Although AI technologies have advanced in recent years, AI still sight into its most effective use in the retail value chain. To further
remains in its infancy (Bughin et al., 2017). A vast majority of re- develop our understanding of the opportunities and address this
tailers have started testing the possibility of AI, yet many retailers gap, we next use Clayton Christensen’s (2003) jobs-to-be-done ap-
are missing the full benefit of scaling the technology throughout proach as a guideline to identify four conceptual dimensions which
the value chain (Standish and Ganapathy, 2020). Many AI applica- highlight how AI can best be applied to the retail value chain.
tions, already available or under development, contribute to retail-
ers’ confusion and frustration with regards to which AI technolo-
gies to invest in. To assess the potential application of AI-enabled 3.3. Reimagining AI in the retail value chain: a jobs-to-be-done
solutions across the various retail value chain activities, this sec- approach
tion first focuses on providing an overview of where AI is currently
being applied in retail. Using a customer-centric innovation view to understanding
To develop an understanding of the current application AI tech- value (Bettencourt and Ulwick, 2008), the jobs-to-be-done ap-
nologies used in retailing, we first reviewed the 2019 Gartner hype proach was developed by Clayton Christensen in his 2003 book,
cycle for artificial intelligence report (Sicular et al., 2019), focus- The Innovator’s Solution, and later expanded upon in Competing
ing on the AI technologies predicted to reach mainstream adop- Against Luck (Christensen et al., 2016a). The theory proposes a
tion in the next five years. The report examines trends and inno- group of principles that explain how to make marketing more ef-
vations in the AI sector and classifies the different AI applications fective and innovation more predictable by focusing on the cus-
(Sicular et al., 2019). These include speech recognition, graphic pro- tomer’s jobs to be done. Christensen et al.’s (2016a) approach is
cessing unit (GPU) accelerators, robotic process automation soft- based on the idea that, in order to stimulate the effective develop-
ware, artificial intelligence (AI)-related consulting and system inte- ment and implementation of innovation, companies should focus
gration (C & SI) services, augmented intelligence, chatbots, machine on the key goals of a product or service. For example, when con-
learning, deep learning, edge AI, intelligent applications, virtual sidering how to best improve a razor blade, companies should be
personal assistant (VPA)-enabled wireless speakers, virtual assis- less concerned with improving the product itself (e.g., adding more
tants, field-programmable gate array (FPGA) accelerators, computer blades), and more concerned with what “job” the razor blades
vision, insight engines, data labelling and annotation services, and do (e.g., quick and easy grooming). As an example, Philips re-
automated machine learning (AutoML) (Sicular et al., 2019). cently presented their OneBlade range of razor blades, that not

Please cite this article as: K. Oosthuizen, E. Botha and J. Robertson et al., Artificial intelligence in retail: The AI-enabled value chain,
Australasian Marketing Journal, https://doi.org/10.1016/j.ausmj.2020.07.007
JID: AMJ
ARTICLE IN PRESS [m5G;August 24, 2020;8:27]

K. Oosthuizen, E. Botha and J. Robertson et al. / Australasian Marketing Journal xxx (xxxx) xxx 5

only shaves, but also trims and styles any length of hair for multi- scalable technology (Wirth, 2018). Therefore, the current linear ap-
ple looks (Philips, 2020). In essence, Christensen et al. (2016a) ar- proach to the retail value chain is not conducive to the advanced
gue that people ‘hire’ products and services to get jobs done, and knowledge and insight management available through AI technolo-
companies can innovate through doing those jobs better. Each job gies.
can be broken down into various steps or stages of execution,
with validating questions to assess the best job fit at each stage
(Bettencourt and Ulwick, 2008).
3.3.2. Inventory management
We argue that this approach can successfully be applied to not
Retailers have two main inventory management objectives:
only final products and services, but also to the application of tech-
first, to buy products to fulfil customers’ requirements; and sec-
nology like AI in the workplace. Better understanding the jobs-to-
ond, to plan the inventory flow to maximise profits (Fairhurst and
be-done by AI technologies will increase the value it delivers to the
Fiorito, 1990). As retailers are always trying to match supply to
company. Following the job mapping approach by Bettencourt and
demand, they are continually revising their sales forecasts to an-
Ulwick (2008), the authors iteratively followed a customer-centric
ticipate demand throughout the value chain (Goworek, 2014). To
validation process, guided by the retail value chain processes as a
achieve the required forecasting capabilities, they need very spe-
reference point, to conceptually cluster what jobs AI can perform
cific sources of knowledge and insight. AI technologies can assist
in the retail value chain. From this perspective four job dimensions
in the process of balancing demand and supply over large assort-
emerged, which are discussed in greater depth in the following
ments to meet the customers’ needs and the company’s financial
sections:
objectives. Current AI applications in this category are those that
(1) Knowledge and insight management AI technologies refer to
can drive lower inventory levels, anticipate future demands and
the ability to provide insights by managing, sharing, using, creating
create localised assortments leading to reduced working capital for
and processing information.
retailers (Chao et al., 2019; Chuprina, 2019; Marr, 2018). These AI
(2) Inventory management AI technologies refer to those that as-
solutions include chatbots, insight engines, intelligent application,
sist in the process of balancing demand to supply over large as-
machine learning and virtual assistants.
sortments to meet customer needs and financial objectives.
AI can furthermore assist retailers in streamlining inventory
(3) Operations optimization AI technologies help retailers operate
management by predicting demand, keeping popular items stocked
effectively and efficiently by minimising costs and maximising op-
on shelves, and using clustering technologies to anticipate future
erational capabilities.
customer requirements. Machine learning, deep learning and intel-
(4) Customer engagement AI technologies enable retailers to build
ligent applications could match supply to demand by using multi-
relationships with their customers.
ple data sources and adjusting demand accordingly (Bughin et al.,
Table 3 details current AI technologies that fulfil these four
2017). Predictive inventory management could drive improvements
‘jobs-to-be-done’ dimensions set against the traditional stages of
in forecast accuracy and optimize the inventory throughout the re-
the retail value chain. Each of these four dimensions of AI appli-
tail value chain, leading to increased profits and cost-saving for
cations in the retail value chain is now discussed in the following
the retailer (Petropoulos et al., 2018). The eCommerce retailer, Otto
section.
Group, has for example reduced their out-of-stock rate by 80% by
using predictive machine learning applications, which also boosted
3.3.1. Knowledge and insight management
revenue, increased margins and assisted to respond to market
Knowledge and insight management AI technologies provide in-
shifts (Trotter, 2018).
sights throughout the value chain by managing, sharing, using, cre-
ating and processing information. Data is one of the foundations
of AI (Haenlein et al., 2019), and the effective translation of that
data into knowledge is key to its success. This dimension includes 3.3.3. Operations optimization
the process of transforming structured and unstructured data in- AI applications assisting with operations optimization are de-
puts into outputs that contribute to the organization’s knowledge signed to improve operations efficiently and effectively by min-
base. Paschen et al. (2019a) refer to this as the building block imising cost and maximising operational capabilities (Li et al.,
of AI. Current examples include deep learning, intelligent appli- 2017). Inefficient operations slow down the movement of products
cations, and insight engines amongst others. The importance of through the value chain, moving the customer further away from
transforming data into knowledge has been stressed by various au- a successful purchase (Rieple and Singh, 2010). AI applications for
thors (Paschen et al., 2019a; Black and Van Esch, 2020). Although this purpose, include AI-related C&SI services, computer vision,
there is more data available than ever before, only a fraction is in- deep learning, Edge AI, intelligent applications, machine learning
tegrated and analyzed within businesses (Chen et al., 2016). While robotic, process automation and virtual assistants which all shorten
some companies use data to create a competitive advantage, many the value chain by improving production speed and managing in-
businesses fall short of gaining real insights from their data. This ventory flow to the customer.
can mainly be ascribed to big data requiring powerful technolo- Various authors agree that streamlining operational processes
gies, computer processing power, skilled personnel and predictive creates efficiencies throughout the retail value chain (Bughin et al.,
models to crunch enormous amounts of data (Djafri et al., 2018; 2017; Li et al., 2017; Marr, 2019; Daugherty and Wilson, 2018). For
Gupta, 2018). example, JD.com, one of China’s largest retailers, has introduced AI
The analysis, processing and interpretation of data is a time- to drive efficiencies in their operations (Marr, 2019). The introduc-
consuming activity in the retail value chain; thus, more sophisti- tion of the AI applications allowed the retailer to deliver 92% of
cated AI technologies can be utilized to reduce the shortcomings their orders on the same or next day (Trotter, 2018). Nike imple-
of human efforts (Chen et al., 2016; Gupta, 2018; Sivarajah et al., mented augmented intelligence to design customised shoes for its
2017). Insight engines can anticipate future customer product customers, and the end-to-end process only takes two weeks from
needs and assist retailers in sourcing optimal assortments for their design to customer delivery (Chao et al., 2019). Optimizing opera-
customers. Gaining knowledge and insights from value chain data tions can offer unexpected benefits by increasing operational effi-
should, therefore, be a key motivator to implement AI technologies ciency, increasing agility and speed across the retail value chain.
in the retail value chain. However, for AI to reach its full poten- These improvements should be the driving force behind imple-
tial, siloed legacy IT systems should be replaced with robust and menting AI in retail operations.

Please cite this article as: K. Oosthuizen, E. Botha and J. Robertson et al., Artificial intelligence in retail: The AI-enabled value chain,
Australasian Marketing Journal, https://doi.org/10.1016/j.ausmj.2020.07.007
 6

JID: AMJ
Australasian Marketing Journal, https://doi.org/10.1016/j.ausmj.2020.07.007
Please cite this article as: K. Oosthuizen, E. Botha and J. Robertson et al., Artificial intelligence in retail: The AI-enabled value chain,

K. Oosthuizen, E. Botha and J. Robertson et al. / Australasian Marketing Journal xxx (xxxx) xxx
Table 3
AI jobs-to-be-done in the Retail Value Chain

Current applications of AI in the retail value chain

Jobs to be done Objective Design Sourcing/ Manufacturing Inventory management Store operations and Sales Fulfilment Customer use and support

ARTICLE IN PRESS
area of Procurement and assembly and distribution
application
Customer To build customer trust Machine Deep learning AI-related C&SI Services; Augmented intelligence;
engagement through personalisation Learning Augmented intelligence; Chatbot; Computer vision;
Chatbot; Computer vision; Deep learning; Edge AI;
Deep learning; Intelligent Insight engines; Machine
applications; Machine learning; Speech recognition;
learning; Virtual assistant Virtual assistant
Inventory Predict demand close to Intelligent Intelligent applications; Chatbot; Virtual assistant Insight engines;
Management supply by anticipating applications Machine learning Intelligent
customer needs and achieving applications;
financial objectives Machine learning
Operations Operating efficiently and Robotic process AI-related C&SI Deep learning; Edge AI; AI-related C&SI services; Edge AI
optimization effectively by minimising cost automation services; Deep Robotic process Computer vision; Edge AI;
and maximising operational software learning; automation software Intelligent applications;
capabilities Intelligent Machine learning; Robotic
applications process automation software;
Virtual assistant
Knowledge and Ability to provide insights by Deep learning; Deep learning; Insight engines Deep learning; Edge AI; GPU AI-related C&SI services;
insight managing, sharing, using, Insight engines Insight engines accelerators; Insight engines Insight engines; Intelligent
management creating and processing applications
information

[m5G;August 24, 2020;8:27]

JID: AMJ
ARTICLE IN PRESS [m5G;August 24, 2020;8:27]

K. Oosthuizen, E. Botha and J. Robertson et al. / Australasian Marketing Journal xxx (xxxx) xxx 7

3.3.4. Customer engagement The dimensions in the AI-enabled retail value chain framework
The value of using AI in customer-facing activities is well doc- are not mutually exclusive. As the backbone of any AI implementa-
umented and Xu et al. (2020) recently provided a definition of AI tion, knowledge and insight management provides insights through-
in the context of customer service as “a technology-enabled sys- out the value chain by managing, sharing, using, creating and pro-
tem for evaluating real-time service scenarios using data collected cessing information. Its objective is to generate knowledge and in-
from digital and/or physical sources in order to provide person- sights in support of all value chain activities. The appropriate use
alized recommendations, alternatives, and solutions to customers’ of AI technologies provides responsive R&D, dynamic price recom-
enquiries or problems”. To create a customer-centric value chain, mendations, and the extensive processing of transactional data. By
retailers need to exceed customer expectations to survive in a com- collecting and analysing data across various data sources, the AI
petitive market. To deliver seamless shopping experiences across applications can anticipate future product and customer needs.
all channels, retailers need to build connections with their cus- AI-enabled customer engagement retail activities include a myr-
tomers (Araujo, 2019). Therefore, AI technologies for customer en- iad of functions geared toward not only optimizing customer in-
gagement in the retail value chain are predominantly focused on teractions, but building customer relationships. For example, these
the customer journey, enabling customer engagement, enhanced functions include supporting customers to navigate the store, an-
customer service, and sales support functions (Kaplan and Haen- swering questions and creating personalized product recommen-
lein, 2019). As per Table 3, the majority of retailers currently focus dations. Retailers investing in customer-facing AI technologies are
on AI applications that would facilitate customer engagement. Re- creating a unique competitive advantage in the market. Techno-
tailers are utilizing AI applications to connect and build relation- logical innovations such as AI help new entrants to seamlessly
ships with customers by personalizing their product recommen- move through the value chain stages in the design, manufacture,
dations and purchases, helping them find their way around the commercialization, distribution and support of products - enabling
store, and answering product-related questions in real time using them to connect with their customers.
apps (Morgan, 2019; Standish and Ganapathy, 2020). A common In inventory management, AI can assist retailers in matching
example of such personalization is the use of chatbots to reduce supply and demand by using multiple data sources and adjust-
customer service costs and speed up customer response time to ing demand accordingly through the implementation of machine
queries (Reddy, 2017). learning, deep learning, and other AI intelligent applications. Pre-
Current AI technologies that assist retailers in engaging with dictive inventory management could drive improvements in fore-
customers across the retail value chain (and not only in the fi- casting accuracy and optimizing inventory throughout the retail
nal stages of the value chain) include speech recognition, robotic value chain, which could lead to increased profits and cost-saving
process automation, AI-related C&SI services, augmented intelli- for the retailer. Retailers that want to benefit from this category
gence, chatbots, computer vision deep learning, Edge AI, insight of AI applications need to work towards identifying employees’ ca-
engines, intelligent applications, machine learning, speech recog- pability to work alongside intelligent applications (Black and Van
nition and virtual assistants. At retailer Sephora, for example, in- Esch, 2020).
store employees are equipped with handheld devices to scan a cus- Optimizing operational efficiencies by streamlining processes
tomer’s face, creating a personalized cosmetic shade to match the with AI applications, can remove silos throughout the value chain.
customer’s complexion. The shade matching creates a unique code Major manufacturers and retailers are already using AI-based tech-
enabling the customer to personalize purchases across all channels nologies throughout their distribution centres to streamline their
(Milnes, 2016). The North Face, a retailer of technical outerwear, operations (Grewal et al., 2017).
utilizes augmented intelligence to help consumers find clothing
and apparel suited to specific weather conditions (Trotter, 2018). 5. Conclusion and managerial implications
These customer engagement AI applications are building customer
loyalty through personalization, moving away from the purely Research asserts that the traditional retail value chain is expe-
transactional towards a more customer-centric approach. riencing a metamorphosis, yet, literature offering managerial guid-
ance on how to respond to these changes is limited (Araujo, 2019;
Van Esch et al., 2019). With added pressure to remain competi-
4. The AI-enabled retail value chain framework tive, many retailers have started to embrace a variety of digital
technologies to engage with their customers (Grewal et al., 2017),
In the previous section, we used the jobs-to-be-done approach and many are utilizing AI applications to establish this connection
(Christensen et al., 2016a; Christensen et al., 2016b) to better un- (Morgan, 2019). To bridge this gap, the application of current AI
derstand how AI can be successfully applied to the retail value technologies to the retail value chain was reviewed, and four di-
chain. We conceptually proposed four AI technology dimensions, mensions of AI applications were conceptualized. AI can best be
which fulfil the majority of the roles in the “traditional” retail value employed in the retail value chain by serving one of the following
chain. The majority of current AI applications are narrow in na- purposes: knowledge and insight management, inventory manage-
ture (Marr, 2017) and are being implemented in some offerings ment, operations optimization and customer engagement. These
and processes (Ransbotham et al., 2017). However, we suggest that four categories of AI technologies in the value chain enabled us
various AI applications, such as machine learning, intelligent appli- to propose a revised AI-enabled retail value chain.
cations, Edge AI and deep learning, can undertake multiple tasks Although extant literature suggests that most AI applications
across the retail value chain. Retail managers would therefore get over the next decade will remain narrow or applied (Kelly, 2017;
the greatest return on investment in investing in these AI technolo- Marr, 2017), we propose that these narrow applications of AI can
gies. be extended to multiple functions in the retail value chain. There-
When applied more generally to the retail value chain, the four fore, retailers should invest in classes of AI technologies (e.g. deep
dimensions identified in section 3.3 can be represented as an im- learning capability) and not just specific applications thereby en-
proved value chain that stands in contrast to the silo mentality suring that these technologies can be used for multiple functions
and linear process proposed in many traditional retail value chains across the value chain. In addition, this framework provides retail-
(see Figure 1). The process contained within the AI-enabled retail ers with a list of priorities for investing in AI. In particular, retail-
value chain framework is iterative and agile, which enables real- ers should start with knowledge and insight management at the
time data flows. foundation. Using this framework, single AI applications can be ap-

Please cite this article as: K. Oosthuizen, E. Botha and J. Robertson et al., Artificial intelligence in retail: The AI-enabled value chain,
Australasian Marketing Journal, https://doi.org/10.1016/j.ausmj.2020.07.007
JID: AMJ
ARTICLE IN PRESS [m5G;August 24, 2020;8:27]

8 K. Oosthuizen, E. Botha and J. Robertson et al. / Australasian Marketing Journal xxx (xxxx) xxx

Figure 1. AI-enabled Retail Value Chain

plied to multiple tasks. Therefore, the framework (in combination for the organization that wishes to implement the AI-enabled retail
with the analysis of the AI jobs-to-be-done dimensions presented value chain.
in Table 3) provides retailers with insight into how to best leverage An AI-enabled retail value chain is heavily reliant on trained
current AI investments. employees who supply high-quality data at each touch point in
Retail managers need to focus on scaling AI technologies across the value chain. If the data is less than optimal, this may create
the value chain to reap the full long-term benefits. Broadening vulnerabilities and areas of risk, as organizations may unintention-
their horizons, retailer management should move away from a ally create biases with accompanying negative outcomes through
narrow focus on technology investments for distribution chan- the data that is being provided for intelligent automation. Future
nels and customer-facing technologies only (Olanrewaju and Will- research could assess how organizations can address these vulner-
mott, 2013). For increased and sustained competitive advantage, abilities and avoid the potential biases. Finally, scaling AI applica-
the suggested conceptual framework can help retailers transform tions across the retail value chain will require the right platforms
their value chains in order to compete and thrive in the changing to be in place, data to be available and employees to support the
retail landscape. initiatives in the long term. Future research should examine the
technological and organizational platforms necessary for the suc-
cessful implementation of an AI-enabled value chain. As the tech-
6. Future research nology, most likely to reshape the retail landscape, retailers that
embrace AI are poised to enhance every link in their value chain.
To further build on the four identified value-adding dimen-
sions that AI solutions can fulfill in the retailers’ value chain, fu-
ture research could explore how these different AI dimensions References
contribute to organizations’ competitive advantage in different
Araujo, M., 2019. The Intelligent Enterprise for the Retail Industry.: 32.
product-market contexts. Furthermore, as many global industries Begley, S., Fox, R., Lunawat, G., et al. (2018) How analytics and dig-
gear up for the widespread adoption of AI technologies, demand ital will drive next-generation retail merchandising. Available at:
and competition will grow for scarce skilled employees with the https://www.mckinsey.com/industries/retail/our-insights/how-analytics-and-
digital- will- drive- next- generation- retail- merchandising.
ability to implement, manage and work alongside the new tech-
Bettencourt, L., Ulwick, A.W., 2008. The Customer-Centered Innovation Map.
nology (Van Esch et al., 2019; Butler-Adam, 2018). It will be crucial Harvard Business Review. 1 MayAvailable at https://hbr.org/2008/05/
for organizations to have a skilled workforce to support the im- the- customer- centered- innovation- map. accessed 31 January 2020.
Black, J.S., Van Esch, P., 2020. AI-enabled recruiting: What is it and how should a
plementation of AI, and there will be an even higher demand for
manager use it. Business Horizons 63 (2), 215–226.
skilled professionals (Van Esch and Black, 2019). While companies Bolton, R.N., Chapman, R.G., Mills, A.J., 2019. Harnessing Digital Disruption With
face external competition in finding skilled employees, low skilled Marketing Simulations. Journal of Marketing Education 41 (1), 15–31. doi:10.
workers could find it challenging to compete with machines and 1177/0273475318803417.
Brynjolfsson, E., McAfee, A., 2016. The Second Machine Age: Work, Progress, and
struggle to be employable in the future (Frey and Osborne, 2017). Prosperity in a Time of Brilliant Technologies, 1 edition W. W. Norton & Com-
Future research can focus on the skills and competencies necessary pany, New York London.

Please cite this article as: K. Oosthuizen, E. Botha and J. Robertson et al., Artificial intelligence in retail: The AI-enabled value chain,
Australasian Marketing Journal, https://doi.org/10.1016/j.ausmj.2020.07.007
JID: AMJ
ARTICLE IN PRESS [m5G;August 24, 2020;8:27]

K. Oosthuizen, E. Botha and J. Robertson et al. / Australasian Marketing Journal xxx (xxxx) xxx 9

Bughin, J., Hazan, E., Ramaswamy, S., Chui, M., Allas, T., Dahlström, P., Henke, N., Kaplan, A., Haenlein, M., 2019. Siri, Siri, in my hand: Who’s the fairest in the land?
Trench, M., 2017. Artificial intelligence the next digital frontier? Mckinsey Global On the interpretations, illustrations, and implications of artificial intelligence.
Institute, p. 80. Business Horizons 62 (1), 15–25. doi:10.1016/j.bushor.2018.08.004.
Butler-Adam, J., 2018. The Fourth Industrial Revolution and education. South African Kaplan, J., 1984. The Industrialization of Artificial Intelligence: From By-Line to Bot-
Journal of Science 114 (5/6). doi:10.17159/sajs.2018/a0271. tom Line.: 7.
Cammett, M., 2006. Development and the Changing Dynamics of Global Production: Kelly, K., 2017. The Inevitable: Understanding the 12 Technological Forces That Will
Global Value Chains and Local Clusters in Apparel Manufacturing. Competition Shape Our Future. Penguin Books.
& Change 10 (1), 23–48. doi:10.1179/102452906X91993. Kietzmann, J., Paschen, J., Treen, E., 2018. Artificial Intelligence in Advertising: How
Campbell, C., Sands, S., Ferraro, C., Tsao, H.Y.J., Mavrommatis, A., 2020. From data to Marketers Can Leverage Artificial Intelligence Along the Consumer Journey. Jour-
action: How marketers can leverage AI. Business Horizons 63 (2), 227–243. nal of Advertising Research 58 (3), 263–267. doi:10.2501/JAR-2018-035.
Carlsson, C., 2018. Decision analytics—Key to digitalisation. Information Sciences Kietzmann, J.H., Hermkens, K., McCarthy, I.P., et al., 2011. Social media? Get serious!
460–461, 424–438. doi:10.1016/j.ins.2017.08.087. Understanding the functional building blocks of social media. Business horizons
Chao, G., Cheung, J., Haller, K., et al., 2019. The coming AI revolution in retail 54 (3), 241–251.
and consumer products: Intelligent automation is transforming both indus- Lee, H., Mendelson, H., Rammoham, S., et al., 2018. Value Chain Innovation: The
tries in unexpected ways. IBM Institute for Business Value 14 January. National Promise of AI. Stanford Business, Stanford, USA Value Chain Innovation Ini-
Retail Federation. Available at https://nrf.com/sites/default/files/2019-01/The% tiativeAvailable at https://www.gsb.stanford.edu/faculty-research/publications/
20coming%20AI%20revolution.pdf. value- chain- innovation- promise- ai.
Chen, Y., Elenee Argentinis, J., Weber, G., 2016. IBM Watson: How Cognitive Com- Levy, M., Weitz, B.A., 2009. Retailing Management. McGraw-Hill Irwin.
puting Can Be Applied to Big Data Challenges in Life Sciences Research. Clinical Li, B., Hou, B., Yu, W., Lu, X., Yang, C., 2017. Applications of artificial intelligence
Therapeutics 38 (4), 688–701. doi:10.1016/j.clinthera.2015.12.001. in intelligent manufacturing: A review. Frontiers of Information Technology &
Christensen, C., Hall, T., Dillon, K., et al., 2016a. Competing against luck. The story Electronic Engineering 18 (1), 86–96. https://doi.org/10.1631/FITEE.1601885.
of innovation and customer choice. First edition. HarperBusiness an imprint of Marr, B., 2017. How Walmart Is Using Machine Learning AI, IoT
HarperCollins Publishers, New York, NY. And Big Data To Boost Retail Performance. Available at: https:
Christensen, C.M., 2003. The Innovator’s Solution: Creating and Sustaining Success- //www.forbes.com/sites/bernardmarr/2017/08/29/how- walmart- is- using-
ful Growth. Harvard Business School Press, Boston, Massachusettes. machine- learning- ai- iot- and- big- data- to- boost- retail- performance/
Christensen, C.M., Hall, T., Dillon, K., et al., 2016b. Know Your Customers’ “Jobs to #35637d5e6cb1.
Be Done”. Harvard Business Review. 1 September. Available at https://hbr.org/ Marr, B., 2019. Artificial Intelligence in Practice: How 50 Successful Companies Used
2016/09/know- your- customers- jobs- to- be- done. accessed 31 January 2020. Artificial Intelligence to Solve Problems, First Edition Wiley, Hoboken.
Chuprina, R., 2019. Artificial Intelligence for Retail in 2020: 12 Real-World Use Cases McChrystal, G.S., Collins, T., Silverman, D., et al., 2015. Team of Teams: New Rules of
December 20 https://spd.group/artificial-intelligence/ai-for-retail/. Engagement for a Complex World. Penguin.
DalleMule, L., Davenport, T.H., 2017. What’s Your Data Strategy. Harvard Business Michael, P., 1985. Competitive Advantage: Creating and Sustaining Superior Perfor-
Review. 1 May. Available at https://hbr.org/2017/05/whats- your- data- strategy. mance. Free Press, USA: NY.
accessed 16 November 2018. Milnes, H., 2016. How Color IQ, Sephora’s shade-matching skin care
Daugherty, P.R., Wilson, J.H., 2018. Human + Machine: Reimagining Work in the Age tool, boosts brand loyalty. Available at: https://digiday.com/marketing/
of AI. Harvard Business Review Press, Kindle. color- iq- sephoras- shade- matching- skin- care- tool- boosts- brand- loyalty/.
Day, G.S., Schoemaker, P.J.H., 2019. See Sooner, Act Faster: How Vigilant Leaders Morgan, B., 2019. The 20 Best Examples Of Using Artificial Intelligence For Retail
Thrive in an Era of Digital Turbulence. The MIT Press, Cambridge, MA. Experiences. Available at: https://www.forbes.com/sites/blakemorgan/2019/03/
de Leon, S.P., 2019. How thredUP is Driving the Circular Fashion Movement with 04/the- 20- best- examples- of- using- artificial- intelligence- for- retail- experiences/
AI August 27 Forbes https://www.forbes.com/sites/cognitiveworld/2019/08/27/ #44d8add44664.
how- thredup- circular- fashion- ai/#36ff4d6079a0. Oh, H., Polidan, M., 2018. Retail Consulting Class: Experiential Learning Platform
Djafri, L., Amar Bensaber, D., Adjoudj, R., 2018. Big Data analytics for prediction: to Develop Future Retail Talents. Journal of Marketing Education 40 (1), 31–46.
parallel processing of the big learning base with the possibility of improving doi:10.1177/0273475317743015.
the final result of the prediction. Information Discovery and Delivery 46 (3), Olanrewaju, T., Willmott, P., 2013. Finding your digital sweet spot. McKinsey and
147–160. doi:10.1108/IDD- 02- 2018- 0 0 02. Co., November available athttp://www. mckinsey. com/insights/business_ technol-
Emont, J., 2018. The Robots Are Coming for Garment Work- ogy/finding_your_digital_sweet_spot.
ers. That’s Good for the U.S., Bad for Poor Countries. The Paschen, J., Kietzmann, J., Kietzmann, T.C., 2016a. Artificial intelligence (AI) and its
Wall Street Journal. https://www.wsj.com/articles/the-robots-are- implications for market knowledge in B2B marketing. Journal of Business & In-
coming- for- garment- workers- thats- good- for- the- u- s- bad- for- poor- countries- dustrial Marketing 34 (7), 1410–1419. doi:10.1108/JBIM-10-2018-0295.
1518797631. Paschen, U., Pitt, C., Kietzmann, J., 2016b. Artificial intelligence: Building blocks and
Fairhurst, A.E., Fiorito, S.S., 1990. Retail buyers’ decision-making process: An an innovation typology. Business Horizons doi:10.1016/j.bushor.2019.10.004.
investigation of contributing variables. The International Review of Retail, Petropoulos, F., Kourentzes, N., Nikolopoulos, K., Siemsen, E., 2018. Judgmental se-
Distribution and Consumer Research 1 (1), 87–100. https://doi.org/10.1080/ lection of forecasting models. Journal of Operations Management 60, 34–46.
095939690 0 0 0 0 0 0 07. Philips., 2020. Philips OneBlade: Trim, Edge, Shave. Available at: https://
Fiorito, S.S., Gable, M., Conseur, A., 2010. Technology: advancing retail buyer perfor- www.philips.co.za/c- m- pe/oneblade- trim- edge- and- shave (accessed 7 February
mance in the twenty-first century. Fernie J (ed. International Journal of Retail & 2020).
Distribution Management 38 (11/12), 879–893. doi:10.1108/09590551011085966. Poole, D.L., Mackworth, A.K., 2010. Artificial Intelligence: Foundations of Computa-
Frey, C.B., Osborne, M.A., 2017. The future of employment: How susceptible are jobs tional Agents. Cambridge University Press.
to computerisation. Technological Forecasting and Social Change 114, 254–280. Porter, M., 1985. The value chain and competitive advantage, Chapter 2. In: Compet-
doi:10.1016/j.techfore.2016.08.019. itive Advantage: Creating and Sustaining Superior Performance. Free Press, New
Galer, S., 2018. Adidas Uses Machine Learning to Deliver Custom Sneakers York, pp. 33–61.
in Just 24 Hours June 13. SAP News Centre https://news.sap.com/2018/06/ Ransbotham, S., Kiron, D., Gerbert, P., et al., 2017. Reshaping business with artificial
adidas- uses- machine- learning- to- deliver- custom- sneakers- in- just- 24- hours/. intelligence - Closing the gap between ambition and action. MIT Sloan Manage-
Goworek, H., 2014. An investigation into retail buying roles and responsibilities for ment review.
own-label clothing: a multiple case-study. The Journal of The Textile Institute Reddy, T., 2017. How chatbots can help reduce customer service costs
105 (7), 760–769. doi:10.1080/0 04050 0 0.2013.848046. by 30%. Available at: https://www.ibm.com/blogs/watson/2017/10/
Grewal, D., Roggeveen, A.L., Nordfält, J., 2017. The Future of Retailing. Journal of Re- how- chatbots- reduce- customer- service- costs- by- 30- percent/.
tailing 93 (1), 1–6. doi:10.1016/j.jretai.2016.12.008. Reinartz, W., Wiegand, N., Imschloss, M., 2019. The Impact of Digital Transformation
Gupta, S., 2018. Driving Digital Strategy: A Guide to Reimagining Your Business. Har- on the Retailing Value Chain.: 49.
vard Business Review Press. Rieple, A., Singh, R., 2010. A value chain analysis of the organic cotton industry: The
Haenlein, M., Kaplan, A., 2019. A Brief History of Artificial Intelligence: On the Past, case of UK retailers and Indian suppliers. Ecological Economics 69 (11), 2292–
Present, and Future of Artificial Intelligence. California Management Review 61 2302. doi:10.1016/j.ecolecon.2010.06.025.
(4), 5–14. doi:10.1177/0 0 08125619864925. Romero, I., Martínez-Román, J.A., 2015. DETERMINANTS OF TECHNOLOGY ADOPTION
Haenlein, M., Kaplan, A., Tan, C.-W., et al., 2019. Artificial intelligence (AI) and man- IN THE RETAIL TRADE INDUSTRY - THE CASE OF SMES IN SPAIN. Amfiteatru
agement analytics. Journal of Management Analytics 6 (4), 341–343. doi:10. Economic 17 (39), 16.
1080/23270012.2019.1699876. Sankaran, G., Knahl, M., Siestrup, G., et al., 2019. Value of Smart Data for Supply
Hagberg, J., Sundstrom, M., Egels-Zandén, N., 2016. The digitalization of retailing: Chain Decisions in a Data Rich, Uncertain World.. CERC 2019, 49–54.
an exploratory framework. International Journal of Retail & Distribution Man- Sephora. (2017). Sephora Virtual Artist. Sephora. https://sephoravirtualartist.com/
agement 44 (7), 694–712. doi:10.1108/IJRDM- 09- 2015- 0140. landing_5.0.php?country=US&lang=en&x=&skintone=&currentModel=
Hagel, J., Seely Brown, J., Wooll, M., et al., 2016. Shorten the value chain: Sicular, S., Hare, J., Brant, K. 2019. Hype Cycle for Artificial Intel-
Transforming the stages of value delivery. 11 January. Deliotte In- ligence, 2019. G00369840, 25 July. Gartner. Available at: https:
sights. Available at https://www2.deloitte.com/insights/us/en/focus/ //www.gartner.com/smarterwithgartner/top-trends-on-the-gartner-
disruptive-strategy-patterns-case-studies/disruptive-strategy-value-chain-models. hype- cycle- for- artificial- intelligence- 2019/.
html .
Ikea. (n.d.). Say hej to IKEA place. https://www.ikea.com/au/en/customer-service/
mobile- apps/say- hej- to- ikea- place- pub1f8af050.

Please cite this article as: K. Oosthuizen, E. Botha and J. Robertson et al., Artificial intelligence in retail: The AI-enabled value chain,
Australasian Marketing Journal, https://doi.org/10.1016/j.ausmj.2020.07.007
JID: AMJ
ARTICLE IN PRESS [m5G;August 24, 2020;8:27]

10 K. Oosthuizen, E. Botha and J. Robertson et al. / Australasian Marketing Journal xxx (xxxx) xxx

Silva, E.S., Hassani, H., Madsen, DØ., 2019. Big Data in fashion: transforming the Van Esch, P., 2019. AI-enabled promiscuity: Wham bam thank you mam.. In: 2019
retail sector. Journal of Business Strategy ahead-of-print(ahead-of-print) doi:10. ANZMAC Conference Proceedings, 2019. Victoria University Wellington, Welling-
1108/JBS- 04- 2019- 0062. ton, pp. 491–494 Available at.
Sivarajah, U., Kamal, M.M., Irani, Z., et al., 2017. Critical analysis of Big Data Verhoef, P.C., Broekhuizen, T., Bart, Y., et al., 2019. Digital transformation: A multi-
challenges and analytical methods. Journal of Business Research 70, 263–286. disciplinary reflection and research agenda. Journal of Business Research doi:10.
doi:10.1016/j.jbusres.2016.08.001. 1016/j.jbusres.2019.09.022.
Standish, J., Ganapathy, V., 2020. AI in Retail: Time to Stop Experimenting and Westerman, G., Bonnet, D., McAfee, A., 2014. Leading Digital: Turning Technol-
Start Delivering Value at Scale. Available at: https://wwd.com/business-news/ ogy into Business Transformation. Harvard Business Review Press, Boston, Mas-
technology/accenture- ai- retail- think- tank- 1203422562/. sachusettes.
Trotter, C., 2018. 50 Best AI Retail Applications. Available at: (Bughin et al., 2017). Wierenga, B., 2010. Marketing and Artificial Intelligence: Great Opportunities, Re-
Van Esch, P., Arli, D., Gheshlaghi, M., Andonopoulos, V., von der Heidt, T., luctant Partners. In: Casillas, J., Martínez-López, F.J. (Eds.), Marketing Intelligent
Northey, G., 2019a. Anthropomorphism and Augmented Reality in the Retail En- Systems Using Soft Computing: Managerial and Research Applications. Springer,
vironment. Journal of Retailing and Consumer Services 49, 35–42. Berlin, Heidelberg, pp. 1–8. doi:10.1007/978- 3- 642- 15606- 9_1 Studies in Fuzzi-
Van Esch, P., Black, S.J., 2019b. Factors That Influence New Generation Candidates to ness and Soft Computing.
Engage With and Complete Digital, AI-Enabled Recruiting. Business Horizons 62 Wirth, N., 2018. Hello marketing, what can artificial intelligence help you
(6), 729–739. with. International Journal of Market Research 60 (5), 435–438. doi:10.1177/
Van Esch, P., Black, S.J., Ferolie, J., 2019. Marketing AI Recruitment: The Next Phase 1470785318776841.
in Job Application and Selection. Computers in Human Behavior 90, 215–222.

Please cite this article as: K. Oosthuizen, E. Botha and J. Robertson et al., Artificial intelligence in retail: The AI-enabled value chain,
Australasian Marketing Journal, https://doi.org/10.1016/j.ausmj.2020.07.007
View publication stats