Organizational Design

In today’s volatile business environment, organizational design is a serious challenge

for any manager, whether of a multinational enterprise or a small team. This book
sets out a step-by-step approach to designing an organization. All the key aspects of
organizational design are covered, including goals, strategy, structure, process,
people, coordination and control, and incentives. The text engages with critical
issues affecting organizations, such as globalization, worldwide competition,
deregulation and ever-new technologies, and contains many helpful features such
as end-of-chapter reviews and unique step-by-step diagrams to orientate the reader
in the design process. Diagnostic questions help the reader to determine the changes
needed in an organization. The action-oriented approach of this text helps
the reader to assess and redesign the complex organizations of today, and plan for
the information-rich organizations of tomorrow.

r i c h a r d m . b u r t o n is Professor of Management at the Fuqua School of

Business, Duke University.

g e r a r d i n e d e s a n c t i s was Thomas F. Keller Professor of Business Adminis-

tration at the Fuqua School of Business, Duke University.

b Ø r g e o b e l is Professor of Organization and Management and Rector of the

Aarhus School of Business, Denmark.
Organizational Design:
A Step-By-Step Approach

by

Richard M. Burton, Gerardine DeSanctis, and Børge Obel

CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo
Cambridge University Press
The Edinburgh Building, Cambridge CB2 2RU, UK
Published in the United States of America by Cambridge University Press, New York

www.cambridge.org
Information on this title: www.cambridge.org/9780521617338

© Richard M. Burton, Gerardine DeSanctis and Børge Obel 2006

This publication is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.

First published 2006

Printed in the United States of America

A catalogue record for this publication is available from the British Library

ISBN-13 978-0-521-85176-3 hardback

ISBN-10 0-521-85176-9 hardback
ISBN-13 978-0-521-61733-8 paperback
ISBN-10 0-521-61733-2 paperback

Cambridge University Press has no responsibility for the persistence or accuracy of URLs
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
 Contents

List of figures page x

List of tables xii
Preface xiii
An outline of the step-by-step approach xv

Step 1: Getting started

1 Define the scope of the organization and assess its goals 3
Introduction: The executive challenge of designing the organization 3
Overview of this book 4
The information-processing view 6
Select an organization for analysis 9
Define the scope of the organization 9
Assess the organization’s goals 11
Diagnostic questions 13
Misfits and balancing competing design dimensions 14
Summary 16
Glossary 17
Where are you in the step-by-step approach? 18

Step 2: Strategy
2 Strategy 23
Introduction 23
Reactor 25
Defender 26
Prospector 28
Analyzer without innovation 29
Analyzer with innovation 30
Diagnostic questions 31
Fit and misfits 33

v
vi Contents

Summary 35
Glossary 35

3 Environment 37
Introduction 37
Calm environment 44
Varied environment 45
Locally stormy environment 45
Turbulent environment 46
Diagnostic questions 47
Fit and misfits 49
Summary 51
Glossary 51
Where are you in the step-by-step approach? 52

Step 3: Structure
4 The configuration and complexity of the firm 57
Introduction 57
Configuration 58
Simple configuration 59
Functional configuration 61
Divisional configuration 63
Matrix configuration 66
Organizational complexity 69
Blob 70
Tall 70
Flat 71
Symmetric 72
Diagnostic questions 73
Fit and misfits 75
Summary 80
Glossary 80

5 Distributed organizations 82
Introduction 82
Structures for spanning geography 84
Global 86
International 87
Multi-domestic 88
Transnational 90
Structures for managing knowledge exchange 91
vii Contents

Ad hoc communications 93
Informated 94
Cellular 95
Network 95
Diagnostic questions 96
Fit and misfits 99
Summary 101
Glossary 102
Where are you in the step-by-step approach? 104

Step 4: Process and people

6 Task design 109
Introduction 109
Orderly 112
Complicated 113
Fragmented 114
Knotty 116
Diagnostic questions 117
Fit and misfits 118
Summary 120
Glossary 121

7 People 122
Introduction 122
Shop 125
Factory 126
Laboratory 127
Office 128
Diagnostic questions 128
Fit and misfits 130
Summary 133
Glossary 133

8 Leadership and organizational climate 135

Introduction 135
Leadership style 136
Maestro 137
Manager 138
Leader 139
Producer 139
Organizational climate 140
viii Contents

Group 142
Internal process 143
Developmental 144
Rational goal 145
Diagnostic questions 146
Fit and misfits 148
Summary 151
Glossary 151
Where are you in the step-by-step approach? 153

Step 5: Coordination and control

9 Coordination, control, and information systems 157
Introduction 157
Coordination and control systems 158
Family 161
Machine 162
Market 163
Clan or mosaic 164
Information systems 166
Event-driven 168
Data-driven 169
People-driven 169
Relationship-driven 171
Diagnostic questions 172
Fit and misfits 175
Summary 179
Glossary 179

10 Incentives 182
Introduction 182
Personal pay 187
Skill pay 188
Bonus-based 189
Profit sharing 191
Diagnostic questions 192
Fit and misfits 194
Summary 197
Glossary 197
Where are you in the step-by-step approach? 199
ix Contents

Applying the step-by-step approach in a dynamic world

11 Design dynamics: managing change and multi-organizations 203
Introduction 203
Where are you in the step-by-step approach? 203
What should you change first? Make changes within each step 206
What should you change second? Make changes between steps 207
Why change? Should we live with some misfits? 207
What are the benefits of change? 208
What are the difficulties of change? 208
Managing misfits over time 212
Multi-organization: multiple units of analysis 215
The multi-organization step-by-step approach 217
Joint venture 219
Merger 220
Strategic alliance or partnership 221
Summary 223
Glossary 223

References 225
Index 229
 List of figures

1.1 Levels in the organizational design process page 10

1.2 The goal space 11
1.3 Making changes in the efficiency/effectiveness space 15
2.1 The strategy space 25
2.2 Locate your organization in the strategy space by rating the
levels of exploration and exploitation 33
3.1 The environment space 43
3.2 Locate your organization in the environment space by
rating the levels of complexity and unpredictability 49
4.1 The alternative organizational configurations of the firm 59
4.2 A simple configuration 60
4.3 A functional configuration 62
4.4 The divisional configuration with product and
information flows 64
4.5 A matrix configuration 66
4.6 The organizational complexity space 70
4.7 Locate your firm in the configuration space 74
4.8 Locate your firm in the organizational complexity space 75
5.1 The organizational design space of structures for
spanning geography 86
5.2 Structures for managing knowledge exchange 93
5.3 Locate your organization in the geographic space by rating
its degree of local responsiveness and optimal sourcing 98
5.4 Locate your organization in the knowledge exchange space
by rating its degree of virtualization and IT-infusion 99
6.1 Task design space 111
6.2 Locate your firm in the task design space 118
7.1 The people space 125
7.2 Locate your organization in the people space 130
8.1 The leadership style space 137
x
xi List of figures

8.2 The climate space 142

8.3 Locate your organization’s leadership style 147
8.4 Locate your firm in the organizational climate space 148
9.1 Coordination and control space 161
9.2 Information systems space 168
9.3 Locate your organization in the coordination and
control space 173
9.4 Locate your chosen organization in the information
systems space 175
10.1 The incentive system design space 187
10.2 Locate your organization in the incentive scheme
design space 193
11.1 The complete set of component types within the 2
2
organizational design space 205
11.2 The difficulty of change 209
11.3 The misfit possibilities curve 211
11.4 The multi-organization 216
 List of tables

2.1 Fit between strategy and organizational goals page 34

3.1 Complexity scores 48
3.2 Fit among organizational environment, strategy, and
organizational goals 50
4.1 Fit among configuration, environment, strategy, and goals 76
5.1 Fit and misfit table for geographic distribution and
knowledge exchange 100
6.1 Fit and misfit for task design 119
7.1 Fit and misfit for people design 130
8.1 Fit and misfit to include leader style and organizational
climate 149
9.1 Fit and misfit to include coordination, control, and
information systems 176
10.1 Fit and misfit table for incentive alignments 194
11.1 My firm’s difficulty of change 210
11.2 Examples of internal, external, and combined sources of
organizational design misfits 213

xii
 Preface

In writing this book, we had specific readers in mind. We dedicate this book
to our executive students at The Fuqua School of Business and at the Aarhus
School of Business, Denmark. Executive students are very special profes-
sionals who come from around the world and bring their varied experience.
They have a goal to acquire new knowledge to act and take decisions that
will make a difference in their world. In particular, they want to improve the
performance of their firm, unit, or organization. From our first detailed
outline to the final editing from Cary, Durham, and Aarhus – whether in
person, video conference or elaborate email – they were ever with us.
Experience and science are two great teachers. Most of us spend most of
our time in organizations at work, home, worship and leisure. Over time, we
amass a wealth of experience through observation and action to apply in the
design of organizations; simply, we use our experience to design. But this
experience is limited and we can enhance it by blending it with the science of
organization design. The science of organization design is an accumulation
of knowledge by many individuals who, over many years, have conducted
research on the performance of organizations under many conditions. It
informs us about how to take action to design an organization. The scien-
tific foundation of this book comes from almost a century of research we call
the multi-contingency approach of Organizational Design.
The executive wants to understand, to diagnose, and take action. Experi-
ence and science are complementary and mutually supportive. In our classes
we try to build upon our executive students’ varied experiences with the
science of organization design to enhance their ability as executives to take
informed decisions and actions. As leaders in their firms they want to know
what is wrong, why it is wrong, and what can be done. In this book we
approach their questions systematically. We begin with the goals of the
organization; then we develop an understanding of the environment; exam-
ine the strategies; tease out the structure and the IT infrastructure; examine
the leadership style; observe the climate; and scrutinize the incentives. Using
xiii
xiv Preface

our experience and the science of good design, we analyze what works well
and what does not work well, or not at all. Good design fits together; poor
design has misfits and the organizational performance suffers. Design is the
diagnosis of misfits and the action to fix them.
What is a good way to read and use this book? The book begins with
organizational goals and builds up a comprehensive integrated model for a
good organizational design. Skip around, examine the figures and tables;
answer the diagnostic questions to get started. For whatever approach you
find comfortable, you should pick an organization such as your own firm to
use for analysis of, and reflection on, the concepts. Along the way, you
should diagnose the organization and think about the actions you want to
take to make your organization perform better.
Many have helped in a number of ways. There are our executive students
who provided the motivation. Over the last two years, Dr Katy Plowright, our
editor at Cambridge, has been the patient yet demanding task master; she has
been our anchor and our guide. Further, we have had the support of a number
of editors at Cambridge; they have been most helpful. Dorthe Døjbak
Håkonsson of the University of Southern Denmark and Min Li of the Fuqua
School read the penultimate draft and made many improvements. Karin Søby
of the Aarhus School of Business read and corrected the manuscript. Finally,
we want to thank our friends and families, who have been there when we
needed them most – all the moments between the blank screen and a book.

R I C H A R D M . B U RTO N
GERARDINE DESANCTIS
B Ø RG E O B E L
AU G U S T 1 2 , 2 0 0 5

On August 16, 2005, Gerry DeSanctis passed away. Gerry was charming,
patient, supportive and kind as well as disciplined, determined and strong
willed. She had passion for her students and was a complete teacher. We
were privileged to work with her in writing this book. She made us better,
both in our work and our spirit. It was a wonderful opportunity and an
experience that bonded our friendship.
Gerry had deep courage. She is our incredible colleague and beloved
friend. In her parting words, ‘‘I’ll always be with you.’’
R I C H A R D M . B U RTO N
B Ø RG E O B E L
SEPTEMPER 10, 2005
 An outline of the step-by-step approach

xv
Step 1 Getting started
1 Define the scope of the organization and
assess its goals

Introduction: The executive challenge of designing the

organization

In today’s volatile world, organizational design is an everyday, ongoing

activity and challenge for every executive, whether managing a global enter-
prise or a small work team. Globalization, worldwide competition, deregu-
lation, and ever-new technologies drive the ongoing reassessment of the
organization. The executive response has been many new forms of organi-
zational design: virtual, learning, modular, cellular, network, alliance, or
spaghetti – to name a few. New organizational forms challenge old ways
of organizing for efficiency and effectiveness. Yet fundamental design prin-
ciples underlie any well-functioning organization. Organizations still re-
quire a formal design. The fundamentals are: what are our goals? What
are the basic tasks? Who makes which decisions? What is the struc-
ture of communication, and what is the incentive structure? Fenton and
Pettigrew (2000, p. 6) state that ‘‘a closer inspection of the literature reveals
that many of the new forms are not entirely new but reminiscent of earlier
typologies, such as Burns and Stalker’s (1961) organic and mechanistic
forms and Galbraith’s preoccupation with lateral relations.’’ Thus fun-
damental concepts and principles of organizational design remain very
important for the modern organization of today and tomorrow.
To address the challenge of designing the organization we adopt a multi-
contingency view (Burton & Obel, 2004). Based on a large body of research,
this view says that an organization’s design should be chosen based on
the particular context, and further that the description of the context
should be multi-dimensional, including both structural and human com-
ponents. Structural components of organizational design include goals,
strategy, and structure. Human components include work processes,

3
4 Burton, DeSanctis, and Obel

people, coordination and control, and incentive mechanisms. Together,

these components provide a holistic approach to the organizational design
challenge.
Organizational design starts with the organization’s goals, and from there
we work from the top to the bottom, considering strategy, structure, process,
people, coordination, and control. This is a top-down approach to design.
We could start the design process using the reverse approach, that is, by
specifying how we want to coordinate and control work tasks and then
designing the organization from the bottom to the top, designing tasks
ahead of strategy; but such an approach would eliminate some possible
good designs because the tasks of the organization can be affected by its
goals and strategy. So we recommend a top-down approach that is comple-
mented by iterative incorporation of lower-level issues on the top-level
design. Political and implementation issues may suggest that the organiza-
tion be designed bottom-up. Here again the top-down approach may have
to be done in an iterative fashion, and further caution has to be exercised to
ensure that lower-level design and choice of tasks do not eliminate some
good alternative designs.

Overview of this book

In this book, we keep to the basics of organizational design. Organizational

design involves two complementary problems: (1) how to partition a big
task of the whole organization into smaller tasks of the subunits; and (2)
how to coordinate these smaller subunit tasks so that they fit together to
efficiently realize the bigger task or organizational goals. By complementary,
we mean that the smaller tasks must be defined and arranged in a way that
allows effective coordination. We consider these issues for ‘‘older,’’ classic
organizational forms as well as ‘‘newer,’’ modern organizational forms.
We present a step-by-step approach which is a ‘‘how to’’ method for
designing an organization. Each step and its subcomponents provide fun-
damental building blocks for any organization, and we guide you through
the process of assessing each building block as well as planning for change.
To simplify and show continuity in our approach, the components of each
building block are mapped onto a series of two-dimensional graphs that
clearly illustrate managerial options. The graphs are interlocking such that a
specific quadrant in any one graph corresponds to the same quadrant in all
5 Define the scope of the organization and assess its goals

other graphs. In this way, you can visualize the relationships among the
organizational design components and readily identify where there are
misfits in your organization’s design. Misfits are misalignments within the
organizational design components that can lead to deterioration in the
firm’s efficiency and effectiveness.
Misfits lead to a decrease in organizational performance, either today or
in the future. Misfits thus are the starting point for the implementation of
change. As such, misfits are the engine of the organizational design process.
If your organization changes in response to design misfits, rather than wait-
ing for financial or other performance problems to arise, goal attainment
is more likely to be achieved.
The graphs that we will provide for each design component will allow you
to visualize and plot the current location of an organization and then
identify the desired point to which you would like the organization to move.
In this way, you can see where you are and where you want the organiza-
tion to be in the organizational design space. While diagnostic questions and
the two-dimensional graphs give you an easy way to get an overview, the
ideas of the book have also been included in the OrgCon® software.1 This
software presents a more elaborate version of the approach presented in this
book and provides a set of analytic and graphical tools that will ease the
process of design. Meanwhile, you can use this book on its own, and the
software is not required to complete the step-by-step approach and design
your organization.
Organizational design is an ongoing executive process that includes both
short-term, routine changes, as well as intermittent, larger-scale changes. We
will address the dynamics of design, including misfit management, for both
routine and larger-scale changes in the context of organizational design
throughout this book.
Our step-by-step approach is based on an information-processing view of
the firm. This provides you with a framework and a toolbox for understand-
ing a wide range of organizations in product and service industries across
global boundaries. The approach helps you to interpret the history of
organizations, assess and redesign complex organizations of today, and plan
for the more information-rich organizations of tomorrow. We next describe
the information-processing view and then move on to defining the scope of
the organization and assessing its goals.

1
OrgCon can be obtained from www.ecomerc.com.
6 Burton, DeSanctis, and Obel

The information-processing view

The information-processing view uses the following logic. An organization

uses information in order to coordinate and control its activities in the face
of uncertainty where uncertainty is an incomplete description of the world
(Arrow, 1974, p. 34). By processing information, the organization observes
what is happening, analyzes problems, and makes choices about what to do,
and communicates to others. Information processing is a way to view an
organization and its design. Information channels ‘‘can be created or aban-
doned and their capacities and the types of signals to be transmitted over
them are subject to choice, a choice based on a comparison of benefits and
costs’’ (Arrow, 1974, p. 37). Both information systems and people possess a
capacity to process information, but ‘‘this capacity is not, however, unlim-
ited and the scarcity of information-handling ability is an essential feature
for the understanding of both individual and organizational behavior’’
(ibid.). Work involves information processing; individuals conduct in-
formation- and knowledge-based activities. They talk, read, write, enter
information in databases, calculate, and analyze. Various media are available
to facilitate information processing – from pens and face-to-face conversa-
tion to computers, networks and video meetings. Innovations in informa-
tion technology affect both the organization’s demand for information
processing and its capacity for processing information.
The step-by-step approach presented in this book is based on the fun-
damental assumption that the work of an organization can be seen as
information processing: observing, transmitting, analyzing, understanding,
deciding, storing, and taking action for implementation. These issues
may be labeled with other words like learning, tacit versus explicit know-
ledge, knowledge management, and data mining, but the basic idea is the
same. Organizations are information-processing entities. Therefore, we want
to design organizations so that they process information effectively and
efficiently.
The basic design problem is to create an organizational design that
matches your organization’s demand for information processing with its
information-processing capacity. Galbraith (1973, 1974), in his seminal
work, put it this way: ‘‘the greater the uncertainty of the task, the greater
the amount of information that has to be processed between decision
makers’’ (Galbraith, 1974, p. 10). Task (or work) uncertainty can arise from
a firm’s technology and the business environment in which the firm operates
7 Define the scope of the organization and assess its goals

(Thompson, 1967) as well as other sources. If the information-processing

demand comes from many routine and predictable tasks with an efficiency
focus, then formalization in the form of rules and programs can increase
the number of tasks that can be handled. As an example, an online retail
store in which the shopping and purchase process is rather routine can use
rules and programs to increase the number of customers it processes per day.
Task uncertainty is low, so the rules and programs are used to manage
exceptions. When there are uncertainties associated with the tasks, then
information processing is referred up the hierarchy to a level where an
overall perspective exists. This is the traditional use of exception-based
hierarchical decision making. Unfortunately, such hierarchical decision
making can handle only a limited amount of uncertainty. If the uncertainty
demands exceed the capacity of the hierarchy, then targets or goals have to
be set for the various tasks, making the tasks somewhat independent.
Coordination of work has moved from an efficiency orientation to an
effectiveness orientation. Organizations thus face a tradeoff: they can either
reduce their need for information processing or increase their capacity to
process information (Galbraith, 1974). These are the two managerial
options.
The first option is to reduce the organization’s need for information
processing by increasing slack resources. For example, if the organization
uses a just-in-time (JIT) inventory approach, which requires precise co-
ordination, then the organization might shift to having buffer inventory.
Buffer inventory replaces the need to process the information required for
JIT. As another example, information-processing needs can be reduced by
creating self-contained tasks that do not require coordination among them
in order to deliver the firm’s product or service. For example, a two-product
firm can create two self-contained single-product divisions that need not
communicate in order to meet their customers’ needs. Of course, this
strategy of reducing the need for information processing may incur high
opportunity costs from loss of coordination of interdependencies. Single-
product divisions may ignore interdependencies in production or
marketing, which may be costly in terms of lost opportunities. Thus, redu-
cing information needs must be balanced with the returns from coordinated
activities.
A second option is to increase the organization’s capacity to process
information. For example, in a hierarchical organization, the hierarchical
processing of information can be increased by investment in a vertical
information system. An information system may increase the speed and
8 Burton, DeSanctis, and Obel

amount of information that can be exchanged. The introduction of satellites,

information computer networks, the Internet and integrated CAD-CAM
systems can increase the information-processing capacity of the organiza-
tion. Upgrading the skills of the workforce, hiring more educated people
with broader abilities, using mobile communication devices, or holding
face-to-face meetings where people can share information are other ways
to increase information capacity. Information-processing capacity can also
be increased by creating lateral communications across the organization.
Direct contact, liaison roles, task forces and permanent teams are other
examples of strategies that will increase the firm’s information-processing
capacity.
The development of new information technologies, methods for organ-
izational learning and technologies for knowledge management require a
revisit of traditional strategies for managing a firm’s information-processing
capacity. Interactive information networks, multimedia systems, and gener-
ally the speed and amount of information that can be processed all have
served to increase the information-processing capacity of firms. At the same
time, the volume of information that firms must process continues to
increase. There are more things we want to know about our customers’
buying behavior, more research to be gathered for product development and
production, more details in the service we want to provide, and so on. So the
challenge of designing the organization in a way that best meets demands for
information processing remains.
Without doubt, organizations are information-processing entities, and
both the information-processing capacity and demands on firms have
surged as the cost of information-processing technology has decreased.
Along with this trend, there has been a reduction in slack resources in most
companies, a slight increase in use of self-contained units, a large invest-
ment in computer-based technologies, and a large increase in lateral com-
munications. All this has led to ‘‘leaner and meaner’’ organizations, less
inventory, less equipment, and fewer employees, particularly middle man-
agers. Those who remain use information much more quickly and effi-
ciently. This introduces the issue of information management by the
human resources in the organization. Many organizations have invested in
the technical side of knowledge-management and other information systems
without getting the benefits, often because the human side was neglected.
For this reason, we will emphasize the human side of organizational design
in our step-by-step approach.
9 Define the scope of the organization and assess its goals

Select an organization for analysis

Let us get started with our step-by-step approach. For the purpose of analysis
you should think about the definition of an organization in theoretical
terms. In such terms an organization can be defined as ‘‘a consciously
coordinated social entity, with a relatively identifiable boundary, which
functions on a relatively continuous basis to achieve a common goal or a
set of goals’’ (Robbins, 1990, p. 4). Thinking about your organization in
these terms will allow you to manage its design and not be overwhelmed
by the many, extensive set of activities involved in managing your orga-
nization every day. As you will see, this definition corresponds well to the
components in our five-step approach.
Now select a specific organization for your use throughout this book.
We will walk through the design of that organization in a step-by-step
fashion. The organization can be a team, department, division, an entire
company, or even a set of companies (such as a holding corporation or a
strategic alliance). Your choice of an organization becomes the unit of
analysis for the entire five-step design process. It is important to stick with
the same unit of analysis as we go through this design process. At the end of
each chapter we will state a number of diagnostic questions for you to
answer that relate to the organization you have chosen. Your answers to
the diagnostic questions will be the basis for the organization’s design.

Def ine the scope of the organization

Let us start with a brief explanation of how you should scope your organiza-
tional design problem. This is a necessary starting point for analysis. We use
the term ‘‘organization’’ or ‘‘firm’’ in the generic sense to refer to the team,
business unit, company, or larger enterprise. For most readers the organiza-
tion is a business firm, but the method we present applies to nonprofit firms,
partnerships, joint ventures, educational institutions, hospitals, churches,
government agencies – any type of organization in practically any kind of
setting.
As stated earlier, organizational design involves two complementary pro-
blems: (1) how to partition a big task into smaller subunit tasks, and (2)
how to coordinate these smaller subunit tasks so that they fit together to
10 Burton, DeSanctis, and Obel

efficiently realize the bigger task and organizational goals. The smaller tasks
must be defined and arranged in a way that allows effective coordination.
For example, the big task of General Motors or IBM is broken down into
divisions and departments. For a project team, the project task must be
broken into individual tasks. These smaller tasks are then integrated so that
the large corporation or project realizes the desired goals. In all organiza-
tions, these fundamental, complementary problems of breaking down big
tasks and putting smaller ones together are repeated again and again in
many forms.
You should think about the design process as a set of cascading organi-
zational design tasks where you go through the step-by-step process for each
task (see figure 1.1). Often the best place for you to start will be at the
corporate level: you should design the upper echelons first. Once that part
has been designed, move on to the next levels, which could be departments
or divisions, as we shall discuss in subsequent chapters. For example, you
first design the divisions in a divisional organization and then you deter-
mine how the divisions should be coordinated with one another. Each
division can be different from the other – one functional, another matrix.
In the cascading process, it is important to consider only one ‘‘organization’’
at a time; do not mix the design of the whole organization as a set of
divisions with the design of any one division. More formally, keep the unit
of analysis consistent. This process may be replicated in an iterative fashion.
The idea of equifinality (Doty et al., 1993) is that for a given situation there
may be more than one feasible design option from which to choose.
Therefore, you may have to go through the design cascade for more than
one option.

Figure 1.1. Levels in the organizational design process.

11 Define the scope of the organization and assess its goals

Assess the organization’s goals

You should start by assessing the relative importance to the organization

of two fundamental goals, efficiency and effectiveness. Efficiency is a primary
focus on inputs, use of resources, and costs. Effectiveness is a focus more
on outputs, products or services, and revenues. These are competing prior-
ities. Some organizations place a higher priority on efficiency, focusing on
minimizing the costs of producing goods or services. Other organizations
emphasize effectiveness, focusing on generating revenues or seizing leading-
edge innovation in the marketplace. All organizations value both efficiency
and effectiveness to some degree, but the question is: which is the dominant
priority? For example, no-frills airlines such as Southwest Airlines and Jet
Blue focus primarily on efficiency. Firms with significant R&D investment,
such as 3M Corporation or a biotech firm, focus primarily on effectiveness.
Some organizations focus simultaneously on high efficiency and high
effectiveness, such as Singapore Airlines and General Electric.
As shown in figure 1.2, efficiency and effectiveness are two dimensions –
not ends on a single scale. You should rate your organization on both goal
dimensions. In this two-dimensional model four different kinds of goal
states are possible. Quadrant A represents the organization with a relatively
low emphasis on both efficiency and effectiveness. It has little focus on using
resources well and it has few or no specific goals related to higher-level ideas
or targets. Such organizations exist and some even with great success. This
could be the case of a monopoly, or it could be an early start-up.

Figure 1.2. The goal space.

12 Burton, DeSanctis, and Obel

A firm in quadrant B has its focus on utilization of the smallest amount of

resources necessary to produce its products or services. Firms here continue
to do what they have done in the past, refining for continued improvement.
Such companies often exist well in stable environments where they can
defend their position with a low cost focus.
Firms in quadrant C are just the opposite. Here the organization has
higher focus on effectiveness but a lower focus on efficiency. This means that
the organization focuses on its goals, but takes less care to attend to the
efficient use of resources. This could be the case in highly volatile environ-
ments or in situations where the organization constantly develops new ideas
and has a first mover advantage and, as such, treats the costs of resources as a
secondary concern.
The final goal position is quadrant D, where there is an emphasis on both
efficiency and effectiveness. Firms in this quadrant confront competitive,
complex and volatile environments that require both product innovations
and low cost in order to compete successfully. Organizations in quadrant D
pursue the dual goals of efficiency and effectiveness with equal vigor.
The goal position of the organization affects its information-processing
requirements. For example, efficiency can be related to first-order learning,
which is ‘‘a routine, incremental conservative process that serves to maintain
stable relations and sustainable existing rules’’ (March, 1991). Effectiveness,
on the other hand, can be associated with second-order learning, where
existing rules are modified and new knowledge in the organization has to be
facilitated. Information requirements are much greater if the organization’s
primary goal is effectiveness rather than efficiency. Efficiency and effective-
ness also require different managerial approaches to environmental scanning
and incentives and thus demand different organizational designs. From this
discussion there seems to be a tension, or competition, between efficiency
and effectiveness.
Most executives want to obtain the right balance between efficiency and
effectiveness, and almost everyone agrees that modern organizations should
focus on both dimensions. But how do you obtain a balance? Some scholars
have argued that organizations focus on efficiency and effectiveness sequen-
tially by going through an evolutionary period with a focus on efficiency
disrupted by revolutionary periods of change where effectiveness is the focus
(e.g., Tushman & Romanelli, 1985). The balance is thus achieved over time
rather than simultaneously. Many managers, on the other hand, argue that
the efficiency–effectiveness foci are ongoing simultaneously, although the
emphasis can vary.
13 Define the scope of the organization and assess its goals

The way the balance is obtained is important. The solution to the balan-
cing could be that one subunit of the organization is efficient and another
effective; one subunit runs the current operations while another focuses on
innovation. But such an approach may not work. A well-known failure is
Xerox’s experience of placing its operations in Rochester, NY, and its re-
search at Xerox PARC in Palo Alto, CA. These were separate business units
that did not coordinate with one another. As a result, other firms, not Xerox,
brought the Windows-based operating system and the Ethernet network
protocol to the market. Although Xerox simultaneously achieved both
effectiveness and efficiency, the company failed to obtain the proper balance.
Recently, organizational scientists have argued that pursuit of efficiency
and effectiveness must be present everywhere in the organization at all times.
In a thorough study of ten multinational firms researchers found that
successful business units were able to simultaneously develop capacities
related to both efficiency and effectiveness (Gibson & Birkinshaw, 2004).
Put in terms of our diagram in figure 1.2, this means that quadrant D is the
ideal state. As we shall see, this is the most complex organizational design to
develop and maintain, and so not all firms are able to take this approach.
Many, many firms find themselves in quadrants B and C for this reason.
Nonetheless, if your organization can be both highly effective and highly
efficient, then you are in the best position to compete successfully in the
marketplace if you are facing a highly volatile environment.
To summarize, the choice of a goal state in relation to efficiency and
effectiveness has profound consequences for the information-processing
demands and capacity of an organization. The efficiency–effectiveness goal
state for your firm significantly affects your choice of the proper organizational
design.

Diagnostic questions

To begin the organizational design process, choose the unit of analysis and
keep that fixed throughout the step-by-step method: the top management
layer of a large firm, a small firm, a division within a large firm, a depart-
ment or a project. Thus, we advise starting with the whole firm by taking a
cascade approach from top to bottom to obtain a complete analysis. Start at
the executive level of the organization, go through the five-step design
process, and then repeat the process for each major department or business
division. You may have to iterate more than once. Of course the task of your
14 Burton, DeSanctis, and Obel

design approach may not be the total firm. But our advice is to start at the
top of the unit you are considering.
Next, assess where the organization is located on the efficiency/effective-
ness diagram of figure 1.2. Write down the arguments for the location using
the vocabulary of your organization. You will need that later when you
consider making changes. Answer the questions below.
1. What is the unit of analysis for the step-by-step approach?
2. What does the organization do? What is its major work activity?
3. How does the organization score on efficiency?
1 2 3 4 5
very low moderate very high
4. How does the organization score on effectiveness?
1 2 3 4 5
very low moderate very high
5. Plot the organization in the efficiency/effectiveness graph of figure 1.2.
6. Where would the organization like to be in the efficiency/effectiveness
graph of figure 1.2?

Misf its and balancing competing design dimensions

As the last two diagnostic questions indicate, the organization design process
consists of two important questions: Where are you, and where do you want
to be? With regard to organizational goals, there are two things for you to
consider about your unit analysis. First, where is the firm in figure 1.2?
Second, where would the organization like to be in this design space?
Let us use figure 1.3 to think through these questions. Suppose that the
organization is currently at point C in the diagram. Your focus is on
effectiveness. Suppose that the competitive environment has changed such
that the firm now must compete more on efficiency. Thus you might desire
to move the organization to the quadrant of point D. However, before
making this change, a more comprehensive review of the organization’s
design is needed. You need to diagnose the consequences of such a change.
This means working through the five steps in our organizational design
approach and determining where each major design dimension is located in
the two-dimensional organizational design space. For example, it may be
15 Define the scope of the organization and assess its goals

Figure 1.3. Making changes in the efficiency/effectiveness space.

that the organization’s structure and work processes, for the most part, lie in
the quadrant of point C (thus explaining your firm’s success in achieving
effectiveness). Suppose the competitive environment and business strategy
(which we will consider in Chapters 2 and 3), on the other hand, lie in the
quadrant of point B. As a result, there are misfits in the organizational
design. That is, the design components do not all lie within the same
quadrant. To address the misfit problem, you have a choice: either move
the structure and work processes toward the quadrant of point B (thus
aligning the organizational dimensions together in the same area of the
design space), or change all of the design dimensions such that they move
toward point D. The latter is a much more significant management change
than the former, and you should carefully evaluate the implications of this
design option before deciding on a plan. Our step-by-step approach will
allow you to assess the consequences of various change strategies and their
effects on goals, strategy, structure, process and people, and coordination
and control.
As noted earlier, the quadrant associated with point D is an ideal location
in the organizational design space. Indeed, much of the managerial hype of
the day suggests to managers that all firms should be located in this place.
But the organizational design space of point D is more costly than a singular
focus of either efficiency or effectiveness and so may not be appropriate for
all firms. Balance is a key theme of this book; organizational design entails
developing design dimensions that are in alignment, thus avoiding misfits
that lead to performance decrement.
Several studies (Burton et al., 2004; Burton & Obel, 2004) have shown
that proper alignments of an organization’s design indeed result in superior
performance. In many instances, this means operating within the quadrants
16 Burton, DeSanctis, and Obel

associated with points B and C and developing organizational design com-

ponents that support an acceptable tradeoff between the dual goals of
efficiency and effectiveness. Though the quadrant associated with point D
may be ideal, it is not always the most suitable goal for management due to
design constraints. Organizations that operate within the quadrants associ-
ated with points B and C can be extremely successful. Only quadrant A is to
be avoided in the long run unless the firm operates in a highly protected
environment, is a very small organization, or is living through the early
period of a start-up venture. Organizations that find themselves in quadrant
A of the design space usually should plan for change, and our five-step
approach, as we shall see, can help identify what the needed changes are and
how to proceed.
Again, many executives may wish their organization to be high in both
efficiency and effectiveness. This is possible and may be desirable, but this
design space is difficult to develop and maintain, especially if the organiza-
tion’s design components currently lie outside of this quadrant. An organ-
ization can more likely move itself toward point D if it has other design
factors that fall into this same quadrant.
It is important for you to work through all of the steps of our design
process and their subcomponents to determine a good fit among the many
components of your organization’s design. A partial approach, completing
only some steps but not others, will be suboptimal. For example, if you
assess the organization’s strategy but not its processes or coordination, you
cannot see what is necessary for the strategy to be effectively realized. Only
when the picture of the organizational design is complete does it become
meaningful.

Summary

This chapter introduced our step-by-step approach for organizational

design based on the multi-contingency approach (Burton & Obel, 2004).
We discussed the scope of the design process, which includes the choice of
the unit of analysis and deciding where your organization is and/or would
like to be located on the efficiency/effectiveness diagram. The basic idea of
viewing the organization from an information-processing perspective was
presented. Further, a series of questions that you should answer for the
organization (unit of analysis) have been provided as the starting point for
the organization’s design.
17 Define the scope of the organization and assess its goals

Glossary

Effectiveness: an organization’s goal priority that contrasts with efficiency; a

focus on outputs, products or services, generating revenues, or seizing
leading-edge innovation in the marketplace.
Efficiency: an organization’s goal priority that contrasts with effectiveness; a
focus on inputs, use of resources, and costs, especially minimizing the
costs of producing goods or services.
Fit: organizational design components that all lie within the same quadrant,
thus balancing the firm’s efficiency and effectiveness. Further, the
information-processing capacity of the firm is balanced with the demand
to enhance performance.
Misfits: organizational design components that do not all lie within the same
quadrant, thus threatening the firm’s efficiency and effectiveness.
Organizational design: the complete specification of strategy, structure,
processes, people, coordination and control, and incentive components of
the firm.
Uncertainty: an incomplete description of the world.
Unit of analysis: the organization that is being designed, whether a team,
business unit, department, division, firm, or larger enterprise; the unit of
analysis must be held constant throughout the step-by-step design
process.
18 Burton, DeSanctis, and Obel

Where are you in the step-by-step approach?

STEP 1
Getting Started
(1) Goals
So far, you have specified the scope of the organization (your unit of
analysis) and the goals in terms of efficiency and effectiveness. Before
continuing, review your firm’s current position in figure 1.3 very carefully
as this position will be the anchor point as you proceed through the next
steps. Does the position correspond well to the vision and mission
statements of your organization? Is there general agreement in the
organization about this position? Are you located in figure 1.3 where you
want to be, or are you in a position that you would like to change? That
is, do you want to change the goals of the organization so that it is
located in a different place in figure 1.3? Please keep in mind the two
situations: where the organization is and where you want the organiza-
tion to be. As we proceed through the upcoming steps, you will see
whether your organization’s design fits where you are, or not, and
whether it fits where you would like the organization to be, or not. Before
you proceed to step 2, make sure that you have decided whether your
analysis is for the current or a future situation and be consistent in your
analysis throughout the book. You may do the analysis for both situations
and make comparisons.
Once you have reviewed your organization’s scope and goals, then you
are ready to go on to step 2.
STEP 2
Strategy
Next, you will examine your firm’s strategy and the firm’s environ-
ment.
(2) Strategy
(3) Environment
19 Define the scope of the organization and assess its goals

STEP 3
Structure
STEP 4
Process and People
STEP 5
Coordination and Control
Step 2 Strategy
2 Strategy

Introduction

In the previous chapter you described an organizational unit of analysis and

its goals. The next step in our step-by-step approach is to describe the
strategy of the organization and the environment in which it operates. In
this chapter we focus on strategy. A firm’s strategy1 helps determine its or-
ganizational design. Our focus here is on describing your firm’s strategy –
not how or why the choice of a strategy is made. Chandler (1962) stated
the fundamental relation in his now famous dictum, ‘‘structure follows
strategy.’’ That is, given a strategy, there are some organizational structures
which can implement that strategy better than others. Strategy is the end;
structure is the means. Strategy is the operationalization of the firm’s goals of
efficiency and/or effectiveness; and the structure is the means to achieve
them. In this chapter we will discuss which strategy should be pursued to
obtain the goals decided in Chapter 1. This has to be done taking into
account the environment in which the firm operates. Thus, ‘‘achieving high
performance in a business results from establishing and maintaining a fit
among three elements: the strategy of the firm, its organizational design, and
the environment in which it operates’’ (Roberts, 2004, p. 12).
A firm’s strategy reflects management’s assessment of the firm’s situation
and its choice of how to pursue the firm’s goals. Strategy can be described in
a number of ways. For example, strategy can be described in terms of five
forces of the firm’s economic situation (Porter, 1985): suppliers, buyers,
substitutes, potential entrants, and the rivalry among existing competitors.
These five forces yield three possible strategies: cost leader, product differen-
tiator, or niche player. In marketing, strategy can be described as the choice

1
We use firm in this chapter as a matter of convenience, but the analysis applies to any unit
of analysis, be it a department or a team.

23
24 Burton, DeSanctis, and Obel

of the four P’s: product, price, promotion, and place, That is, which product
should the firm produce; what is its price; how should it be promoted and
advertised; and how should it be distributed (Kottler, 2000). On the input
side of the firm, operations strategy is the choice of the firm’s supply chain
including the chain management and outsourcing. This involves the choice
and management of the resources and capabilities (Makadok, 2001). Your
choice of a strategy for your firm is always a question of what the firm
should do in its situation to meet its goals of efficiency and effectiveness.
A simple and powerful way to describe a firm’s strategy is in terms of
the following typology (Miles & Snow, 1978): (1) reactor, (2) defender,
(3) prospector, and (4) analyzer without or with innovation. This typology
has proved to be very robust and is frequently used today (Hambrick, 2003).
The dominant strategic approach is reflected in such actions as capital
investment, concern for quality, price level compared to competitors, pref-
erence for product innovation, and preference for process innovation. Many
factors make up a firm’s strategy, but the most important thing is its
approach to innovation, whether it exploits its current situation and
whether it adopts the strategy of exploring new innovations (March, 1991).
Exploration includes search, variation, risk taking, and innovation. Ex-
ploration is the process of seeking new technologies or new ways of doing
things. Exploitation includes refinement, efficiency, selection, and imple-
mentation (March, 1991). Exploitation is taking advantage of current or
known technologies to do things in a new or novel way. Originally, explor-
ation and exploitation were developed to analyze organizational learning
and the nature of knowledge, both of which are related to firm strategy.
Strategy is the application of knowledge, and learning is a change in the
knowledge base to develop new strategies. Fundamentally, strategy choice,
knowledge usage and learning are all concerned with how the firm chooses
which actions to take based upon limited information.
Exploration and exploitation are dimensions of strategy that can be used to
form the basis for categorization of a firm’s strategy into one of four types
(Haakonson et al., 2004). If your firm is a reactor it is low on both exploration
and exploitation; it lacks an intentional strategy toward innovation. It makes
adjustments when forced or when there is an urgent need or problem. If your
firm is a defender it is high on exploitation and low on exploration; it is
innovative only in narrow, limited areas. Its innovation is confined and highly
focused. If your firm is a prospector it is high on exploration and low on
exploitation; it takes an aggressive approach to innovation, systematically
searching for new opportunities. It experiments regularly with change.
25 Strategy

Figure 2.1. The strategy space.

Analyzers take a mixed mode approach to innovation. If your firm is an

analyzer without innovation your strategy is similar to a defender but with
more emphasis on exploration. If your firm is an analyzer with innovation
your strategy is similar to a prospector but with more emphasis on exploit-
ation. Figure 2.1 displays the strategic space of exploration and exploitation
along with the four basic types of strategy that relate to these dimensions.
Now let’s consider these strategic types in more detail. In figure 2.1 , begin
in the lower left corner, move to the upper left, then to the lower right and
finally to the upper right corner. This is a convenient way to compare and
contrast the strategies. Although the descriptions and examples given here
are provided for the firm as a whole, we emphasize that strategies can be
used to describe a business unit, division, department, or team – that is, for
smaller units of analysis. In the case of smaller units, exploration and
exploitation must be considered relative to other, equivalent units of analysis
(e.g., other departments or teams, whether inside or outside the firm) that
compete with your unit for success in the ‘‘marketplace.’’

Reactor

The reactor is neither an explorer nor an exploiter of the firm’s opportun-

ities. Generally, the reactor strategy is neither efficient nor effective in terms
of achieving the firm’s goals. The reactor acts without a focus on exploration
or exploitation; instead, the reactor tries to adjust to the situation after it is
possible to capture any opportunities that may have been present but are
now lost. There is no innovation. The executive does not systematically
anticipate, plan, or project into the future. The organization does not take
26 Burton, DeSanctis, and Obel

a deliberate position to become efficient or effective. At the other extreme,

the reactor may be a dreamer that pursues innovation without any focus; the
dreamer is neither efficient nor effective.
If your firm is a reactor, you make decisions based largely upon bad news
as it becomes known to you, whether this is poor performance such as
decreased profits or earnings, events such as a loss of a major customer, or
internal problems such as conflicts or inappropriate utilization of resources.
Problems emerge as surprises and are dealt with as they occur. A reactor
strategy is often observed in organizations that are in transition. A good
example is a firm after a merger, where the focus is on the internal reorgan-
ization and management power struggles; the interest in both exploitation
and exploration can be lost.
If the reactor strategy is followed for the long term your firm ultimately
will be caught with more bad news, most likely in the form of poor
performance in the marketplace or internal processes that cannot be man-
aged in the available time. The organization will go out of existence as it
will eventually not be able to obtain sales or capital and perhaps human
resources. Thus, it is difficult to give examples of existing companies that
can be categorized as following the reactor strategy. They often die if
they have a reactor strategy for too long. Inefficient and ineffective govern-
ment organizations and firms in the midst of bankruptcy are known for
following reactor strategies. The one-time computer giant Digital Corpor-
ation could in its later years be categorized as a reactor. It basically did not
develop new technology after its famous VAX computer, and it completely
missed the entrance of the PC into the market. It reacted too late with a too
high cost structure. The pieces of Digital that are left are now part of
Hewlett-Packard.

Defender

Now move to the upper left corner of figure 2.1. If your firm has a greater
focus on exploitation than on exploration, then your strategy is a defender.
The defender is high on exploiting its resources and situation but low on
exploring anything new or being innovative.
The executives inside a defender firm are focused on keeping the organ-
ization’s position in the market. There is less emphasis on developing new
ideas, products or services. Instead, there is an emphasis on maintaining a
competitive position, which may be measured in terms of market share or
27 Strategy

profitability. Sales forecasting, as an extrapolation of the past, is a frequent

tool used to support the defender strategy; that is, the past is projected as the
future. Plans are developed to keep the position and fend off competitors, or
at least keep them from encroaching into the well-defined territory of the
defender. The defender usually has very competitive prices or a product
niche that it works hard to keep others from penetrating.
The defender maintains its position by being efficient in the utilization of
resources. This strategy keeps the defender invulnerable to less efficient
competitors. The defender can make changes in existing processes for
existing products and services, but the goal is to be efficient and maintain
its position. Thus the defender firm focuses on process innovation and has
efficiency as the primary goal, as discussed in Chapter 1. On the other hand,
effectiveness is low (relatively speaking) in a defender firm.
If your firm is a defender you will find that you cannot change much or
change quickly. A high capital requirement is often a barrier to entry in the
defender’s industry. Thus there is a steady strategy of repeatedly doing the
same thing efficiently. The emphasis on quality may be high as a means to
prevent new entrants from coming into the market or to prevent existing
competitors from taking over the firm’s market share.
The defender can do well for a long time. Its vulnerability comes when its
products or services are no longer desired in the market. Another threat is if
new technology reduces high capital requirements, thus allowing competi-
tion from new entrants. Similar threats may come from new regulation or
deregulation, as we have seen in industries such as transportation and
telecommunications.
A defender is slow to make significant change. When buyers stop pur-
chasing its products, the defender is not in a good position to develop new
products or new markets. Slowness to change combined with a high focus
on efficiency makes the defender vulnerable in the long run.
One good example is LEGO. It has kept its focus on toy bricks for decades.
Prior to 2000 LEGO was making a handsome profit and was a growth
company for many years. It protected its position with aggressive marketing,
defending patents, copyrights and trademarks, and with an ongoing process
of automating the production as much as possible. During the period 2000–
2004 LEGO struggled to make changes to meet new demands for more
electronically based toys, but it was not well situated to move from adopting
a defender strategy to a more innovative strategy. Another example is Coca-
Cola, Inc., which has been defending its COKE brand fiercely, investing
more resources in defensive moves than in new product development.
28 Burton, DeSanctis, and Obel

Prospector

The prospector is located in the lower right corner of figure 2.1. The
prospector is high on exploration of its opportunities but low on exploiting
its current situation. Thus if you have a very high focus on exploration but a
very low focus on exploitation, then your firm’s strategy is a prospector. The
prospector focuses on innovation of new things to the detriment of being
efficient and exploitative of existing opportunities. It searches continually
for new market opportunities and experiments regularly with new ideas,
new technology, and new processes. The prospector firm is the creator of
change, and so other firms must adjust to its actions. But the prospector is
not much concerned with exploiting its situation or developing efficiencies
in its use of resources.
The prospector maintains its competitive position by being new and
making changes to the competitive situation that others must adjust to.
The prospector constantly questions the status quo, and this puts it in its
own niche and allows it to enjoy first-mover advantages. The prospector can
make large changes in products and services much more readily than the
defender. It is driven by the effectiveness goal as discussed in Chapter 1. The
prospector’s efficiency is low, but due to the competitive situation it may
survive with a higher cost structure by demanding high prices.
The prospector firm is change-oriented, preferring the new over the status
quo. Quality is not the primary concern, nor is being price competitive.
Instead, these issues are subsumed by the novelty of new products or
services. Of course, quality and price will become important to the customer,
especially when other firms enter to match the prospector’s latest innov-
ation. The prospector’s reaction is to seek a new product or service, thus
leading the market in innovation. A new product or service is continually
required for the prospector to prosper.
This strategy is risky. On the upside, a new product or service may have
enormous payoffs. On the downside, the prospector firm can quickly ex-
haust its resources since usually it operates within a limited time frame for
success; in other words, new product (or service) development life cycles
must be relatively short. The prospector can do well for a long time. Its
vulnerability comes when it fails to innovate and provide new products and
services to the market within life cycle demands.
Many start-up ventures, for example, in the biotechnology industry today
can be categorized as prospectors. 3M is often cited as having a prospector
29 Strategy

strategy with its constant development of new and innovative products.

Google is taking a similar tack with its continual innovation of new
Internet-based information services.

Analyzer without innovation

If you have a strong focus on exploitation and weak focus on exploration,

then your strategy is an analyzer without innovation. The analyzer without
innovation is very similar to the defender except that it does have a passive
innovation strategy or a copy strategy. The analyzer without innovation
looks to what other firms are doing that is successful and then imitates with
similar products or services to meet customer needs. The analyzer without
innovation is high on exploitation of its resources and situation but very
moderate on exploration while copying innovations of others. It is the upper
middle of figure 2.1.
Similar to the defender, the analyzer without innovation is focused on
keeping the organization’s position in the market; meanwhile, it is attentive
to what others are doing. Some firms are quite good at being quick followers
by observing what others do successfully and then moving quickly to do the
same, or very similar. This strategy can be in part defender as the follower
moves quickly to keep its position; or it can be more projective where the
firm actually goes beyond what the originator did. In either case, it is an
imitation or copying strategy where it is important to recognize what others
are doing successfully and then follow.
The analyzer without innovation maintains its position by being efficient
in the utilization of resources and following others; this maintains a degree
of invulnerability to the moves of competitors. The analyzer without innov-
ation can make small changes in existing processes for existing products and
services, but the purpose is to be efficient and maintain its position. It can
innovate in its defense of market position by following others if necessary.
The analyzer without innovation primarily pursues the efficiency goal as
discussed in Chapter 1. Its effectiveness goal is moderate.
Usually the analyzer without innovation does not change much, but the
point is that it can change by following others. The analyzer without
innovation is adept at doing much the same thing efficiently, but with a
few changes from time to time.
The analyzer without innovation can do well for a long time, particularly
by following what others do with aggressive pursuit. The challenge for the
30 Burton, DeSanctis, and Obel

analyzer without innovation is to select well what to follow. Its vulnerability

comes when it follows the wrong trend or fails to imitate quickly enough,
so that its products or services are no longer desired in the market. If
customers stop purchasing its products or services, the firm that pursues
this strategy is not in a good position to develop its own new products or
new markets. Because it is not easy to follow quickly, the firm that selects the
analyzer without innovation strategy must be organized to detect and
imitate quickly.
In the fashion industry you see numerous analyzers without innovation.
They go to Paris to spot the trend of the year and then copy the Haute
Couture in a mass-market fashion. Magazine and television productions
often do the same thing. They succeed by aggressively and adeptly imitating
the latest approach to finding customers that have been established by the
market leaders.

Analyzer with innovation

If your firm has a focus on both exploration and exploitation, then your
strategy is an analyzer with innovation. The analyzer with innovation is a
dual strategy combining aspects of both a defender and a prospector. It
exploits its current position of resource utilization and market position, and
at the same time it adopts an active innovation strategy of developing
new products, services, and their delivery processes. The analyzer with
innovation is both efficient and effective. It is the upper right of figure 2.1.
The analyzer with innovation is active in exploration as it purposefully
innovates and searches for new products and services. It goes beyond just
looking at what others do and instead surveys more widely in technology
and markets to look for opportunities that it can develop into new products
and services. Some firms have a market-driven approach to innovation as
they look at market or customer needs and then try to innovate to meet
those needs. They may limit themselves to markets they know well, or they
may look for new markets as well. Other firms take a more technology-
driven strategy in which they invest in more basic technology and try to
capitalize on the results of the technological developments.
If you pursue the strategy of analyzer with innovation, you must have a
dual focus on defending your firm’s position in its markets while at the same
time innovating with new products and services. This is a difficult balance
31 Strategy

requiring great skill and managerial expertise. The firm must emphasize
developing new ideas, products, and processes. Success means producing
new products and services on a regular basis. At the same time there is an
emphasis on keeping the firm’s position in terms of market share or profit-
ability. Plans are developed to defend the firm’s position and fend off
competitors while at the same time exploring new frontiers. The analyzer
with innovation requires the best, most complex of organizational designs.
There is a downside to this strategy. The analyzer with innovation is
vulnerable in that it can fail to maintain the combination of exploration
and exploitation needed to keep its existing markets for short-term eco-
nomic performance and innovation in new products to meet future oppor-
tunities. The dual goals of efficiency and effectiveness create conflict that
must be cleverly reconciled within a flexible and robust organizational
design.
Xerox might be categorized as one of the less successful analyzers with
innovation. It has both exploitation and exploration in its strategy, but these
strategies are pursued separately, by different divisions located very far apart
with different management structures. Despite the fact that Xerox has
developed very novel and innovative products over the years, the firm has
rarely managed to successfully transfer new ideas from development to
production to sales. A successful analyzer with innovation strategy requires
both exploitation and exploration embedded in the whole organization as
part of the design (Gibson & Birkinshaw, 2004).
IBM seems to be a more successful example. The firm invests in new
product development and new forms of service offerings but constantly
looks at what is going on in the market. One example of this is the use of
the Linux operating system in many of their new products and systems,
as well as their recent acquisition of a major consulting firm. These moves
are undertaken with great care, following careful analysis of market trends,
but they are aggressive, risk-taking moves that bring innovation into the
company.

Diagnostic questions

Now continue with the analysis of your chosen organization. Recall that
in Chapter 1 you chose a unit of analysis and assessed its goals on the
dimensions of efficiency and effectiveness. Locate this same organization
32 Burton, DeSanctis, and Obel

(the same unit of analysis) on the exploration–exploitation dimensions of

figure 2.1. Then you can categorize the firm’s strategy as a reactor, defender,
analyzer without innovation, analyzer with innovation, or prospector.
Answer the diagnostic questions below. By working through these diagnostic
questions, you can locate where the firm is in the strategy space.
1. What is the unit of analysis that you chose in Chapter 1? Use this unit of
analysis as the organization when answering the questions below.
The questions below will help you locate your organization on the
exploration and the exploitation dimensions. For each item within
question 2 and question 3, use a 1 to 5 rating scale to score your chosen
organization as follows:
1 2 3 4 5
very low moderate very high

2. Exploration:
a. How innovative are the organization’s products, (1)–(5)? ___
b. What is the price compared to the value of the product, (1)–(5)? ___
c. What is the price level compared to the quality level, (1)–(5)? ___
d. How frequently does the firm develop new products (1)–(5)? ___
e. How difficult is it for other firms to develop related products?
(1)–(5) ___
Now mark the organization’s location on the exploration axis in figure
2.2.2 If the score you gave is greater than 3, then the organization is high
on exploration. If the score you gave is less than 3, then it is low on
exploration.
3. Exploitation:
a. What is the organization’s degree of process innovation, (1)–(5)?___
b. What are its prices compared to the competition (1)–(5)?___

2
The detailed questions that we include throughout this book come from either research
instruments used to measure the particular concept or a dissection of the definition of the
concept. The detailed questions will help you focus on how to score exploration and
exploitation. We suggest that you use an averaging or weighted averaging procedure of the
detailed scores to get to the overall score. If you do not agree with the average, then you may
use your own judgment to adjust it. You may use other detailed questions to get the score if
that fits your chosen firm or industry better. For example, in some industries the number of
patents is used to compare the degree of innovation across firms, so if that is appropriate for
your firm, that question could augment or replace a question listed here.
33 Strategy

Figure 2.2. Locate your organization in the strategy space by rating the levels of exploration and
exploitation.

c. What is its quality in terms of its standardization and reliability

(1)–(5)?___
d. What is the number of products the organization has compared to its
competitors (1)–(5)?___
e. What are the barriers to entry in its industry (1)–(5)?___
Now mark the organization’s location on the exploitation axis in figure 2.2.
If the score you gave is greater than 3, then the organization is high on
exploitation. If the score you gave is less than 3, then it is low on exploitation.
4. Now with these values of exploration and exploitation, locate the
organization on the graph. What is its strategy?
Next we want you to examine how your organization’s strategy fits with
the goals that you chose in Chapter 1.

Fit and Misf its

What goals fit well with your strategy?3 If the organization’s strategy does
not align with its goals, then there are misfits that need to be addressed.
What are the misfits and what can you do about them? As an example,
suppose your chosen organization has a defender strategy which has a focus

3
The balancing of the information-processing demand with capacity has been discussed in
the research literature in terms of fit and misfit (Venkatraman, 1989; Burton et al., 2002).
34 Burton, DeSanctis, and Obel

Table 2.1. Fit between strategy and organizational goals

Corresponding
quadrant in
organizational
design space A B C D
Strategy types Reactor Defender Prospector Analyzer Analyzer
with without
innovation innovation

Organizational Neither Efficiency Effectiveness Efficiency and effectiveness

goals

on exploitation. This strategy is a fit with the efficiency goal. Now go back to
Chapter 1 to verify that your goals were primarily efficiency. If so, your
organizational strategy and goals fit. There is no misfit. On the other hand, if
your goals do not fit the defender strategy, then you have a misfit which
calls for some change to bring your organization into alignment. Suppose
your goals are to be both efficient and effective. Then a better strategy would
be an analyzer, either with innovation or without innovation. Now you have
a choice either to change your organization’s goals to fit the defender
strategy, or to change the strategy to an analyzer to fit your goals.
Table 2.1 shows the mapping of strategy types and organizational goals.
These correspond to the four quadrants of our organizational design space.
Each of the four columns in the table marked A, B, C, and D shows
situations of fit among the strategy and goal components of the organiza-
tional design space. In other words, for an organization to have good fit,
your strategy and goals should fall within the same column in this table.
First, take the strategy you identified in this chapter as given and see what
matching goal is acceptable. If your goal matches your strategy, then the
organization has no misfits; however, if strategy and goals are not in the
same column of the table, then there is a misfit between the organization’s
strategy and goals, and one or the other should change in order to bring the
organization into alignment. What would be required by your organization
to adopt a revised goal?
Second, try the reverse approach. Take the goal you identified in Chapter 1
as given and see what matching strategy is acceptable, given the information
in table 2.1. If there is a misfit, what would be required by your organization to
adopt this revised strategy? For example, is a greater emphasis on exploration
35 Strategy

feasible? Or a switch from exploitation to exploration? Think through the

possibilities in light of your firm’s particular situation.
Finally, think about whether it would be preferable for your organization
to change the strategy or change the goal. How would you implement such
changes? As you move ahead in the book, we will develop a more complete
picture of your organization and examine alternatives for managing and
changing your organization as needed.
In later chapters, we will add organizational design alternatives which will
expand the table presented here. The organizational design space will
become more complex, but you will be able to develop more complete
and better alternatives for achieving an organization’s goals. In the next
chapter, we will complete Step 2 by examining the environment of the
organization.

Summary

This chapter has continued our step-by-step approach by examining the

strategy of an organization. You first described the strategy of an organiza-
tion in terms of degree of exploration and exploitation, categorized as:
reactor, defender, analyzer without innovation, analyzer with innovation,
or prospector. You then examined the fit of the strategy with your goals as
assessed in Chapter 1. If there are misfits between the organization’s strategy
and goals, you should think about actions you might take to either adjust
the strategy or adjust the goals so that these can be aligned.
In the next chapter, we will examine the organizational environment.

Glossary

Analyzer with innovation: a strategy that is similar to a prospector but with

more emphasis on exploitation.
Analyzer without innovation: a strategy that is similar to a defender but with
more emphasis on exploration.
Defender: a strategy that focuses on exploitation and innovation only in
narrow, limited areas.
Environment: the marketplace, the regulatory and legal situation, the
opportunities and other aspects of the context in which the firm operates.
Exploration: search, variation, risk taking, and innovation by the firm.
36 Burton, DeSanctis, and Obel

Exploitation: refinement, efficiency, selection, and implementation by the

firm.
Organization: a collection of people identified socially as a firm or one of its
subunits; deliberately formed, goal-directed, bounded, and functions on a
relatively continuous basis.
Prospector: a strategy that takes an aggressive approach to innovation,
systematically searching for new opportunities. It experiments regularly
with change.
Reactor: a strategy that lacks an intentional strategy toward innovation. It
makes adjustments when forced or when there is an urgent need or
problem.
Strategy: the firm’s position on exploration and exploitation.
Structure: the partition of tasks by work roles and the reporting
relationships among the work roles.
3 Environment

Introduction

In our step-by-step approach, you have described the goals and strategy for
your unit of analysis. In this chapter, we focus on the environment that
surrounds an organization. The environment is everything outside the
boundary of the organizational unit of analysis. When you think about the
environment for a firm, think about what could have an effect on the way
the organization performs. It could be: its customers, its competitors, its
suppliers, the financial market, or the political system. If your unit of
analysis is a department, then the other departments in the organization
are a part of its environment. The environmental imperative states that the
organizational design is determined in large part by the environment of an
organization. In brief, the environment is for the most part given for a firm,
and the firm should then adjust its design to fit the environment. The
performance of the firm depends upon how a firm makes the organization fit
with the environment. Scott (1998) calls this the rational view of organizing.
This also fits with the open systems view of organizational design:
Open systems theory can be defined as a theory of organization that views
organizations not as simple ‘‘closed’’ bureaucratic structures separate from
their surroundings, but as highly complex entities, facing considerable
uncertainties in their operations and constantly interacting with their envir-
onment. This system also assumes that organizational components will seek
‘‘equilibrium’’ among the forces pressing on them and their own responses
to their forces. (Milakovich & Gordon, 2001, p. 165)
There are a number of ways to describe an organization’s environment.
Early on, Ashby (1956) described the environmental variety as the number
of distinct elements, whereas Lawrence and Lorsch (1967) used uncertainty.
Burton and Obel (2004) used a four-dimensional description: complexity,

37
38 Burton, DeSanctis, and Obel

which is the number of factors in the environment and their interdepend-

ency; uncertainty, which is the variance among the factors; equivocality,
which is the ignorance and confusion about the existence of some factors;
and hostility, which is the extent of malicious external threats. These four
factors are modifications of earlier descriptions. Lawrence (1981) began with
four descriptors: instability, ignorance of data, number of variables, and
interdependence of variables. He then reduced these four to two: unpredict-
ability and complexity. Duncan (1972) used environmental change or dyna-
mism, and environmental complexity. Later, Bourgeois and Eisenhardt
(1988) defined a high velocity environment as one where changes are rapid
and discontinuous so that information is often unavailable. Recently,
Siggelkow and Rivkin (2005) also described the environment in terms of
turbulence and complexity.
Among all of these descriptions, there are some common aspects. First,
they are general properties of an organization’s environment, not a detailed
listing of all of the elemental factors themselves. Second, the measures are
perceptions made by the management of the firm; they are not necessarily
objective. This does not mean they are inaccurate, but that management
creates its own understanding of the environment and the implications for
design. Third, whatever the particular environmental description used, the
environment is a large determinant of the organizational design, i.e., the
environmental imperative means that the environment is a major determin-
ant of how an organization should be designed. Contingency theory and the
principle that structure follows strategy discussed in Chapter 2 – all follow
the common theme that there must be a fit between the environment and
the organization.
The environment creates both limits and opportunities for a firm’s strat-
egy and, subsequently, its structure. Lawrence and Lorsch (1967), for
example, argued that increased environmental uncertainty leads to increased
organizational differentiation. They define differentiation to mean that an
organization has departments that are different in both tasks and orienta-
tion. Lawrence and Lorsch studied three well-defined industries that they
categorized as ranging from low to high uncertainty. They found that
increased uncertainty in the environment required increased differentiation
in the organizational structure in order for the organization to be efficient.
Then integration is required to make the different departments work in
coordination. Integration devices typically include rules and procedures,
configurational plans, the authority of the hierarchy, and decision-making
committees.
39 Environment

The way you should define the environment is in terms of what you know
affects your organization. If a firm is a monopolist, it does not have any
relevant competitors. On the other hand, if a firm is in a very strong
competitive market, the most significant dimension in its environment
may be its competitors. If a firm sells goods in a seasonal industry, then
cycles of consumer demand are an important dimension of its environment.
Thus some dimensions describe the environment of one firm whereas other
dimensions describe the environment of another firm. Some organizations
have many important and somewhat interrelated factors in their environ-
ment, whereas others have much simpler environments with few and unre-
lated factors. Further, some factors have a direct effect on firm performance
and some have an indirect effect. A change in exchange rates may directly
affect the costs and revenue of particular activities or products. Similarly, a
change in government subsidies may directly determine the viability of an
industry. For example, sales of windmills in the US are directly dependent
on US government subsidies. In the early days of the Bush administration it
was unclear whether the administration would continue to support the
windmill industry or not. The choice of the government in this case had a
direct and significant impact on the companies producing windmills. Other
factors have more indirect effects. For example, the deregulation of the
aviation industry had the effect that new competitors could enter the
market, but it was not clear how they would enter, when they would enter
and which strategies a mature airline could initialize to prevent or postpone
particular types of competitors from entering the market. Here the deregu-
lation affected the environment but in a more indirect way, and there was
uncertainty about what would happen.
The environment thus refers to the forces surrounding an organization
that impact its performance. For the firm as a whole, the environment
usually is the competitive marketplace. For a department or business unit,
it includes upper management and the other units of the firm that affect the
business of the department. For a team, the environment is the department,
other organizational units in which the team operates, and possibly other
teams that influence the team’s workload and its success in carrying out its
tasks. It is important to assess an organization in terms of its immediate
environment and to do so as part of the ongoing process of organizational
design. If a firm switches industries, its environment will change (as it moves
into a new marketplace). If a firm undergoes internal reorganization, the
environment for a given department, business unit, or team may change.
Likewise, if one business unit operates in country A while another operates
40 Burton, DeSanctis, and Obel

in country B, there are likely to be different environments due to differences

in customs, regulations, and so forth, across the two countries. Again, the
environment of an organization should be assessed in terms of the forces
affecting the organization, whether these forces lie within the larger organ-
ization in which the focal organization operates, or whether these are forces
that lie outside in the external marketplace.
When you think about the forces that describe a firm’s environment, can
you predict how they will affect the firm? Do you know what competitors
will do? Can you predict what new regulations the political system will
initiate in the future? Sometimes you can and sometimes you cannot. If
the European Parliament has agreed upon a new framework, you can
estimate to a certain extent how local regulations will change. But you
may not know when the local governments will pass the new laws. If you
are a vertically integrated company you may be able to control the value
chain to some degree, but in other situations you may be very dependent on
your suppliers and you may not know their reaction. Thus the environment
has some degree of uncertainty.
It may be rather straightforward to state which factors in the environment
are likely to affect the actions and performance of a firm, but it may be much
more difficult to estimate the degree of uncertainty associated with those
factors. Some of the uncertainty may be stated in probabilistic terms,
whereas other parts of the uncertainty may be much more difficult to
estimate probabilistically. For example, there may be new aspects of the
environment that your chosen organization has never before experienced.
This could be a new technology or a new type of regulation.
The characterization of an organization’s environment in terms of com-
plexity and uncertainty is important and relevant because an increase in
both complexity of the environment and the uncertainty of the environment
increases the demand for information processing in the organization. If
there is a high degree of complexity, more elements have to be monitored
and the effect of change has to be estimated. If there is a high degree of
uncertainty, more plans may have to be established and a higher degree
of flexibility may be needed.
Should the description of an organization’s environment be objective or
subjective? This issue was addressed many years ago and is still a difficult
problem (Bourgeois, 1980). We often talk about environmental forces as if
they are objectively determined when in fact they may not be so. You will
often find that firms in the same industry, confronting the same environ-
ment, behave differently. Some companies in the industry perform badly
41 Environment

while other companies show an excellent performance. One reason why this
occurs is that they perceive and categorize the same environment very
differently. As an example, within the US airline industry Southwest Airlines
has defined its competitors as customers who drive to their destinations,
whereas other airlines have defined their competitors purely in terms of
other airlines. In this way, Southwest Airlines perceives and categorizes its
environment very differently than, say, Delta or American Airlines.
Why is that? One reason could be the cognitive capacity of the individuals
in the firm that allows one firm to understand the environment much better
than the other. Or it could be that the environmental scanning is done much
better in one company than in the other. Or the differences could be
deliberate intentions of management. As an example of the latter, Southwest
has always assumed that it must compete with the low price of driving to a
destination; thus, the airline has undercut fares of competing airlines by
huge margins, even when the industry was expanding. In the 1980s when
other airlines were offering discount fares from Dallas, Texas to Houston,
Texas for $76, Southwest priced their fare at $17, because this fare was less
than the cost of driving. Southwest management deliberately defined its
competitive environment in terms of the low cost of driving from Dallas to
Houston; whereas other airlines defined their environment in terms of the
prices of what other airlines charged to fly the same route, which was $76 or
more. Thus, different airline companies in the same market defined their
environments very differently and pursued different strategies.
In order to survive, organizations continually monitor their environment.
You may be able to predict much more precisely a firm’s environment by
talking to customers, or suppliers, or politicians, or specialized research
firms. By going to tradeshows or following basic research activities you
may be able to predict technological developments. By tracking industry
information you may be able to predict industry trends. By meeting with
government officials you may be able to anticipate or influence political
events. One thing is for sure: knowing more allows you to better understand
your firm’s environment and anticipate its impact on the firm.
To describe an organization’s environment, we use two dimensions:
complexity and unpredictability. Complexity is measured as the number of
factors in an organization’s environment and their interdependency. Envir-
onmental complexity increases as the number of factors increases and/or the
interdependency among the factors increases. Unpredictability is lack of
understanding or ignorance of the environment in terms of the nature of
the factors and their variance; greater variance means less predictability.
42 Burton, DeSanctis, and Obel

Consider the example of General Electric (GE), where the environ-

mental factors for its thirteen product groups are relatively independent.
(For instance, the market for jet engines is independent of the market for
lighting.) In addition, some markets are more predictable than others. (For
example, lighting is easier to forecast than the market for jet engines which is
subject to new airplane orders and the global market for air travel.) GE’s
environment has a large number of relatively independent environmental
factors, some of which are difficult to predict.
The two dimensions of complexity and unpredictability were chosen
because they can be related to a vast literature of empirical studies of
organizations, and they fit well with our information-processing view of
organizational design. An increase in each of the environmental dimensions
increases the demand for information-processing capacity in a firm but in
different ways. Greater environmental complexity increases the amount of
information to process, as there are more issues of importance to the
organization. Greater unpredictability requires greater capability to forecast
or adjust to the changing environment. Neither necessarily increases the
amount of information, but each does require a different response from an
organization. An organization must either project what will happen or
adjust quickly to the environment. The former is forecasting, and the latter
is adapting to feedback. Many organizations use a combination of both; for
example, a firm with uncertain sales will forecast and also adjust quickly to
actual sales.
The two environmental characteristics are general attributes. Complexity
refers to the number of powerful forces affecting an organization. If a firm
has only one or two major competitors it faces low complexity; whereas if a
firm must continually adjust to numerous conditions – competitors, prices,
labor pool, new products – it faces high complexity. Unpredictability is
the degree of uncertainty about the forces that impact a firm. The higher
the environmental unpredictability, the less accurate the forecasts are and the
more uncertain management can be about the future. Consider a consumer
products firm with operations in the United States and Russia. The number
of variables, or forces, influencing the firm’s strategy may be similar across
the two locations. Hence, complexity is about the same. But there is likely to
be more environmental uncertainty today for the units that operate in Russia
than those that operate in the US. This is due to the nature of the emerging
market and political context of Russia today, as compared to the US.
Applying the complexity and unpredictability dimensions to describe the
environment, we get four types of environments: a calm environment, a
43 Environment

Figure 3.1. The environment space.

varied environment, a locally stormy environment, and, finally, a turbulent

environment. We will discuss each of these environments.
Figure 3.1 shows the complexity and unpredictability dimensions with the
four environmental categories: calm, varied, stormy, and turbulent. Gener-
ally, there are increased information-processing demands on the organiza-
tion as we move from a calm environment to a turbulent environment. An
organization has more issues to consider and coordinate as the complexity
and unpredictability of the environment increase.
Each of the four environmental categories is a different combination of
complexity and unpredictability. If a firm is in a calm environment, then the
environment is low in complexity and is predictable. The firm has few
factors to consider and they are predictable; you know what is important
in the environment with a good deal of certainty. There are no surprises and
few adjustments are required. If a firm is in a varied environment, then the
environment is high in complexity but is predictable. There are many
interdependent factors, but these are well known and predictable. If a firm
is in a locally stormy environment, then the environment is low in complex-
ity, but unpredictable. There are few and usually independent factors, but
they are not predictable. Unrelated surprises require many adjustments
which can be dealt with one by one. If a firm is in a turbulent environment,
then the environment is high in complexity and is unpredictable. There are
many interdependent factors which are unpredictable. This is the most
demanding environment for information processing, requiring many
short-term adjustments and coordinated responses on the part of the
organization.
Now, let’s consider these environments in more detail. Generally, we move
from a less demanding environment to a very demanding one. In figure 3.1,
44 Burton, DeSanctis, and Obel

begin in the lower left corner, move to the upper left, then to the lower right,
and finally to the upper right corner.

Calm Environment

A calm environment has low complexity and low unpredictability (i.e., it is

highly predictable). It is simple and known with few surprises. If a firm only
has a few products and sells them into markets where the markets are
predictable, we say it has a calm environment. The political and financial
issues usually are not major challenges for management except if the firm
is in a monopoly situation protected by the political system. Some public
organizations think that they are in a calm environment. Utility companies –
at least those that have not yet been deregulated – may find themselves in a
calm environment. Calm environments occur less and less frequently as
more industries are deregulated. Further, the deregulation of the financial
system, the creation of a single European market, the NAFTA agreement and
similar agreements have done away with many calm environments.
If you are an executive in a calm environment you do not need to spend
much time assessing your organization’s environment, either to forecast
what will happen or to adjust to surprises. Today’s environment will be
tomorrow’s as well. There will be few surprises. So, you can focus on other
organizational design issues, addressing more internal concerns.
It is important to recognize the risks of an executive’s perception of a calm
environment. First, the executive’s perception can be wrong. Second, the
environment can change, and with an assumption of a calm environment, it
is likely that any change will be ignored or missed. So, the presumption of a
calm environment by an executive is potentially risky. The CEO of Intel, Andy
Grove, has a famous dictum that only the paranoid survive. This is a useful
warning, especially for firms that perceive they are in a calm environment.
If you perceive that your organization is in a calm environment in the
short run, you may be shocked or surprised without warning. With this
perception and mindset, the time lag for needed change is likely to be long –
perhaps too long for the survival of the firm. Bon Gout, an importer of
specialty goods in Denmark, was in a calm environment for many years,
where it had a good working relationship with Samsonite, one of its
suppliers. Bon Gout sold Samsonite goods to retail outlets throughout
Denmark. When Samsonite decided without consultation that it would sell
45 Environment

directly to the retailers, the Bon Gout organization was in shock; the
environment for Bon Gout instantly changed from calm to turbulent. Bon
Gout was no longer in a calm environment and new action was required.

Varied Environment

The varied environment is complex as there are many factors to take into
consideration and they can be interdependent (i.e., they influence one
another), but these factors are relatively predictable and/or they tend to
change within known limits.
If a firm has many products and sells them into markets where the
markets are predictable, we say it has a varied environment. Further, polit-
ical and financial issues can add to the number of factors in the environ-
ment. If the markets, the politics and financial factors are all interdependent,
as they are for many governmental suppliers, the environment is varied. In
such a varied environment there are many factors for an organization to
consider, but it is possible to predict what will occur. Market forecasts,
analysis of political trends (e.g., road construction or environmental protec-
tion) are frequently applied techniques to predict the future environment. It
is possible to project into the future with reasonable accuracy and under-
standing. The focus of the executive in a varied environment is on planning
and coordination that will allow the organization to manage in the face of
the interdependencies among the factors that are in its environment.
In earlier times, the toy manufacturer LEGO was in a varied environment.
It operated in many countries with many different variations of its products,
and there were many legal, financial, and logistic issues to take into consid-
eration. Demand had seasonal variations but it was quite predictable. But
that environment has changed to a locally stormy environment, where the
sales are rather unpredictable.

Locally Stormy Environment

The locally stormy environment is highly unpredictable but not very

complex. That is, there are a few factors in the environment which are
relatively independent, but they are unpredictable. The locally stormy
46 Burton, DeSanctis, and Obel

business environment is analogous to the prediction of rain with .5

probability for a farmer. Amount of rainfall may be one of the few factors
that determine a crop’s growth rate, but the predictability of rain may be
extremely low. Start-up companies that are dependent on a patent right
or the result of a particular outcome of a clinical trial are in stormy
environments.
In the locally stormy environment, executives are most concerned about
the unpredictability of environmental factors that affect their firm. Many
years ago Ashby (1956) proposed the Law of Requisite Variety, which states
that a system’s internal flexibility must meet the outside uncertainty for the
system to survive. For purposes of organizational design, the Law of Requis-
ite Variety means that a firm needs to be flexible so that it can meet the
unpredictability of its environment. Put another way, the information pro-
cessing capacity of the firm should be able to adjust when unpredictable
events occur. Unpredictability means that the response time in which the
firm has to react is much shorter than if the environment were predictable.
In a predictable environment a firm has time to plan for the future; but in
an unpredictable situation the peak information requirement is much
higher as you have to do many things when the unpredictable factors in
the environment suddenly become known. In a locally stormy environment
the advantage is that only a few factors have to be monitored and they are
relatively independent. Thus, the adjustments can be made one by one – a
much simpler problem than if the environment is turbulent, as we shall
see below.
Unpredictability without complexity can be dealt with on a local basis,
not requiring the coordination for the whole firm. As mentioned before, GE
has a different environment for its several relatively independent product
groups. These markets can be adjusted to one by one without concern for
each other since the markets are independent of each other.

Turbulent Environment

The turbulent environment has both high complexity and high unpredict-
ability. There are many interdependent factors which are not predictable.
This environment is analogous to that faced by the farmer who has not only
the rain to consider but also the market price for grain, and the rain and
price may be correlated; further, both are difficult to predict. For you as a
47 Environment

manager, this is the most difficult environment in which to operate as it

requires limited forecasting and also the flexibility of quick and coordinated
adjustments as events become known. A turbulent environment requires a
firm to have a large and fast information-processing capacity so that the firm
can choose quickly among alternative courses of action for the organization.
Thus, the adjustments must be made together and made quickly.
Today’s global airlines confront a turbulent environment. Having
emerged from a relatively calm environment of regulation, they now have
many factors to consider: the ticket prices of other global competitors, the
emergence of low price niche competitors, the global price of jet fuel, the
global security situation, the competition from nontravel substitutes such as
video conferencing, to name a few. Further, these factors may be interde-
pendent. E.g., the use of video conferencing may be related to the price
of travel and also to the global security situation; in addition, global
security affects the price of jet fuel. There are many interdependent factors;
furthermore, it is difficult to predict their behavior.

Diagnostic Questions

In the first chapter, you located your unit of analysis on the efficiency and
effectiveness dimensions and thus categorized your organization’s goals. In
Chapter 2, you located your unit of analysis on the exploration and exploit-
ation dimensions for your strategy. Here you should do the same for your
chosen unit of analysis on the environmental complexity dimension and
unpredictability dimension in figure 3.1. Then you can categorize your
chosen firm’s environment as calm, varied, locally stormy, or turbulent.
1. First, assess the degree of complexity of the environment for your unit of
analysis. Complexity is the number of variables in the environment and
their interdependency. It refers to factors that can influence the
operations and outcome of your organization. These may include the
industry, the competitors, suppliers, the financial system, the human
resource talent pool, new technology, prices, quality requirements,
financial conditions, governmental relations, and political conditions
among many other factors. Identify external conditions that in a
significant way could affect your chosen organization. These should be
conditions that you constantly feel you need to scan and monitor.
Examples of external conditions are:
48 Burton, DeSanctis, and Obel

Table 3.1. Complexity scores

Interdependency of factors Number of factors in the environment

1–3 4– 6 7–9 10–12 More than 12

Low 1 1 2 2 3
Medium 1 2 3 4 5
High 3 4 4 5 5

a. What is the number of critical factors in the organization’s

environment? List each factor below and then count to estimate the
total number of critical factors.
1:_____________________
2:_____________________
3:_____________________
4:_____________________
5:_____________________
etc.
b. What is the overall interdependency among these factors? That is, to
what extent in this set of factors interrelated, or correlated with one
another? Select an overall rating of: low, medium, or high.
Now, using the table 3.1, find the complexity score on the scale of 1 to 5.
Find the column corresponding to the number of factors in the
environment. Find the row corresponding to the extent of interdepend-
ency among the factors. Your complexity score is in the cell
corresponding to the column and row for your firm.
2. Unpredictability: For each factor that you have included in the firm’s
environment in Part 1a above, score its unpredictability on the scale from
1 to 5 as follows:
1 2 3 4 5
very low moderate very high
Critical Factors in the Unpredictability score
Environment (1 ¼ low, 5 ¼ high)
1:___________________________ ___________________
2:___________________________ ___________________
3:___________________________ ___________________
4:___________________________ ___________________
5:___________________________ ___________________
etc.
49 Environment

Figure 3.2. Locate your chosen organization in the environment space by rating the levels of
complexity and unpredictability.

Assess the total environmental unpredictability from the individual unpre-

dictability factors.1 Now with the firm’s scores on environmental complexity
and unpredictability, locate its environment in figure 3.2.
What is the environmental category of your chosen organization?
Next, you will examine how your chosen organization’s environment fits
with the goals and strategy that you identified in Chapters 1 and 2.

Fit and Misf its

Which goals and strategies fit well with your chosen organization’s environ-
ment? Table 3.2 summarizes the best fit combinations.
As in Chapter 2 you can view the fit combinations by examining the
columns. Start with column A. You see that in a calm environment a reactor
strategy and nonspecific goals fit well together. In a calm environment little
information processing is needed. You know what is happening; tomorrow
will be like today. We can further point out that in a calm environment it
does not make a lot of difference which strategy you have. In a calm
environment a defender strategy will also be appropriate, although the effort
to defend the position will not be as needed as in a varied environment.
Research has shown that a proper alignment of the business strategy with
the environment is much more important in dynamic environments than

1
We suggest that you use an averaging procedure of the detailed scores to get to the overall
score, but you may make an overall estimate of the score if you prefer.
50 Burton, DeSanctis, and Obel

Table 3.2. Fit among organizational environment, strategy, and organizational goals

Corresponding
quadrant in
organizational
design space A B C D
Environment Calm Varied Locally Stormy Turbulent

Strategy types Reactor Defender Prospector Analyzer Analyzer

with without
innovation innovation

Organizational Neither Efficiency Effectiveness Efficiency and Effectiveness

goals

in stable environments (Obel, 1993; Siggelkow & Rivkin, 2005). So the

importance of an alignment increases as we move from column A to
column D.
The varied environment in column B is still very predictable, but more
factors can affect the organization. Planning and forecasting are the name of
the game. Little new is happening, and when it happens it is with a pace that
the organization can accommodate. A focus on exploitation is appropriate
for this type of environment, and a defender strategy fits well here.
In column C the environment is less complex but more unpredictable. It
fits well with a prospector strategy with a high degree of exploration. Here
the focus has changed from efficiency as we saw it in column B to a focus on
effectiveness.
In column D the environment is turbulent. The organization is affected
by many factors in the environment and some, if not most, of them change
in an unpredictable way. The focus in such an environment fits well with a
combination of effectiveness and efficiency goals and an analyzer strategy.
Misfits occur if your assessment in Chapters 1, 2, and 3 end up in different
columns. It is a common impression that the business environment changes
more often and more dramatically today than earlier. As was stated above,
the importance of a proper alignment is higher in unpredictable environ-
ments than in predictable environments. Thus it is a potentially dangerous
situation for a firm if it has been located in a calm environment and
suddenly finds itself in a locally stormy or turbulent environment. In such
a case there is a need either to change the strategy or to somehow calm down
the environment. Usually, the latter is not possible, or it is possible only for
very large companies and is difficult to do.
51 Environment

If you are located in different columns based on your answers to the

diagnostic questions from Chapter 1, 2 and 3, then think about how you
might modify your organization so that you can obtain an alignment, with
all answers in the same column of table 3.2. Would you change the goals, the
strategy, or work on the environment, or would you try to move all three to
a new common quadrant?

Summary

In this chapter we have assessed an organization’s environment and dis-

cussed the impact on its strategy. The elements of the environment have
been discussed where complexity and unpredictability are measures of the
environment. This leads to the four types of environment: a calm environ-
ment, a varied environment, a locally stormy environment, and a turbulent
environment. Finally, misfits among the goals, environment, and strategy
were presented.

Glossary

Calm environment: an environment with low complexity and low

unpredictability.
Environmental complexity: the number of factors in the environment and
their interdependency.
Interdependency: interrelatedness; the extent to which different factors
influence one another or are correlated with one another.
Interrelatedness: see interdependency.
Locally stormy environment: an environment with low complexity and high
unpredictability.
Turbulent environment: an environment with high complexity and high
unpredictability.
Unpredictability: degree of uncertainty about the forces that impact a firm;
lack of understanding or ignorance of the environment in terms of the
nature of the factors and their variance.
Varied environment: an environment with high complexity and low
unpredictability.
52 Burton, DeSanctis, and Obel

Where are you in the step-by-step approach?

STEP 1
Getting Started
(1) Goals
STEP 2
Strategy
(2) Strategy
(3) Environment
You have now examined your firm’s strategy and its environment. Just to
review, there are two issues here. First, does your strategy fit with your
goals? If it does not fit, reconsider your goal statement or adjust your
strategy to fit with your goals. Second, does your strategy fit the
environment? If not, what changes should you consider? Usually, it is more
difficult to change the environment than to change the strategy. Generally
you will want to move the organization’s strategy to fit the environment.
But you can also try to move both the environment and the strategy to a
new position. To change your current environment may involve lobbying,
buying your competitor, heavy advertising, or other rather difficult
changes. However, you may also change your environment by moving the
firm to a new locale with a more appealing environment, by exiting a
particular market, or by moving into a new market.
It is very important to assure that the organization’s goals, strategy,
and environment are located where you want them to be. These three
items are the fundamental anchors for designing the organization and
thus will strongly affect the design that you will end up with in your
step-by-step analysis. Once you have your goals–strategy–environment
relationships in order, you can move on to design the firm’s structure.
STEP 3
Structure
The next step is to review how the organization is configured in
terms of its assignment of subtasks and coordinating relationships
53 Environment

among all the subtasks. You will then assess how this configuration
operates across time and space boundaries. Further, you will want to
ensure that the structure is compatible with your strategy, the
environment where you operate, and your goals.
(4) Configuration complexity
(5) Geographic distribution and knowledge exchange
STEP 4
Process and People
STEP 5
Coordination and Control
Step 3 Structure
4 The configuration and complexity of
the firm

Introduction

The choice of a firm’s configuration – sometimes called its structure or

architecture – is a critical decision for the executive (Burton & Obel, 2004).
The next step in the step-by-step approach is for you to choose a configur-
ation and decide on the organizational complexity which will enable your
firm to perform well for its given goals and strategy and in its environment.
A poor choice of configuration leads to opportunity losses which can be a
threat to the organization’s short-term efficiency and effectiveness as well as
its long-term viability. As in the earlier chapters, for convenience of expres-
sion, we use the term ‘‘firm’’ or ‘‘organization’’ as the unit of analysis, but
you can substitute team, department or division if that is the unit of analysis
you chose in Chapters 1, 2 and 3.
A firm’s configuration, frequently pictured as an organizational chart, tells
us how the firm partitions big tasks into smaller tasks either by specializa-
tion or product, and then also indicates the formal communication patterns.
As discussed in Chapter 1, organizational design of configuration involves
two complementary problems: (1) how to partition a big task of the whole
organization into smaller basic tasks of the subunits, and (2) how to coord-
inate these smaller subunit tasks so that they fit together to efficiently realize
the bigger task or organizational goals (Mintzberg, 1983). Once tasks are
partitioned and the smaller subunits specified, the information-processing
questions for organizational design are: Who makes what decisions based
upon what information? Who talks with whom about what, or what is the
structure of communication (March and Simon, 1958; Galbraith, 1973,
1974)? When you have answered these design questions for your firm, you
know a good deal about how it will work and can proceed to realize the
goals. There are a number of choices for the configuration. Here we will

57
58 Burton, DeSanctis, and Obel

consider the basic choices and how they address the fundamental questions
of design.
A firm’s organizational complexity further specifies the organizational
design. For example, how many subunits are there? The number of depart-
ments should be specified as well as the number of vertical levels in the
hierarchy. These are called the horizontal and vertical differentiation, which
together are called the organizational complexity of the firm. More ‘‘modern’’
dimensions of the classic configurations include the degree of virtualization
and the extent to which IT systems are embedded in the configuration. These
new types of configurations are considered in Chapter 5.

Conf iguration

What are the dimensions for partitioning the larger task of the firm into
smaller tasks such that the smaller tasks can be coordinated and work well
together? In the literature on organization design two fundamental dimen-
sions have been used to distinguish the basic configurations: Product/service/
customer orientation and functional specialization. The product/service/
customer dimension suggests that the total firm task will be partitioned by
the outputs of the firm, which give it an external focus. If the firm has
divisions or departments with product or customer names, then it has an
external focus and is high on this dimension. The functional specialization
dimension indicates that the work will be divided by specialized activities. If
the firm has departments with function names, such as production and
marketing, then it has a more internal focus and is high on this dimension.
These two product and functional dimensions indicate the focus of how the
work will be divided and then, given this breakdown, how it must be
coordinated. These two dimensions of configuration generate four basic
configurations as shown in figure 4.1. The four basic configurations are:
simple, functional, divisional, and matrix (Miles & Snow, 1978).
These four basic configurations can be combined in different patterns,
and thus are the building blocks of more complicated structures depending
on whether your unit of analysis is a team, a department, a division, or even
the whole organization. For example, in the case of a divisional configur-
ation, each of the several divisions of a divisional structure can be func-
tional, matrix, or another divisional. Thus these basic configurations may all
be present inside one large organization. They should be designed using the
top-down approach described in Chapter 1.
59 The configuration and complexity of the firm

Figure 4.1. The alternative organizational configurations of the firm.

The four basic configurations of simple, functional, divisional, and matrix

score differently on the product/service/customer and functional specializa-
tion dimensions. In figure 4.1, we show the four configurations as they are
located on the two dimensions: product/service/customer orientation on the
horizontal axis and functional specialization on the vertical axis. The simple
configuration is low on the product/service/customer dimension and also
low on the functional specialization dimension. The functional configur-
ation is also low on the product/service/customer dimension, but it is high
on functional orientation. The divisional configuration is the opposite: high
on product/service/customer dimension, but low on specialization. And
finally, the matrix configuration is high on both dimensions, suggesting a
need for a high information-processing capacity to achieve the twin goals of
efficiency and effectiveness.
Next, we discuss the four configurations, beginning with the simple
configuration. In figure 4.1, we begin in the lower left corner, move to the
upper left, then to the lower right, and finally to the upper right hand corner.

Simple conf iguration

The simple configuration is: low on the product/service/customer dimension

and low on the functional specialization dimension. The simple configur-
ation is usually a small organization, consisting of an executive and perhaps
a few other individuals. The executive tells the others what to do and
manages the ongoing operations. The individual employees do not have
specific tasks or activities to perform, nor are there well-defined job descrip-
tions. The total task of the firm is broken down into smaller tasks and
60 Burton, DeSanctis, and Obel

assigned to the employees by the executive on an as-needed basis; the

coordination of the activities is also done by the executive. Both the task
assignments and the coordination are accomplished by the executive in an
ongoing and continuous manner. Little is fixed; things are very fluid and can
be very flexible to adjust to the situation at hand. The executive is at the
center of the information flows, makes the decisions, coordinates the activ-
ities of the employees, and controls the operations – telling others what to
do. The executive is also the main contact with the market, customers,
suppliers, and clients of the firm. In figure 4.2, there is an organizational
chart of the simple configuration where Sophia is the executive.
The simple configuration is usual for small firms – whether new small
start-ups or older firms. In some rare situations, the executive of the larger
firm may choose the simple configuration – particularly for an owner-
managed firm which has grown from a small start-up to a larger firm. For
all of these firms, the executive is in charge of or oversees almost everything
that goes on. In terms of information processing, it can be a very demanding
task.
The simple configuration is flexible but not usually efficient or effective.
The efficiencies of specialization are not realized, as the employees are asked
to do many tasks for which they may not be fully skilled. The simple
configuration depends heavily upon the vision of the executive for its
effectiveness orientation. A danger is that the focus can be narrow and not
very effective for the firm’s customers if needs change and vary over time.

Figure 4.2. A simple configuration.

61 The configuration and complexity of the firm

Because the executive is the focal point of all information processing in

the firm, it can be difficult for this one person to take time to adjust the
firm’s direction or seek innovative opportunities. In brief, the simple con-
figuration does not take advantage of the efficiencies of specialization, and
its effectiveness depends heavily upon the actions of one person – the top
executive.
The executive is at the center of all that happens in the simple configur-
ation. It is the executive’s show. If the executive uses time well, makes good
decisions, and coordinates activities well, then the simple configuration
leads to good performance. But if the executive becomes overloaded and
fails on any of these tasks, the firm performance will suffer.

Functional conf iguration

The functional configuration is: low on the product/service/customer di-

mension and high on functional specialization. The focus of work is based
on the functional specialization – hence the name.
The functional configuration is more complex with respect to informa-
tion processing than the simple configuration. In the functional configur-
ation, there are department managers with specified subunits, each of which
has well-defined jobs. The total firm task is broken down and assigned to
subunits; the coordination, or putting together, is accomplished by the
hierarchy, which uses a combination of rules and directives. In contrast to
the simple configuration, where little is fixed (i.e., task assignment and the
organizational structure can change frequently), much is fixed for the
functional configuration. It is more machine-like and can accommodate
large-scale organization as well as a high degree of information processing.
The production flow is to hand off work from one subunit to the next, e.g.,
operations to marketing, which requires coordination. The executive is
again at the center of the organization for information flows to and from
the top, making decisions, and coordinating activities of the subunits.
Figure 4.3 shows an organizational chart that illustrates the functional
configuration where the functional departments are: supply, manufacturing,
and sales. There could also be functions such as operations, marketing,
finance, and human resources. Operations and marketing are usually called
line functions and departments like finance and human resources are
called staff functions. The executive level coordinates the manufacturing,
sales, and other major efforts and is concerned with matters such as
62 Burton, DeSanctis, and Obel

Figure 4.3. A functional configuration.

planning and realized versus projected expectations of firm productivity.

Although information flows through the top of the organization, coordin-
ation is required among the subunit activities, and each department pro-
cesses information on its own, offloading some of the information-
processing demand that was pushed to the executive level in the simple
configuration.
The major advantage of the functional configuration is that specialization
provides the rationale to assign individuals and subunits to specific tasks
which they learn to do efficiently. From Adam Smith (1776) onward to this
day, the economies of specialization make the functional configuration the
most common configuration. The rationale is an efficient organization
which is directed by the executive level. There is a strong reliance on the
skill of the executive – both for the short-term coordination of ongoing
operations and for the long-run choice of specialization.
A frequent question about the functional configuration is, how many
subunit functions should there be? It is a question of (1) the limited time
of the executive who must make decisions and coordinate the subunits
(departments), and (2) the capacity of the subunits to process information.
The time demands on the executive grow as the number of subunits
increases, but also the coordination demands increase as the number of
products increases. Unfortunately, as either the number of subunits or the
63 The configuration and complexity of the firm

number of products increase, the coordination demands increase non-

linearly with the implication that only a few functional subunits are possible
(Burton & Obel, 1984). Most firms have about five subunits and rarely
should they have more than seven. More formally, NK complexity theory
shows that where N is the number of subunits and K is the degree of
interdependent information flows going through the top, e.g., K ¼ N1,
the coordination demands become overwhelming quickly as the number of
subunits increases. (See Siggelkow & Rivkin, 2005 for discussion of NK
theory.)
As in the simple configuration, the executive in the functional configur-
ation is at the center of the organization and may become overloaded if the
environment is not predictable. Where adjustment and change in work tasks
are required, the situation can become overwhelming. The functional con-
figuration is efficient for unchanging activities; however, that efficiency is
lost when rapid change is required. The functional configuration is a good
choice if you want the organization to operate with high efficiency and
precision. The configuration works well for tasks that are repeated frequently
and in high volume.

Divisional conf iguration

The divisional configuration is high on the product/service/customer di-

mension and low on specialization. Here the focus is not so much on the
internal specialization but more on the outside products and services that
the firm produces, or on the customers it serves.
There is an executive level that oversees subunits which are relatively
independent of each other and have limits on their contact with the head-
quarters. Each subunit, which may be called a division, an SBU (strategic
business unit), product business, customer business, or country business, is
its own business, frequently organized as a simple or functional configur-
ation within the subunit. Each division is externally focused and has its own
markets and customers. It pursues its own destiny within the constraints and
policies of the headquarters. The most important relationship is financial,
where each division has its financial goals, receives its operating monies as
well as its long-term investment funds (Williamson, 1975). The top execu-
tive sets policy for the divisions. These policies can be quite general, such as,
‘‘operate within the law,’’ or they can be quite detailed including financial
reporting standards, human resource policies, and innovation directives for
64 Burton, DeSanctis, and Obel

new processes and products. The extent of involvement by the top executive
can vary. At the one extreme, the headquarters is a ‘‘bank’’ that provides
financial oversight and not much else, and at the other extreme, each
division can be driven from the top. For the latter case, the headquarters
is likely to become overloaded with large information flows and many
decisions to make; performance suffers (Chandler, 1962). So, the divisional
configuration works best when there is limited coordination from the top
and each division is left to run its own business where it has resources and
can coordinate its activities to focus on the market for its products, its
customers or in its region (Burton & Obel, 1984). As noted earlier, within
each division, the organization can be configured as a simple or functional
configuration or even another division. In figure 4.4, the divisional configur-
ation is shown where the product flows and information flows have been
added.
The advantage of the rationale for the divisional configuration is that it
aims to be effective with its external focus on the product, customer, or
region. The divisional configuration is more market-responsive than the

Figure 4.4. The divisional configuration with product and information flows.
65 The configuration and complexity of the firm

functional configuration. Because the divisions are relatively autonomous,

they can make decisions on their own, meet the needs of the marketplace in
creative ways, and thus foster opportunity for growth. Many firms treat new
acquisitions as divisions, allowing them to operate relatively independently
of headquarters so that they can continue with the success experienced prior
to acquisition. Dividing the firm into product or brand groups is another
way to create divisions that can grow or die depending on their success in
the marketplace.
The disadvantage of the divisional form is that each division is relatively
independent of the other in its operations and markets. The divisional
configuration does not handle interdivisional dependencies well. For
example, two divisions selling competing products to the same customer
or developing a new technology which requires their joint efforts will find it
difficult to coordinate and/or avoide duplication of efforts. Interdependent
efficiencies are lost. If a customer goes to IBM to buy both consulting
services and network computers, they must deal separately with the two
divisions that manage these products. The divisions have separate sales and
support groups, and the customer must bear the coordination cost of
dealing with the two divisions. But each division can be efficient using a
functional configuration. If a division becomes less effective, it can be sold
off in the marketplace, as IBM recently has done with its PC division.
At the top, the executive of the divisional configuration is responsible for
the choice of the divisions and their level of activity, and more generally the
overall firm performance. There is a strong reliance on the divisions and
their own executives to relate to their own markets with products and
services. The divisional configuration requires a top management level in
which the head of each division has strong executive capability beyond those
of the Chief Executive Officer.
How many divisions should a divisional firm have? Unlike the functional
configuration where more functions or more products increase the coordin-
ation demands nonlinearly, additional divisions do not (Burton & Obel,
1984). In the extreme, an additional division only means choosing the
divisional executive, adding one more column to the financial reports and
affirming that the existing policies and information systems apply to the
new division. In terms of NK complexity theory, K is very low or the
divisions are only loosely connected. So, the number of divisions can be
quite large, up to twenty or so. Considerably more subunits are possible in
the divisional configuration than in the functional configuration.
66 Burton, DeSanctis, and Obel

For the divisional configuration, the top executive is the center for
corporate finance and policy. If divisional interdependencies are abundant,
then the executive can become overloaded resolving interdivisional issues.
The goal is to have divisions with minimal interdependency.

Matrix conf iguration

The matrix configuration is high on both dimensions: product/service/

customer and functional specialization, suggesting a need for a high infor-
mation-processing capacity to achieve the twin goals of efficiency and
effectiveness. There is both the functional hierarchy and the divisional
hierarchy for the same firm. The top executive is responsible for both the
functional and divisional dimensions – to set policy, set priorities and
resolve conflicts among the subunits. The top executive is not involved in
the details of operations, but does oversee the entire firm. Most of the
difficult coordination problems are handled by the matrix managers, i.e.,
those that link between the lateral divisions and the functional hierarchy.
Matrix managers make multiple variable tradeoffs which involve both the
function and the division. Figure 4.5 illustrates a matrix in which functional
specialization is combined with product orientation to yield coordination of
functions across the product groups. As shown in figure 4.5, the matrix

Figure 4.5. A matrix configuration.

67 The configuration and complexity of the firm

configuration requires simultaneous coordination of the functional special-

ties across the projects, products, services, or customers that the firm
services. When there is a change in the timing of an activity, it ripples across
the whole of the firm – called the jello effect.
The matrix can be very flexible, dealing with new information and
adjusting to the new situations quickly to utilize limited resources to meet
firm priorities. The advantage is that the matrix can realize both the effi-
ciency of the functional form and the effectiveness of the divisional form –
overcoming the limitations of both forms. When it works well, both effi-
ciency and effectiveness result. Complicated tradeoffs are considered; deci-
sions are made; and the firm moves on. But when the matrix is not well
managed, it can be neither efficient nor effective. The challenges of man-
aging a matrix include reconciling conflicts between the lateral and vertical
subunits, information overload, excessive meetings, and decision delay.
The firm does not move. The matrix configuration requires managerial skills
that include a focus on the entire firm as well as one’s own function or
division, the acceptance of uncertainty, the willingness to consider compli-
cated tradeoffs and negotiate realistic solutions, and a focus on results. These
benefits must exceed the additional costs of coordination if the matrix is to
be justified beyond the functional or divisional configurations.
The matrix configuration has many two-dimensional names in practice:
function and product, function and project, specialty and industry cus-
tomer, product and customer, product and region or country, basic tech-
nology and product – to name a few. There are also three three-dimensional
matrices as many multinational firms have function, product, and country
or regional dimensions. Some add a fourth dimension of customers as
well. Matrix relations go beyond setting up the matrix configuration
reporting relationships. Management must invest in developing cross-
organizational coordination. This is realized in many ways: lateral rela-
tions, liaison roles, various coordinating committees – all of which consider
issues which are not dealt with well or quickly within the hierarchy. Com-
bining the matrix configuration together with its various other cross-
organizational relations, it is quite possible to have eight dimensions:
function, product, region, customer, technology development, basic re-
search, human resources, and inter-country finances – perhaps more to be
managed and involving complicated tradeoffs and coordination. Most firms
are not so complicated; yet, many modern firms have an array of matrix
relations, or cross-organizational mechanisms to coordinate across the
dominant hierarchy in the firm.
68 Burton, DeSanctis, and Obel

How big can a matrix be? The matrix has both a functional and divisional
dimension to manage simultaneously, so the size is the number of functions
multiplied by the number of divisions. Given the jello effect, or more
formally in NK theory where N is the number of subunits and K, the degree
of interdependency, we suggest that the matrix can include only a small
number of subunits, say four or five on each dimension. However, the big
Swedish-Swiss multinational firm ABB at one time had a matrix configur-
ation where there were over 100 separate SBUs along one dimension. They
used an additional middle level of management in the matrix to support the
complexity of interdependency to be coordinated. Still, the matrix was too
complex to manage and was eventually dismantled and replaced with a
simpler configuration.
When both efficiency and effectiveness are needed, the matrix configur-
ation is an appropriate choice. The matrix is usually more costly to manage
than a hierarchy as there are more managers, more information, and more
complicated coordination to be done. Further, the managers must consider
and deal with many considerations that simultaneously have overall effects
as well as effects on the subunit. Individuals who have been successful in
a hierarchy may not be comfortable or successful in a matrix configura-
tion. The matrix must be justified in terms of the strategy and the firm’s
environment.
The top management in the matrix has a focus on both efficiency and
effectiveness – attempting to obtain both. The top management cannot
direct the organization but must rely heavily upon the functional and
divisional managers for the detailed, ongoing coordination adjustments in
order to meet the firm priorities. Yet, the executive has much to do: set
priorities, resolve differences among the subunits, and generally oversee the
entire firm.
It is important to emphasize that the matrix can also lead to poor
performance. The dual coordination across the functions and the divisions
can lead to conflicts of priorities between the managers of subunits. If
conflict management requires great involvement by the top executive, a
major advantage of the matrix has been lost. The telltale signs of a matrix
in trouble are, again: overload of decisions at the top as the managers
are not able to solve problems; problems are not dealt with at all and
opportunities are lost; budgets are exceeded; operations are not coordi-
nated and resource utilization is lost or inefficient; employees are unhappy
and confused; subunits are spending excessive time on coordinating
69 The configuration and complexity of the firm

with other subunits to the detriment of subunit performance; and oppor-

tunities are lost. When the matrix works well, it can achieve efficiency and
effectiveness. But when things go badly, they can be very bad.
The organizational configuration is not the whole story; the organiza-
tional complexity is another property of the organization which can be
designed to meet the goals of efficiency and effectiveness.

Organizational complexity

The organization’s complexity (not to be confused with environmental

complexity discussed in Chapter 3) is a property or characteristic of the
organizational structure. Organizational complexity is the vertical and hori-
zontal differentiation of task management in the firm. It is how the firm’s
configuration is broken down into its several subunits. Roughly, it is the
width and height of the hierarchy. The two organizational complexity
dimensions are: the horizontal differentiation, or width, and the vertical
differentiation, or height. The horizontal differentiation is the degree of task
specialization across the hierarchy; the vertical differentiation is the depth
of the hierarchy – top to bottom. Each of the configurations we discussed
above can have high or low horizontal and vertical differentiation; it is a
property of the organization. The practical design questions relating to
organizational complexity include: the width of specialization for the firm,
the span of control for the firm, the delayering of the firm to eliminate
middle management, the scope in a divisional configuration, and the limi-
tation on the number of functions and divisions in the matrix configuration.
Again, these design choices are to be made in terms of efficiency and
effectiveness goals.
As shown in figure 4.6, choices regarding the degree of a firm’s horizontal
differentiation and vertical differentiation result in four types of organiza-
tional complexity. The organizational complexity is classified as: blob, tall,
flat, and symmetric. These are not ideal types of configurations but rather
archetypes that show how information processing will be conducted in the
firm based on how work is allocated among subunits.
Next, we discuss four complexity types on the horizontal and vertical
dimensions: blob for low and low; tall for low and high; flat for high and
low; and symmetric for high and high, respectively.
70 Burton, DeSanctis, and Obel

Figure 4.6. The organizational complexity space.

Blob

If the firm does not formally divide its work into subunits, then it is like a
blob. It is undifferentiated; it is low on both horizontal and vertical differen-
tiation. The blob has little specialization of task; the firm can be quite
flexible and quick to respond to ongoing changes. Job descriptions are very
loose, or may not exist. This lack of definition of who is to do what is very
demanding on the executive, requiring decision making for new situations
on a continuing basis, where the executive can become overloaded and not
be able to give adequate attention to the activities. The blob can also be
confusing to customers or to newcomers who join the organization, since it
is not clear who does what, or where one should go for specific types of
information.

Tall

The tall firm is low on horizontal differentiation and high on vertical

differentiation. The tall firm has a large middle management which focuses
on information processing – taking directions and information from the
top and making it precise for lower levels in the hierarchy; and taking
detailed information from the bottom and summarizing and interpreting
it for the top executives. The multilayered middle management connects the
executive to the specialized task level, e.g., the top executive to the factory or
service worker; the CEO to the programmer. The middle management takes
71 The configuration and complexity of the firm

directions and orders from the executive and breaks them down into smaller
task implications which then must be coordinated across the subunits.
For example, the executive of an auto plant may set the production plan
at 100 automobiles, which must be broken down into plans for many
subunits and coordinated among the subunits so that all of the functions
work together to meet the plans.
From the bottom, the middle management summarizes what is
happening at the bottom and passes it up the hierarchy, where the infor-
mation is aggregated as it goes up, so that the top can deal with simpler, but
relevant, information for decision making and control. Both the down and
up processes involve a good deal of information processing; it takes man-
agerial time and can lead to delay. The span of control is limited as the
information processing demands on the middle managers can be quite
high. The inter-level vertical information transferal is usually large, involv-
ing frequent interaction of detailed information. If an additional function is
added, it must be coordinated across all the functions, and thus the addition
of one more function increases the information processing demands
nonlinearly. This limits the number of direct reports to a few; most firms
have five to seven functions.
Recently, many firms have shortened their hierarchy, eliminating middle-
management levels in the firm. This is frequently called ‘‘delayering.’’ On the
organization chart, it can be simply the removal of a level, but much more is
involved. A simple removal creates a mismatch and miscommunication
between the two remaining levels. When a level is removed, the connections
between the level above and the level below must also be changed. So, the
information and communications must be redesigned, usually from top to
bottom. Without the informational assessment and modification, the newly
delayered firm will initially struggle. With more advanced information
technology, it is now possible to achieve quickly the vertical coordination
with a shorter middle management, but it still requires a redesign of the
organization and its use of information. It is not simply a matter of remov-
ing a layer in the hierarchy and seeing what happens.

Flat

The flat firm is high on horizontal differentiation and low on vertical

differentiation. There are fewer middle managers (or subunits) to coordin-
ate between the top executives and the lower levels in the organization.
72 Burton, DeSanctis, and Obel

Usually, these middle managers do not focus on detailed operations

which take lots of time and attention. They focus instead on resource
allocation, general policy, and finance. Other issues can be innovation,
R&D, human resources – all of which involve policy and strategy. The
information is aggregated and minimal. Short-term information exchanges
focus on financial goals and cash flows. The long-term information ex-
changes focus on capital budgets and technology planning. If your firm is
flat in its structure, the scope of the firm work across subunits can be quite
varied, especially if there are no operational connections among them. The
span of control can be wide if the focus of information flow is on policy,
not detailed operations. Put another way, the information flows are mini-
mal if exchanges focus on general policy but grow quickly for detailed
operations.
A major advantage of the flat organizational structure is that each unit has
autonomy to focus on its own work. Subunits can attend to the needs of
customers or suppliers or new products – whatever is their particular charge,
in terms of focus of work. On the other hand, the executive level of the
organization bears the burden of coordinating among these subunits, and
they can get out of synch, lack coordination, leading to inefficiencies for the
firm as a whole.

Symmetric

The symmetric firm is high on horizontal differentiation and high on

vertical differentiation. This means that the organization’s work is broken
down into many task specialties as well as many vertical reporting levels.
Horizontally breaking down tasks into smaller tasks means that work can
be done simultaneously in the horizontal subunits. Parallel processing of
work, ability of each to deal with customers or others in the marketplace,
and the opportunity to work independently all help to facilitate organiza-
tional effectiveness. To overcome the problem of the flat organization, in
which the executive level must process information generated by each of
the task-based subunits, a middle level (or perhaps multiple middle levels)
are created that aggregate work from bottom to top and facilitate infor-
mation flow from top to bottom. The symmetric organization tries to hit
the ideal balance of vertical and horizontal breakdown of work into sub-
units. Middle levels help to coordinate work to yield efficiencies and so
that each unit can concentrate on its activities for high effectiveness. The
73 The configuration and complexity of the firm

information-processing requirements of the symmetric organization are

very high because the coordination demands are high both horizontally
and vertically throughout the firm.

Diagnostic questions

For your firm, you can examine the two dimensions: product/service/cus-
tomer orientation and functional specialization in figure 4.7. Locate where
you are in the figure. Then you can categorize your firm as: simple, func-
tional, divisional, or matrix. To begin, answer the diagnostic questions
below.
1. What is your unit of analysis that you chose in Chapters 1, 2, and 3? Use
this unit of analysis as the organization when answering the questions
below.
The questions below will help you locate your organization on the
functional orientation dimension and the product/service/customer
dimension. For each item within question 2 and question 3, use a 1 to
5 rating scale to score your organization as follows:

1 2 3 4 5
very low moderate very high

2. Product/service/customer orientation
a. Is the focus of your activities on internal operations of the
organization (1) or, are you more focused toward products, services,
and/or customer (5)? —
b. To what extent do you form subunits to organize work around your
customers, (1)–(5)? —
c. To what extent do you form subunits to organize work around your
firm’s products or services, (1)–(5)? —
Now mark your organization’s location on the horizontal axis in figure
4.7.1
3. Functional specialization
a. To what extent does your organization create subunits based on
specialties or skills of its workers, (1)–(5)? —

1
We suggest that you use an averaging procedure of the detailed scores to get to the overall
score.
74 Burton, DeSanctis, and Obel

b. Is your orientation toward specialization of functions and clear

definition of jobs, i.e., high on functional specialization, or is your
orientation low on specialization, (1)–(5)? —
c. To what extent does the firm have clearly defined roles and
responsibilities for various subunits, (1)–(5)? —
Now mark your organization’s location on the vertical axis in figure 4.7.
4. You can now locate your organization on figure 4.7. What is your firm’s
configuration?
Next consider the organizational complexity. In Figure 4.8, the firm’s
horizontal differentiation and vertical differentiation are the dimensions.
Here are questions which will help you locate your firm as blob, tall, flat,
and symmetric:
5. Horizontal differentiation
At the lowest level of your firm, how many subunits are there across the firm
(where many subunits indicate a high horizontal differentiation and a few
indicates low horizontal differentiation), few (1) or many (5)? —
If there are more than 5 subunits, your organization is high in horizontal
differentiation.
6. Vertical differentiation
How many levels are there from the top of the firm to the bottom, few
(1) or many (5)? —
How many levels are there between the lowest-level employee and the
CEO, few (1) or many (5)? —
Roughly, if there are more than 5 levels, assign a rating of 5 (high vertical
differentiation). If there are 4 or 5 levels, assign a rating of 4. If there are 3
levels, assign a rating of 3. If there are 2 levels, assign a level of 2. If there
is 1 level, assign a rating of 1(low in vertical differentiation).

Figure 4.7. Locate your firm in the configuration space.

75 The configuration and complexity of the firm

Figure 4.8. Locate your firm in the organizational complexity space.

You can now locate your organization on the figure; what is your firm’s
complexity?

Fit and misf its

What is a good configuration for your firm, and what is an appropriate

organizational complexity? What is a good fit? Here we add fit for the
configuration and organizational complexity to the goals, strategy, and
environment you assessed in the prior chapters. This is shown in table 4.1.
In each of the columns A, B, C, and D, the fit relations can be read vertically
from top to bottom. Misfits are any set of relations which do not fall within
one column. There are a very large number of possible misfits. The critical
issue is the limited attention and time of the top management. Any of these
misfits can overload the executive management and lead to diminished
performance for the firm. The telltale signs are: decision backlogs where
the adjustments are not timely; individuals are either not told what to do or
not given activities that are coordinated one with the other; or the executive
works excessively long hours, but it is not enough.
In column A there is a fit among: the blob, a simple configuration, a calm
environment, a reactor strategy and ill-defined goals. The executive has the
time and can devote attention to handling the relatively low information-
processing demands of the configuration. If the strategy or the environment
requires more information processing, then the executive can quickly
become overloaded with undesired consequences. The simple configuration
has a misfit with any condition or combination of conditions, which gener-
ates excessive information-processing demands for the executive: the firm is
large; the strategy is a defender or analyzer which requires time-consuming
76 Burton, DeSanctis, and Obel

Table 4.1. Fit among configuration, environment, strategy, and goals

Corresponding
quadrant in
organization
design space A B C D
Organizational Blob Tall Flat Symmetric
complexity

Configuration Simple Functional Divisional Matrix

Environment Calm Varied Locally Stormy Turbulent

Strategy types Reactor Defender Prospector Analyzer Analyzer

with without
innovation innovation

Organizational Neither Efficiency Effectiveness Efficiency and

goals Effectiveness

planning and control; and the environment is not calm, which also requires
adjustments and changes.
For a simple configuration and blob type of complexity, the reactor strategy
is a good fit. This strategy lacks focus on the goals and usually is reactive to the
situation at hand. The simple configuration is well matched for a reactor
strategy as the executive can read the situation and then adjust to the market
and the customers. The simple configuration could also work well for the
prospector strategy where there is a narrow projective strategy. Here, rapid
adjustment is not required until a survival threat emerges. The simple config-
uration is not a good fit for a defender strategy where efficiency is very im-
portant or for an analyzer strategy where longer-term planning is the norm.
In column A the corresponding environment is the calm environment. At
first glance, the opposite turbulent environment may seem the better fit. The
simple configuration can adjust quickly, which may not be needed for a
calm environment which is predictable and not complex. But with all the
information-processing demand centered on one individual, the demand on
the executive quickly becomes overwhelming – especially when both the
operational, tactical, as well as the strategic decisions are located with one
individual. Nonetheless, the simple configuration can quickly adapt to
operational changes but usually has great difficulties adapting to strategic
changes. Therefore, the simple configuration works well for the calm envir-
onment, as there are fewer adjustments required. With limited attention and
77 The configuration and complexity of the firm

time, the executive finds the calm environment less demanding. The critical
factor is the attention and time limitation of the executive.
Moving to column B, there is a fit among: the tall complexity, the
functional configuration, the varied environment, the defender strategy
and the efficiency goal. The information-processing demands have
increased considerably, but the functional configuration and tall structure
can handle a large amount of information which is predictable within
known variations. Detailed and involved coordination can be realized.
However, if the environmental unpredictability increases requiring add-
itional information processing, then the functional configuration is not
suited to making large coordinated changes quickly – a misfit situation.
For the functional configuration and tall complexity, the defender strategy
with a focus on efficiency is a good fit where the defender strategy maintains
the firm’s position in its product/service market. Customers and clients are
usually continuing, frequently with deep relationships. The marketing/sales
function maintains strong ties with the customer. There is little emphasis on
innovation in new products or services; however, there can be an emphasis
on process innovation to reduce costs. The tall firm with a large middle
management can focus on the efficient use of resources with detailed
coordination of operations. The focus is continuity of operations based
upon detailed plans. The firm can vary its production quantities efficiently
within narrow limits, but larger variations in quantity or new activities
compromise that efficiency. The tall firm is a misfit with the innovation-
focused strategies. The defender wants to keep things as they are and the
functional configuration fits it well. For other strategies, the functional
configuration fits less well. It fits the analyzer without innovation reasonably
well as planned change is possible. But for the analyzer with innovation or
prospector, the functional configuration is too slow to be effective.
The corresponding environment for the functional configuration and
tall complexity is the varied environment which is predictable and has high
complexity. The emphasis is on predictability where the executive can
anticipate what will happen in the future. The executive can deal with the
complexity of the environment via rules and standard procedures that have
been developed and learned over time. But when the environment becomes
unpredictable – turbulent or locally stormy – then the functional configur-
ation loses its efficiency as quick adjustments are needed and it cannot
adjust quickly; it can perform the wrong activities as well. Frequently
executive managers will try to take on this additional burden of driving
change and become overloaded. The telltale signs of overload are backlogs of
78 Burton, DeSanctis, and Obel

decisions and changes that are not coordinated across the whole of the
organization.
Column C describes the firm that has a flat complexity, a divisional
configuration, a locally stormy environment, a prospector strategy and an
effectiveness goal. Each division has its own environment, which is stormy
but largely independent of the others. The executive can create independent
divisions to deal with the local conditions. Then the high-level executive can
deal with policy and financial issues – limiting the information-processing
demands. If, however, the environment becomes interdependent, such as
two divisions competing for the same customer, the executive can become
overloaded with coordinating details – again, creating a misfit. The tall firm
is a misfit for the divisional configuration. It would be impossible for the top
executive to become involved in the detailed operations of the firm, which
would only lead to overload and its inefficient consequences.
For the divisional configuration and flat complexity, the prospector
strategy with a focus on effectiveness is a good fit. The divisional configur-
ation fits well the corporate strategy of prospector, where each division’s
own strategy is not specified from the top but is developed within the
opportunities of the division’s own markets and resources. The flat firm
with a minimal middle management can focus on the efficient allocation of
resources across the divisions using financial criteria. The divisional mix
may include a defender strategy for one division and a prospector for
another. The flat firm is a misfit with the intense information and high
subunit coordination needed by a defender or analyzer.
The divisional configuration and flat complexity are a good fit with a locally
stormy environment which can be highly unpredictable, sub-environment by
sub-environment. Each sub-environment should closely match the division
boundary. That is, for a product division the environment for each product
can be unpredictable but not interdependent among the products. For Gen-
eral Electric (GE), the jet engine market is quite independent of the appliance
market, but each market and technology might be unpredictable. Here, the
top executive has partitioned the environment into independent segments,
where the divisional executive can deal with it. The flat firm does not deal well
with the highly coordinated operations which involve great detail; here there
will be opportunity losses from the poor coordination.
In column D, the firm has a symmetric shape, a matrix configuration, a
turbulent environment, an analyzer strategy and the dual goals of efficiency
and effectiveness. The information-processing demands are very large, and
detailed coordination of new situations is required (Galbraith, 1973). The
79 The configuration and complexity of the firm

firm cannot be broken down into relatively independent divisions, nor can a
tall hierarchy handle all of the changes required.
The matrix configuration is a misfit with the following conditions: there is
not a dual focus on efficiency and effectiveness which justifies the costs of
additional management with particular managerial skills for coordinating
up and across the hierarchy; the strategy is defender or prospector, which
can be achieved with a single hierarchy; the environment is not complex or
is predictable, which can be managed with a single hierarchy. When a single
hierarchy configuration is sufficient for the desired goals, then the matrix
adds cost without justification. When the strategy has a dual focus and the
environment is turbulent, requiring large information processing to deal
with the difficult ongoing coordination to achieve efficient use of resources,
then the matrix configuration is appropriate.
The analyzer strategy, both with and without innovation, is a good fit for
the matrix configuration and symmetric complexity. The analyzer operates
in two domains of symmetric complexity. The first is the exploitation of the
existing business, where efficiency is important; and, the second is explor-
ation of innovation or technology enhancement, where the effectiveness of
new ideas and products are paramount. The matrix configuration brings the
two together to realize the needed changes in a timely manner. The matrix
configuration fosters a dual focus on exploitation and exploration. It can
work well in a turbulent environment where there are a large number of
issues and problems to consider which are highly unpredictable, requiring
the firm to adjust quickly. Together, there is a very large requirement for
information processing to coordinate on an ongoing manner so that timely
decisions can be made and action taken. The matrix configuration can deal
well with the turbulent environment, and where there are many activities to
coordinate across the entire firm. The functional managers and the div-
isional (product, project, or customer) managers are the coordinators both
within their own subunit and across all of the subunits of the firm. There are
ongoing adjustments and task changes due to the unpredictability of the
environment. The information exchanges are very large and detailed. The
span of control can be wide when the matrix managers work together and
make the needed adjustments themselves, i.e., when the matrix is working
well. When the matrix is not working well and many decisions move up to
the top management, then the span of control can be quite limited. The key
is the amount of information processing that the top management must deal
with to achieve the coordination. The growth of the matrix is quite limited if
the top management is involved in the details of operations. If the top
80 Burton, DeSanctis, and Obel

management is not so involved, growth can be larger but is limited because

the coordination difficulties grow nonlinearly as each new matrix dimension
adds many coordination interfaces (Burton & Obel, 1984). The symmetric
complexity is not a good fit unless the environment is both unpredictable
and complex. If one of these conditions holds and not the other, the simpler
tall or flat firm is sufficient.
If your firm is located in different columns based upon your answers to
the diagnostic questions of these four chapters, then you should think about
what you might do to bring all elements of the firm’s organizational design
into fit in the column that meets the firm’s goals. But also think about what
is involved to move to a different goal and thus a different column.

Summary

In this chapter, you have assessed your organization’s configuration and

organizational complexity and considered how these relate to your firm’s
strategy, environment, and goals. There are four configurations: simple,
functional, divisional, and matrix; and there are four complexity types: blob,
tall, flat, and symmetric. We discussed how your firm’s configuration and
complexity should be designed so as to improve the likelihood of the firm
reaching goals, given the environment and your firm’s strategy. If there are
misfits among the design components, then you should consider how these
might be changed in order to bring them into alignment.

Glossary

Blob complexity: an undefined organization in the sense that there are no

formally specified subunits.
Configuration: assignment of work to units of the organization; indicates the
hierarchy of responsibility; represented pictorially as an organization
chart; an icon of the organization; sometimes called the structure or
architecture of the organization.
Divisional configuration: tasks are assigned to relatively independent
divisional units by product, customer, region; or other externally
oriented focus; each division relatively self contained; executives make
policy and financial decisions; organizational chart represents the
divisions reporting to the headquarters.
81 The configuration and complexity of the firm

Flat complexity: an organization with many jobs at the bottom and few
levels bottom to top; high horizontal differentiation and low vertical
differentiation.
Functional configuration: tasks are assigned by specialization of work; tasks
are grouped by skills requirements.
Functional specialization: the primary partitioning of the firm’s task into
smaller specialized tasks and the efficient completion of each task.
Horizontal differentiation: the degree of task specialization across the
hierarchy.
Jello effect: the situation where a small change in one part of the
organization requires change and adjustment throughout the organiza-
tion. E.g., in a matrix configuration, a change in one function or project
frequently requires adjustments in a number of other functions and
projects.
Matrix configuration: a combination of a functional and divisional form; a
dual focus; a dual hierarchy.
Organizational complexity: the vertical and horizontal differentiation
dimensions of the organization; the shape of the organization.
Product/service/customer orientation: the primary partitioning of the firm’s
task into smaller tasks by the output orientation and their effective
completion.
Simple configuration: tasks or work activities are specified on an ongoing
basis rather than in advance; an organization where one individual, the
boss, is responsible for all activities; the organization chart usually shows
all employees reporting to one person.
Span of control: the horizontal complexity or width of the organization,
usually measured as the number of individuals reporting to a manager.
Symmetric complexity: an organization with a balance of specific jobs and
levels; neither tall nor flat; high on both horizontal and vertical
differentiation.
Tall complexity: an organization with a large number of levels from bottom
to top; low horizontal differentiation and high vertical differentiation.
Vertical differentiation: the depth of the hierarchy; total number of levels,
top to bottom.
5 Distributed organizations

Introduction

In the previous chapter you described the organization in terms of its basic
structure. Today many organizations operate in multiple locations and use
variants of the basic organizational structures presented in Chapter 4.
Distributed organizational designs enable firms to coordinate work across
national and other geographic boundaries and meet the knowledge needs of
our increasingly service-oriented economy.
This segment of our step-by-step approach consists of two analyses. First,
we will assess your organization’s approach to organizing across geographic
boundaries. Second, we will assess the organization’s approach to organizing
for knowledge exchange. The variants in structure described in this chap-
ter are sometimes called ‘‘new organizational forms’’ because they repre-
sent new twists on classic ways of organizing (DeSanctis & Fulk, 1999;
Heydebrand, 1989). The variants described here represent options for
designing your firm to meet the demands of today’s global, information-
intense organization.
When an organization is founded, it typically starts by doing business in
one locale. That locale services a particular neighborhood, city, or even an
entire country. As the executive managing a business centered in one locale,
you become an expert in doing business in that environment. You know the
people, the culture, and the general setting in which your firm operates. Your
firm becomes highly knowledgeable about that one locale. This allows you to
build efficiency and effectiveness based on local knowledge. In addition,
doing business with partners, that is, managing relationships with suppliers,
distributors, government regulators, and other entities, is relatively straight-
forward when all are co-located. You share a common language, laws,
customs, and ways of doing business.

82
83 Distributed organizations

As your business expands to include operations or sales activity to regions

that are geographically distant from your home locale, you face significant
organizational challenges. These relate to managing people and processes
across distance and culture. The challenges are important whether you are
managing a large organization or a small work team. Should you organize
around the talent (expertise) or function, regardless of locale, or should you
organize based on geographic location? To what extent should you allow
various locales to operate independently versus enforce standard ways of
doing things across the firm? Should you own or directly manage operations
in distal locales, or would it be better to form partnerships or alliances that
permit local units to organize as they see fit? To what extent should you rely
on information technology, versus travel and face-to-face meetings, to link
people together?
As an example, consider the European appliance manufacturer, Merloni
Elettrodomestici, an Italian-owned company with headquarters in Paris.
Initially limited to regional operations in Italy, the company began exporting
its products in 1972. The company soon had a sales force distributed across
Europe and later acquired or built facilities in Germany, France, the UK,
Spain, Portugal, Holland, and Belgium (Bower, 2001). The firm then faced
the challenge of how to organize a growing, distributed enterprise. Merloni
first created country-based organizations, all coordinated through the Paris
office; later the company consolidated by product line, creating several
centers of excellence that coordinated all operations for a given product
throughout Europe. By 1994 a matrix reporting structure was added to link
research and development activities with brand-based activities. In more
recent years the firm has created a ‘‘developing markets division,’’ with
country-based organizations in Russia and China while maintaining prod-
uct-line organizations in Europe. Merloni’s experience illustrates the kinds
of design choices that firms have to make as they expand their businesses
around the world.
Similar design decisions must be made by managers of small work teams.
Consider an energy engineering firm operating within a large multina-
tional enterprise. The large company may operate in a divisional fashion,
but any one division may have engineering teams with research expertise in
Germany, design expertise in the US, and production expertise in Malaysia.
How should the team organize? All work could be coordinated through
one locale, such as the US facility, or, instead, team members in each of
the three countries could do their work autonomously and simply link
tasks together on an as-needed, project-by-project basis. Team members
84 Burton, DeSanctis, and Obel

could complete all task demands on their own, or they could hire external
parties to work as contractors, passing routine or specialized work to the
contractors and leaving core work tasks for the regular team members. The
team could designate liaison representatives in each country to coordinate
with team members in the other countries; or teams could set up a matrix
system in which specialists from each locale regularly communicate with all
parallel specialists in the other locales.
Every day you face these kinds of design decisions in your organization.
As an executive, you decide how work is to be allocated and coordinated
across place and space. When workers are in different time zones, speak
different languages, have different holiday schedules, work habits, and
skill sets, the work allocation and coordination challenges escalate. World-
wide distribution of work creates access to resources to accomplish the
organization’s mission, and it can bring your company close to the custom-
ers and suppliers with whom you do business; but distribution of work also
creates immense challenges in how to design work processes for the greatest
efficiency and effectiveness.
Let us begin by assessing an organization’s approach to organizing across
geographic boundaries. Next, we will assess the organization’s approach to
organizing for knowledge exchange.

Structures for spanning geography

A simple and powerful way to describe a firm’s approach to managing across

distance is in terms of the extent to which it locates based on optimal
sourcing versus a particular geographic boundary, and the extent to which
it locates to yield local responsiveness versus global standards and economies
of scale. Other dimensions could be used as well; however, we focus on these
because they relate to theory and research and provide simplicity within our
step-by-step approach.
Optimal sourcing refers to the decision to locate operations in the place
in the world that brings the greatest advantage to the firm in terms of
customer contact, cost efficiency, human resource skill need, or other ob-
jective. For example, a software company may choose to locate in the Silicon
Valley of the US or Sophia Antipolis on the Côte d’Azur of France in order
to be close to programmers and engineers with needed technical skills
for new product development. Alternatively, if the firm seeks low-cost,
high-skilled labor, it may locate one or more facilities in Hyderabad or
85 Distributed organizations

New Delhi. In order to reach customers in developing markets of Russia, the

company may locate offices in Novgorod, Kursk, or Vladivostok. If a firm
locates work based on placing it as close as possible to the resource supply,
then the firm is placing high value on optimal sourcing. Alternatively, if a
firm locates work based on other factors, such as being conveniently close to
headquarters or in a particular city, country, or geographic region where the
company has existing business or operations, then optimal sourcing is low.
Local responsiveness refers to the decision to distribute work in many
locales versus consolidating work in one or a few centralized locations.
Distributing work to many locales maximizes your firm’s flexibility to
complete work tasks any time, any place. A highly distributed workforce,
perhaps consisting of an army of mobile salespersons, software program-
mers, or service operators, exemplifies this extreme. These workers may be
organized by country or region if the firm is very large; but the key is that the
locally responsive firm tries to distribute work as much as possible, thus
enabling close contact with customers or suppliers and the ability to antici-
pate and respond to their needs. Consolidation of operations reduces local
responsiveness, although it brings economies of scale, the opportunity to
standardize work practices, and a general increase in managerial control over
the work.
Optimal sourcing and local responsiveness represent tradeoffs in organ-
izational design that correspond to the firm strategies of exploitation and
exploration discussed in Chapter 2. To the extent that your firm pursues
exploitation, you should organize work to be high in optimal sourcing.
To the extent that the firm pursues exploration, you should organize work
to be high in local responsiveness.
The tradeoffs associated with organizing for optimal sourcing and local
responsiveness yield four basic options for structuring the distributed or-
ganization. The following typology is based on the seminal work of Bartlett
and Ghoshal (1998), who studied how multinational companies develop
strategy and marketing worldwide. Bartlett and Ghoshal described their
categories in broad terms and multiple dimensions. The typology we present
here is strictly based on two dimensions and focuses on organizational
structure (and not the broader strategy and marketing issues). Our typol-
ogy descriptions diverge somewhat from theirs, but we stick with the
fundamental assumptions of their theory.
Figure 5.1 displays the organizational design space of optimal sourcing and
local responsiveness along with the four types of distributed organizational
86 Burton, DeSanctis, and Obel

Figure 5.1. The organizational design space of structures for spanning geography.

design that relate to these dimensions: (1) global, (2) international, (3) multi-
domestic, and (4) transnational.
These four approaches to organizational design across geography cor-
respond to the four classic organization configurations described in
Chapter 4, with global corresponding to the simple configuration, inter-
national corresponding to the functional configuration, multi-domestic
corresponding to the divisional configuration, and transnational corres-
ponding to the matrix. As with the classic configurations discussed in
Chapter 4, the typology illustrates major design approaches. Variations are
possible, and firms sometimes combine these approaches, perhaps using
different designs at different levels of the firm; or there may be variation
across business units, with some units designing globally, for example, and
other units taking an international or transnational approach. Consistency
in design up-and-down or across the multinational firm is not imperative.
More important is that you choose a configuration that is consistent with
your organization’s goals, strategy and environment.

Global

If your organization is not organized to be locally responsive or to yield

optimal sourcing but instead concentrates its work activities in one locale
(usually the place of your firm’s founding, i.e., its corporate headquarters),
then it is global in design. The global design is a centralized approach to
organizing and thus yields the advantage of high centralization of decision
making and work practices that are established by the ‘‘home base’’ of
operations. A global organizational design is consistent with a strategy of
87 Distributed organizations

offering similar products or services worldwide, no matter where products

are ultimately sold. Work practices (the way work is done) are controlled by,
and consistent with, the way corporate headquarters wants things to be
done. Organizations that offer identical products or services to all markets
and concentrate their upstream and support activities in one country are
global in design. The global design scores low on both optimal sourcing and
local responsiveness. The global design is akin to the simple configuration
outlined in Chapter 4.
Consider the case of a software company that offers project management
software solutions to customers all over the world. The company is head-
quartered in the US, in Arlington, Virginia. The management, developers,
and support staff are located there. Sales representatives may call on cus-
tomers throughout the world, and contract workers from locales other than
Virginia may be hired to complete specific tasks. Support staff may make site
visits to customers throughout the world. But the management of all these
work activities is centered at the Virginia headquarters. There is little
customization of work as a function of geography. Instead, work is managed
from the corporate center. There may be customization of products and
services to meet the needs of specific customers, but the arrangement of this
work is controlled at the center, that is, in the US locale. This organization
has a global design.

International

International firms move beyond the home base of operations to create

‘‘centers of excellence,’’ or hubs for each major product or service. These
hubs are located so as to maximize optimal sourcing; that is, work is located
as close as possible to the resource the organization needs to do the work,
wherever that may be in the world. These hubs then service worldwide
markets. Local responsiveness is low as the products and services tend to
be standardized rather than customized to location.
The upstream segment of the oil business traditionally has been managed
using the international approach. Drilling operations are placed anywhere in
the world where there is an accessible, fertile oil field. This may mean that
the firm has many operations in Alaska, Malaysia, or off the coast of western
Africa, and very few operations in Europe or North America. Thus the
upstream business is not balanced or bounded by geography; instead, it
diffuses to ‘‘where the work can best be done.’’
88 Burton, DeSanctis, and Obel

Let’s consider a modern, service-oriented business, such as the example of

the software company described above. Suppose our software development
firm, located in Arlington, Virginia, has grown such that its project manage-
ment software products have sold to customers throughout the US, Western
Europe, Japan, Taiwan, and Australia. The company now services thousands
of customers in these locales. This has created a huge workload for the
corporate operations; further, the company cannot find enough program-
mers in Virginia at sufficiently low cost to meet the software support needs
of customers who are now spread around the globe. To meet customer
support needs, the company may open two operations in India, perhaps
one in New Delhi and one in Hyderabad, where high-quality customer
relations and technical skills are available at a relatively low cost of labor.
Further, these operations have sufficient labor pools to replenish workers
who leave the firm in good economic times when turnover tends to be high.
The New Delhi operation serves as a customer call center. The Hyderabad
operation serves as a technical support center where programmers work on
software fixes and upgrades. This is an international type of organizational
design.
The international organizational design is like a functional form in that
configuration is based on concentrating expertise, or skills, as a way of
managing work. The organization distributes work geographically, away
from corporate headquarters, but corporate headquarters maintains a great
deal of authority. The organization is ‘‘tall’’ rather than flat; that is, the
vertical differentiation is high.

Multi-domestic

If a firm takes a decentralized, geography-based approach to organizing

work then its configuration is multi-domestic. The multi-domestic organiza-
tion customizes operations to specific countries or regions, offering unique
products or services to meet local preferences. The multi-domestic is like a
divisional form. Work is organized to exploit local opportunities, especially
the cultural, political, and geographic opportunities associated with a given
locale. Divisions or other operations are located in different locales to yield
local responsiveness to customers, not to yield optimal sourcing. The multi-
domestic organizational design can be very effective for the firm that is
entering markets which are very different from the home locale, where
management has little experience in the new locale and wants to benefit
89 Distributed organizations

from learning about customer needs or possible ways of doing work in that
locale. This type of organizational design is flat in shape rather than tall,
meaning that work practices will vary as a function of locale. The locale
could be a region or a country.
Consider the case of 3M in the US. As the firm entered Europe in the mid-
twentieth century, it took a multi-domestic approach. At that time the
company was interested in acquiring research and development expertise
in Europe and in generating new business there; however, the American
management had relatively low knowledge about local ways of doing work,
customer needs, and so on, across the European market. Further, the
European market was fragmented, with language, currencies, employee
work practices, and customer tastes all varying across countries. Given
these environmental conditions, along with a strategy of exploration in the
European market, 3M established country-based organizations, and gave
each a high degree of autonomy to manage work and grow the business in
these countries (INSEAD, 1994). By the end of the twentieth century,
conditions were different. The EU was established, work practices were
more similar across Western Europe, and, in general, the European market
became more (though not entirely!) homogeneous. With this environmental
change, 3M moved to a regional-centered design, consolidating the country
operations. Locale remains a cornerstone of 3M’s organizational design
today, with operations organized as a function of region, including the
Americas, Western Europe, Eastern Europe, Russia, China, Japan, Southeast
Asia, Africa, and the Middle East. Organizing by locale supports high local
responsiveness. Products and services, as well as management of employees
and work tasks, vary as a function of region to provide maximum local
responsiveness.
Like the divisional configuration, the multi-domestic approach to struc-
turing work supports growth via exploration of new products and services.
The multi-domestic configuration is a good choice of organizational design
if the source of growth is geographically based. Consider our example of the
software development firm, headquartered in Arlington, Virginia. Suppose
the firm seeks to grow the service side of its business, offering a variety of
project management services to customers that use its project management
software. If the firm assesses the environment and determines that service
needs are quite different in, say, developed countries versus developing
countries, and, further, that there are major differences in service needs in
Asian countries versus Western countries, the firm could adopt a multi-
domestic organizational design for management of its service business. The
90 Burton, DeSanctis, and Obel

firm might establish regional operations in Japan or Singapore to service the

many developed economies of Asia. It might then establish a country-based
operation in mainland China to explore the unique needs of that market.
Similarly, it might locate an operation in London from which it would
service the Western European market, with additional offices in Moscow
and Budapest, to service the unique needs of these more volatile, growing
economies. The multi-domestic approach involves significantly more invest-
ment for the otherwise corporate-centered firm. This approach to organiza-
tional design requires that the firm establish local presence, develop
employees to manage in that locale, and so on. But the potential payoffs
for growth are great, if the locales are selected and managed well. Over time,
less successful local operations can be shut down, while those that are more
successful can be allowed to grow.

Transnational

The transnational organization blends the international and multi-domestic

structures to yield both the location advantages of regional or country-based
design and the economic efficiencies of optimal sourcing. In the trans-
national design some operations are located close to needed resources; but
location decisions also are made such that the firm has presence in all areas
of the world that are of strategic importance. In this way, the organization
develops customized offerings by region while at the same time gaining
efficiencies through worldwide centers of operation. The transnational or-
ganization takes a sophisticated approach to locating its operations. Some
are centralized in the home locale; some are optimally sourced, wherever
those sources are located; and others are distributed among country or
regional operations.
Unilever, Proctor and Gamble, and NEC are examples of companies that
have adopted transnational designs (Bartlett & Ghoshal, 1998). Consider the
detergent business within Unilever. Research and product development
activities are located based on optimal sourcing. Basic research facilities
are located in the US and Europe, in centers close to universities and ample
supplies of chemists and chemical engineers. Product development groups,
on the other hand, are located close to the business units that they serve,
wherever that may be in the world. There are manufacturing facilities in Asia
and Latin America, where natural resources are available and labor costs are
relatively low, but sales, distribution, and service operations are localized, in
91 Distributed organizations

some cases by country, or even region within a country, to reach and

respond to the needs of particular customer groups.
Why choose a transnational design? This design makes sense if the
organization has diverse and conflicting strategic needs and/or high vari-
ation in the business environments the firm confronts as it moves outside of
its home country. Suppose you manage the software company described
earlier, with corporate headquarters in Arlington, Virginia. You might take
an optimal sourcing approach to locating a technical support or call center,
perhaps placing both facilities in India despite low presence of customers
in that country. At the same time you might locate sales and service
operations in London, Moscow, Budapest, Japan, and China, so that you
reach your major strategic markets. These facilities might operate some-
what differently, depending on the needs of customers in the regions they
serve and the degree of investment your company intends to make in the
particular region. The transnational approach to your organizational de-
sign means that you disperse subunits, develop specialized centers of oper-
ation, and then link these through effective management of interdependent
relationships (Cullen, 2002).
The transnational is the most complex distributed design to manage
because some aspects of the firm’s work are region-centric whereas other
aspects of the firm’s work are resource-centric. Like the matrix organization,
management of the transnational organization requires a combination of
centralized and decentralized decision making. Managed well, the trans-
national design can bring the benefits of high efficiency and effectiveness
to a firm. The corporate headquarters must be very adept at knowing
what work is best located as a function of sourcing and what work is best
located to yield local responsiveness. Most important, once the location
decisions have been made, management must be adept at coordinating
among the firm’s distributed operations. This means that there must be
structures to support the exchange of knowledge among the various
geographic locales. We now turn to the possibilities for organization of
knowledge exchange.

Structures for managing knowledge exchange

A major challenge for the distributed organization is to structure so as to

maximize efficiency and effectiveness of knowledge exchange. We focus on
knowledge exchange for this component of organizational design, rather
92 Burton, DeSanctis, and Obel

than the more general issue of information exchange. Knowledge is infor-

mation that corresponds to a particular context. Knowledge exchange is the
sharing of information that requires interpretation, or intelligence, to fully
understand and apply. Configuring your organization for knowledge ex-
change is important, because how you design for knowledge exchange will
affect other aspects of your organization’s design, especially the design of
coordination and control systems. Structures for managing knowledge ex-
change constitute the infrastructure on which the basic organization config-
urations discussed in Chapter 4 rely as they become more distributed in
design.
Structures for knowledge exchange can help the organization to increase
its information capacity, that is, the amount of information that the organ-
ization can process. Consider the case of a pharmaceutical company that is
continually under pressure to develop new products while controlling costs
in an intense, highly competitive industry. Suppose you are the chief exe-
cutive of this company. The company has subunits working on tasks related
to primary care products, oncology, vaccines, neurological agents, meta-
bolic, and gastrointestinal products. Each of these product areas has its
unique research and production requirements, and each may face special
scientific, legal, or other constraints as it conducts its work. Each area is
continually gathering and generating knowledge to do its work, but at
the same time some or all areas may benefit from knowledge processed by
other groups. The potential for shared knowledge may be chemistry-based,
or it may relate to the development of research platforms, testing methods,
statistical analyses, production methods, sales opportunities, and the like.
How can you structure your organization for knowledge exchange? What
are your options? In today’s modern enterprise, you can rely on two im-
portant mechanisms, virtualization and information technology, to manage
knowledge exchanges.
Virtualization refers to the degree of boundary-spanning or organiza-
tional ‘‘reach’’ that a company uses as the basis for knowledge exchange
(Davidow & Malone, 1992). Organizations that are high in virtualization
look outward, linking teams, business units, or even the firm itself with
parties outside the organizational boundary in order to gain knowledge.
Organizations that are low in virtualization take a more inward focus,
gaining knowledge by developing it inside corporate boundaries, inside
specialized groups, or by acquiring knowledge externally and then har-
boring it inside the firm. An example of the latter occurs when one com-
pany makes an acquisition of another company in order to capture new
93 Distributed organizations

Figure 5.2. Structures for managing knowledge exchange.

capabilities from the marketplace. IT infusion refers to the extent to which a

firm relies on information technology-based systems, including data pro-
cessing and computer-based communication systems, to manage knowledge
exchange. Although nearly all organizations today rely on IT to acquire and
transfer knowledge, some firms rely more heavily on IT-based systems for
knowledge exchange, whereas others rely more heavily on face-to-face or
manual systems to support knowledge exchange.
These two dimensions suggest four major types of organizational designs
for knowledge exchange, as shown in figure 5.2.

Ad hoc communications

Ad hoc communications are the primary knowledge exchange approach

used by firms that score low on both the dimension of virtualization
and IT-infusion. These organizations rely on person-to-person contacts or
small groups (2–10 people) whose members are all from inside the organi-
zation (or organizational unit) to share knowledge on an as-needed basis.
Small, loosely created groups of people provide knowledge-based innova-
tion that is ground-up, meaning that it comes from the people closest to
the work at hand. These informal groups typically are temporary. They
come together to meet a specific task need. They usually are appointed
by management, work with a fair amount of autonomy and may vary in
the methods used to track their progress and share results with those out-
side of the group. Ad hoc communications manage knowledge exchange on
an as-needed basis; members adjust the way they organize work and report
94 Burton, DeSanctis, and Obel

to one another and their boss depending on the nature of the particular
task at hand. In ad hoc communications, a leader’s preferences for mana-
ging work may have a high degree of influence in exactly how the work is
organized. So, there are not pre-established routines for conducting work.
Instead, the organization of work – how knowledge is shared and tasks are
completed – is put together on the fly, to meet the particular work needs. Ad
hoc communications can be a very effective way to generate and transfer
information to meet the specific knowledge needs of a given project, event,
or client.

Informated

Shoshana Zuboff (1988) coined the term the informated organization to

refer to the use of computer and related systems to manage information up
and down the organization. Informated organizations are low in virtualiza-
tion and high in IT infusion. The label harkens back to an earlier era of
automation, or machine-based design of work, but informated implies more
than automated work. Today computer technology is heavily embedded in
the design and monitoring of work processes such that tasks can be stream-
lined, closely linked with one another, and continually managed for im-
provement in quality and cost control (Keen & McDonald, 2000). Most
modern banks use information systems to monitor customer inquiries and
call center operators to continually improve customer support and yield
greater worker productivity. In a similar way, UPS, the worldwide package
delivery service, links information systems inside trucks with package-
tracking systems that allow customers, workers, and management to view
systems operations from the time of order placement all the way to package
delivery. These are examples of informated organizations. Via computer
technology, events and processes are made visible and measurable, making
redesign and customization of work possible. The informated organization
increases the intellective demands of work, as well as the possibilities for
creative ways of rearranging and linking work activities. Informated organ-
izational design is possible in a wide range of industries from manufacturing
to the service sector. Going forward, we can anticipate more informated
organizational design in businesses such as hospitals, consulting, and edu-
cation as they increasingly rely on intense knowledge exchange to meet
client needs.
95 Distributed organizations

Cellular

Cellular organizational forms are characterized by small, autonomous

groups or business units that largely self-govern and can grow, reproduce,
and form relations with other units as needed (Miles et al., 1997). The label
is based on the biological metaphor of a living cell. The cellular organization
has an inward focus for knowledge exchange within teams or units, coupled
with boundary-spanning across cells and into the marketplace in order to
gain access to external sources of knowledge. Compared to the informated
organization, which is high-tech and competes on its ability to rapidly
arrange and rearrange work processes, the cellular organization competes
on the ability of cells to import and export knowledge while harboring
information inside for creative or competitive benefit. Furthermore, com-
pared to the informated organizations, cellular organizations will have more
variability in the process of knowledge exchange within subunits. Compared
to ad hoc communication-based organizations, cellular organizations de-
velop extensive relationships with external parties for knowledge sustenance.
Thus the cellular organization scores high on virtualization but relatively
low on IT-infusion.
Like ad hoc communications, cellular organizations rely heavily on rich
forms of interaction between people for knowledge exchange. Information
technology may be used to coordinate work, but systems will not be
consistent or perfected in the sense of being fully engineered for smoothness
and quality control. Instead, flexibility in the way knowledge is exchanged is
important. High interpersonal contact, such as conferences, seminars and
face-to-face meetings, may be used to facilitate tacit knowledge sharing. The
goal is to promote an entrepreneurial mindset with a focus on innovation
and growth within the cell and the larger organization of which it is a part.
Cellular forms tend to be suitable for research firms and divisional forms
where highly varied approaches to knowledge development and exchange
are needed across the subunits.

Network

The network organization links units within the firm with one another
and, further, develops active linkages between internal units and external
organizations to meet the organization’s knowledge needs. It scores high
96 Burton, DeSanctis, and Obel

both on virtualization and IT-infusion. Networks often take the form of

strategic alliances, research partnerships, and consortia. Like the cellular
form, the network configuration would be nearly impossible to create
without the use of modern information technology. The network is similar
to the traditional matrix configuration in that it is both product and
specialty intense; however, unlike the matrix, it lacks symmetry or balance
in its structure (Miles & Snow, 1986). Information technology is used to link
units in multiple directions, not just vertically or horizontally (Boudreau
et al., 1998). Theoretically, the network can operate as a ring, a star, or a web
of ties among the subunits of the firm. Resources, people, and ideas flow in
all directions. Specific points of exchange can be made between the subunits
and with external parties ‘‘that matter’’ to the business need. In this way, the
network does not necessarily get out of control, creating information over-
load. Instead, network ties are formed and managed intelligently, putting
knowledge exchange when and where it’s needed (Hansen & Nohria, 2004;
Velstring et al., 2004). The network organization combines the information-
intensity of the informated firm with the boundary-spanning approach to
knowledge exchange found in the cellular organization.

Diagnostic questions

1. What approach does your chosen organization use to span geography

and support knowledge exchange? Answer the questions below.1 Use the
same unit of analysis when answering these questions as you have used in
the prior chapters.
For each item within questions 2 through 4, use a 1 to 5 rating scale to
score your organization as follows:
1 2 3 4 5
very low moderate very high

2. What is the organization’s degree of local responsiveness (1)–(5)?

To answer this question you may think about the following:

1
As in prior chapters, you can use an averaging procedure of the detailed scores to get to the
overall score for each dimension, or you can use the questions as a general guide to estimate
your firm’s score on each dimension.
97 Distributed organizations

a. To what extent are the units of your firm located close to corporate
headquarters (1) or far from corporate headquarters (5)?—
b. To what extent does your firm consolidate work in one region of the
world (1) or distribute its work to many locales (5)?—
c. To what extent are the important business decisions in your
organization made with a corporate perspective in mind (1) versus
a local perspective (5)?—
d. Overall, is the firm organized to assure centralization of decision
making and consistency of work practices across subunits (1) or
autonomy of local units and customization to meet local needs
(5)?—
How does your chosen organization score on local responsiveness?
Now mark the organization’s location on the local responsiveness axis in
figure 5.3.
3. What is your organization’s degree of optimal sourcing (1)–(5)?
To answer this question you may think about the following:
To what extent does your firm make decisions about where to locate its
operations based on the following criteria:
a. close proximity to customers (1)–(5)?—
b. close proximity to human resources, whether skilled or unskilled (1)–
(5)?—
c. close proximity to suppliers (1)–(5)?—
d. close proximity to business partners (1)–(5)?—
e. close proximity to resources to its ongoing business (1)–(5)?—
How does the organization score on optimal sourcing, low or high?
Mark your organization’s location on the local responsiveness axis in
figure 5.3. Now with your scores on local responsiveness and optimal
sourcing, locate the organization in the geographic design space shown
in figure 5.3.
4. What is your organization’s degree of virtualization (1)–(5)?
To answer this question you may think about the following:
a. To what extent does the organization rely on internal (1) versus
external (5) sources of knowledge to do its work? ___
b. To what extent does the organization develop specialized groups, or
centers, of expertise that are free to grow on their own (1)–(5)? ___
c. To what extent does the organization develop unique partnerships
with other units or organizations for purposes of fostering
innovation, product development, or innovative forms of service
support (1)–(5)? ___
98 Burton, DeSanctis, and Obel

Figure 5.3. Locate your organization in the geographic space by rating its degree of local
responsiveness and optimal sourcing.

d. To what extent is the organization dependent on information

supplied by other organizations or units in order to do its work
(1)–(5)? ___
e. To what extent are other organizations or units dependent on
information supplied by the organization or unit in order to do their
work (1)–(5)? ___
Based on your rating for items a through e above, determine the score for
virtualization. Now mark the organization’s location on the virtualiza-
tion axis in figure 5.4.
5. What is your organization’s degree of IT-infusion (1)–(5)?
To answer this question you may think about the following:
a. To what extent does the organization rely on computer-based systems
to manage its most critical business activities (1)–(5)?—
b. To what extent does the organization invest in continual improve-
ment of business processes (1)–(5)?—
c. Which tends to be more frequently used to manage interpersonal
communication, face-to-face contact between people (1), or elec-
tronic communication systems (5)? —
d. Do systems of communication in the organization tend to be
informal and changing (1) or more formalized and prescribed (5)? —
Now mark your organization’s location in the organizational design
space for knowledge exchange shown in figure 5.4.
Next we want you to examine how your firm’s approach to distributed
design and knowledge exchange fits with its configuration, as well as
goals, strategy, and the other design components that you identified in
the prior chapters.
99 Distributed organizations

Figure 5.4. Locate your organization in the knowledge exchange space by rating its degree of
virtualization and IT-infusion.

Fit and misf its

Table 5.1 summarizes the four columns in the organizational design space
thus far. For each row in this table, circle the firm’s type (A, B, C, or D), as
selected in this and the prior chapters. Next, review the columns. Are all of
the circled items in the same column? To the extent that the circled items fall
in the same column, the organization has good alignment, or fit, among its
goals, strategy, and structure. As before, design types in the same column
tend to fit well together.
Firms in column A are simple in structure and manage knowledge on an ad
hoc, informal basis. These organizations don’t have a strong organizational
form; instead, the task design is changed to meet the needs of the work at
hand. When firms in column A ‘‘go global,’’ that is, when they venture outside
of their own country, they tend to do well with centralized, headquarters-
based management. Hence, a global approach to geographic distribution
makes sense. Headquarters can ‘‘call the shots’’ and manage on the fly to
meet the demands of the business. There is great flexibility in this organiza-
tional design, but its efficiency and effectiveness are limited. Management in
the home country locale must be very adept at processing information for the
firm as a whole. As we have noted in earlier chapters, this type of organiza-
tional design works well when the environment is stable or the organization is
small. If the environment shifts in some significant way or becomes more
turbulent, the global organizational design can be very slow to adapt.
Firms in column B organize knowledge by specialty, i.e., by area of
expertise, and then use information processing to gain high efficiencies in
100 Burton, DeSanctis, and Obel

Table 5.1. Fit and misfit table for geographic distribution and knowledge exchange

Corresponding
quadrant in
organizational
design space A B C D
Knowledge Ad hoc Informated Cellular Network
exchange communications

Geographic Global International Multi-domestic Transnational

distribution

Complexity Blob Tall Flat Symmetric

Configuration Simple Functional Divisional Matrix

Environment Calm Varied Locally stormy Turbulent

Strategy types Reactor Defender Prospector Analyzer Analyzer

with without
innovation innovation

Organizational Neither Efficiency Effectiveness Efficiency and Effectiveness

goals

business processes. These are functionally designed firms that distribute

work based on optimal sourcing. The functional organizational structure
does well to organize for geographic distribution using an international
model, creating centers of excellence and informating to gain improvements
for managing business processes and enabling cost control. Alternatively, if
the firm has a divisional structure, then when it distributes globally, a multi-
domestic form of organizing presents a better fit. In the latter case, each
domestic locale can operate as a cell, developing creative methods to meet its
unique knowledge needs. The column C approach to organizational design
supports a prospector strategy.
Column D is perhaps the ideal approach to organizational design for the
geographically distributed firm with high information-processing demands.
This organization form should develop network forms of knowledge ex-
change to meet the demands of a turbulent environment. The network
approach to organizing facilitates both exploration and exploitation strat-
egies, thus facilitating goals of both effectiveness and efficiency. As discussed
earlier in this chapter, the transnational organizational design is the trickiest
to manage; a clever combination of local and global operations, and the use
of IT-infused and rich human interactions must be applied to meet business
101 Distributed organizations

needs. If a firm has been successful in managing a matrix design in its home
market, then it is more likely to be able to manage a transnational design in
global markets.
More than likely, not all of your organization’s design components are in
the same column (i.e., they do not all fall within the same column of table
5.1). Where you see misfits you should ask yourself whether these misfits are
causing decrements in your firm’s ability to meet its goals. If not, it may be
fine to live with the misfits. However, as you look to the future, you should
consider whether any misfits should be corrected. A large number of misfits
should give you pause, and you should consider adjusting your firm’s misfits
such that you bring the organizational design components of table 5.1 into
the same column. Note that misfits that are in adjacent columns are less of a
concern than those that are in opposing (distal) columns.
As we end Step 3 in our step-by-step approach, it is important to keep in
mind that organizational design is an ongoing process, not a goal state to
achieve. Design components and their fit with one another must be de-
veloped on a continuing basis. A firm may be informated today, but if it
does not invest in development of information systems on an ongoing basis,
the firm may lag in this design component in the future. Similarly, a firm
might implement systems but employees resist their use or find over time
that they rely on more flexible, ad hoc systems to do their work. In this way,
your firm can drift from one column to another as time goes by. Sometimes
this drift is intentional on the part of management and other times it
happens due to managerial neglect or due to changes in how people choose
to do their work. For example, if the firm is growing or is acquired, yielding
a fresh labor pool, IT-infused work that was prevalent in the past may
become more people-based if newcomers are not trained in, or accepting
of, established systems. A new manager may foster networks of relationships
to exchange knowledge, even if the organizational strategy calls for a defen-
sive, more inward focus on exploitation. The point is that what you see as a
fit today can migrate to a misfit tomorrow, whether planned or unplanned.
It is important that management continually assess and develop fit among
the organizational design components.

Summary

The modern organization embellishes basic configurations of simple, func-

tional, divisional, and matrix to yield distributed arrangements, including
102 Burton, DeSanctis, and Obel

global, international, multi-domestic, and transnational. Firms that are

consistent in their basic (Chapter 4) and distributed (Chapter 5) organiza-
tional structures are more likely to achieve their efficiency and effectiveness
goals. In this chapter we reviewed the design alternatives for geographic
distribution and we described how organization structure can be used to
support different forms of knowledge exchange. We now move on to the
next step of organizational design, which is the design of process and people
systems to support organization structure.

Glossary

Ad hoc communications: use of temporary, informal pairs of workers or

work groups from inside the firm as the primary design for knowledge
exchange. Virtualization and IT infusion are low.
Cellular organization: small, autonomous subunits that largely self-govern
and can grow, reproduce, and form relations with other units, including
those outside the firm, as needed; information technology facilitates team
interaction and connections among the various ‘‘cells.’’
Global: a distributed organizational design in which work activities are
centrally organized to yield the advantage of consistency of work
practices, as established by the ‘‘home base’’ of operations (usually the
corporate headquarters). Optimal sourcing is low.
Informated: embedding of computer technology in the design and
monitoring of work processes such that tasks can be streamlined, closely
linked with one another, and continually managed for improvement in
quality and cost control. Virtualization and IT infusion are high.
International: a distributed organizational design in which work is located as
close as possible to the resources the organization needs to do the work,
wherever they may be in the world. Work is located close to resource
inputs, not as a function of customer location or to spread operations
across locales or regions of the world.
Knowledge: a higher-order notion of information; knowledge is information
that corresponds to a particular context and requires interpretation, or
intelligence, to fully understand.
IT-infused: the extent to which an organization relies on information
technology, including data processing and computer-based communi-
cation systems, to support knowledge exchange.
103 Distributed organizations

Local responsiveness: the decision to distribute work in many locales versus

consolidating work in a centralized location.
Multi-domestic: a decentralized, geography-based approach to organizing in
which operations are customized to specific countries or regions, offering
unique products or services to meet local preferences.
Networked organization: a distributed organizational design in which links
of information exchange are IT-infused and established between units
within the firm and between internal units and external organizations to
meet the organization’s knowledge needs.
Optimal sourcing: the decision to locate operations in the place in the world
that brings the greatest advantage to the firm in terms of customer
contact, cost efficiency, human resource skill need, or other objective.
Transnational: an organizational design that blends the international and
multi-domestic structures to yield both the location advantages of
regional or country-based design and the economic efficiencies of
optimal sourcing.
Virtualization: the degree of boundary-spanning (reaching across the
borders of the organization), that a firm uses as the basis for knowledge
exchange.
104 Burton, DeSanctis, and Obel

Where are you in the step-by-step approach?

STEP 1
Getting Started
(1) Goals
STEP 2
Strategy
(2) Strategy
(3) Environment
STEP 3
Structure
(4) Configuration and complexity
(5) Geographic distribution and knowledge exchange
In Step 3 you described the organization’s structure in terms of:
configuration, complexity, geographical location and knowledge
exchange. Do these four parts of the firm’s structure fit together? If
not, which should be changed, and how would you go about it?
Think carefully about this question. Further, review where your firm
is located in the spatial graphs presented in Step 3. Do you think
that the description in the figures gives a fair and reasonable
representation of your unit of analysis? If not, which changes would
you like to make? Would you like to see your organization move
from one place on the graph to another place? Consider the
accuracy and the internal consistency of your organization and,
again, think about where your organization is and where you would
like it to be, and your plans for changing organizational components
so that they are in fit with one another.
Once you are satisfied with the organization’s structure as analyzed
in Step 3, then you should review the goals, strategy and
environment. Are they aligned or are there misfits with structure?
If you have misfits, think about what you can and might do. Usually,
105 Distributed organizations

if there are misfits between the strategy and the structure, the
structure should be adjusted to fit the strategy, i.e., structure follows
strategy.
Step 3 is complete, and you are ready to move on to Step 4.
STEP 4
Process and People
We will examine the task design in your firm, that is, how the
organization carries out its work. Then you will assess your firm’s
people, leadership, and climate.
(6) Task design
(7) People
(8) Leadership and organizational climate
STEP 5
Coordination and Control
Step 4 Process and people
6 Task Design

Introduction

How should an organization be designed to perform its work? At the most

basic level, one can think of an organization as performing a very large task
which must be broken down into smaller and smaller tasks in order to get
the work done. Suppose you manage a software design company. Should
you divide the work into processes such as design, development, sales, and
service; or might it be better to divide the work according to client type:
individuals, small business, large business, and government? Of course,
other options are possible too. Once a firm selects a way to organize the
work at the highest level (the big task), there is the question of how work
should be divided inside of each of the subtasks. Within subtasks, the work
is further divided, until it reaches the lowest task-level of the organization.
Task design is decomposing work into subtasks while considering the
coordination among the subtasks to meet organizational goals. Prior to
the information age, task design was sometimes called ‘‘technology design’’
by organizational designers. In the traditional setting of manufacturing,
technology design was a matter of figuring out whether work should be
arranged sequentially (as in assembly lines), in parallel (as in custom
building), via teams that continually passed work back and forth among
members, or in some other way. Today the word ‘‘technology’’ has broader
meaning, so we use the simpler label of task design; but the essential design
question remains the same, which is how the big task of a firm is broken
down into smaller tasks and how these smaller tasks interconnect with one
another so that the big task is successfully completed.1 As we shall see, a

1
There are two complementary task-design approaches. We begin with the total organizational
task and examine its decomposition into smaller tasks. The complementary approach is to begin

109
110 Burton, DeSanctis, and Obel

firm’s approach to task design is related to its choice of efficiency and

effectiveness goals, as well as the structure and strategy of the firm. Task
design determines the coordination requirements for the firm’s work, and
thus it is vital that there is fit between task design and the other components
of organizational design. Given a firm’s strategy and structure, some
approaches to task design will fit better than others.
Researchers have described the approaches to task design in several differ-
ent ways. Woodward (1965) in her classic studies of organizations categorized
tasks as: unit, mass, and process production, where each had a different task
design. Unit is more craft-like; mass is assembly line; and process is auto-
mated. These represent distinctly different ways of organizing and managing
work. She found a nonlinear relationship between task design and other
components of the organization’s design. The unit and process approaches
had many components in common, whereas the mass production approach
was different. Compared to mass production, the unit and process produc-
tion had higher skilled workers, lower organizational complexity, lower
formalization, and lower centralization. The work pace of mass production
was very precise and so required more detailed coordination than the less
clock-driven unit production and the maintenance-oriented process produc-
tion. Woodward’s studies were the first to link task design to other dimensions
of organizational design. Thompson (1967) categorized the relationship
between tasks as: sequential, pooled, or reciprocal. Sequential tasks are co-
ordinated by standardization of the tasks; pooled tasks are coordinated via
planning and task allocation; and reciprocal tasks are coordinated by mutual
adjustment. Carroll et al. (2005) examined the dynamics of changing these
task relationships and found in a project setting that transforming sequential
tasks into parallel and reciprocal tasks decreases project time initially, but
increases project time over time as more reciprocal tasks demand much
greater coordination and costly mutual adjustments.
Scott (1998) described task design along three dimensions: complexity of
items requiring simultaneous consideration, uncertainty, or unpredictability,
and interdependency, where a change in one requires a change in another
item. (Note that these task design characteristics are similar to the environ-
mental characteristics described in Chapter 3.) Greater complexity, greater
uncertainty, and greater interdependency all require greater information
processing to obtain the coordination required to get the work completed.

with the individual works tasks and aggregate them into larger tasks (Sinha & Van de Ven, 2005).
For both approaches, coordination of the subtasks is a central issue.
111 Task Design

Summarizing the above discussion, an organization’s task design can be

categorized along two important dimensions: repetitiveness and divisibility.
If a task is well defined such that it is undertaken again and again, then it
has higher repetitiveness. Notice that standardization in execution of the
task enables repetitiveness. If the task is not standardized and varies in how
it is done, then it has low repetitiveness. A highly repetitive task has low
uncertainty, whereas a task low in repetitiveness has a higher uncertainty.
When a bigger task is broken down into subtasks which require little
coordination (i.e., the subtasks are independent), it is highly divisible. On
the other hand, if the subtasks require high coordination with one an-
other (i.e., they are interdependent), then the task has low divisibility. Note
that this definition of divisibility is related to Thompson’s categories of
sequential, pooled and reciprocal interdependency. With these two di-
mensions, we have four basic task designs which we call: orderly, compli-
cated, fragmented, and knotty. The four task design categories are shown
in figure 6.1.
We will discuss the four categories briefly before considering each in
detail. The orderly task design is highly divisible and highly repetitive; it
requires relatively little coordination among the subtasks to accomplish the
work. The complicated task design is not very divisible but is highly repeti-
tive. It requires more coordination of the connected and repetitive tasks. The
fragmented task design is highly divisible, but not very repetitive. It requires a
different kind of coordination to adjust to ongoing variations across the
subtasks, but adjustments for connectedness among subtasks is not re-
quired. The knotty task design is neither divisible nor repetitive; it is the
most difficult to coordinate as adjustments to both connectedness and
non-repetitiveness are required simultaneously.

Figure 6.1. Task design space.

112 Burton, DeSanctis, and Obel

For your unit of analysis try to think about the highest-level task that
must be designed in your organization. How is it defined? How is it
currently designed? For example, if your chosen organization is a bank,
the work of the bank could be divided into subtasks based on specialization
such as handling investments, money transfer, lending, etc. Alternatively, the
work of the bank could be defined in relation to customer groups: private
customers, institutional investors, small business customers, etc. The bank
may define the work so that it can be repeated, by standardizing transactions
for all specializations or customer groups. Alternatively, it may take pride in
customizing the work, deliberately avoiding standardization so that inter-
actions with customers or others are managed uniquely. When we talk about
task design we are thus talking about the overall design, not just the design
of the individual subtasks. The coordination requirements are very different
in the two task designs just presented. Task design is also related to business
process re-engineering as well as process management methods and
philosophies, such as just-in-time and supply-chain-management.
Next, we consider the four task designs in more detail. In figure 6.1, we
begin in the lower left corner for an orderly task design and move to
complicated, fragmented, and finally knotty task design.
As we go through these task types, it is important to keep in mind where
we are in the organizational design process. Task design follows design of
strategy and structure. As an executive, you have a choice about how to
design the work of your firm. The choices described below may be more or
less appropriate, depending on your firm’s strategy and structure.

Orderly

If you choose an orderly task design, then you are organizing your firm’s
work so that it is highly divisible and highly repetitive. You break up the
work into pieces so that you can direct each work unit to perform independ-
ently of other units. When each unit completes its work, the results flow
back up to the executive level, and then you assign a new piece of work to
that unit. Units that experience problems or difficulties turn to you to
resolve problems. An obvious advantage of this task design is that slow-
downs or other difficulties in one unit don’t prevent other units from
continuing progress on their tasks. Within each unit the tasks are standard-
ized as much as possible so that they can be readily repeated. To the extent
113 Task Design

that workers in each unit are able to develop skills to do the tasks assigned to
their unit, the specialization of tasks can yield very high efficiency.
When tasks are designed using the orderly approach, there is almost no
coordination required between units performing the subtasks of the
organization and no need for them to adjust to one another. Piece work,
whether in manufacturing or service, has these characteristics. The work of a
law firm could be organized in this fashion too. As clients contact the firm,
they may be assigned to an attorney who handles their case independently.
Once a case is closed, the attorney is assigned another case. The attorneys
operate independently, processing cases, taking as much time as needed
before moving onto the next case. As another variation on the orderly
approach, the attorneys might be grouped by specialty such that customers
with family law needs are assigned to the family law group, clients with
criminal law needs are assigned to the criminal litigation group, and so on.
Again, the work of the organization is divided across units such that
individual units perform their work independently of the others, completing
the entire task assigned to them (low divisibility). Completing the ‘‘big task’’
work of the firm is accomplished as the work is more or less standardized
and the individual units gain expertise to do their assigned work in an
efficient manner. As another example, consider mill workers who do hand
sewing and are given an inventory or list of things to do. Each worker takes
an assigned garment to sew and places finished items in an out-basket. These
in-process inventories help create the divisibility of work. The worker may
have fixed productivity targets to meet, and these are monitored at the
executive level. The executive has little to coordinate, except to assure that
assignment and completion of work is done in a satisfactory manner.
Further, in the orderly approach it is the executive level that brings in the
work to the organization; the workers or subunits do not typically
solicit work on their own. It is the executive-level responsibility to assure
that each unit or worker has something to work on.

Complicated

If you choose to design your organization’s task so that it is low on

divisibility but yet remains highly repetitive, then you have a complicated
task design. Complicated tasks require a high degree of coordination due to
low divisibility; that is, the subtasks can be performed by different units of
114 Burton, DeSanctis, and Obel

the firm, but they are interdependent to get the work done. As an example,
suppose you manage a hospital emergency room. You might divide the work
into four subtasks: (1) admissions, (2) triage screening, (3) focused care, and
(4) release. Patients move sequentially through these processes, with differ-
ent groups of people (subunits) responsible for each of the four subtasks.
The work processes are repetitive and the services remain quite standardized
(at least at the level of the ‘‘big task’’ design). The complicated task design
suits processing of large volumes of work. There are many examples of a
complicated task design in manufacturing, the most classic being the
automobile assembly line. McDonald’s is an example in the restaurant
industry. The subtasks of order processing are highly repetitive but not
divisible, as the completion of an order for a customer requires that each
part of a meal is assembled correctly. Every order is unique within a limited
set of possibilities so that the tasks become very repetitive.
Mass production requires not only the skills of orderly production but
also precise coordination among the units responsible for the subtasks. The
production processes must be timed to avoid bottlenecks and to meet
efficiency goals in which inventories between processes are minimized.
A well-designed complicated task requires that these work processes are
repetitive and ongoing. The executive level overseeing the firm’s work
focuses on the coordination of the connected processes, which require
continuous attention. Given the low divisibility, a breakdown in any one
small task can shut down the whole operation, which can be very costly.
Detailed and ongoing coordination requires a high level of information
processing. Advances in operations research, along with the embedding of
information technology into manufacturing processes, have increased man-
agerial success in using complicated task designs. Firms that use these
designs can compete based on their ability to process work with great speed
and sophistication.

Fragmented

If you choose to design your organization’s task so that it is high on

divisibility yet low on repetitiveness then you are using a fragmented task
design. Fragmented tasks require less coordination than complicated tasks
due to their high divisibility. By reducing coordination needs, each subunit
can process work at its own pace; it doesn’t have to wait for other units
115 Task Design

to complete their work in order to proceed. Further, the subunits can take
creative approaches to completing their tasks, perhaps soliciting their own
customers or clients, since bottlenecks are not a concern. By breaking down
the big task of the firm, the subunits are likely to be more innovative and
aggressive. Some may outperform others or contribute more to the firm’s
overall work completion.
Consider a technology development firm, such as a software developer,
that is trying to grow its business. The needs of its customer base (individ-
uals, small business, large business, and government) are quite different, that
is, they are not repetitive. The work of each subunit is conducted independ-
ently. Alternatively, the firm might divide the work according to type of
software, such as desktop software and network-based software. In either
case, if the big task is treated as highly divisible and with low repetitiveness,
then the task design is fragmented. Within each subtask, the work could be
further fragmented, or another task design might be selected.
The fragmented task design means that the firm divides its work so as to
accommodate the varied nature of its business. Although it is tempting
to think that task design is inherent in the work itself, it is important to
recognize that in many cases the same work might be designed in different
ways. Thus task design is a matter of managerial choice. Suppose your
organization is an investment bank. You might choose a fragmented
approach, dividing your big task into subtasks such as investment coun-
seling, trust services, and estate planning. Each group is free to solicit its
own customers and design its services to meet customer needs. There may
be repetitive work within each of these subunits, but at the level of the
big task design of the bank, there is low repetitiveness; that is, customers
are directed to one group or another, and work is accommodated to meet
their unique needs. The nonrepetitiveness approach to task design requires
lots of adjustments (i.e., execution of work is not standardized); but as
these adjustments are not connected, the coordination requirements are
quite minimal. To manage a fragmented task design, the executive needs to
assure that the subtasks (i.e., the subunits) have resources and a reading
on the environment, but the executive need not be involved in detailed
coordination. In the case of the investment bank, the fragmented
design may not be the ideal choice, especially if customers prefer that the
subtasks be coordinated, e.g., if they want their estate planning to involve
their trust accounts. This is the downside of designing tasks to have high
divisibility. The investment banker might consider a knotty task design
instead.
116 Burton, DeSanctis, and Obel

Knotty

The knotty task design is low on both divisibility and repetitiveness. If you
choose a knotty task design for your organization, then you will have to
invest in ways to coordinate work among the subtasks and at the same
time support the nonrepetitive approach to doing the work. Knotty tasks are
not standardized. This approach to task design encourages those responsible
for subtasks to develop innovative ways to do their work, accommodating
the unique demands of each customer, while at the same time those per-
forming subtasks must integrate their work with other units in the firm.
Knotty tasks are likely to lead to the greatest customer satisfaction since
production is customized, but they are the most demanding type of task
to manage.
When products are new, the knotty approach to task design is often
favored by managers. High technology innovative products and services
are illustrations – such as a new and short-lived video game, a biotech
entity, or a new global financial instrument. The executive focuses on the
coordination of the connected processes, which are continually changing.
Given the low divisibility, a breakdown in any one small task can shut down
the whole operation, which can be very costly. Given the nonrepetitive
approach to task design, the information-processing demands increase
greatly. Taken together, the information-processing demands go up non-
linearly with executive overload. Therefore, this task design is the most
demanding on management.
New product development (NPD) in automobiles at Toyota or Renault,
pharmaceuticals at Eli Lilly, or household products at Unilever, requires
high coordination and adjustment of the tasks to the emerging tech-
nology. NPD tasks are often designed according to a knotty approach,
but a knotty approach can be applied in more routine industries for
competitive advantage. For example, a gourmet restaurant may create
new food offerings each day, with each new offering requiring unique
production and high coordination among the kitchen staff. Since the
task is designed to be nonrepetitive – providing a new dining experience
each time customers visit the restaurant – the organization must have
highly skilled staff that can continually innovate and coordinate with
perfection.
117 Task Design

Diagnostic Questions

How is the task designed in the organization you have chosen to design? As
in prior chapters, use the same unit of analysis that you selected in Chapter 1
to answer the following questions. In answering these questions, it is very
important to take a top-down approach and limit the analysis of the task to
the ‘‘big task’’ of your unit analysis. (Remember, subtasks, once created, have
their own designs.)
Note that the rating scales for task design are reversed, so that 1 ¼ high
and 5 ¼ low.
1. What is the degree of repetitiveness of the task in the firm, i.e., high to
low?
a. Does the firm treat each work task as unique (low)?
b. Does it execute the task today much as it did yesterday (high), or is
there a good deal of variation (low)?
c. To what extent does it standardize the task (high) rather than
customize it (low)?
Score the repetitiveness on a scale from 1 to 5 as follows:
1 2 3 4 5
very high moderate very low

2. What is the degree of divisibility of the task in the firm, i.e., high to low?
a. Does the firm divide its big task into subtasks that are independent of
one another (high), or are the subtasks connected, requiring a lot of
coordination (low)?
b. Does it manage the task as a set of specialized independent functions
(high) or as a process flow (low)?
c. To what extent are the units that perform the subtasks free to design
their work as they wish (high) rather than as instructed (low)?
Score the degree of divisibility on a scale from 1 to 5 as follows:
1 2 3 4 5
very high moderate very low

You can now locate your firm on the graph in figure 6.2. What is the firm’s
task design?
118 Burton, DeSanctis, and Obel

Figure 6.2. Locate your chosen firm in the task design space.

Fit and Misf its

The table below is the table from Chapter 5 with the task design row added.
Again, there is fit among the design elements of your chosen firm if the
entries for each row fall into the same column. Misfits are deviations from a
common column.
The orderly task design is appropriate if your firm’s goal is neither
efficiency nor effectiveness. There can be some efficiencies in the orderly
task design due to its repetitiveness, and for this reason, managers may
find this approach to task design to be appealing. The orderly task
design approach works well so long as the environment is calm and the
corresponding strategy is a reactor which is also unfocused. A simple
configuration works well for the orderly task design as it breaks the total
task into smaller tasks which require very little coordination from the
executive. Knowledge can be exchanged on an ad hoc basis. So long as
things are calm, the organization with the simple configuration using the
orderly task design creates minimal information-processing requirements.
The executive is not overloaded with detailed coordination problems –
unless the environment changes. The risk for the firm occurs if new business
causes a shift in the type of work needed such that high repetitiveness is not
possible. Then the orderly approach is a misfit and the task design inappro-
priate. Organizing work so that it is divisible and can be executed as
independent subtasks puts a high load on the manager if there is any change
in the environment.
If your organization adopts an orderly task design approach then you
should be aware that this is a misfit with an efficiency strategy of a defender
119 Task Design

Table 6.1. Fit and misfit for task design

Corresponding
quadrant in
organizational
design space A B C D
Task design Orderly Complicated Fragmented Knotty

Knowledge Ad hoc Informated Cellular Network

exchange communications

Geographic Global International Multi-domestic Transnational

distribution

Complexity Blob Tall Flat Symmetric

Configuration Simple Functional Divisional Matrix

Environment Calm Varied Locally stormy Turbulent

Strategy types Reactor Defender Prospector Analyzer Analyzer

with without
innovation innovation

Organizational Neither Efficiency Effectiveness Efficiency and Effectiveness

goals

or an effectiveness strategy of a prospector. Any deviation from a calm

environment creates difficulty as adjustments will be required. Functional,
matrix, and divisional configurations are more costly and are not needed to
achieve the required coordination for the highly divisible and highly repeti-
tive task. For most organizations, an orderly task design is not sustainable
except for the most routine operations, making these types of tasks good
candidates for automation or outsourcing.
The complicated task design is focused more on efficiency than on
effectiveness. The corresponding strategy is a defender where the efficiency
of repetitiveness helps achieve profitability through low cost. The varied
environment, which is complex but predictable, is a good fit for the compli-
cated task design. So is the functional configuration since it has the capacity
to coordinate detailed and standardized processes which rely heavily on rules
and procedures.
The complicated task design is a misfit with an analyzer strategy which
requires innovation and introduces nonrepetitive processes. Similarly, tur-
bulent and stormy environments require adjustments which are extremely
120 Burton, DeSanctis, and Obel

difficult to make if you have designed your organization using a compli-

cated task design. The executive will be overloaded with the coordination
details.
The fragmented task design is focused more on effectiveness than on
efficiency. This approach to task design works well if you are pursuing a
prospector strategy, seeking high degrees of effectiveness and continual
innovation. If your firm faces a locally stormy business environment, i.e.,
high unpredictability, then it makes sense to design work so that it is high in
divisibility and low in repetitiveness. The divisional configuration is a good
fit; here management focuses on providing resources and policy but not
detailed coordination. The cellular configuration is also a good fit.
The fragmented task design is a misfit if your firm has the dual goals of
both efficiency and effectiveness. Fragmented task design breaks the big task
into subtasks which are relatively independent and optimal in the use of
resources. It is therefore hard to achieve efficiencies for the big task if the
fragmented task design is adopted.
The knotty task design is appropriate if your firm has the dual goals of
both effectiveness and efficiency. The turbulent environment which is com-
plex and unpredictable is a good fit. The corresponding strategy is an
analyzer with innovation. The matrix configuration is a good fit because it
emphasizes coordination across multiple dimensions and ongoing coordin-
ated adjustments of the work to meet organizational goals. The knotty task
design customizes work and so, if done well, can yield high customer
satisfaction for a range of customer demands. As we shall see in the next
chapter, conducting the knotty task requires highly skilled employees and
management that can simultaneously support autonomy, control, and learn-
ing as tasks are executed.
The knotty task design is a misfit with any strategy, environment or
configuration which has a dominant focus on either efficiency or effective-
ness. If your chosen firm is pursuing a defender or prospector strategy, then
you should avoid the knotty approach to task design because it is too
complex and expensive and so not the best fit for your goals.

Summary

This chapter on task design further completes the description of a firm to

find a design that fits with its goals, strategy, and structure. In this chapter
121 Task Design

you described an organization’s task design in terms of repetitiveness and

divisibility, and categorized it as: orderly, complicated, fragmented or
knotty. If the organization’s current approach to task design does not fit
its organization’s goals, strategy or structure, it should consider adjusting the
task design so that the task design is aligned with the other dimensions. Next
we turn to the human resource requirements that are needed to support a
firm’s task design.

Glossary

Complicated task design: an organizational task design in which work is

organized in a way that it is not very divisible but highly repetitive;
usually requires a high degree of coordination among the subtasks.
Divisible task design: a task which can be broken into subtasks that are
relatively independent of one another with respect to resource utilization
and dependency of operations.
Fragmented task design: an organizational task design in which work is
organized to be highly divisible but not repetitive; usually requires less
coordination compared to complicated task design.
Knotty task design: an organizational task design in which work is organized
in a way that it is neither divisible nor repetitive; usually requires not only
coordination among subtasks but also support for the nonrepetitive
nature of subtasks.
Orderly task design: an organizational task design in which work is organized
in a way that it is highly divisible and highly repetitive; usually requires
relatively little coordination among the subtasks.
Repetitive task: a task which is well-defined (i.e., standardized) so that it is
executed again and again.
Task design: decomposing work (the big task) into subtasks and then
coordinating among the subtasks to meet organizational goals.
7 People

Introduction

‘‘An enterprise is its people,’’ so stated the famous Japanese industrialist

Matsushita Konosuke. Like many successful business leaders, Matsushita
believed that effective management of people was an essential ingredient
to a firm’s ability to reach its goals (PHP Institute, 1994). From an organiza-
tional design perspective the question for the executive is: what is the best
way to manage people, given the organization’s goals, strategy, structure, and
task design? As is the case for other dimensions of design, multiple ap-
proaches are possible. Deciding among these approaches depends on two
critical factors: the number of people in the organization and their profes-
sionalization. The organization and the people must fit together. Depending
on the design choices you have made with regard to goals, strategy, struc-
ture, and task, different approaches to managing people are recommended.
We will deal with the people dimension of organizational design in two
ways. In this chapter you will describe the size and capabilities of a firm’s
workforce. These dimensions affect the firm’s information-processing cap-
acity. In the next chapter you will assess the leadership and organizational
climate of the firm. These dimensions affect the ability of people to handle
information and make decisions.
We defined the organization in Chapter 1 as a social entity, so it follows
that how you manage the people in your firm depends on how many people
you employ and the kinds of capabilities they bring to the organization.
Managing people is a complex matter, and many factors might be con-
sidered. Here we want to focus on the most fundamental factors that relate
to organizational design. So we take a minimal approach in this chapter,
focusing on (1) the size of the labor pool, and (2) the degree of professional-
ization. In the next chapter we will address the attitudes of top management

122
123 People

and employees as they are important as well with respect to understanding

the relationship between people and the organizational design.
Depending on the relative size of the firm’s workforce and its professional
capabilities, different managerial approaches are appropriate. Our attention
to these two factors fits with our information-processing view of the firm.
Most of the information processing in a firm is done by the individuals in
the firm. People represent the intellectual capacity of the firm. This is
especially true in today’s knowledge-intense enterprise.
Although people bring skills and intellectual resources, having more
people is not necessarily a better state from an organizational design stand-
point. Large organizations (i.e., those with more people) must be designed
differently than smaller ones. For example, larger firms are usually more
decentralized. If there are only two people in a firm, then the decision
making, communications and coordination are easy. As the number of
people increases, communication becomes very problematic. People gener-
ate information as they do their work, and they also require information as
inputs to their work. So there is a growing need for information exchange as
the size of the firm increases. If each person talks to everyone else, then the
communication links grow quickly and exponentially with the number of
people (Burton & Obel, 2004). It is not a practical solution for General
Motors to have all of its 300,000 employees talking with each other. Even for
much smaller firms or subunits, communications are limited. Electronic
communication systems may make transmission of information relatively
easy, but, as we know from information-processing theory (Chapter 1), each
person’s attention is limited and costly. This is also true for the subunits in
which people are organized. Information-processing capacity is limited. So
we need ways to limit information and focus it on the goals and tasks. The
configurations we discussed in Chapters 4 and 5 limit communications and
direct them along the hierarchy in the functional and divisional configur-
ations or across units in the matrix. Indeed, one of the major reasons to
form configurations is to manage the otherwise extraordinarily high infor-
mation flow in the firm. Depending on the organizational configuration,
one person communicates with a small fraction of the total workforce in the
firm. The firm is able to coordinate its activities across a very large number
of people.
As you design your organization, you must decide not only whether to
employ many people or a few but also what types of people are needed,
given your strategy and structure. The professionalization of the workforce is
a measure of its skills, knowledge, and capacity to both generate and
124 Burton, DeSanctis, and Obel

produce information. If your organization has a more professional work-

force, then individuals can perform tasks that are more complicated, lengthy
and are more cognitively difficult. Education, training, and experience
increase the professionalization. In turn, how people are managed can affect
their professional development. Organizations that are not designed to
benefit from and enhance the professionalization of their workers are
less likely to meet their efficiency and effectiveness goals. The individual’s
knowledge is the basis for what he/she can do. In this sense, it is the
realized skill. This knowledge may be explicit, which means it can be
codified. Or it may be tacit, which means it is not readily documented.
Explicit knowledge is easier to capture and transfer around the organization
to others; tacit knowledge is far more difficult to transfer and requires rich
forms of social interaction in order to be shared. The more professional-
ization that exists inside the firm, the greater is its capacity to exchange tacit
knowledge.
Whether explicit or tacit, knowledge is the basis for the skills and other
aspects of the professionalization level, as well as the routines and other cap-
abilities people apply in doing the work of the organization. It is important
to note that individuals are ‘‘boundedly rational,’’ which means that we are
limited in our capacity to process information (March & Simon, 1958). It
seems obvious that each of us cannot do all things perfectly and instantan-
eously. We have imperfect information, which we interpret reasonably but
imperfectly; and we communicate only a fraction of what we would like to
communicate and, again, imperfectly. The bounded rationality of people is
at the heart of why we need an organization. At the most fundamental level,
we need configurations, task designs and information systems to permit us
to reach large goals in the face of our bounded rationality. Even with high
professionalization, individuals are boundedly rational, and the organiza-
tion is a way to cope with that limitation while at the same time harnessing
the skills and capabilities that people collectively offer for performing work
tasks.
The number of people in an organization and their professionalization
measure the basic characteristics of the people dimension of the organiza-
tion’s design. The number of people is simply a count of all individuals in the
firm (i.e., unit of analysis). Professionalization is the collective skill level of
the individuals and a measure of their capabilities for the work tasks at hand.
Professionalization depends upon employees’ education, training, and ex-
perience, i.e., their knowledge base, whether accumulated prior to their hire
or on the job.
125 People

Figure 7.1. The people space.

In Figure 7.1, there are two dimensions: professionalization on the hori-

zontal axis and the number of individuals on the vertical axis. Depending on
these two dimensions, there are four general approaches to people manage-
ment. Beginning in the lower left corner, there is the shop where the number
of people is low and professionalization is low. Moving to the upper left
corner, we have the factory, which has a large number of people, but
relatively low professionalization. In the lower right corner professionaliza-
tion is high, but there are few people, which we call a laboratory. Finally, in
the upper right corner there are many people with high professionalization,
which we call an office. Each category describes a different approach to
designing the people component of organizational design. We will now look
into each of the four categories.

Shop

The shop approach to managing people involves employing few people

who are low on professionalization. The shop design works well if the
individuals have not had specialized training or if their experience has not
given them extensive skills. Examples of such organizations are small stores
that employ people who are only given a few days of training. In this design,
the information-processing capacity of the employees is low. Routines must
be simple with only a few steps; they must be easily understood and easy to
learn. The manager of the shop must be ‘‘hands on’’ in directing people in
order for the shop to run smoothly. Coordination requirements are low so
long as there are few people, particularly if the task design is fragmented.
126 Burton, DeSanctis, and Obel

The shop is not focused on efficiency or effectiveness. People are generally

not efficient due to low skill and are expected to wait for direction from
management. People are not encouraged to develop skills, make decisions,
or advance significant change to improve the organization.
The shop design is appropriate if tasks are orderly and the available
workforce is small and low-skilled. In this case, the manager can give
individualized instructions and change the instructions as work comes
into the organization. Difficulties arise, however, if a large number of people
or those with professional skills are managed via the shop approach. If
management tries to ‘‘micro manage’’ in the sense of directing a large
number of people on what to do on an individualized basis, then the
organization will be inefficient. If management tries to micro manage
professionals on what to do on an individualized basis, then the organiza-
tion will be ineffective. For larger organizations, or for those with a profes-
sional workforce, management should consider a factory, laboratory, or
office approach.

Factory

The factory approach to managing people involves employing many indi-

viduals who are low in professionalization. Here the assumption is that
people have relatively little specialized expertise and the routines are rela-
tively simple, meaning that their work tasks can be executed repetitively
following training. Employing a large number of individuals means that
there are very high coordination requirements. Thus, the factory is focused
on efficiency, which requires detailed coordination for a large number of
individuals. To run a factory, you need many people, and you need focused
skill sets rather than broad professionalization.
The assembly line is a classic example of the factory. To build an automo-
bile, the total task is broken down into hundreds of small tasks each of which
is relatively low-skilled. But the overall coordination is extremely high as the
matching fender, wheel, engine, and hundreds of other items must come
together exactly on the assembly, where task design makes it routine to put
the pieces together. Today’s modern call center manages people as a factory.
Large numbers of people are employed and given focused, repetitive tasks to
perform as they place or receive large numbers of phone calls. People with
low professionalization are ideal for such work. Another example is a large
127 People

hotel, where there are a large number of low-skill jobs, and tasks are
designed in an orderly fashion with high repetitiveness.
If workers have high professionalization, then the factory approach to
managing people is less desirable, since the design does not take advantage
of the knowledge and skill capacities of professionals.

Laboratory

The laboratory approach to managing people involves employing a few

people, each with high professionalization. Professional routines which are
obtained through extensive education, training, and experience permit rela-
tive independence among the individuals or among small collections of
individuals (such as those working together in a cellular configuration). High
professionalization facilitates worker autonomy rather than strict supervi-
sion, so each one can work alone and the manager’s job is to support the
individual. A good example could be a university faculty where independent
scholarship is the norm. Other examples include a financial research group,
salespeople who seek their own clients, and high-tech software design groups.
In the laboratory design detailed coordination of activities is not required; in
fact, it may be considered intrusive in the sense that directive management
can stifle knowledge production and discovery of innovative ways of doing
things. Motivation and incentives are the important means for coordination.
The laboratory is focused more on effectiveness and high quality than on
efficiency. Each individual, or subunit of professionals, can work with relative
autonomy and achieve the purpose for the organization. There are many high-
level work routines for the individuals, but these routines are varied and
largely under the control of the individuals, not the organization. The routines
are very complex and include many tacit aspects which cannot be codified but
have high-quality requirements. Detailed coordination is usually not required.
Instead, workers coordinate in small groups for intense information sharing,
and these groups then build directed knowledge exchange with the specific
clients, subunits, or other sources needed to accomplish their tasks.
3M is a good example of a company that uses a laboratory approach to
managing people. Individuals and subunits are encouraged to create new
products and services with relative autonomy. It is a cellular configuration
where there are relatively few individuals in each cell, but it replicates itself
into a very large corporation.
128 Burton, DeSanctis, and Obel

Off ice

The office approach to managing people involves employing many people

with high professionalization. Again, professionalization comes from edu-
cation, training and experience. Due to the large number of people to
manage, the need for coordination is very high. The organization must
process lots of information and support extensive communication. High-
level work routines are very important as they help define and manage work
for efficiency and effectiveness.
Large consulting organizations have highly skilled professionals who must
be coordinated in complicated detail to meet the needs of the client. These
organizations often manage people as an office. Employees are given auton-
omy to do their work but they also engage in heavy communication with
others as they develop client projects and meet customer needs. Consultants
may be organized into subunits which, in turn, are managed as an office,
meaning that the subunits operate with a combination of autonomy and
inter-unit coordination. Many large-scale engineering construction firms
and new product development projects within pharmaceutical firms are
managed in this way.
Deciding whether to organize as an office – like deciding about other
components of organizational design – is a matter of managerial choice.
The office approach is recommended if a firm has large numbers of people
with high professionalization. Note that the communication demands of the
office are larger than in the laboratory. Workers are given autonomy to find,
process, and produce knowledge; but they also are expected to engage in
intense knowledge sharing with others who are dependent on them to do the
work of the organization. The interdependent nature of work is higher in an
office than in a laboratory. Subunits have more people, though they are
managed as professionals. Again, the office consists of a large number of
highly skilled people who work together to accomplish the simultaneous goals
of efficiency in using work routines to execute tasks and effectiveness in
meeting organizational goals.

Diagnostic Questions

For your firm, you can examine the two dimensions, number of people and
their professionalization, and locate where the firm is in figure 7.2. Then you
129 People

can categorize the firm’s design of people as: shop, factory, laboratory, or
office. To begin, answer the diagnostic questions below.
1. What is your unit of analysis that you chose in earlier chapters? Use this
unit of analysis as the organization when answering the questions below.
The questions below will help you locate your chosen organization on
the number of people and professionalization dimensions.
2. Number of individuals
How many people are there in the firm? Normally, we mean the number
of employees. The measure is the actual count of individuals who are
working in the organization, whether full time or part time.1
Here is the mapping for the actual number into the scale in figure 7.1.
Less than 100 employees – 1
101–500 employees – 2
501–1000 employees – 3
1001–2000 employees – 4
More than 2001 employees – 5
3. Professionalization
What proportion of employees hold advanced (university) degrees or
have many years of specialized training and experience?
0 to 10 % – 1
11 to 20 % – 2
21 to 50 % – 3
51 to 75 % – 4
76 to 100 % – 5
4. You can now locate your organization in figure 7.2. What would you call
the people mapping of the firm?

Fit and Misf its

What is a good fit between management of people and other dimensions of

an organization’s design? Here, we discuss the fit and misfit relations. In table
7.1, we add fit for the people to earlier dimensions. In each of the columns A,
B, C, and D, the fit relations can be read vertically from top to bottom.

1
It is important to include all people, not just full-time equivalents, as it is the number of
individuals that determine the coordination requirements.
130 Burton, DeSanctis, and Obel

Figure 7.2. Locate your chosen organization in the people space.

Table 7.1. Fit and misfit for people design

Corresponding
quadrant in
organizational
design space A B C D
People Shop Factory Laboratory Office

Task design Orderly Complicated Fragmented Knotty

Knowledge Ad hoc Informated Cellular Network

exchange communications

Geographic Global International Multi-domestic Transnational

distribution

Complexity Blob Tall Flat Symmetric

Configuration Simple Functional Divisional Matrix

Environment Calm Varied Locally stormy Turbulent

Strategy types Reactor Defender Prospector Analyzer with Analyzer without

innovation innovation

Organizational Neither Efficiency Effectiveness Efficiency and Effectiveness

goals

As before, misfits are any set of relations which do not fall within one
column. There are a very large number of possible misfits. The misfits arise
from the size of the workforce or their professionalization. A critical issue is
the limited time and skill of the individual as well as the limited attention
131 People

and time of management. These misfits can overload the executive oversight of
people and lead to diminished performance for the firm. The telltale signs are:
individuals are not aware of what to do, or are given directives that hamper
rather than facilitate their coordination; decision and communications
backlogs increase; adjustments are not timely; or all work excessively long
hours without results.
For column A, there is a fit for the shop with a reactor strategy, a calm
environment, a simple configuration, and orderly task design. In the shop
setting people can do their jobs following a rather limited direction and
coordination. There are relatively low information-processing demands on
everyone, except on the manager who oversees the shop. If the strategy or
the environment requires more attention and time, then the individual’s
tasks need to change and the executive can quickly become overloaded. The
individual’s activities quickly become misaligned with the new challenges.
One approach to handle these new challenges is significant professionalization
of the employees.
The potential misfits are numerous in the shop setting. If the environment
becomes more uncertain and requires changes involving a new strategy that
is not just reacting to the events in the environment, it is very likely that the
nonprofessional people inside the shop will either resist change or lack the
skills or experience for change. Alternatively, if the workforce is professional
or very large and managed via one-to-one directives by the manager, the
shop becomes very ineffective and inefficient.
Moving to B, the factory is a fit for an efficiency goal, defender strategy,
varied environment, a functional configuration, and a complicated task
design. The information-processing demands have increased considerably
as there is a large number of individuals to coordinate. For many variations,
a tall functional configuration can handle a large amount of information
needed for detailed and involved coordination. However, if the environ-
mental unpredictability increases and thus requires additional information
processing, then the functional configuration is not suited to make large
coordinated changes quickly. In such a misfit situation, the individuals will
require greater professionalization. Use of automated systems can help
reduce the need for greater professionalization, as computer-based routines
and information substitute for adding skills and experience to the workforce.
It is for this reason that call centers are sometimes criticized as technology-
based sweatshops. Technology substitutes for increasing the professionaliza-
tion of the workforce. Efficiency is enhanced, but the knowledge capacity of
the people is not enriched.
132 Burton, DeSanctis, and Obel

The higher-level routines and formalized rules between and among jobs
and individuals help to coordinate the total set of activities for the factory.
High organizational complexity with a large number of low-skill jobs and a
tall organization for coordination works well, but change is difficult. The
functional configuration is a good fit for the detailed operational coordin-
ation of the factory. A varied environment where changes are small and
anticipated is a good match as well. The complicated task design of low
divisibility and high repetitiveness fits the factory, because there are a large
number of people who are managed at low skill level.
The potential threat for managing people in a factory setting is change
that may involve a higher degree of decentralization and a flexible response
to changes.
For column C, the laboratory is a good fit for: a flat complexity, a
divisional or cellular configuration, a locally stormy environment, a pro-
spector strategy, and an effectiveness goal. In a laboratory the individuals are
very skilled and can deal with variation derived from environment and
innovation. The executive can create independent divisions to deal with
the local conditions. Each division has its own environment which is stormy
but largely independent of the others. As the number of subunits increases,
the coordination issues will become problematic and will eventually create
a misfit. Alternatively, if the environment changes such that two divisions
compete for the same customer, the executive can become overloaded with
coordinating details – again, creating a misfit. The tall firm is a misfit for the
divisional configuration. If the top executive becomes involved in the
detailed operations of the firm, cognitive overload, neglect of important
issues, and poor firm performance will occur.
The cellular configuration is a good fit where there is relative independ-
ence between cells – not necessarily within a cell. On a larger scale, the
divisional configuration is similar; there can be relative independence be-
tween divisions – not within divisions. A prospector strategy which develops
new things in a locally stormy environment is a good match. High skill
permits exploration. These tasks can be fragmented with high divisibility
and low repetitiveness.
For column D, the office is a fit for a symmetric complexity, a matrix
configuration, a turbulent environment, an analyzer strategy and dual goals.
This requires highly skilled individuals to work together so as to realize the
needed coordination. The information-processing demands are very large as
detailed coordination is required by new situations (Galbraith, 1973). The
133 People

firm simply cannot be broken down into independent divisions, nor can a
tall hierarchy handle all of the changes required to effectively adapt to a
turbulent environment. The executive can create divisions to deal with the
local conditions, but the matrix configuration is a good match when the task
design is knotty and the environment is turbulent. A large number of skilled
individuals will help to execute knotty tasks and deal with continual change
and coordination across the matrix. Departures from the alignment lead to
costly misfits.
The matrix configuration where there are two or more reporting dimen-
sions (but fewer than where all individuals can talk with each other) is a mid-
range solution to limiting information processing and obtaining the needed
coordination. The focus here is both efficiency and effectiveness. An analyzer
strategy with innovation in a turbulent environment is a good match. Further,
the task design can be knotty with high divisibility and low repetitiveness.
If your chosen firm is located in different columns based upon your
answers from these four chapters, then you should think about what you
might do to bring the firm into fit in the column that meets your goals. But
also think about what is involved to move to a different goal and thus a
different column.

Summary

In this chapter, we have outlined the people component of organizational

design, which should fit together with other design components to meet
your firm’s goals. There are four categories: shop, factory, laboratory, and
office, which are located on two dimensions: the number of people and
their professionalization. The idea of choosing the two dimensions is related
to the task of processing information, where the degree of professionaliza-
tion is a proxy measure for the individual’s ability to process information.
Taken together with the number of people, we thus have a measure of
the information-processing capabilities embedded in the human capital of
the firm. We also discussed misfits. These misfits were developed from
the idea of information processing given the setup of the 2
2 model.
For example, a defender strategy with its high volume of standardized
activities does not fit a laboratory with few people with a high degree of
professionalization.
134 Burton, DeSanctis, and Obel

Glossary

Factory: supporting the firm’s strategy and structure with many people who
have low skills. Their work routines are designed to be as simple as
possible and easily learned.
Laboratory: an approach to people grouping which has few people, where
each one has high professionalization or a high level of skill which was
obtained through extensive education, training, and experience.
Office: an approach to supporting the firm’s strategy and structure with many
people with high professionalization, where the skills incorporate a high
level of knowledge obtained from education, training, and experience.
Professionalization: the skill capability of the individuals in the firm, depen-
ding on employees’ education, training, and experience (i.e., knowledge
base) accumulated either prior to their hire or on the job.
Routine: a set of tasks processed which together accomplish a higher level
unit of work or task.
Shop: supporting the firm’s strategy and structure with few people who are
not highly skilled. People have had little specialized training or
experience.
Size: for organizational design purposes, the total number of people in the
firm who are working on the firm’s big task; includes full-time, part-time,
temporary, and volunteer workers.
8 Leadership and organizational climate

Introduction

Leadership style and organizational climate are two of the most widely
used, debated and researched concepts in management. Everyone can
make a list of great leaders. We know good leadership when we see it.
What is a leader? Does a good leader stand alone, or must there be a good
fit with the firm’s culture and climate?1 What is a good climate? What
climate is needed to be successful in an organizational change process?
Leadership and organizational climate are important issues to take into
account when you design an organization. A firm’s leadership style and
organizational climate are the two sides of how the people in the organ-
ization think and act. In your approach to designing an organization you
should focus on analyzing the leadership style and organizational climate.
The leadership style is the predominant mode used by the top manage-
ment of your unit of analysis to manage employees. This is the top
management of the entire organization if your unit of analysis is an entire
company or firm. It is the department head or team leader(s) if your unit
of analysis is a department or team. Top management is the individual or
group of people at the highest level of your unit of analysis. The organiza-
tional climate is the internal environment or working atmosphere as
experienced by organizational employees. The organizational climate for
your unit of analysis may or may not be consistent with the climate of
the broader organization. So let us get on with these two important
concepts. We start with the leadership style and then continue with the
organizational climate.

1
We focus on the organizational climate in this book. There has been a long discussion in
the literature about the difference between culture and climate. (E.g., see Denison, 1996).

135
136 Burton, DeSanctis, and Obel

Leadership style

Theory X and theory Y leadership descriptions (McGregor, 1969) are widely

used in management conversation as contrasting styles. A theory X leader is
directive, short term and control-oriented whereas a theory Y leader dele-
gates, is long term and motivates through inspiration. Autocratic versus
democratic leaders as described by Likert (1967) and managers versus
leaders as described by Kotter (1988) capture contrasting styles. Building
upon Cyert and March (1963), Burton and Obel (2004) argue that these
contrasting styles can be summarized as decision-making preferences that
are a function of a leader’s preference for delegation, on the one hand, and
the tendency to avoid uncertainty, on the other hand. Håkonsson et al.
(2004) found empirical support for such a categorization. Preference for
delegation follows from Cyert and March (1963)’s idea of problemistic
search. The managerial propensity to delegate serves as a decision-making
heuristic whenever the executive finds delegation to be efficient due to their
limited attention and time availability. Similarly, Cyert and March’s notion
of uncertainty avoidance incorporates several executive desires: preference
for detail, tendency to be reactive rather than proactive, short-term versus
long-term decision making, and ability to motivate via control rather than
inspiration. To illustrate, one way an executive can avoid the uncertainty of
long-run anticipation and commitments is to provide detailed directions to
employees based on short-run feedback. This means solving pressing prob-
lems rather than developing long-run strategies. It also means avoiding
having to anticipate the business environment or otherwise negotiate change
within the organization to meet major environmental shifts. Some execu-
tives tend to provide detailed instructions to employees and avoid the
uncertainty of managing for the future. Other executives are the opposite
– they embrace the ‘‘big picture,’’ let employees find their own direction, and
take risks for the future despite the uncertainties involved. Of course, there
are gradations in between, as we shall see.
We use the two dimensions, preference for delegation and uncertainty
avoidance, to analyze leadership style. Together, these two dimensions meas-
ure how managers influence organizational efficiency and effectiveness, i.e.,
how managers contribute directly to organizational performance through
their leadership. Preference for delegation is the degree to which the top
management encourages lower-level managers or other employees who are
their direct reports to make decisions about what and how work is to be
137 Leadership and organizational climate

Figure 8.1. The leadership style space.

done in the organization. Preference for delegation is high if top manage-

ment relies on lower-level managers and employees to work autonomously
and make decisions without top-management approval. Preference for dele-
gation is low if top management prefers to make decisions about how and
what work is done and to direct activities in a close-handed way. Uncertainty
avoidance is the degree to which the top management shuns taking actions
or making choices that involve major risk. Uncertainty avoidance is low if
your top management tends to be risk taking, whereas uncertainty avoid-
ance is high if your top management tends to be risk averse.
The two leadership dimensions are shown in figure 8.1. Uncertainty
avoidance is on the vertical axis and preference for delegation is on the
horizontal axis. This provides us with four leadership style categories:
maestro, manager, leader, and producer. The maestro prefers little delegation
and accepts uncertainty. The manager, similar to theory X, prefers little
delegation and avoids uncertainty. The leader, opposite of the manager and
similar to theory Y, accepts uncertainty and delegates decision making to
subordinates. And finally, the producer avoids uncertainty and has a high
preference for delegation. The manager and leader are well-known contrast-
ing styles; the maestro and producer are new style descriptions. We now
describe each in more detail.

Maestro

The maestro has a low preference for delegation and low uncertainty avoid-
ance. The maestro will intervene directly to assure that decisions are made
congruent with his or her own desires. At the same time, the maestro does
not avoid the uncertainty of long-term decisions and their implications for
the firm.
138 Burton, DeSanctis, and Obel

The maestro can become overly involved and overly burdened with too
much to do when the lack of delegation creates a bottleneck for decision
making and a barrier to action. Decisions are not made; projects are not
started; products are developed too late for the market. Further, we can see
that the effective maestro requires great expertise – expertise for knowing
how and when to take risks and how to lead people to make great progress
for the organization.
The maestro leadership style fits well with the small start-up company,
while for the large mature corporation a reactive strategy taking unnecessary
risk may be the situation. The maestro leadership style may be appropriate
in case of crises or in case of major changes like a merger.
The maestro’s attention on ongoing decisions and uncertainty avoidance
leaves top management open to longer-term vulnerabilities. If the top
management takes a maestro approach, then it reacts better than it antici-
pates. Environmental and innovation changes can be missed or observed
too late for reaction. Making change in an organization that is led with
a maestro style will be difficult, particularly in the time frame when
change is needed. Thus there is an explicit focus on neither efficiency nor
effectiveness.

Manager

The manager has high uncertainty avoidance and a low preference for
delegation. Avoiding uncertainty is realized again by making reactive and
short-term decisions with a fine level of detail. The manager focuses more
on the control of operations than on strategic decisions. The manager does
not delegate decision-making authority but instead uses formalized rules
to manage subordinates. The manager knows what is happening in detail
and can react quickly to undesired activities, i.e., bring things back into
control. The manager achieves the goal of efficiency in operations where the
utilization of resources is very important.
If the top management adopts a manager style of leadership, then it
has excessive attention to detail that can make an organization vulnerable
to those issues that, for one reason or another, are overlooked or receive
little managerial attention. There can be little attention to the longer-term
strategy of what to do or the ‘‘bigger picture’’ of work to be done. Thus,
some vital details may be missed. And if the environment becomes less
predictable, the firm is likely to miss the opportunity for change, which
139 Leadership and organizational climate

can harm overall effectiveness. Further, the manager approach to leadership

gives little attention to innovation, except for efficiency-related innovation
that threatens the technology base of the firm. In general, the manager
leadership style has a short-term orientation that tends to overlook issues
that make the firm viable for the longer term. The primary focus of a
manager is on efficiency.

Leader

The leader has a high preference for delegation and low uncertainty avoid-
ance. The leader is confident that others can make good decisions for the
firm and thus finds delegation an efficient way to save time. Moreover, the
leader does not avoid long-term uncertainty but instead embraces its chal-
lenges by attending to more strategic decisions. If the top management takes
a leader approach to managing people, then it spends much time thinking
about the long term, taking risks and avoiding the time-consuming task of
detailed control. The leader encourages new ideas, initiatives and projects,
both its own and those of subordinates. The leader explores new ideas and
actions. And with the confidence to let subordinates make decisions and
take actions, the leader can focus on more strategic considerations of the
long term.
A leader has a focus on effectiveness and is willing to take substantial risk
in order to achieve ambitious goals. The leader is vulnerable to weak
following behavior in the process of implementation. If the subordinates
do not live up to the confidence of their leader, then organizational per-
formance can suffer – perhaps for an extended time period. Lack of atten-
tion to detail can create large problems for the organization. Further, the
leader can take on risky projects which turn out badly for the firm and
the leader.

Producer

The producer has a high preference for delegation and scores high on
uncertainty avoidance. The producer focuses on both efficiency and effect-
iveness. If your firm’s top management adopts a producer style of leadership,
then the organization is likely to be well positioned vis-à-vis its competitors.
The producer assures that new products and services are developed and
140 Burton, DeSanctis, and Obel

introduced. The focus of attention is a dual one: short term and long term;
operations and strategy; products and innovation; internal activities and
the environment reading; hands-on management and letting others act
independently; and efficiency and effectiveness.
The producer wants to know what is going on, assigns work to others, but
does not need to make each and every decision the organization confronts.
To avoid uncertainty, the producer has a long-term forecasting and planning
focus. The producer exploits the subordinates’ managerial resources well,
delegating to be efficient in use of time, especially when others make
decisions consistent with his or her preferences. The producer style is
delegation with detailed oversight and a focus on the short term to avoid
uncertainty. The strength of the producer’s leadership style is the delegation
to others, but the producer does this with an oversight that can assure
decisions are made according to his or her preferences and that those actions
are coordinated across the subordinates.

Organizational climate

Organizational climate is the ‘‘relatively enduring quality of the internal

environment of an organization that a) is experienced by its members,
b) influences their behavior, and c) can be described in terms of the values
of a particular set of characteristics (or attitudes) of the organization’’
(Tagiuri & Litwin, 1968, p. 27). Climate is a characteristic of an organization
which is experienced by its members. It is a psychological measure of the
organization. Whereas leadership style refers exclusively to top management,
organizational climate refers to all members of the organization, including
superiors and subordinates.
Zammuto and Krakower (1991) measured organizational climate using
many dimensions: trust, conflict, morale, rewards, resistance to change,
leader credibility, and scapegoating (that is, blaming others for mistakes or
problems in the organization). In a study of 246 Danish service firms,
Burton et al. (2004) found that these seven dimensions could be reduced to
two: tension and resistance to change. Here we treat tension and readiness to
change (the opposite of resistance to change) as the two most fundamental
design dimensions for organizational climate.
Tension is the degree to which there is a sense of stress or a psychological
‘‘edge’’ in the work atmosphere. Tension incorporates a combination of
organizational factors as experienced by insiders, including trust, conflict,
141 Leadership and organizational climate

morale, rewards, leader credibility, and scapegoating. When tension is high,

trust is low, conflict is high, morale is low, rewards are perceived as inequit-
able, leader credibility is low and there is a tendency toward scapegoating.
Low tension is the opposite; trust is high, conflict is low, morale is high,
rewards are perceived as equitable, leader credibility is high, and there is
little or no scapegoating. At first glance, high tension sounds like a bad state
for an organization. How could it be healthy for an organization to have low
trust, high conflict, low morale, etc.? Although any one of these dimensions
may have negative consequences, in combination they can bring an intensity
and vigor to the organization – especially if they do not occur in the
extreme. Extremely high conflict and low morale, etc., may be disastrous,
but some degree of these in combination with the other factors mentioned
above can spur effectiveness, especially if they occur in combination with the
other design factors for managing people and processes, as we discussed in
Chapters 6 and 7. Some degree of tension in the organizational climate is
stressful, yet it increases the pace of work and movement toward efficiencies.
Imagine the transition period of IBM in the late 1980s as Lou Gerstner took
charge of moving the company from being a leading, established but old-
style computer company to a more high-tech savvy software and service-
based enterprise. Tension was high during this transition, but tension
enabled the insiders to confront the challenges at hand and to mobilize to
cut costs and move toward efficiency.
Readiness to change is the degree to which the people in the organization
are likely to shift direction or adjust their work habits to meet new, un-
anticipated challenges. Ongoing norms and practices, or routine ways of
doing things, can be an asset to an organization, in that they provide a
sort of social skill set for getting work done. But ongoing ways of doing
things also can be a liability if people are set in their existing routines of
work and resist change. Readiness to change, to discover, and adopt new
work habits and practices is vital if an organization is to be effective over
time.
Climate can then be categorized into four climate types: group, internal
process, developmental goal, and rational goal, as shown in figure 8.2. The
group climate has a low tension and low readiness for change; it is a quiet
place. The internal process climate tends to be more mechanical with a low
readiness for change and relatively high tension. The developmental climate
is more externally oriented with relatively low tension and a high readiness
for change. The rational goal climate is also externally oriented to succeed
with high readiness for change, but with relatively high tension.
142 Burton, DeSanctis, and Obel

Figure 8.2. The climate space.

Group

The group climate has low tension and low readiness to change. It is a
pleasant place to work where individuals trust each other; conflict is low;
rewards are perceived as equitable, and there is little readiness to change.
Here, the individuals are comfortable with the situation as well as with each
other and see no need to consider any change. The group climate can be
stressful for the leader, who must make decisions about how work is done
and maintain the status quo, but it is pleasant for employees generally. A
group climate usually has a low degree of conflict. If conflict exists, it is
constructive and tends to strengthen the organization, rather than destroy it,
i.e., there can be disagreement on the group purpose itself. This is usually
coupled with a high or moderately high degree of employee morale. Indi-
viduals feel that they belong to and are part of the organization. Rewards
need not be equally distributed, but there must be a sense of fairness where
the basis for the distribution is understood and accepted by the individuals
in the organization.
If an organization has a group climate then it will find that managing
information flow is relatively easy. Information is more likely to be ‘‘broad-
cast’’ than ‘‘channeled.’’ ‘‘Need to know’’ is replaced by ‘‘everybody knows,’’
or informal communication among specific parties who need to share
knowledge. There are few secrets. The group climate can handle complex
sets of information.
143 Leadership and organizational climate

Although things are pleasant in the group climate, people do not have
a high readiness for change. There is a high degree of trust and little
scapegoating and the leader likely enjoys a high degree of credibility with
the subordinates. However, getting people to embrace change is nonetheless
a challenge in that the group climate has a consistent pattern of beliefs and
attitudes about desirable behavior that are not readily adjusted as circum-
stances change.

Internal process

The internal process climate has high tension and low readiness to change. It
is a less pleasant place to work where individuals are less trusting, have more
conflict, perceive rewards as inequitable, and yet there is little readiness to
change.
Compared to the group climate, the internal process climate is character-
ized by lower trust. There is not a sharing and open atmosphere among the
individuals as each is more inward and guarded. Conflict can be high in the
organization and so people may disagree over both means and ends (i.e.,
work methods and goals). Rewards are perceived to be given inequitably. As
a result, employee morale can be low.
In the internal process climate people tend to focus inwardly on how work
is done, i.e., the work methods or processes. This can be very important to
gaining organizational efficiency. So, such a climate is not necessarily de-
structive for the organization. The managerial challenge is to keep people
focused on work processes without letting trust, conflict, perceived inequi-
ties, and so on, become so low that they obstruct organizational success.
Managed carefully, an internal process climate can bring organizational
benefit. As an example, consider the popular Six Sigma programs (Hahn,
1999), which emphasize a culture of measurement, excellence, confrontation
of conflict, and rewards based on continual error reduction. A Six Sigma
program will tend to promote an internal process climate. The managerial
challenge is to nurture the internal process climate in such a way that it does
not spin into a downward spiral of negativity by employees but instead
promotes the value of excellence, achieving organizational efficiencies and
error control.
If your organization has an internal process climate then it will observe
that there tends to be a low readiness to change. Perhaps this is not intuitive
as it might be argued that a change, or any change, would be welcome
144 Burton, DeSanctis, and Obel

in such a climate. But the evidence suggests that an intense process-

orientation on the part of people goes along with a preference to keeping
that orientation rather than engaging in activities that could lead to a
different situation. Perhaps it is the reduced level of trust in these organiza-
tions that helps explain this reluctance. There is less faith in the leadership
and more faith in processes themselves, so resistance to change tends to be
high. A higher than ideal level of scapegoating seems consistent with this
story about the internal process organization.
The internal process climate does not possess the capacity to process a
lot of information through informal means. There is not a norm of sharing
and openness. Instead, the organization structure must supply the requisite
information-processing capacity. Information tends to be private and within
the role or specific job scope of those who need the information. Infor-
mation is passed on within prescriptions and according to procedures. It is
closely associated with the job or task, or ‘‘a need to know.’’ Spontaneous
information links are largely missing, or not utilized.

Developmental

The developmental climate has low tension and a high readiness to change.
It is a pleasant place to work, where people generally trust each other;
conflict is relatively low; rewards are perceived as equitable, and people are
quite willing to engage in change. People in the developmental climate are
comfortable with each other and welcome new opportunities.
Some of the characteristics for the developmental climate are similar to
those of the group climate. For both, trust is high, conflict is low, and morale
is high, with relatively equitable rewards. The significant difference is the
readiness to change, which tends to be low in a group climate but high in
a developmental climate. If an organization has a developmental climate,
you will find that there generally is a great focus on the growth of the
individuals and their quality of work life. So although there is a focus
on growth of the organization the tension is rather low. This is the basis
for the high readiness to change. In the developmental climate, rewards can
be more individually based than in the group process climate with less
attention to the impacts on perceived equity. Individual contribution to
the organization is more important and, in a well-functioning developmen-
tal climate, this is accepted by employees. Compared to the group and
internal process climates, the developmental climate is more externally
145 Leadership and organizational climate

oriented. People believe and act based on an assumption that success is

realized more outside the organization.
There are also small differences with respect to leader credibility and the
level of scapegoating. The developmental climate has different information
characteristics as compared to the group climate. The group climate will
focus more heavily on internal information, whereas the developmental
climate focuses more on external environmental information. Environ-
mental information is likely to have more value for development and growth.
Additionally, compromise is important (Quinn and Kimberley, 1984).

Rational goal

The rational goal climate has high tension and a high readiness to change.
It is goal-driven and the individuals are a bit on edge as the tension is high,
but at the same time tension is not so high that it is detrimental to
performance. In fact, tension helps to drive performance as people deal with
fluctuations in trust, conflict, and so on. People are willing to change and
accept new challenges and opportunities if they believe goals can be met.
The rational goal climate is closer to the internal process climate than to
the developmental climates, although they are different. The main difference
is the readiness to change. The rational goal climate is structured with an
emphasis on planning, productivity, and efficiency (Quinn and Kimberley,
1984).
Information processing in the rational goal climate is similar to internal
process climate but with a greater emphasis on environmental/external
information. The low level of trust, high conflict, etc., leads to a private,
or customized view of information; sharing and exchange of information
does not occur spontaneously, but information is shared if it is goal-
oriented. Put another way, people do not share information for its own
sake but rather to meet specific needs related to their work tasks. In this
way, information sharing is tempered rather than fully open. The rational
goal climate is a very competitive environment to work in. It is not to be
expected that the employees will be loyal to the organization in the sense
that high turnover can be expected. Rewards are performance-based. The
organization may work hard to keep the most valued or skilled employees
but not worry too much if others are unhappy and leave. With the high
readiness to change, reorganization of personnel level can be expected with a
very tough competition for the prestigious jobs.
146 Burton, DeSanctis, and Obel

Diagnostic questions

For your organization you should first examine the two dimensions in
figure 8.3: preference for delegation and uncertainty avoidance. Locate where
the top management leadership style is along these two dimensions and then
categorize the leadership style as: maestro, manager, leader, or producer. To
begin, answer the diagnostic questions below.
1. For your unit of analysis, what is the top management that you are
describing here? It may be a single executive or a set of people (such as an
executive group or board) who oversee your unit of analysis. Use this top
management level when answering the questions 2 and 3 below. Note
that if you are the executive in charge of your unit of analysis, then these
questions are about your leadership style.2
The questions below will help you locate your top management on the
preference for delegation and uncertainty avoidance dimensions.
2. Preference for delegation
a. To what extent does top management maintain control themselves
(1), or encourage others to take on responsibility for managing work
tasks (5)?
b. To what extent does top management allow its direct reports to make
important decisions and take actions for the organization (1¼low,
5¼high)?
c. Overall, for your unit of analysis, what is top management’s
preference for delegation, (1) low or (5) high?
Score the preference for delegation on a scale from 1 to 5 as follows:

1 2 3 4 5
very low moderate very high

3. Uncertainty avoidance
a. To what extent does top management concern itself with the ‘‘big
picture’’ (1), rather than the detail (5), in decision making?
b. Does top management tend to be aggressive (1), or cautious (5), in its
decision making?

2
As before, you can average your scores for the items within each question to create an
overall score for each design dimension, or you can use the questions as a guide to assign
an overall score for each design dimension.
147 Leadership and organizational climate

c. How risk embracing (1), versus risk avoiding (5), is the top
management?
d. To what extent is top management control oriented in the
management of its direct reports, low (1) or high (5)?
Score the uncertainty avoidance on a scale from 1 to 5 as follows:

1 2 3 4 5
very low moderate very high

4. You can now locate your organization on figure 8.3. What is its
leadership style?
Now, consider the organizational climate. Remember to include the entire
unit of analysis as you answer these questions. In figure 8.4, the firm’s
readiness to change and tension are the dimensions, and the organizational
climate is then categorized as: group, internal process, developmental or
rational goal. Here are questions which will help you locate your chosen firm.
5. Readiness for change
a. To what extent do people prefer old ways of thinking and doing
things (1) versus embrace new ways of thinking and doing things
(5)?
b. To what extent do people tend to shift direction or adjust their work
habits to meet new, unanticipated challenges, low (1) or high (5)?
c. Overall, what is the organization’s level of readiness to change, low (1)
or high (5)?
Use a 1 to 5 rating scale to score your organization as follows, 1 ¼ low to
5 ¼ high:

Figure 8.3. Locate your organization’s leadership style.

148 Burton, DeSanctis, and Obel

Figure 8.4. Locate your firm in the organizational climate space.

1 2 3 4 5
very low moderate very high

6. Tension
a. What is the level of distrust in the firm, low (1) or high (5)?
b. What is the level of conflict in the firm, low (1) or high (5)?
c. To what extent do people perceive rewards not to be equal across
employees, low (1) or high (5)?
d. To what degree do people question the credibility of the organiza-
tion’s leaders, low (1) or high (5)?
e. What is the level of scapegoating, or blaming, of people for problems,
low (1) or high (5)?
Use a 1 to 5 rating scale to score your organization as follows, 1 ¼ low to
5 ¼ high:

1 2 3 4 5
very low moderate very high

You can now locate your chosen organization on the figure 8.4. What is
your firm’s climate?

Fit and misf its

What is a good leadership style for a firm, and what is an appropriate

organizational climate? What is a good fit? In table 8.1, we add fit for the
149 Leadership and organizational climate

Table 8.1. Fit and misfit to include leader style and organizational climate.

Corresponding
quadrant in
organizational
design space A B C D
Organizational Group Internal process Developmental Rational goal
climate

Leader style Maestro Manager Leader Producer

People Shop Factory Laboratory Office

Task design Orderly Complicated Fragmented Knotty

Knowledge Ad hoc Informated Cellular Network

exchange communications

Geographic Global International Multi-domestic Transnational

distribution

Organizational Blob Tall Flat Symmetric

complexity

Configuration Simple Functional Divisional Matrix

Environment Calm Varied Locally stormy Turbulent

Strategy types Reactor Defender Prospector Analyzer with Analyzer without

innovation innovation

Organizational Neither Efficiency Effectiveness Efficiency and Effectiveness

goals

leadership and organizational climate to the goals, strategy, environment,

configuration, and task design for your chosen firm. In each of the columns
A, B, C, and D, the fit relations can be read vertically from top to bottom.
Misfits for leadership and climate pose a particular difficulty from the
point of view of organizational design. Although you may be able to change
the goals, strategy, or the configuration of your chosen organization, it may
be very difficult for an executive to change the leadership style. You may have
no control over this design factor. Therefore, managing the fit between
leadership style and other design components can be problematic. To change
the leadership style may require a new executive and/or other members in
the top management roles. Similarly, organizational climate is a relatively
enduring property of the organization and cannot be easily changed in the
short run. So if there are misfits with the leadership style and climate, it may
150 Burton, DeSanctis, and Obel

be easier to adjust to them rather than to change them in a significant way.

Of course, if this means changing to a different and less satisfactory strategy
for your chosen organization, it may be necessary to take a long-term view
and take on the difficult actions necessary to bring the organization’s leader-
ship style and climate into alignment with goals, strategy, and configuration.
As the table suggests, in column A there is a fit among the maestro, the
group climate, the blob organizational complexity, a simple configuration, a
calm environment, a reactor strategy, and ill-defined goals. The organi-
zational climate is pleasant and non-threatening. It usually is not very
fast-paced. A new executive with a new style can quickly become a threat
to the individuals and create a misfit with the group climate. If the firm is
not performing well, there may be good reason to create a misfit, spurring
the opportunity to redesign the organization. A new organizational design
can be introduced and brought into alignment, bringing the various com-
ponents together into a new quadrant of the organizational design space
over time. In this way the organization can achieve firm goals of efficiency
and effectiveness. We will discuss the process of misfits and change manage-
ment in more detail in Chapter 11.
Moving to column B, there is a fit among the manager, the internal
process climate, the tall, functional configuration, the varied environment,
the defender strategy, and the efficiency goal. The information-processing
demands have increased considerably, but the manager takes a more hands-
on approach with less delegation and detailed monitoring. The commensur-
ate climate has high tension with less trust and leadership credibility, and is
less pleasant. Generally, this climate is less difficult to establish but it takes a
longer time to reduce the tension. At the same time, the firm’s efficiency
goals can be realized, but innovation is less likely.
For column C, the firm has a top management with a leader style, a
developmental climate, a flat, divisional configuration, a locally stormy en-
vironment, a prospector strategy, and an effectiveness goal. The executive lets
others make decisions but accepts the uncertainty. The climate has low
tension and a high readiness for change. Many individuals would find the
organization with the profile of column C to be an exciting place to work due
to high trust and executive support. It fits well with a prospector strategy and
an effectiveness goal. If there is a desire to focus on short-term efficiency, the
executive may become more control-oriented and directive, which is a threat
to the developmental climate. Then the executive can become quickly over-
loaded with the details, which may further threaten the developmental
climate rather than resolve it. The leader style and developmental climate
151 Leadership and organizational climate

work best when innovation is valued and the organization pursues a pro-
spector strategy with subunits organized as independent divisions or cells.
For column D, the firm has a leadership style that acts as a producer,
a rational goal climate, a symmetric, matrix configuration, a turbulent
environment, an analyzer strategy, and pursuit of the dual goals of efficiency
and effectiveness. The producer leadership style means that top management
delegates with high information processing but also tries to avoid uncer-
tainty. The climate has high tension but also has a readiness for change. The
organization in this quadrant is performance-driven, aiming to achieve both
efficiency and effectiveness of innovation. It is a demanding place to work,
where tension is high, but some individuals find it exciting and embrace a
high readiness for change. The organization in column D is a good fit with
the turbulent environment and analyzer strategy of innovation and change.
Coordination needs are high in this type of organization and quick change is
required to meet organizational goals. As such, the goal-driven matrix
configuration with large information-processing capacity is a good fit.
If your chosen firm is located in different columns based upon your
answers to the diagnostic questions in this chapter, then you should think
about what you might do to bring the organization into fit in the column
that meets your goals. But also think about what is involved to move to a
different goal and thus a different column.

Summary

In this chapter, we have included the leadership style measured as prefer-

ence for delegation and uncertainty avoidance and organizational climate
measured as tension and readiness for change to the set of relations which
should fit together to meet a firm’s goals. There are four leader styles:
maestro, manager, leader, and producer; and there are four climates: group,
internal process, developmental, and rational goal. We then discussed fit and
misfit possibilities. Next we move on to consider approaches for managing
coordination and control in your organization.

Glossary

Developmental climate: an organizational climate characterized by low

tension and a high readiness to change.
152 Burton, DeSanctis, and Obel

Group climate: an organizational climate characterized by low tension and

low readiness to change.
Internal process climate: an organizational climate characterized by high
tension and low readiness to change.
Leader: a leadership style in which top management accepts uncertainty and
delegates decision-making to subordinates (similar to theory Y).
Leadership style: the predominant mode used by the top management of
your unit of analysis to manage subordinates, which is measured in terms
of preference for delegation and uncertainty avoidance.
Maestro: a leadership style in which top management orchestrates the work
of others through a combination of direct involvement and high
tolerance for uncertainty.
Manager: a leadership style in which top management prefers little
delegation and avoids uncertainty (similar to theory X).
Organizational climate: the internal environment or working atmosphere of
the organization as experienced by all employees, including the leader and
subordinates.
Preference for delegation: the degree to which the top management of the
organization encourages lower-level managers or other employees who
are their direct reports to make decisions about what and how work is to
be done in the organization.
Producer: a leadership style in which top management avoids uncertainty
through short- and long-term planning and has a high preference for
delegation, but with detailed oversight.
Rational goal climate: an organizational climate characterized by high
tension and a high readiness to change.
Readiness for change: the degree to which the people in the organization are
likely to shift direction or adjust their work habits to meet new,
unanticipated challenges.
Tension: the degree to which there is a sense of stress or a psychological
‘‘edge’’ in the work atmosphere; it incorporates a combination of
organizational factors as experienced by insiders, including trust, conflict,
morale, rewards, leader credibility, and scapegoating.
Uncertainty avoidance: the degree to which the top management shuns
taking actions or making choices that involve major risk with short-term,
reactive decision making; low uncertainty avoidance means management
is risk taking with longer-term focus.
153 Leadership and organizational climate

Where are you in the step-by-step approach?

STEP 1
Getting Started
(1) Goals
STEP 2
Strategy
(2) Strategy
(3) Environment
STEP 3
Structure
(4) Configuration and complexity
(5) Geographic distribution knowledge exchange
STEP 4
Process and People
(6) Task design
(7) People
(8) Leadership and organizational climate
Here, you focused on the work of the firm and the people who do
the work. First, you designed the tasks, then you considered the
number of people and their skills, and, finally, you examined the
leadership style of the top management and the climate where
everyone works. A good organizational design includes fit among
all of these elements: the tasks, people, leadership, and climate.
Before moving on, it is important to review how well these fit
together in your organization. Identify misfits and consider what
steps you might take to improve fit. When you are satisfied that
there is a fit among the processes and the people, you should
review how they fit with the goals, strategy, and structure of your
firm. If there are misfits, think about what you can and will do to
154 Burton, DeSanctis, and Obel

bring these organizational design components into alignment.

Then move on to the next step, design of coordination and control
systems.
STEP 5
Coordination and Control
Next we examine the coordination and control components
of your firm including the information systems and the
incentives.
(9) Coordination, control, and information systems
(10) Incentives
Step 5 Coordination and Control
9 Coordination, control, and information
systems

Introduction

Once you have designed the processes and people-based systems to support
your chosen organization’s structure, the next step is to design coordination,
control, and information systems to manage the linkages between the various
parts of the organization. Along with people and processes, coordination,
control, and information systems are important to assure smooth working-
together among the organizational components, so that all move in a common
direction toward strategic goals. Coordination, control, and information
systems support integration of the organization, and they also provide moni-
toring and support for decision making so that managers can anticipate and
react to internal and external changes that require organizational adjustment.
In this chapter we consider the range of devices that managers can use to
coordinate and control the organization’s work. Prior to the development of
computer systems, coordination and control systems were entirely manually
based. Today, of course, these systems are both computer-based and manu-
ally based. Some systems are visible, in the sense that they are stated as tangible
rules or can be seen in the form of reports or established routines. Others are
invisible and operate in the informal ways that people think and act. These
systems may even be created ‘‘on the fly,’’ i.e., on an as-needed basis to meet
unanticipated needs. Considered together, coordination, control, and infor-
mation systems constitute the infrastructure of the firm, that is, the underlying
pathways for information sharing. A vibrant, well-designed infrastructure can
facilitate healthy integration of the organization’s structure, the functioning of
its business processes, and interactions among its people.
We will consider two major aspects of organizational infrastructure in
this chapter: coordination and control systems, and information systems.
Coordination and control systems are methods for linking together the

157
158 Burton, DeSanctis, and Obel

otherwise disparate elements of the organization’s structure and supporting

responsiveness to changes in the environment or task demands. In other
words, these are systems that integrate, or tie together, the various subunits
of the organization. Information systems are methods for providing meaningful
data to decision makers. As noted above, information systems may be com-
puter-based, but this is not necessary. Information systems also can be based
on paper memos, conversations, or informal meetings. Whether manually or
computer-based, information systems serve as conduits for the flow and
processing of meaningful data throughout the organization. Similarly, coord-
ination and control systems can be computer-based and/or manually based.
Information systems provide the data necessary for coordination and
control systems to operate. In fact, the distinction between information
systems and coordination and control systems is largely conceptual. The
two are (or should be) intricately intertwined to facilitate the work of
management. They should not be designed separately but rather together.
As we discuss these systems it is important to keep in mind that we are
talking in very general terms about the guiding principles, or philosophy, that
drive systems design in an organization rather than the specifics involved in
setting up a particular coordination/control system or an information system.
Specific design of systems is a major undertaking requiring much detailed
attention. The executive’s highest-level concern is with choosing basic prin-
ciples, or guiding dimensions, for systems design. This is our concern here.
We lay out the major options along important dimensions for systems design.
Once you decide where you want your organization to be along these major
dimensions, then you can proceed to the detailed design of your organiza-
tion’s infrastructure. Though our discussion is general and high-level, choos-
ing where you want your organization to be on the dimensions outlined in
this chapter is not a trivial exercise. These are some of the most important
design decisions you must make as an executive, and so they deserve careful
attention. We urge you to think through each dimension described below
with care, as the choices you make will set the tone for the development of
your firm’s coordination, control, and information systems infrastructure.

Coordination and control systems

Early on in the development of professional business organizations, mana-

gers gave dominant attention to developing control systems, or methods for
159 Coordination, control, and information systems

assuring quality and efficiency of information flow between the highest and
lowest level of the firm. The role of middle management was to assure
control – to pass information up and down the hierarchy between top
management and workers. Control systems monitor and measure the per-
formances of subunits and their people, providing feedback to managers
about compliance of these units. Budgets, production measurement systems,
and performance reviews are examples of control systems. As organizations
have become flatter and more distributed, there has been increasing em-
phasis on coordination, including the lateral flow of information among the
subunits. Coordination systems support flexibility and adaptiveness within
and across departmental or divisional boundaries. Coordination systems
‘‘let the right hand know what the left hand is doing,’’ so to speak. Cross-
functional teams and committees, as well as project management systems,
are examples of coordination systems. Today, control and coordination
systems are inextricably intertwined. Systems such as Movex, Navision,
SAP and People Soft, which provide support for management of inventory,
people, and projects, are examples of large-scale coordination and control
systems. Smaller-scale coordination and control systems include liaison
roles, committees, formal and informal rules, job descriptions, statements
of procedures, codes of ethics, employee or customer survey systems, statis-
tical sampling systems, and generally accepted ‘‘ways of doing things’’ in
the organization. As you can see, coordination and control systems embrace
a myriad of possible methods for directing, monitoring, and assuring
adaptiveness and flexibility of the firm.
To design coordination and control systems it is useful to begin by
making two fundamental choices. First, how formalized do you want these
systems to be? Second, how centralized should control and coordination
be? Formalization and centralization are the two fundamental design
dimensions that underlie the design of coordination and control systems.
Formalization is the degree to which the organization specifies a set of
rules or codes to govern how work is done. One of the simplest ways to
coordinate work is through formal rules and regulations that govern how
work is to be done, who is to do it, and under what circumstances or
constraints. Formalization is high if these rules are very detailed and con-
sistently communicated to organizational members. Formalization is high if
rules are recorded in policy statements, such as in classic bureaucracies,
where written codes are formulated and published for all to see and share; or
if computer or other programs are written to monitor and provide feedback.
Monitoring and feedback systems serve to reinforce the formalization
160 Burton, DeSanctis, and Obel

process. As we shall see, it is possible for formalization to be high even if

rules are not ‘‘written down.’’ Rules can be communicated through training
procedures, modeling of behavior, or verbalized codes of working that
people are expected to learn over time. As an example of the latter, consider
an organization with high ethical standards that everyone knows and
follows, even if the ethical guidelines are not written into a document. The
guidelines are taught, practiced, and followed; everyone knows about them,
and so they are visible even if not on paper. The important thing to note
about formalization is that it bases coordination and control in very strong
expectations of how work should be done, with monitoring and feedback
mechanisms in place. In highly formal organizations there are penalties for
breaking rules.
Formalization is low if there is not a set of strongly written or accepted
rules or codes of conduct. Where formalization is low, there is high variance,
and hence flexibility, in the methods and procedures used to govern the
organization’s work. Rules are likely to change over time and vary across
circumstances. In the extreme, an organization with no formalization is
chaotic, and an organization with very high formalization is bureaucratic
and stifling of creativity. Most organizations operate somewhere in between,
with relatively high or relatively low formalization.
Centralization is the degree to which coordination and control are man-
aged by a core person or level of the organization, usually corporate head-
quarters. In the small start-up firm or the traditional bureaucracy,
centralization is usually high. Many modern firms have moved toward more
decentralized approaches to coordination and control. Hence, it can be
more meaningful to think about the degree of decentralization you want
in your firm’s coordination and control systems. Decentralization is the
degree to which responsibility for coordination and control lies in the
subunits of the firm and individual managers, rather than corporate head-
quarters or one specific level of the hierarchy. Decentralized systems accom-
modate the diverse needs of the more distributed enterprise and allow more
local responsibility for firm actions. The centralization–decentralization
distinction especially applies to operational kinds of decisions. If strategic
decisions are made at the top level in the firm but operational decisions are
made by the subunits, the organization is more decentralized than a firm in
which operational decision making is located at the top. Put another way,
coordination and control systems are more concerned with the design of
work process than setting strategy or policy. In today’s knowledge economy
managing work processes means managing how people interface with one
161 Coordination, control, and information systems

Figure 9.1. Coordination and control space.

another to complete the organization’s tasks. There is a strong human aspect

to systems design.
We use the dimensions of formalization and decentralization to summar-
ize the design options for coordination and control systems. Together, these
two dimensions suggest five basic approaches to the design of coordination
and control systems. You can think of these as different guiding paradigms,
or models of managing work processes. They range from being very simple
to being quite elaborate.
Figure 9.1 summarizes the five design options using the dimensions of
formalization and decentralization.

Family

If both formalization and decentralization are low, then coordination and

control systems are designed to rely on informal and centralized means of
control. We refer to this as a family-based model. There are few written rules
and procedures, and people know what to do based on what they are told by
a centralized source, probably the CEO or, in the case of a business unit or
department, the head manager. The organization operates like a family
where the head(s) of the household dictate(s) what is to be expected and
how work is to be done. The head of the household directs the show and
others following along because they believe in him/her. In the family model,
control and coordination are ad hoc, in the sense that rules are developed as
needed. So long as people comply with directives from the central source,
this approach to systems design works reasonably well. Further, low decen-
tralization and low formalization allow flexibility. However, troubles arise if
162 Burton, DeSanctis, and Obel

new members are added (e.g., the family grows), the head of household
changes (e.g., a new CEO is appointed), or other disruptions occur that make
reliance on informality and centralization ineffective for managing the work
of the organization. The well-known Walt Disney, founder of the Disney
Corporation, was famous for managing his company as a family operation,
with Walt acting as the family patriarch. Employees were expected to work
together like family members, helping each other out as needed, following
directives from the top, holding strong loyalty to the founder.
Entrepreneurs often run start-up ventures using a family model. Coordin-
ation is informal and centralized. People pull together to get work done,
relying on the boss for direction. Coordination using a family model can
be effective if the leader is competent, and the members are cooperative,
but this form of coordination also lends itself to becoming dysfunctional, if
members or the family leader are incompetent, the leader changes and is
not viewed with the same respect as the prior leader, or people don’t ‘‘get
along.’’ These types of challenges tend to require more formality and
decentralization of control in order for the organization to do its work.

Machine

If an organization adopts a machine model for coordination and control,

then the emphasis is on a high degree of formalization and a high degree of
centralization. Unlike in the family model, where authority rests in a central,
core place, such as corporate headquarters, the CEO, the auditing depart-
ment, or some other center of power in the organization, the machine model
systems are designed with documentation of rules and procedures in mind.
Considerable attention must be given to specifying how work should be
done, how it is to be monitored, and how the feedback and correction
systems should be designed. Machine-based organizations can tend toward
bureaucracy with many rules and procedures to govern work processes, but
this does not mean that such organizations are inefficient. On the contrary,
coordination and control systems can help the organization to operate as a
smart/intelligent machine. Just-in-time inventory management, Six Sigma
quality control systems, and 360-degree feedback are examples of control
systems that are designed with a machine-based model in mind. The organ-
ization makes high use of information to build efficiencies and adapt to
changing demands by modifying rules so as to make the organization
163 Coordination, control, and information systems

dynamic, not fixed. As we shall see, excellent data processing will facilitate
the success of the machine approach to coordination and control.
Wal-Mart is a good example of a company that has built excellent coord-
ination and control systems that allow the company to monitor all aspects of
its business, from hiring and employee development to inventory manage-
ment, sales, distribution, and forecasting. Many hospitals likewise adopt a
machine type of model when designing coordination and control systems. In
a hospital setting, a disciplined approach to information tracking is needed
to assure a high level of patient care. Job descriptions and detailed proced-
ures for doing work are critical. Reporting and accountability systems are
important for managing both large and small tasks.
Of course, the downside of the machine model is that it can lack creativity
and flexibility. Old systems have to be replaced with newer ones on a regular
basis; otherwise, the existing systems are likely to keep the organization
entrenched in the past rather than changing to meet new environmental or
other demands.

Market

The market model for the design of coordination and control systems
emphasizes low formalization and high decentralization. Some coordination
and control systems may be formalized, such as budgeting and performance
reviews, but overall there is an emphasis on more informal sources of
control, such as the value of sharing information or a culture in which
people are encouraged to ‘‘speak up’’ and report problems. Informal ap-
proaches to setting expectations and detecting difficulties occur through
training, custom, and everyday interaction of people. A key aspect of the
market model is that there are variations in coordination and control across
different departments or subunits of the organization, because it is difficult
in a decentralized, informal approach to develop consistent ways of doing
work and monitoring effectiveness. Standardized approaches to systems
design and use are shunned in favor of letting subunits police themselves
and/or work with other subunits as driven by informal norms.
If done well, the market approach can be an effective approach for
managing coordination and control, especially for promoting innovation
and customizing coordination and control needs to particular subunits of
the enterprise. Nokia (ECCH, 2003) has taken a market-based approach to
164 Burton, DeSanctis, and Obel

innovation, relying heavily on informal norms to foster innovation through-

out the company. Governance is similar to a confederacy or federation, with
pockets of control residing in product lines or regional groups rather than
corporate headquarters.
In the market model governance is relatively decentralized, meaning that
groups or business units oversee themselves with high autonomy relative to
corporate headquarters. A market organization is risk taking, tactical, and
innovative. With few stated rules, things may seem chaotic to the newcomer
who is trying to figure out ‘‘the rules’’ and ‘‘how to get things done.’’ But for
those who work inside the organization, low formalization and high decen-
tralization foster innovation. Of course, the downside of the market ap-
proach is that all subunits may not police themselves equally well, and there
can be a tendency toward conflict if the various units develop quite different
ways of executing work tasks.

Clan or mosaic

When decentralization is high and formalization is high, there are two

possible approaches for designing coordination and control systems.
Though these have common attributes, they are sufficiently different ap-
proaches to coordination to warrant separate discussion. The clan model
tends toward somewhat greater formalization and less decentralization. The
clan model uses strong norms to guide how work is done, and these norms
are deeply embedded in the hearts and minds of employees no matter where
they reside in the organization. Employees are selected based on their
likelihood of conforming to norms, which then are communicated through
training manuals and other formal means. In addition, rules for coordin-
ation and control area are communicated via extensive modeling by both
workers and managers and in discussions of ‘‘the way we do things’’ during
the everyday life of the organization (Ouchi, 1980). There are strong expect-
ations, and attention is given to designing systems that communicate these
norms on an everyday basis. Written rules and procedures establish a
minimal set of necessary standards from which people can design work
routines on an as-needed basis to meet changing work demands. In this
way, the clan model tends to be more flexible than the machine model.
Southwest Airlines with its colorful CEO, Herb Kellerher, developed a
very successful clan type of model on which highly efficient and effective
165 Coordination, control, and information systems

coordination and control systems have been built. The firm is famous for
selecting happy, fun-loving employees who interact well with customers
and value high efficiency and high quality of service that continually im-
proves over time. The airline is widely distributed across the US and operates
in a disciplined but not bureaucratic fashion. Employees are well versed in
a minimal, basic set of guidelines from which they can then make decisions
that meet customer needs. The firm insists on high quality standards, and
there is also a high degree of consistency that results from formalization of
rules and procedures. People work together as a strong community with
common values (Gittell, 2003). These common values provide the basis for
coordination and control systems.
The success of a clan depends heavily on having leaders who communi-
cate a strong set of norms and values that underlie how work is to be
accomplished, and on selecting and training employees who are versed in
those norms. At the same time, the people inside the clan don’t feel trapped
or suppressed. Instead, they are loyal to the organization and work together
for high efficiency and effectiveness. The Swedish furniture giant IKEA is
another example of an effective use of a clan model for coordination and
control. It developed ‘‘the IKEA way’’ of doing business, which consists of a
written set of principles for doing business. These are standard throughout
the world and strictly enforced, yet there is also managerial and employee
freedom to take needed steps to meet customer needs that vary across
countries and around the world.
The mosaic model for coordination and control tends toward somewhat
greater decentralization and less formalization than the clan model, al-
though it remains high on formalization and decentralization relative to
the models reviewed earlier. In the mosaic model there is a greater tendency
for heterogeneity (rather than similarity) of systems than in the clan model.
Coordination and control systems – including the rules that they embed –
are not identical throughout the organization. Instead, they vary as a
function of the subunit. Again, this is a matter of degree. The company
may have one inventory system for all its operations worldwide, but if it is a
mosaic, then it does not attempt to have all of its coordination and control
systems (e.g., accounting, human resources, performance measurement,
knowledge-management systems) standardized throughout the firm. In-
stead, common standards are minimized, and all remaining standards and
methods of monitoring are customized to meet the needs of subunits. To
allow effective coordination across the firm as a whole, the organization
attempts to keep disparate systems as compatible as possible; the various
166 Burton, DeSanctis, and Obel

coordination and control systems are not loose or scattered; instead, they fit
together into a meaningful whole, much like a mosaic.
Unilever takes a mosaic approach as it fosters heterogeneity across coun-
tries/regions and product lines, yet it integrates its disparate systems with
a strong managerial infrastructure of interdependency. Hence there is a
holistic quality to the organization – it is more than a collection of cells
or business units that operate with their own coordination and control
systems.
Using the mosaic model, if the organization decides to change its coordin-
ation or control systems in one area of the firm, it is not necessary that the
entire firm change its systems too. The mosaic approach allows disparate
systems across the enterprise to change much like a kaleidoscope. The pieces
move together. Although the change process may appear blurry, it is none-
theless systematic. As you can see, a mosaic model for systems design is
difficult to achieve. The approach requires what Powley et al. (2004) de-
scribed as a dialogic democracy, that is, extensive two-way communication
across subunits of the firm to assure that the needs of the whole are met even
as the customized coordination and control systems are developed for
subunits.

Information systems

In addition to coordination and control, information systems are vital

components of the infrastructure of the organization. Whereas coordination
and control systems are methods for governing how work is done, infor-
mation systems are methods for providing meaningful data to decision-
makers. We use the term information systems in the broad sense to include
all systems that collect, store, and process information within the enterprise.
The supplier or user of information systems may be outside the firm (e.g.,
customers or a government agency), but the systems themselves operate
under the control of the organization. Thus, the design of information
systems is integral to the larger process of organizational design. Informa-
tion systems may be computer- or telecommunications-based, but this is
not necessary. In fact, person-to-person passing of information is vital to
some organizational designs, as we shall see.
Many types of information systems are possible. Which are the most vital
for an organization? What types of systems should be the priority? Although
167 Coordination, control, and information systems

many factors can be used to determine the design of an organization’s

information systems, here we focus on two critical dimensions: the amount
of information and the tacit nature of information. Think about an organ-
ization’s dominant, overall information-processing needs when considering
these dimensions. The goal is to design information systems that can
optimize an organization’s information capacity.
Amount of information is the overall volume of data that an organization
must collect, process, and store on a regular basis. To some extent, this
dimension is a function of firm size; larger organizations tend to have
greater information-processing demands. But amount of information is
more closely related to the kind of work that a firm does, and to the design
of work tasks (per Chapter 6), than to organizational size. If tasks are
repetitive and executed hundreds or thousands of times throughout the
day (such as in a large retail chain or a bank), then the information-
processing demands are huge. Here we would say that the amount of in-
formation that must be processed is large. On the other hand, if tasks are
one-time tasks and there are relatively few and they may be general, frag-
mented or knotty, that is, there is variety in how tasks are done, then the
amount of information to be processed is smaller.
Note that a low amount of information to be managed does not imply
that information processing is an easy matter! Collecting, processing, and
disseminating information in a low-volume setting can be just as difficult as
in a high-volume setting. The amount of information has implications for
the approach management takes to information systems design rather than
for degree of difficulty of developing the systems.
The second critical dimension for information systems design is the tacit
nature of information that is exchanged within the organization. Tacit know-
ledge is characterized by causal ambiguity and difficulty of codification
(Choo, 1998; Polyani, 1966). Tacit information is not readily articulated as
a set of facts or rules, and so is difficult to transfer (Sorenson et al., 2004).
This is in contrast to explicit knowledge, which can be expressed formally
as a system of symbols and facts, and therefore more readily communica-
ted (Nonaka & Takeuchi, 1995). Of course, all organizations must process
both tacit and explicit knowledge. The question is which type is more critical
to the everyday functioning of the organization; that is, which is more
important to executing tasks and getting work done. If exchange of high
amounts of tacit information is critical to an organization’s everyday work,
then its approach to information systems design will be more people- or
relationship-based rather than event- or data-based.
168 Burton, DeSanctis, and Obel

Figure 9.2. Information systems space.

The two dimensions of amount of information and tacit nature of infor-

mation suggest four general approaches to information systems design.
These are summarized in figure 9.2

Event-driven

If the overall amount of information processing in a firm is low and the tacit
nature of information is low, then the information systems of the firm can
be designed using an event-driven model. This means that systems are
designed to process information associated with specific occasions or results
as they occur. Event-driven systems are reactive to needs as they arise.
Examples of event-driven systems are factual meetings, announcements,
and communication of directives from one location in the firm (usually a
manager or CEO) to another. Email and telephone calls communication, or
other systems, can also be event-driven. Systems are set up to communicate
information as needs arrive, with information flowing from the source to
the destination, hopefully in the smoothest possible way and with maximum
clarity.
Event-driven systems require little forethought or planning to implement,
except for purchasing and installing the technologies (e.g., an email system)
that provide the necessary conduits for passing information and alerting
employees to be responsive to information that flows through the system.
Investment in sophisticated search and retrieval systems is not a priority,
since the amount of information to be processed is relatively low, and the
primary objective of systems design is to pass information on on an as-needed
169 Coordination, control, and information systems

basis. Use of systems for sophisticated data analysis, interpretation, or long-

term planning is not a priority. So long as the tacit nature of information
is low (that is, the meaning of information is clear and readily interpre-
table), event-based systems do not present a problem for the organization.
Event-driven systems make sense in small, reactive organizations.

Data-driven

As the amount of information to be processed increases, information

systems design should become more data-driven. Systems can no longer
be based on a reactive model and instead require ongoing capture, analysis
and transfer of vital information. The data-driven approach is appropriate
for organizations that must process high volumes of information, and do so
in a systematic and intelligent manner in order to increase the firm’s infor-
mation-processing capacity. The data-driven approach assumes that infor-
mation is codifiable, that is, it can be readily captured and stored. The low
tacit nature of information is key to the efficiency of the data-driven
approach. Data-driven information systems increase the information pro-
cessing in the firm by bringing timely, detailed information to decision
makers, who can then act quickly and precisely to meet organizational goals.
If a firm adopts a data-driven model for information systems, then it can
expect to invest in large transaction-based systems, databases, and enter-
prise-wide systems such as SAP or PeopleSoft. These types of systems are
critical to the data-driven organizational design as they make it possible to
manage huge amounts of information for purposes of inventory manage-
ment, performance measurement, forecasting, quality control, and so on.
The data-driven approach goes along with the machine-based model for
developing the firm’s coordination and control infrastructure.

People-driven

Firms that process highly tacit information that is relatively low in volume
should rely on a people-driven model for information systems design. A
people-driven approach emphasizes capture, processing, and transfer of data
that is embedded in the minds and actions of people. It presumes that the
170 Burton, DeSanctis, and Obel

vital information of the organization is difficult to codify in a routine way,

and therefore the priority of systems should be either to bring people
together face-to-face so that they can share tacit knowledge, or to use
computer- or telecommunications-based systems that readily support
subtle, rich knowledge transfer. Face-to-face meetings are a classic type of
information system for sharing of tacit knowledge. If face-to-face meetings
are not possible, then intense phone conversations can suffice, or decision-
support systems and other rich online media, such as video conferencing,
are options for information systems.
Although you may not think of meetings or conferences as information
systems, they are just that – if they are thoughtfully designed with effective
knowledge transfer in mind. Organizations can take systematic steps to
design people-based information systems, and they should do so if the
amount of information to be exchanged is not extremely high and the tacit
nature of the information to be shared makes structured, routine types of
systems inappropriate or impossible to implement.
People-driven systems work well so long as the volume of information to
be exchanged is not too high. Consider a company that has developed a new
venture in a line of business that is unfamiliar to most involved, such as the
recent acquisition of the IBM PC product line by the Lenovo Group, a
Chinese company. As Lenovo and IBM undertake their merger, the newness
of the business model and its setting require exchange of information that is
not readily processed through a database, simple announcements, or other
standard information systems. Knowledge of new markets, products, cul-
ture, and the creation of new ways of doing business together in a novel
context demand many one-to-one personal exchanges among members of
both organizations. Online discussion groups and other unstructured ways
of exchanging ideas and designing methods of working together as a new
company are needed. A people-driven approach to information systems
design is appropriate in the formative stages of this new organization.
The people-driven approach makes sense wherever the tacit and relatively
unique nature of information requires a high degree of interpersonal inter-
action in order for information to be effectively transmitted. Development
of more routine, or standardized, information systems are not worthwhile
because information is not readily codifiable and can be expected to change
over time. Laboratory organizations and customized consulting are other
examples of settings where a people-driven model is appropriate. The
people-driven approach fits well with the market model for developing the
firm’s coordination and control infrastructure.
171 Coordination, control, and information systems

Relationship-driven

The most complex model for information systems design is also the one
with the greatest potential for promoting firm efficiency and effectiveness.
The relationship-driven approach to information systems design emphasizes
capture, processing, and transfer of data that is embedded in the links, or
relationships, between people and data. This is an appropriate design model
if the overall amount of information to be processed is high and the tacit
nature of information is high. Relationship-driven systems integrate hard
(codifiable) data with soft (interpretation) data to yield rich results for
organizational decision making.
The most well-known relationship-driven systems today are so-called
customer relationship management (CRM) systems. CRM systems capture
large, quantifiable data about customers but also provide interactive cap-
abilities so that two salespeople, for example, can exchange unstructured
observations or comments about their experiences and implications for
meeting new customer needs. Video conferences in which, for example,
physicians can talk to one another at a distance while both view and interact
with a patient’s MRI or CT images, are another example of relationship-
driven systems; the physicians may add comments or suggestions to the
medical record which are then visible, along with the more quantifiable data,
later on in the patient care process.
Relationship-driven systems are complex to develop because they include
both data-driven and people-driven elements. Well-designed relationship
systems include up-to-date transaction and database information as well as
softer, interpretive information that arises as people use the quantifiable
data. In this way the systems are not simply ‘‘updated’’ over time but instead
continually grow in their knowledge capacity as they are used. Sophisticated
searching algorithms and natural language interfaces are important to the
ongoing success of relationship-driven systems.
Organizations that process high volumes of information but also highly
tacit information cannot rely on a machine-based model because too much
of the information that they process is non-codifiable. Similarly, they cannot
rely on a people-driven model because, although effective for processing
tacit information, the people-driven approach is inefficient when the
volume of information to be processed is high. The relationship-driven
approach fits well with either the clan or mosaic model for developing the
firm’s coordination and control infrastructure. If there is a clan form of
172 Burton, DeSanctis, and Obel

governance, then the information systems will tend to be similar in design

throughout the enterprise. If there is a mosaic form of governance, then the
relationship-management model should vary to meet particular subunit
needs.

Diagnostic questions

Consider your unit of analysis as a whole in answering these questions.

Think in terms of the dominant, driving needs of your organization’s work
as it pursues its business goals. For each diagnostic question, use a 1 to 5
rating scale to score the organization as follows:
1 2 3 4 5
very low moderate very high

1. First, examine the two dimensions in figure 9.3: formalization and

decentralization. Rate your organization in terms of reliance on
formalization and decentralization as mechanisms for coordination
and control of work. Next, you can categorize the organization’s
governance approach as: family, machine, market, or clan/mosaic. To
begin, answer the diagnostic questions below.1
2. Decentralization
To what extent does your organization govern work tasks, people, and
processes using a decentralized approach: very little (1) or to a great
extent (5)? To answer this question you may think about the following:
a. To what extent are subunit decisions and actions directed by
corporate headquarters or another sole authority (1) versus managed
independently by the subunits (5)?
b. To what degree does the top management leave control of operational
decisions to managers or others in charge of those operations (1)–(5)?
c. How much discretion do subunit managers have in establishing their
budgets (1)–(5)?
d. How much discretion do subunit managers have in determining how
his or her unit will be evaluated (1)–(5)?

1
As before, you can average your scores for the items within each question to create an
overall score for each design dimension, or you can use the questions as a guide to assign an
overall score for each design dimension.
173 Coordination, control, and information systems

e. How much discretion do subunit managers have over how work

exceptions are to be handled (1)–(5)?
3. Formalization
To what extent does your organization use formalized methods of
coordination and control: very little (1) or to a great extent (5)? To
answer this question you may think about the following:
a. To what extent does the organization rely on rules, codes, or policies
to manage how work is done (1)–(5)?
b. Are there precise job descriptions to describe who does what and how
(1)–(5)?
c. Are there well-known expectations about what is ‘‘correct,’’
‘‘acceptable,’’ or ‘‘expected’’ of employees (1)–(5)?
d. Are there well-known penalties for violating rules or not meeting
expectations of on-the-job behavior (1)–(5)?
e. To what extent are employee actions monitored, recorded and/or
provided as feedback to either the workers themselves or to managers
(1)–(5)?
You can now locate your firm on figure 9.3. What is its design model for
coordination and control systems?
4. Next, examine the two dimensions in figure 9.4: tacit nature of
information and overall amount of information. Rate the organization
in terms of these aspects of information processing. From there, you can
identify its information systems design model: event-driven, data-driven,
people-driven, or relationship-driven. To begin, answer the diagnostic
questions below.

Figure 9.3. Locate your organization in the coordination and control space.
174 Burton, DeSanctis, and Obel

5. Tacit nature of information

Think about the kind of knowledge that is most critical to your
organization’s success in doing its everyday work. Rate your answer to
each question using the scale from very little (1) or to a great extent (5).
a. To what extent is the vital knowledge of the firm codifiable (1) or non
codifiable (5), in the sense that it does not lend itself to being
explicitly captured, processed and stored?
b. Could most of the important information that is exchanged within
the firm be readily recorded on paper or in a computer system (1 ¼
yes, 5 ¼ no)?
c. Does the information exchanged within the organization require
interpretation in order to be meaningful (1 ¼ not much interpretation
required, 5 ¼ high interpretation required)?
d. Is information relatively easy to understand and explain (1), or are
there subtleties to understanding the information, requiring special-
ized experience or expertise to fully ‘‘make sense’’ of the information
(5)?
6. Amount of information
What is the overall volume of data that the organization must collect,
process, and store on a regular basis, low or high? To answer this
question you may think about the following, answering each question
with a rating from very little (1) or to a great extent (5)?
a. Are most of the organization’s work tasks unique (1), or are the work
tasks repeated in large volume (5)?
b. Does execution of work tasks rely on having a relatively small (1) or
large (5) amount of data available to the worker or decision maker
responsible for the task?
c. To what extent are there common types of data that can be captured
and made useful for many transactions and tasks (1 ¼ low extent, 5 ¼
high extent)?
You can now locate the organization in the figure 9.4. What is its model
for information systems design?

Fit and misf its

What models for coordination, control, and information systems should be

used to design an organization’s infrastructure? What is a good fit? In table
9.1, we add the coordination, control, and information-systems models to
175 Coordination, control, and information systems

Figure 9.4. Locate your chosen organization in the information systems space.

the goals, strategy, structure, and process and people elements of organiza-
tional design. In each of the columns A, B, C, and D, the fit relations can be
read vertically from top to bottom.
The table shows how the infrastructure of coordination, control, and
information systems for a firm should be designed so as to be in alignment
with goals, strategy, structure, people and processes. In today’s world, there
is a tendency for managers to think that infrastructure, especially technol-
ogy, should be up to date and similar in features and operations across all
firms. The popularity of SAP, databases, videoconferencing, and the like
illustrates this trend. Although these technologies may be useful to any firm,
our analysis emphasizes that a thoughtful approach to infrastructure design
means taking time to map system priorities to those of the organization’s
overall design needs. A data-driven approach, for example, is not appropri-
ate for all organizations. An event-driven, people-driven, or relationship
approach may be more appropriate. It is useful to think in terms of first
selecting among the coordination and control designs and next among the
information systems designs, although these two go hand-in-hand as you
design your firm’s infrastructure.
If a firm is in column A in terms of goals, strategy, environment, and so
on, then coordination and control systems should be designed with a family
approach in mind. This means that there is great amount of informality,
high centralization of authority and control, a shop-oriented approach to
managing people, and a maestro leadership style. The configuration is
176 Burton, DeSanctis, and Obel

Table 9.1. Fit and misfit to include coordination, control, and information systems

Corresponding
quadrant in
organizational
design space A B C D
Information systems Event-driven Data-driven People-driven Relationship-driven

Coordination and Family Machine Market Clan/Mosaic

control

Climate Group Internal Developmental Rational goal

process

Leadership Maestro Manager Leader Producer

People Shop Factory Laboratory Office

Task design Orderly Complicated Fragmented Knotty

Knowledge Ad hoc Informated Cellular Network

exchange communications

Geographic Global International Multi-domestic Transnational

distribution

Organizational Blob Tall Flat Symmetric

complexity

Configuration Simple Functional Divisional Matrix

Environment Calm Varied Locally stormy Turbulent

Strategy types Reactor Defender Prospector Analyzer with Analyzer without

innovation innovation

Organizational Neither Efficiency Effectiveness Efficiency and Effectiveness

goals

simple and the business environment is calm, so elaborate coordination and

control mechanisms, as well as high-tech information systems, are not
necessary. In fact, the formality and high volume of information processing
brought on by elaborate systems can distract a firm in column A from
smoothly pursuing its goals. The mantra here for infrastructure develop-
ment should be ‘‘keep it simple.’’ The development of the more sophisticated
infrastructure can wait until the strategy or the environment demand it. The
family approach to coordination and control requires the executive to
provide timely and detailed oversight of work and workers. Further, people
177 Coordination, control, and information systems

must be ready to respond to announcements or directives as they are issued.

Keeping people informed of what is expected on a regular basis is critical,
since automated systems are not available to do this. The column A ap-
proach to infrastructure management will be very ineffective if the firm’s
goals, strategy, environment, configuration, or people and work processes lie
in any other column. Misfits arise quickly in those cases. So, if a firm is
currently using a family, event-driven approach to managing infrastructure,
but its goals or other key design attributes lie in another column, then the
firm should move to develop a more sophisticated infrastructure.
Moving to column B, we see that a machine-based model for coordin-
ation and control is consistent with a data-driven approach to information
systems design, and these go along with a factory model of people manage-
ment, a complicated task design, use of a manager style of leadership, and
development of an internal process climate. The successful firm in column B
values efficiency, is functional and tall in configuration, adopts a defender
strategy, and manages knowledge exchange using an intelligent, informated
approach. The information-processing capacity in column B is considerably
higher than in column A, assuming the machine- and data-driven ap-
proaches to infrastructure development are implemented and all other
design dimensions are in alignment (that is, they fall together within column
B). The more complex data-driven approach to information systems
design is necessary in order to support the high information-processing
needs of the column B firm. Detailed job descriptions, policy statements,
and methods for executing work tasks are critical, as are sophisticated
monitoring and control systems. The commensurate climate has high ten-
sion with less trust and leadership credibility; but productivity and efficiency
will tend to be high so long as new innovations are not needed on a regular
basis.
If firm goals require greater innovation due to a locally stormy business
environment and/or task designs are fragmented and the firm is arranged in
a divisional or multi-domestic structure, then the organization’s infrastruc-
ture cannot be as standardized as it is in column B. Instead, a column C
approach is more appropriate. Firms in column C pursue innovation and
foster autonomy of subunits, so variety in the infrastructure across subunits
is needed along with people-based information systems that support high
amounts of tacit knowledge sharing. Standardization of infrastructure is
minimized in favor of more customized approaches to coordination,
control, and information processing across subunits. This is a market–based
model of coordination and control, and it is less efficient than the machine
178 Burton, DeSanctis, and Obel

approach, but it is more likely to meet the firm’s goals for continual
innovation in the face of a locally stormy environment. As noted in earlier
chapters, column C firms do best if top management adopts a leader style,
a developmental climate, and a laboratory approach to managing people.
The executive lets others make decisions and tolerates inconsistency in
coordination and control infrastructure but accepts the entailed uncertainty
and high coordination cost for the organization as a whole. The climate
in column C has low tension and a high readiness for change. The downside,
of course, is that the infrastructure within column C is not conducive to
huge volumes of information processing. If both high volumes of infor-
mation and high amounts of tacit information sharing are desired, then the
firm should look to move to the column D approach to infrastructure
design.
For column D, the firm has a leadership style that acts as a producer, a
rational goal climate, a symmetric, matrix configuration, a turbulent envir-
onment, an analyzer strategy and pursuit of the dual goals of efficiency and
effectiveness. Infrastructure development for column D is undoubtedly the
most difficult and requires continual updating and adjustment in order to
be successful. Both people and data form the basis of information systems
design, and the firm infrastructure combines high amounts of subunit
autonomy with high amounts of formalization in control over people and
work processes. The firm in column D is simultaneously disciplined and
highly innovative. Consequently, design of coordination and control systems
must be done with extreme care, and they must be continually improved
and nurtured in order to remain supportive of the firm’s information-
processing needs. Flexibility is assured by either relying on developing a
cooperative, clan mentality among employees, in which people tolerate high
formalization while at the same time coordinating based on professional
sharing of tacit knowledge, or developing multiple but compatible methods
of coordination and control across subunits, with care taken to be sure that
systems fit together across the firm into a meaningful mosaic. The clan and
mosaic models allow the diverse subunits of the organization in column D
to act in concert despite high volumes of information sharing and high
decentralization of managerial control.
As we have observed before, the firm in column D relies on a producer
leadership style in which top management delegates decision making but
also controls uncertainty, e.g., through a high degree of formalization. The
climate has high tension but also has readiness for change. The organization
in this column is performance-oriented, aiming to achieve both efficiency
179 Coordination, control, and information systems

and effectiveness. It is a demanding place to work and operates with a high

degree of professionalism and continual capture and sharing of tacit know-
ledge. Formalization of rules is dynamic, which means that rules are strong
yet changeable as environmental conditions require the organization to
respond to or anticipate changes.
If your chosen firm is located in different columns based upon your
answers to the diagnostic questions in this chapter, then you should think
about what you might do to bring your organization into fit in the column
that meets your goals.

Summary

Coordination, control, and information systems collectively provide path-

ways for information sharing in the firm. Detailed design of these systems
requires specification of how work is to be done, who is to do it, how
monitoring and feedback will be managed, what information is to be
captured, stored, and processed, how knowledge is to be shared, and so
on. Job descriptions, inventory management systems, customer relationship
systems, databases, meetings, rich media systems, Internet portals, and many
other types of systems can be implemented as part of an organization’s
infrastructure for coordination, control, and information processing. This
chapter has provided basic approaches, or models, for deciding design
priorities and selecting among the many possible systems that might be
implemented. We outlined two critical dimensions for the design of coord-
ination and control systems and two critical dimensions for information
systems design. Applying these dimensions to your firm, you should take
steps to assure that the basic models that underlie the firm’s infrastructure
are in alignment with other aspects of its organizational design.

Glossary

Amount of information: overall volume of data that the organization must

collect, process, and store on a regular basis.
Centralization: the degree to which coordination and control of the orga-
nization’s work are managed by a core person or level of the organization,
usually corporate headquarters.
180 Burton, DeSanctis, and Obel

Clan model: design of coordination and control systems that rely on high
formalization, especially strong behavioral norms, and high decentral-
ization; there is a greater tendency for homogeneity (rather than variety)
of systems than in the mosaic model.
Control systems: methods for assuring quality and efficiency of information
flow between the highest and lowest level of the firm.
Coordination systems: methods for linking together the otherwise disparate
elements of an organization’s structure and supporting flexibility and
adaptiveness within and across departmental or divisional boundaries.
Coordination and control systems: systems that integrate the various parts of
the organization to support goal achievement and responsiveness to the
environment or task demands.
Data-driven systems: an information-systems design approach that empha-
sizes capture, processing, and transfer of high volumes of data that is
explicit in nature; an appropriate design model if the overall amount of
information processing in the firm is high and the tacit nature of
information is low.
Decentralization: the degree to which responsibility for coordination and
control lies in the subunits of the firm and individual managers, rather
than corporate headquarters or one specific level of the hierarchy.
Event-driven systems: an information systems design approach that
emphasizes transfer of meaningful data associated with specific occasions
or results as they occur; an appropriate design model if the overall
amount of information processing in the firm is low and the tacit nature
of information is low.
Family model: design of coordination and control systems that rely on
informal and centralized means of control.
Formalization: the degree to which the organization specifies a set of rules or
codes to govern how work is done.
Information system: methods for providing meaningful data to decision
makers both vertically and horizontally in the organization; information
systems may be computer-based, but this is not necessary.
Machine model: design of coordination and control systems that rely on
formal and centralized means of control.
Market model: design of coordination and control systems that rely on
informal and decentralized means of control.
Mosaic model: design of coordination and control systems that rely on high
formalization and high decentralization; there is a greater tendency for
heterogeneity (rather than similarity) of systems than in the clan model.
181 Coordination, control, and information systems

Organizational infrastructure: the collection of coordination, control, and

information systems that provide pathways for information sharing in
the firm.
People-driven systems: an information-systems design approach that
emphasizes capture, processing, and transfer of data that is embedded
in the minds and actions of people and so is difficult to codify in a
routine way; an appropriate design model if the overall amount of
information to be processed is low and the tacit nature of information is
high.
Relationship-driven systems: an information-systems design approach that
emphasizes capture, processing, and transfer of data that is embedded in
the links, or relationships, between people and data; an appropriate
design model if the overall amount of information to be processed is high
and the tacit nature of information is high.
Tacit nature of information: the degree to which the critical information of
the organization is causally ambiguous, difficult to codify, and so is
difficult to transfer from one person or locale to another.
10 Incentives

Introduction

In the previous chapters, you began with goals and analyzed your chosen
organization (unit of analysis) with respect to strategy, environment, config-
uration, task design, people, leadership, climate, coordination and control,
and information systems. Now we add another vital component to the
design of the firm: what incentives do you give to individuals and groups
so that they make decisions and take actions that fit well with the other
design characteristics and help to meet organizational goals? Incentives
support the firm’s infrastructure of coordination, control, and information
systems, and thus help to assure that work tasks are executed and organiza-
tional goals can be achieved. Design of incentives goes hand-in-hand with all
the other components of organizational design.
What do we mean by incentives? Incentives are means or instruments
designed to encourage certain actions or behavior on the part of employees,
or groups of employees (e.g., subunits of employees). Incentives are not just
the objective set of rewards that the firm offers but also the way people
interpret the rewards and act upon the rewards that the firm provides.
Monetary rewards in the form of salaries, wages and benefits are incentives,
but they are not the only ones. People respond to praise, acceptance,
belongingness, and recognition of self worth. A pat on the back as well as
a paycheck can be an incentive. A promotion or title can be an incentive.
In the end, the incentives must be internalized by people so that they
accept the incentives, consider the incentives to be fair, and are motivated
to do well. Roughly, an incentive will be viewed as fair if people think the
reward is reasonable for the level of effort, and also if the reward is consistent
with the rewards offered to other individuals or groups nearby.

182
183 Incentives

The alignment of incentives with organizational goals can be problema-

tic and many organizations get it wrong. In his famous paper entitled,
‘‘On the Folly of Rewarding A while Hoping for B,’’ Kerr (1975) descri-
bed many situations in which organizations misalign incentives, desired
behavior and outcomes. A common example is, as an executive, you reward
what you can measure, not what you want to achieve. Executives do this
when they hope for team work while rewarding individual performance.
As another example, consider a company that is trying to be fair to its
employees in a downturn and so devises a scheme whereby the employees
get six months’ pay if fired, but nothing if they leave voluntarily. In such a
situation employees will begin to work in such a way so as to be fired.
The intention to be very fair becomes an incentive for poor work and
shirking.1 These are unintended consequences of good intentions on the
part of management. A heuristic you can use to help guard against the folly
of rewarding B while hoping for A is to put yourself in the position of
the employee and think about all of the things you might do in response
to an incentive offered by the organization – make a list. For the situation
described above, it seems reasonable to anticipate that some employees,
though not all, would become poor performers in order to be fired. Being
fired can be in the individual’s self-interest, but it is not in the firm’s best
interest.
A traditional view of incentive design has taken a control perspective,
that is, designing incentives so that they control either the behavior of
employees or the results of what managers or employees do (i.e., how they
make decisions or what actions they take). The distinction between control-
ling behavior and observing results is fundamental (Stinchcome, 1965).
Controlling behavior involves monitoring the individual, i.e., how work is
done. Controlling results involves monitoring the outcomes once work is
completed. One reason for making the distinction between controlling
behavior and results is that it can be difficult, if not impossible, to see what
the employees actually do from a behavior or process perspective. In a virtual

1
For our executive MBA classes, we ask each student to describe an incentive situation
where the firm is ‘‘Hoping for A, while rewarding B.’’ The examples are easy to find. They can
be categorized as: reward sales, when profit is the goal; reward what is easy to measure (in
numbers), not what is important; reward what is measured when the real objectives are costly
to measure; reward a false equity of treating everyone the same when differential results are
desired and known; a false sense of fairness; executive fear of assigning differences. In the best
situations, the incentive scheme supports the firm’s goals; in the worst situation, it drives the
individuals in the opposite direction of the desired behavior.
184 Burton, DeSanctis, and Obel

organization the employees may be far apart and it does not make sense to
try to control the actual behavior. Further, it may not be possible to know
what a person is doing by observation. For example, it is difficult to know if
an individual is thinking deeply on a firm problem or loafing. For decades, it
has been a norm to compensate the sales force on the road by an individual
bonus, as it is difficult to control their effort directly.
In other situations it may not be clear what the appropriate behavior is.
The employee may have better information or better skills to make a decision
or take an action, so he or she can judge better with regard to what to do.
Here, to assure goal achievement, the control of the outcome is much better.
The worker has better information and knowledge regarding behavior, so
management bases rewards on outcomes instead.
You can influence the activities of the employees or subunits in an
organization in basically three ways: you can tell them what to do; you can
have a set of behavioral rules that specify what they should do in certain
situations; and/or you can create an incentive system that indirectly influ-
ences your employees to do what you want them to do. These three ways of
controlling the activities of the employees should be seen as a package.
For example, there may be a rule that says what to do and an incentive
system whereby the employees are rewarded if they follow the rule well.
There may also be a central decision authority that tells the employees
which set of rules to invoke in a particular situation. Or the employees
can be given a freedom to make decisions about how to do their job, but
their performance is evaluated on the basis of the outcome on which you
are rewarded.
One further complication is that the employees may not determine the
outcomes of their work. The uncontrolled environment could be a major
factor in determining the outcome as well as the employee’s decisions and
actions. For an uncertain environment you have to sort out how much of
the outcome is based on the employees’ performance and how much is the
result of uncontrollable events. Were the employees working hard or just
lucky? Thus, an important issue is who bears the risk – the individual or the
organization? Results-based incentives put the risk on the employee (or
subunit) and thus may include elements outside the control of people in
the organization. This is the basis for agency theory where the principal
‘‘gives’’ some of the risk to the agent, who has more, though imperfect,
information (Kowtha, 1997). Agents are rewarded based on results, even
though some of those results are outside of their control. Corporate share-
holders who evaluate management based on stock value, or owners of sports
185 Incentives

teams who evaluate coaches based on their winning records (regardless of

injuries or the intensity of competition) are examples.
For an incentive based on behavior (rather than outcome), employees
bear no risk on the results, and thus the organization assumes the business
risk. People are not responsible for bad decisions, bad luck, or unforeseen
events in the environment that they cannot control. As an example, soldiers
might be rewarded based on following military protocol and executing the
orders of their superiors, regardless of whether their actions lead to winning
one or another particular battle. Employees in a call center might be
rewarded for the number of calls they process and the quality with which
they treat customers, regardless of sales actually made. Of course, the
illustration above is in a relative sense. If the organization fails for an
extended period of time, it will disappear and the individual will lose
employment, so the individual always bears some risk.
Why would an individual assume some of the risk? One reason could
be that the amount of compensation is higher than it would otherwise be.
Top managers are given large compensation packages in high-risk firms
for this reason. Also, people vary in their risk preferences, and one’s cul-
tural background may affect willingness to take risks. So it may be ap-
propriate to let some employees or business units assume more risk than
others. For example, younger workers or entrepreneurs may prefer large
risks, and research and product development groups may prefer large risks
relative to other departments.
The level of employee education is an important element that can help to
estimate whether an individual is willing to assume some risk (Kowtha,
1997). In general, the higher the skill level, the more risk the individual is
likely to assume – perhaps due to greater understanding of the risk. If an
individual is to assume some risk, the individual should have the necessary
skills and knowledge to understand and adapt to the uncertainty. Also, the
individual should be given authority to make the appropriate decisions
in the course of carrying out his or her work tasks (Kowtha, 1997). This
introduces the relationship between the degree of delegation and the de-
sign of the incentives. If decisions are delegated, the incentives should be
aligned so that they support the organization’s goals of efficiency and
effectiveness.
There is little doubt that incentives affect employee behavior. This notion
is supported by research in psychology and economics (Gibbons, 1998), as
well as in everyday observations. The individual in the organization is
affected by the particular incentives that the organization offers. The same
186 Burton, DeSanctis, and Obel

is true for groups or other subunits. If people are provided incentives to

work in concert with others, versus to work independently, behavior will be
affected accordingly. Do we do it alone, or do we work in concert with other
groups or subunits to achieve organizational goals? The answer depends on
the design of the incentive system.
In relation to organizational design, we examine incentives from a stra-
tegic organizational design point of view. This means that we look at the
basic principles for designing specific incentive systems for the groups or
individuals that make up the unit of analysis and the nature of their
activities in the organization. We will not discuss the details of setting up
a specific incentive contract or plan for a particular individual or subunit.
The incentives may be related to monetary compensation or other types of
rewards. Many issues have to be taken into account when the incentive
system is designed.
From the above discussion we can see that a fundamental design choice is
whether to base incentives on behavior or results. This design dimension is
the basis of evaluation of the work. This way of looking at incentives goes
back to the sixties. A survey of some of the theoretical issues related to this
view can be found in Kowtha (1997). At one extreme, behavioral incentives
focus on procedures: compliance with standards, rules and routines. At the
other extreme, results incentives focus on outcomes, i.e., the effectiveness of
meeting the goals of the organization. An important issue is whether
behavior (i.e., procedures) or results (i.e., outcomes) can be monitored
and evaluated and whether the monitoring is possible on an individual or
a group basis. Information is always costly, and the measurement of activ-
ities related to incentives – whether individual or group – is costly. The
choice of controlling behavior versus results has also been analyzed from a
transaction cost view. Hennart (1993) argues that the choice of controlling
behavior or results is not a pure choice; the optimal form is a mixture of
the two. A mixture of the two is often the norm.
A second fundamental design choice is the target of incentives, that is,
whether to base incentives on individual or group work performance.
Whether work incentives are individual or group-based depends on your
unit of analysis. If your unit of analysis is a team, for example, then you can
design incentives for the individual team members, or you can design
incentives for the team as a whole. In the latter case, you could reward the
team based on its collective behavior or results, rather than on the results
or behavior of any one person. If your unit of analysis is a large organiza-
tion with a collection of divisions or business units, then you could reward
187 Incentives

Figure 10.1. The incentive system design space.

each division or subunit based on its behavior or results (individual), or you

could reward the divisions based on their collective (group) behavior or
results. Individual and group represent two different targets in the design of
incentive systems.
Using the two dimensions of incentive system design, basis of evaluation,
and target of incentives, we can outline a typology of four distinct types of
incentive systems as shown in figure 10.1: personal pay, skill pay, bonus-
based, and profit sharing. This typology provides the overall principles on
which the incentive system for an organization should be designed.
We will now discuss each of the four basic types of incentive systems:
personal pay, skill pay, bonus-based, and profit sharing.

Personal pay

Personal pay is particular to the individual people or subunits within the

unit of analysis and their formal or psychological contract with the em-
ployer. A personal pay approach to incentive design is based on the behavior
of the individual, as shown in lower left quadrant of figure 10.1. It is an
agreement between the individual and the boss or organization.
The personal pay approach does not emphasize results or outcomes, but
instead emphasizes individual compliance with rules or directives. There
may be some evaluation based on performance (such as an occasional
bonus), but this is not the dominant concern. Instead, rewards are designed
with ‘‘doing work the right way’’ in mind. Usually, personal pay-based
systems take the form of an incentive agreement or contract whereby
188 Burton, DeSanctis, and Obel

employee behavior is measured in terms of people showing up at a particu-

lar time and then working for a number of hours. The measurement of the
behavior is often done by ‘‘clocking in’’ and ‘‘clocking out.’’ There may be
job descriptions with particular rules that the individual has to follow.
Within the rules given, the individual follows the directions of the boss
and does things as requested. The actual pay is negotiated, and there can
be significant differences between persons or groups doing the same job.
There is a focus neither on efficiency nor on effectiveness, as the most
important thing for employees is to follow rules or directives. The effective-
ness and efficiency of such a reward system requires that the environment
does not change too often and that the organization has chosen the right
rules to follow. Further, to some degree this approach assumes that the
people being evaluated have the kind of personality that can accept and
function within such a reward system. Dynamic risk seekers will have a
difficult time with such incentives. Further, the actual pay level will tend to
be lower than that in other incentive systems as it is the organization or firm
that assumes all the risk.

Skill pay

Skill–based pay is perhaps the most widely used type of incentive system
in the modern world. It is the basis for pay in most organizations, whether
private firms or governmental bureaucracies. Within this design approach,
pay differentials are skill- or position-based rather than results-based. The
salaries, or wages for a normal week, are determined by individual skills or
particular position or job in the organization, that is, the group to which the
individual belongs. In many organizations, particularly public ones, skill is
measured in terms of formal education and seniority, which incorporates
the ability to perform certain tasks. Thus, one can receive an increase in
one’s salary by getting more formal education, staying in the same organiza-
tion for years, or moving up the ranks while becoming more skilled within
the rules of the organization. Although rewards are related to the individual,
the pay is based on belonging to a particular group described by seniority,
rank, and education – the upper left quadrant of figure 10.1.
This incentive system is based on the idea that people should utilize the
requisite skills and follow job descriptions, rules, and the policies of the
organization. The assumption is that it is possible to describe in some detail
189 Incentives

what the people should do in the course of their everyday work. In modern
organizations, these details are often embedded in computer systems that
provide information on how to execute tasks and, in turn, monitor the
speed and accuracy of work performed. Further, it is assumed that people
will accept direction from the organization’s authority, usually the hierarchy,
on how to do their work.

Bonus-based

We now move to the lower right quadrant in figure 10.1. Here, contrary to
the two categories above, a bonus-based incentive system changes the focus
from the behavior to the results – i.e., from process to outcomes. The
organizational results can be mapped back to the accountable party, who
then is given the bonus. Usually, the bonus is ‘‘in addition to’’ the normal
pay as skill-based. However, there are some employees such as salespersons
who receive the totality of their compensation based upon the resulting
sales. We consider both of these as bonus-based. The bonus-based view of
incentive system design is rooted in a management-by-objectives philoso-
phy. Goals are set for employees (which are derived from the organizational
goals), and rewards (e.g., pay, promotions, travel, stock options, etc.) are
distributed based on performance which is compared to pre-established
targets or goals. It could be a sales or production target, but it could also
be the outcome of an organizational unit (e.g., business unit profit). The
use of results-based incentives is appropriate if the organization has the
ability to clearly link the performance and outcome to behavior. Some
uncertainty in this link is acceptable, so long as people generally view the
behavior–outcome link to be fair and largely under their control.
An example of such bonus-based incentive scheme is when a top man-
ager’s salary is dependent on the profit of the organization or, if it is a public
traded company, on the stock-value. The idea here is to align the incen-
tives of the top manager with the owners, such that decisions are taken to
increase the value of the company. Such an incentive approach has been
highly recommended by many stock analysts, but there are also many
unwanted side effects of these schemes. They may trigger unethical or even
illegal behavior on the part of executives anxious to get the expected results
when results do not come out as desired. There can also be a tendency to
withhold information – hiding bad news – from superiors providing
190 Burton, DeSanctis, and Obel

rewards. Thus, when you design an incentive scheme for an organization, it

is important to not only consider the results you want from employees
but also to anticipate negative side effects that might occur if the scheme
is implemented (Hennart, 1993).
Bonus-based incentives can also be used in informal and short-term
relationships, such as day work for unskilled labor or compensation of
skilled individuals engaged in specialized tasks (e.g., project consulting or
repairing high technology equipment). For the unskilled individual, the
boss may give very detailed directions. But for high-skilled workers, the
request may be as simple as ‘‘It does not work, fix it,’’ and the employees are
compensated if fixes are made. Contract lawyers might be paid based on
winning litigation. Real estate agents or brokers might be paid on closing
important sales deals. In the consulting business there are ‘‘no cure – no
pay’’ contracts. In either situation, the relationship is individualized and
customized to meet the needs of the organization. Relationships may be
long-term, as in long-term IT outsourcing contracts in which service-level
agreements determine compensation level, but bonus-based systems more
often are short-term focused and linked to a particular task.
The advantage of a bonus-based incentive system is its flexibility for the
organization. First, the organization can ask the individual to do whatever
is needed to be done. Further, bonus-based requests can be created by the
organization on an ongoing basis and little planning is required. Second, the
bonus aspect of the contract normally does not have a long-term commit-
ment by the organization. Thus, the organization can change its requests on
short notice, terminate the relationship, or renegotiate the bonus aspect of
the contract. Such relationships were common in the old days for manual
work. In Denmark today, it only exists for the people who work in the
Danish harbors loading and unloading ships. However, it is coming back for
a number of freelance workers, e.g., workers in virtual networks without
long-term contracts, call-center workers, and some factory workers. For the
individual, the relationship provides some freedom but is less desirable in
the sense that there is little security in the relationship and the organization
can be quite arbitrary in creating and terminating incentives. Cisco, the
maker of network routers, has used bonus-based pay to increase productiv-
ity of router assembly groups during periods of peak demand. The incen-
tives then are removed during periods of low customer demand (with
skill-based pay remaining as the base pay for the unit). Jack Welch, CEO
of GE for many years, told his business units that they had to be top-
performing in their respective business, or he would sell them off. Incentives
191 Incentives

were tied to results, along with the threat of breaking off the relationship if
results did not meet expectations.
To summarize, bonus-based incentives are related to contractual relation-
ships between the individual and the firm, and they can take many different
forms in today’s modern organization.

Profit sharing

The fourth category in the upper right quadrant of figure 10.1 is profit
sharing, which is group-based, either among a group of individuals or a
collection of subunits. The basic idea is similar to bonus-based incentives
that tie compensation to results rather than behavior. People are rewarded
on the basis of effective collaboration with others to yield high perfor-
mance on the part of the group. Profit sharing is not only giving a fixed
bonus to the unit of analysis, but also a share in the profits (revenue less
costs) to all members of the unit. As such it is more group-based than
individual-based. The group could be a team, a division, department, or
other subunit; it could also be the total organization – the firm. The idea
behind a profit-sharing incentive system is that it should enhance group
performance in a developmental mode where it is not possible to antici-
pate or control the actual outcome by controlling behavior. For the profit-
sharing scheme to work, people should feel that individual performance
can make a difference for the group outcome. The task itself must depend
upon the joint efforts of everyone in the target group. The profit-sharing
incentive system is appropriate when results are based on different individ-
uals coordinating their skills and knowledge, and where excellent individual
performance is not sufficient to yield high group performance. In other
words, this incentive system presumes interdependency among the work of
the individuals who make up the group that is the target.
Team-based sports illustrate how this type of incentive system can be
beneficial. Successful teams require not only highly talented players but also
effective collaboration among teammates to yield high performance on
the part of the group. Michael Jordan may have been the best National
Basketball Association player in his time, but that was not enough to ensure
top performance by the Washington Warriors. In a similar way, teams,
departments, or other subunits within your firm may depend on not only
excellent human resource talent but also high group performance for
192 Burton, DeSanctis, and Obel

maximum effectiveness. Whole Foods, a large US gourmet grocery chain,

divides its stores and the departments within them (meats, fish, dairy,
prepared foods, etc.) into profit centers that receive rewards on a regular
basis for P/L performance at both the store and departmental levels.
The smaller the target group the more likely the profit-sharing scheme
will have the anticipated effect. If the firm is large then profit sharing based
on the total outcome of the organization is less likely to be effective, since
individuals cannot see the effect of their efforts on the organizational
performance. A so-called free rider problem can result, with some people
relying on the skills and success of others to carry the group to success. On
the other hand, the profit-sharing approach can be very effective if people
believe their contributions to group efforts ‘‘matter,’’ so they are committed
to working together with colleagues, and they view the incentive scheme to
be fair.

Diagnostic questions

The above four categories are prototype incentive schemes. Often, firms
design incentive systems with a combination of approaches, such as a
behavior-based incentive program that is augmented somewhat with
results-based incentives. This yields a lower risk for the individual, but the
basic philosophy is the focus on results. So the important issue from an
organization design perspective is to establish the driving philosophy. From
there you can determine the specifics of the incentive scheme for your
organization.
Now it is time to look at your unit of analysis and determine where your
chosen organization is located in figure 10.2 and where you would like it to
be located.
1. What is your unit of analysis that you chose in Chapter 1 and have used
throughout our step-by-step process? Use this unit of analysis as the
organization when answering the questions below.
The questions below will help you locate your organization on the
behavior–results and individual–group dimensions. For each question
use a 1 to 5 rating scale to score the organization as follows:

1 2 3 4 5
very low moderate very high
193 Incentives

2. To what degree are results-based incentives used in your organization

(1)–(5)? To help you answer this question you may consider the
following:
a. Are people rewarded using a fixed salary or hourly/daily wage (1), or
are they rewarded based solely on the quality/quantity of their work
output (5)?
b. To what degree are salaries or other rewards based on the skills,
experiences, on-the-job efforts, or cooperativeness of the employee
(1), versus the quantity or quality of their work results (5)?
c. In evaluating people for promotions, benefits, or other rewards,
whether tangible or intangible, does management emphasize how
work is done (1) or the results of the work (5)?
3. To what degree are group-based incentives used in the organization
(1)–(5)?
To help you answer this question you may consider the following:
a. Do the activities in your organization require collaboration by
individuals (1) or can they be done by a group (5)?
b. To what extent are teams compensated based only on individual
performance (1) versus performance as an organization (5)?
c. In evaluating people for promotions, benefits, or other rewards,
whether tangible or intangible, does management emphasize the
individual’s work (1) or the total organization’s (unit of analysis)
work (5)?
4. Now with the values of behavior–results and individual–group
dimensions, locate your chosen organization on the graph. What is its
incentive scheme?

Figure 10.2. Locate your organization in the incentive scheme design space.
194 Burton, DeSanctis, and Obel

Fit and misfits

Table 10.1 shows the fit and misfit relations for incentive scheme design
relative to the other aspects of organizational design that we have considered
in our step-by-step approach. Again the columns represent fit with the
design dimensions. Misfits occur if the organizational design elements do
not all fall within the same column. The four columns correspond to the
four main quadrants of the organizational design space.

Table 10.1. Fit and misfit table for incentive alignments

Corresponding
quadrant in
organizational
design space A B C D

Incentives Personal pay Skill pay Bonus-based Profit sharing

Information systems Event-driven Data-driven People-driven Relationship-driven
Coordination and Family Machine Market Clan/Mosaic
control
Organizational Group Internal process Developmental Rational goal
climate
Leadership Maestro Manager Leader Producer
People Shop Factory Laboratory Office
Task design Orderly Complicated Fragmented Knotty
Knowledge Ad hoc Informated Cellular Network
exchange communications
Geographic Global International Multi-domestic Transnational
distribution
Organizational Blob Tall Flat Symmetric
complexity
Configuration Simple Functional Divisional Matrix
Environment Calm Varied Locally stormy Turbulent
Strategy types Reactor Defender Prospector Analyzer with Analyzer
innovation without
innovation
Dimensions Neither Exploit Explore Exploit Explore
of strategy
Organizational Neither Efficiency Effectiveness Efficiency and Effectiveness
goals
195 Incentives

The personal pay approach in column A fits well with the simple con-
figuration where the boss can change tasks quickly. Employees are paid
basically for being there and doing what they are instructed to do. This
incentive scheme works well if there are not a lot of information-processing
demands for the organization, such as in a calm environment or for a
reactor strategy. The personal pay approach is focused neither on efficiency
nor effectiveness. It will not work well in situations with few rules and a high
degree of decentralization which will be required in non-static environments
with a prospector strategy.
The skill-based incentive system in column B fits well with a bureaucracy
or machine type of organization with a high degree of specialization that
can form the basis for group-based incentives. With a focus on skills, the
desire is to obtain efficiency to defend the organization’s position. This
incentive approach works well in stable environments and can handle a
great deal of complexity both in the environment and in the organization.
‘‘What to do’’ is based on a set of rules and standardized behavior. The
skills are often associated with a high degree of specialization that comes
with training and experience. This fits with situations with long periods
of stability, job specialization, and a high degree of formalization. Thus,
this incentive structure fits an organization with a relatively stable envi-
ronment, a defender strategy, and a high degree of specialization in func-
tional configuration. A machine approach to coordination and control,
complemented with a data-driven approach to information-systems design,
will serve to reinforce the success of the skill-based incentive system. Such
an incentive system is an integral part of a bureaucracy. Here the individual
runs no risk vis-à-vis the actual outcome of the organization. This could also
be the reason why such a system is often the preferred incentive structure in
organizations where it may be difficult to specify what the outcome is (as is
the case in many public organizations).
The skill-based incentive approach is a misfit with a prospector strategy in
stormy environments with a focus on product innovation and a quick
reaction to changes in, for example, customer requirements. In such situ-
ations a results-based incentive is preferable. The skill-based incentive
system is often one part of an incentive structure that is then topped off
with a results-based part (such as bonuses).
The bonus incentive system in column C requires that people are willing
to assume the risk where the reward will depend on some issues beyond their
control. Usually such people like to understand the risk they are taking.
Depending on the situation, this may mean that the skill level of people
196 Burton, DeSanctis, and Obel

involved in this incentive system is relatively high and/or the personality or

cultural expectations of people are such that they accept risk taking. The
bonus incentive system fits well with a developmental climate that has low
tension and a low resistance to change. The market style of coordination and
control, along with people-based information systems, serves to reinforce
the success of the bonus system. A locally stormy environment calls for fast
response to novel solutions; in that case you may not have the appropriate
information at the top level. Thus you need to delegate. Once you delegate,
more incentives have to be based on results rather than behavior (Nagar,
2002). In this way, the bonus-based incentive system fits well with a pro-
spector strategy in a divisional or cellular configuration with a high degree of
decentralization for the locally stormy environment.
The profit-sharing incentive scheme in column D works well in a matrix
or transnational organization where there is an emphasis on coordination
among the people in the organization to achieve successful projects with
limited resources. It is also a good fit with an analyzer strategy in a turbulent
environment as well as a rational goal climate which strives for change with a
cooperative spirit and high tension. Along with the clan or mosaic approach
to coordination and control, a leadership style with controlled delegation,
and relationship-based information systems, the profit-sharing incentive
approach is part of the cultural glue that holds the organization together,
integrating disparate subunits that confront a turbulent environment.
Some firm outcomes are more sensitive to incentive schemes than others,
and thus the design of incentive schemes is more important for the former.
For example, the divisional configuration is rather insensitive to a behavior-
versus a results-based incentive system. On the other hand, a functional
configuration is very sensitive to choice of incentive system. To introduce a
results-based incentive system for a departmental outcome in a functional
structure will introduce a negative competition that most likely will destroy
the required cooperation between the departments (Burton & Obel, 1988).
It may seem like a good idea to put the department on a profit scheme, but it
can lead to conflict and poor firm performance if not managed such that the
other organizational design components fit with such an approach.
To get an incentive system to provide the behavior you want, the individ-
ual must accept it. People must perceive it as motivating. Further, it must be
related to dimensions that you want the incentives to affect. Here you must
remember that incentives affect the other organization dimensions that are
not a part of the incentive system. If the pay is based on volume but not on
quality, then you get volume, but not necessarily quality. Thus, there are
197 Incentives

often undesired consequences of an incentive scheme. As a heuristic, it is a

good test to ask the question: what might I do in this situation? Some
possibilities are likely to be desired, whereas others are not. Hoping for A
while rewarding B is a frequent mistake.
In the long run the various dimensions in the step-by-step model may
influence each other. The equity question is very important as an incentive
scheme that is seen by people as being unfair will affect the climate and may
weaken the coordination and control infrastructure. From the theory of
motivation we have a simple model which says employees compare their
compensation to a peer group and that the comparison has to be viewed as
fair. If not, it will very likely create dissatisfaction and thus influence the
climate. Perceptions of non-equity may not lower organizational perform-
ance (Bartol & Locke, 2000), but they can indirectly lead to misfits in the
climate, leadership, and coordination and control aspects of organizational
design. In the long run, misfits will tend to reduce organizational efficiency
and effectiveness.

Summary

In this chapter we have described the basic incentive dimensions and

categories from a strategic organizational diagnosis and design point of
view. The focus has been on whether the incentives system has a primary
focus on individual versus group and whether the incentives are based on
affecting behavior or results. There are numerous examples where the
chosen incentive system does not support the firm’s goals. We concluded
by relating the four basic incentive categories: personal pay, skill pay, bonus-
based and profit sharing, to the design dimensions presented in the previous
chapters. A firm that designs its incentive system so as to align with other
aspects of organizational design is more likely to achieve its effectiveness
and/or efficiency goals.

Glossary

Basis of evaluation: a fundamental design choice of whether to base

incentives on behavior or results.
198 Burton, DeSanctis, and Obel

Behavior evaluation: measure of the adherence to work orders, standards,

and/or rules; contrasts with results evaluation.
Bonus-based pay: an incentive system where the rewards are based on results
evaluation as executed by an individual.
Group rewards: pay or other rewards which are given to the individual based
upon a group’s behavior or results.
Incentives: means or instruments designed to encourage certain actions or
behavior on the part of employees, or groups of employees.
Individual rewards: pay or other rewards which are given based upon the
individual’s behavior or results.
Personal pay: an incentive system where the rewards are based on behavior
evaluation as executed by an individual.
Profit sharing: an incentive system where the rewards are based on results
evaluation as executed by the group of people within the unit of analysis.
Results evaluation: measure of the outcomes of work or performance;
contrasts with a behavior evaluation.
Skill pay: an incentive system where the rewards are based on behavior
evaluation as executed by the particular group of people within the unit
of analysis.
Target of incentives: a fundamental design choice of whether to base
incentives on individual or group work.
199 Incentives

Where are you in the step-by-step approach?

STEP 1
Getting Started
(1) Goals
STEP 2
Strategy
(2) Strategy
(3) Environment
STEP 3
Structure
(4) Configuration complexity
(5) Geographic distribution knowledge exchange
STEP 4
Process and People
(6) Task design
(7) People
(8) Leadership and organizational climate
STEP 5
Coordination and Control
(9) Coordination, control, and information systems
(10) Incentives
You have now completed the step-by-step approach to
organizational design. By completing all steps, you have
taken a holistic approach, rather than a piecemeal approach,
to assessing and planning for change in your organization.
The holistic approach is more complete and will help the
organization confront the change and move toward its goals.
200 Burton, DeSanctis, and Obel

In this last step, Step 5, you assessed and planned for change
in the firm’s coordination and control systems, information
systems and incentives. These systems are just as critical as
the design components considered in the earlier steps, and it
is important that you have completed Step 5 with care to
help your firm realize its goals. As previously, these
coordination, control, information systems, and incentives
should fit with the strategy, structure, processes, and people
in the firm. If there are misfits, they should be addressed.
Think about and plan for changes that will need to be made
to reduce misfits and bring all components of the organiza-
tional design into alignment. First, consider alignment
within the coordination, control, information systems and
incentives, and then alignment among all the design
components of the previous steps for a holistic approach.
In the next and final chapter on design dynamics we consider
the process of organizational change in more detail.
Applying the step-by-step
approach in a dynamic world
11 Design dynamics: managing change and
multi-organizations

Introduction

In this book, we have developed a step-by-step approach to organizational

design, focusing on analysis of an existing organization and identification of
misfits among organizational design components. We have outlined a way to
diagnose whether a firm has strategic organizational design misfits and
considered what you might change to align the basic design components
so as to fix misfits and meet organizational goals. In this final chapter we
address what to change and why. It is a summary of the design process with
special attention to the question, what aspects of your organization should
be changed? First, we will briefly restate the step-by-step approach and then
we will examine the change process in more detail. Finally, we will discuss
more complex design situations, that is, multi-organizations, including joint
ventures, mergers, and strategic alliances/partnerships.

Where are you in the step-by-step approach?

If you have systematically followed the steps outlined in Chapters 1 through

10, then you have completed the step-by-step approach for your chosen
firm. You have followed each step below by, first, assessing your firm’s
current organizational design; second, identifying misfits (i.e., detecting
components that are not well aligned to meet organizational goals); and,
third, deciding what you might do to either bring the design components
into the same quadrant of the organizational design space or tolerate the
misfits. To summarize, the step-by-step approach is:

203
204 Burton, DeSanctis, and Obel

Step 1: GETTING STARTED: Define the scope of the organization and assess
its goals.
Chapter 1
Step 2: STRATEGY: Review the organization’s strategy, and assess the
environment in which the organization operates.
Chapters 2 and 3
Step 3: STRUCTURE: Review how the organization is configured in terms of
its reporting relationships. Assess how a configuration operates across
time and space boundaries.
Chapters 4 and 5
Step 4: PROCESS AND PEOPLE: Review the organization’s work and how it
executes its tasks. Assess the organization’s tasks, people, leadership, and
climate.
Chapters 6, 7, and 8
Step 5: COORDINATION AND CONTROL: Assess the range of devices that
make up the organization’s infrastructure, including coordination,
control and information systems, as well as design of incentive systems.
Chapters 9 and 10
The approach is a holistic approach where you consider all fourteen design
components in a step-by-step manner. Research has shown that serious
performance losses are quite likely if a non-holistic approach is taken, that
is, if some design components are changed but not others such that the
organization is not in alignment (Burton et al., 2002).
Beginning with goals and strategy, environment, configuration etc., and
proceeding through to the design of incentives in Chapter 10, we developed
a complete set of typologies for organizational design components, as well as
fit and misfit statements, which were summarized in table 10.1. In table 10.1
there is fit within a column and misfit between columns.
In figure 11.1, we summarize these fit and misfit relations. There is fit
within each quadrant and misfit between the quadrants. This is an equiva-
lent statement to table 10.1. Note that there is a legend to the left
of quadrant A indicating the fourteen major design components in this
book.
To review quickly, there is fit within each quadrant A, B, C, and D. In
quadrant C, for example, there is fit for a prospector strategy in a locally
stormy environment with a divisional configuration, and a flat type of
organizational complexity with a bonus-based incentive structure. A similar
matching of components exists within each of quadrants A, B, and D. But
there is a misfit for the firm when one or more organizational design
205 Design dynamics: managing change and multi-organizations

Figure 11.1. The complete set of component types within the 2
2 organizational design space.

components fall in different quadrants. For example, a prospector strategy

in C is a misfit with calm environment in A. Because there are so many
design components, a very large number of misfits is possible. One way to
quickly assess the extent of misfits for your chosen firm is to circle the design
type you identified as you answered the diagnostic questions at the end of
each of the ten chapters. If all, or nearly all, of the firm’s design components
fall within the typologies of the same quadrant, then there are no or few
misfits, and change may not be necessary at this time (unless you anticipate
changes in the near future). On the other hand, if your firm has components
that fall across all four quadrants, then there is a very large number of
misfits, and you must give some serious thought as to how to redesign the
organization so as to reduce misfits and bring the design components into
alignment.
206 Burton, DeSanctis, and Obel

The design and change process is a holistic consideration involving many

possible changes. In the step-by-step approach we have developed an orderly
and systematic way to take on this complex managerial problem.

What should you change first? Make changes within each step

Beginning with step 1, choose goals of efficiency and/or effectiveness that

match what you, as the executive, want for the firm. Then, in step 2, be sure
that the strategy and environment fit the goal(s) selected in step 1. Next, in
step 3, the organizational configuration should match the strategy. In step 4,
the task design, the people, the leadership style, and climate should fit
together so as to facilitate the success of the strategy and structure. And
finally, the control, coordination, information, and incentive systems should
be designed so as to be compatible with the design of people and process
systems.
To summarize, the ideal approach is to fix any misfits within a step before
proceeding to the next step. At the end of each step check the consistency
with each of the previous steps. It may be necessary to make a number of
backward iterations to obtain a holistic solution. The step-by-step approach
is an analytical solution approach and not necessarily the best sequence
for implementation of a new design. Nonetheless, the step-by-step order
provides an approach to develop a managerial action plan.
Opportunity losses for misfits within a step can be quite large; and
further, changes within steps are relatively less difficult to make than be-
tween-step changes. Within step 2, for example, a misfit between the strategy
and environment can be devastating for your chosen firm. Consider the
prospector strategy, which, if used in a calm environment, incurs two losses.
First, the prospector will develop new products and services which are not
needed; and second, the prospector costs will be high. Within step 3,
consider a functional configuration in a multi-domestic firm. Conflict is
very likely between the functions and the country/regional locales, which
will lead to gross inefficiencies and lack of responsiveness to the local
conditions – potentially bringing significant losses for the firm. Within step
4, consider a leadership style of a manager who operates in a developmental
climate. The manager may well impose restrictions that will destroy
the benefits of the developmental climate (Burton et al., 2004). Within step
5, a bonus-based incentive system within a clan type of governance may
207 Design dynamics: managing change and multi-organizations

encourage costly, non-clan behavior. These are only a very few of the large
number of possible costly misfits that can occur within a step.

What should you change second? Make changes between steps

Now, turn to changes between steps. If your goals in step 1 are a misfit with
your strategy in step 2, the goals will most likely not be realized. For
example, adopting a defender strategy will not allow the organization to
meet its effectiveness goals. For steps 2 and 3, if the strategy does not fit with
the structure, then the structure will dominate and the desired strategy will
not be realized. This situation is a variation on Chandler’s (1962) famous
dictum, ‘‘structure follows strategy.’’ A functional configuration does not
support a prospector strategy. For example, consider steps 2 and 4: if you
have a defender strategy and adopt a leader-oriented leadership style, the
result may lead to conflict and losses as the leader will tend to let subunits
move away from executing current activities with machine-like discipline
in favor of spurring innovation. If you are able to fix misfits within a step
and between steps, then you can yield a comprehensive solution for the
organizational design of your firm.

Why change? Should we live with some misfits?

We know that any misfit has an opportunity loss in that performance suffers
(Burton et al., 2002). So at first glance it seems that all misfits should be
changed – immediately. If the misfits could be fixed without any cost of
change, then it would make sense to fix all misfits as quickly as possible. But
there are difficulties to changing organizational design components. The
fixing of misfits is not free and can involve considerable cost. Should we live
with some misfits? Yes, we can live with some misfits, and it is helpful to
weigh the benefits of fixing misfits against the costs and difficulties of
undertaking the change process. For example, suppose the leadership style
is a misfit with other components of organizational design, but to fire
existing senior management and recruit new executives will take time and
yield short-term morale problems that must be managed. An alternative
208 Burton, DeSanctis, and Obel

approach may be to live with the misfit for a while; work with senior
managers to create awareness of the leadership style problem. Work with
them to adjust their style to reduce the impact of the misfit. Replacing senior
management may be expensive and time-consuming, thus outweighing the
potential benefits.
Next, we want to examine the balance between the benefits and difficulties
of change.

What are the benefits of change?

The benefits of changing organizational design components to reduce or

eliminate misfits will help improve your firm’s performance and help you to
realize goals of efficiency and effectiveness. Misfits decrease a firm’s per-
formance. This is clear (Burton & Obel, 2004). There is a long tradition of
research demonstrating the opportunity losses from misfits (Gresov, 1989;
Miller, 1992; Naman & Slevin, 1993; Donaldson, 2001).
What is less clear is the exact nature of the opportunity-loss function, i.e.,
whether more misfits significantly decrease performance or not. In a study
of 222 Danish medium-sized firms, Burton et al. (2002) found that any
deviation from total fit, i.e., even one misfit, diminished performance,
but that additional misfits did not further diminish the loss significantly.
The implication is that all misfits should be fixed, if any are to be fixed. In
contrast, Håkonsson et al. (2005) focused on climate and leadership where
they found that additional misfits, beyond one misfit, diminished perform-
ance significantly. Donaldson (2001) argues that it is the degree of deviation
from the fit line that determines the effect on performance. The implication
here is that it makes good sense to fix some misfits even if you cannot fix
them all.
A further consideration is that some misfits are more problematic than
others. We know that, if a firm is in a turbulent environment, misfits have a
more significant effect on performance than in a calm environment (Obel,
1993; Håkonsson et al. 2005; Siggelkow & Rivkin, 2005). This suggests that if
you are in a turbulent environment it is more important that you take a
holistic approach (i.e., attend to as many of the 14 design components as
possible rather than just a few) to fixing misfits than if you are located in a
calm environment. In a calm environment you can live with more misfits
than in a volatile situation.
209 Design dynamics: managing change and multi-organizations

What are the difficulties of change?

There are two difficulties in making changes and fixing misfits. First, some
misfits are easier to fix than others – the ease of change. Second, fixing
misfits is not a simple linear procedure; namely, when you fix one misfit you
may create another one. We want to consider these two issues in turn.
Which misfit is the easiest to change? We asked our executive MBA
students who are practicing managers. In figure 11.2 below, their results
are summarized. Formalization of the organization is the easiest to change;
and the environment is the most difficult. Formalization reflects the firm’s
rules; it is relatively easy to add more rules or change a rule. The next thing
is to get compliance with the rules, which may be more difficult and is
related to the design of the incentives. The environment is the most difficult
to change; in practical terms, it means new markets, products, or services
with different levels of uncertainty and environmental complexity. Your
firm might have to be very different in its strategy and structure and
perhaps its people to move to a new environment. This is likely a very
difficult change process. These difficulties are consistent with the step-by-
step approach. The environment is difficult to change, so for step 2, the
strategy can be adjusted to it. Then in step 3, the complexity, geographic
distribution, and knowledge sharing can be changed and matched with the

Figure 11.2. The difficulty of change (Modified from Burton & Obel, 2004, p. 406).
210 Burton, DeSanctis, and Obel

strategy. The components of step 4, task design, people, leadership, and

climate tend to be very difficult to change and can be addressed following
the other changes.
The ratings of difficulty of change shown in figure 11.2 are suggestive for
your firm but not definitive. You must develop a similar chart for your own
firm. Your chart depends upon your particular situation and its challenges.
For example, there may be insurmountable political barriers to your chan-
ging the strategy or task design in your firm. If the top management owns
the firm, changing the leadership style may not be an option at all. For your
firm, fill in numbers from 1 to 5 for each design component listed in
table 11.1. For each component use a 1 to 5 rating scale to score your
organization as follows:
1 2 3 4 5
very easy to change moderate very difficult to change
For each component, you can then indicate more specifically what your firm
would have to do to make desired changes.
Knowing what misfits your firm has, you can now balance them against
the difficulty of making any changes, and what to do using the step-by-step
approach as a guide.

Table 11.1. My firm’s difficulty of change

Difficulty of change for your

Design component firm (1¼low, 5¼high)

Goals
Strategy
Environment
Configuration
Organizational complexity
Geographic distribution
Knowledge exchange
Task design
People
Leadership style
Organizational climate
Coordination and control systems
Information systems
Incentives
211 Design dynamics: managing change and multi-organizations

Figure 11.3. The misfit possibilities curve (Modified from Burton & Obel, 2004, p. 402).

Unfortunately, change is not a simple process whereby fixing one misfit

reduces the total number of misfits. A fixed misfit may create other misfits.
Examine figure 11.1 above. Let us assume you have a total fit in quadrant B.
Now, let us assume a new senior executive is put in place who adopts a
leader style rather than a manager style. With this one change in leadership
style, you now have even more misfits as there is a misfit with all of the other
characteristics, i.e., leadership is now in quadrant C and everything else is in
quadrant B. Then assume the strategy is changed to a prospector to fit the
new leader; now you have those two in fit, but each of those two is in misfit
with everything else, resulting in more total misfits.
In figure 11.3 above, we map out the implications where we begin with
total fit of design components on the left side and work through the possible
number of misfits as they are fixed one by one. First, there are more misfits
as suggested above, then eventually the number decreases as the firm is
brought back into total fit.
The implications of this discussion for managing change are enormous.
First, just fixing one misfit may not make the situation notably better. It
probably requires fixing at least a few so that you stay near the bottom of the
curve in figure 11.3. Otherwise, if you only fix one, you may well make
things worse. Second, if you are thinking of fixing one misfit, you should
map out a total plan of a sequence of fixes – a holistic approach. You may
find yourself in a rather unplanned drift of fix and indication of new misfits
which require fixing. A myopic fix approach will not work well. As suggested
earlier, fixing the misfits within a step is not sufficient; the fixing of misfits
between steps is also needed.
212 Burton, DeSanctis, and Obel

It is a difficult balance of identifying misfits, choosing which ones to fix,

and in what order, and balancing the benefits of fewer misfits with the
difficulties of changing design components. Taking a holistic and systematic
approach to organizational design is vital.

Managing misfits over time

Misfits arise from three sources: external to the organization, internally from
managerial action, and jointly from a combination of external and internal
sources. Table 11.2 provides illustrations of these types of misfits. Internal
sources of misfits are on the left, external ones on the right, and joint ones in
the middle. Generally, the environment evolves on its own with very little
effect from managerial action. The competition can introduce new products
and services. The government can impose new regulations or form new
trade agreements. The economy can spur business activity or stifle it.
Political conditions can create turbulence or promote calm. On the other
hand, internal changes are largely under managerial control. You can de-
velop a new strategy – step 2. You can change senior management to fit a
desired leadership style and you can redesign how tasks are done – step 4,
but this may be more difficult. You can adopt a new IT system or a new
incentive system, and so on – step 5. In the mid-range are alternatives that
are partially under managerial control but not fully. For example, the
organizational climate may evolve without your intention, although man-
agement certainly can influence it – step 4. You can change the structure of
incentives, but how employees view them may not be something you can
fully anticipate – step 5. You can specify an approach for knowledge ex-
change – step 3, but things may shift over time if information that once was
tacit should become more explicit (e.g., subtle knowledge about customers
may become more explicit if it is documented and tracked, moving out of
the heads of a few and into an accessible format for many to use), possibly
moving your organization’s knowledge system from one type to another.
Since managers have more control over managing misfits that arise from
internal sources, let us examine these in more detail. You have the best
possibility for managing organizational design changes with success when
you attend to those components that are under your control.
You have many ways to potentially change your organization and thus
to create or fix misfits. If your firm is in reasonably good alignment but
213 Design dynamics: managing change and multi-organizations

Table 11.2. Examples of internal, external, and combined sources of organizational design misfits (Modified from
Burton & Obel, 2004, p. 395)

External: Misfits created

Internal: Misfits created through from environment
managerial action Combined Internal & External sources

• A change in organizational • Management tries to improve • Competitors introduce new,

strategy the organizational climate, unanticipated products or
• A divisional structure is but despite efforts to increase services, so that the
consolidated into a matrix trust and morale, tension environment is no longer
structure remains high due to calm
• Separate multi-domestic units uncertainty in the industry • The government imposes new
are consolidated into an • The incentive system in an regulations requiring more
international design international locale is changed stringent control systems
• Promotion or hiring brings to bonus-based, but • The government forms new
about a new leadership style employees perceive the new trade agreements that open up
• A large number of employees system to be unfair given the new market opportunities
are laid off, changing the nature of their local culture • Political conditions become
people dimension • Knowledge exchange systems highly turbulent and
• Employees neglect continued designed on a people-driven unusually difficult to forecast
skill development, thus model become obsolete as
decreasing their customers presume systems
professionalization over time are relationship-based
• Formalization is increased to
move from a family
governance approach to a
more disciplined machine
approach
• The organization merges with
a firm that is quite different in
structure, process, and people

yet not achieving its long-term potential and anticipates that it will not
perform well in the future with its current fit, then you can and should
create short-term misfits for a better long-term performance. You are an-
ticipating that the current, good situation will create opportunity losses in
the future, even if you have a current fit. Thus, you should act now to create
misfits. You have a choice which will create a misfit today, but it may
make good sense for the longer run. Examples of changes that you might
make in organizational design components include the following within
the step-by-step approach:
 a new strategy for the firm,
 a new product or new product line,
 a new way of designing work tasks to satisfy new customer needs,
 a new information system,
214 Burton, DeSanctis, and Obel

 a different approach to people management and incentive system design,

 hiring or promoting a new top management team with a different
leadership style,
 a new approach to optimal sourcing,
 greater reliance on virtualization for knowledge exchange,
 a major expansion or growth,
 a merger with another firm,
 a strategic alliance or partnership with another firm, and
 generally, the introduction of continuing change.
Any one of the above may create a misfit now. Once it has been created,
then you should work to bring the organization back into fit as we have
discussed throughout this book. If the misfit persists, then there are signifi-
cant opportunity losses to incur. Most of the situations listed above we have
discussed earlier in the book. Later in this chapter, we will analyze the multi-
organization situation of a joint venture, merger, and strategic alliance/
partnership.
Can you do anything to manage external sources of misfits? External
shocks can be gradual, evolving, and predictable. You may have time to
anticipate and adjust, and fix the misfits in a timely manner. But, external
shocks can be large, sudden, and unpredicted. The latter situation has been
called a punctuated equilibrium by some scholars (Tushman & Romanelli,
1985). A period of environmental predictability is suddenly disrupted in a
way that managers could not detect in advance; or, even if the shock is
detected, the organization is not able to react in sufficient time. In such a
setting, the environment optimizes – not the firm, and even analyzer firms
may find themselves in defender mode. So what can you do when the
environment changes very suddenly and rapidly? One approach is time-
paced innovative action (Eisenhardt & Brown, 1998). It is a strategy of
period-based innovation or change. The organization engages in redesign
on a continual (time-determined) schedule, so that it becomes nimble and
adept at reorganizing (experiencing misfits and adjusting for them). Intel
builds new factories in a clockwork fashion even before they know what will
be housed there. Gillette introduces a new razor regularly. Some public
European companies replace their executive leaders as a function of their
age, not their performance (i.e., forced retirement); the organization wants
to move on and try a fresh leadership style, even if things may be proceeding
smoothly. Carlsberg has a new Christmas beer each year. Some high-tech
start-ups reorganize every six months to keep workers skill adept and ready
for change. Even universities begin classes each year according to a fixed
215 Design dynamics: managing change and multi-organizations

calendar. In brief, you are in command of what you do without waiting for
events to dictate your actions. You create your own environment and
situation (to some extent) rather than waiting. Using time-paced innov-
ation, there is the risk that you can become very disconnected with the
environment, leaving it behind while the organization changes, which can
lead to poor performance. On the other hand, you are more likely to
maintain internal fit for your firm despite occasional disruptions, which
has certain advantages.

Multi-organization: multiple units of analysis

Thus far, we have considered a single unit of analysis: a firm, a division, a

department, or a team. Now, we examine multiple units of analysis, such as
the joint venture, the merger, and the strategic alliance or partnership. These
organizational designs have come to the forefront of interest today as firms
are continually involved in forming relationships with other firms. At first
glance, we might expect that there is something new or different about how
these multi-organizations should be designed when compared to their
simpler single organization counterparts. But this is not the case. The basics
of organizational design still hold and should be applied when managing
multi-organizations. Why? First, multi-organizations are not a new phe-
nomenon. Firms have been engaged in these activities for many years. We
know from experience that the majority of multi-organizational initiatives
fail; yet, the firms that are able to manage the change process associated with
joint ventures, mergers, and alliances to build an integrated, aligned organ-
ization are more likely to be successful (compared to those that fail to
integrate and align a newly formed multi-organization) (Carey et al.,
2004). During the twentieth century, General Motors, IBM, and Cisco grew,
in part, as the result of their success in integrating acquired firms into
their companies. Many firms have been badly hurt by their inability to inte-
grate acquisitions or work in alignment with partners. HP’s failure to
integrate Compaq into its larger enterprise provides a recent example of a
troublesome multi-organization experience.
Second and perhaps more importantly, these multi-organizations are
organizations in their own right, where each has its own strategy, environ-
ment, configuration, distributed organization, people, leadership, climate,
coordination, control, information systems, and incentives. That is, the
216 Burton, DeSanctis, and Obel

Figure 11.4. The multi-organization.

fundamentals that we have developed throughout this book apply here as

well. The unit of analysis is different, but the step-by-step approach applies.
You can analyze a joint venture, merger, or alliance in the same way as you
do a single organization. You can also examine misfits and plan to change
the organization. The tools are the same. The complication is that you
must consider more than one unit of analysis when designing these
ventures.
The joint venture, the merger, the strategic alliance or partnership – each
one is its own unit of analysis, but each is closely related to its parent
organizations in ways that are not present in stand-alone organizations.
Here, we have three organizations (units of analysis) to consider in concert:
the organization itself and its two parents. In figure 11.4 we show the three
organizations. The two parents, A and B, come together to form C, the focal
organization, which may take the form of a joint venture, merger, alliance,
or partnership. When considering a new organization of this type, you want
to examine all three organizations, and the relationships of the parents to the
new organization, C.
Sometimes A and B form C to be similar in design to the two parents –
similar in goals, strategy, people, leadership, and so on. But many times
A and B form C to create something quite different from themselves. A
and B may think of themselves as having complementarities, or different
(yet compatible) design components, rather than similarities in design
217 Design dynamics: managing change and multi-organizations

components. For example, A may have market access to a new region but
not the required product know-how, which B holds, that is needed to
produce the product for the region. A and B thus bring complementary
capabilities to the formation of the C organization. Complementarities may
make great strategic sense as a reason for creating C, but these also may re-
flect design differences that yield misfits. If so, then the misfits must be
identified and managed. Many misfits put the success of C at risk. Indeed,
many multi-organizations fail precisely because strategists have embraced
complementarities between A and B but overlooked the downsides of
organizational design misfits that carried into the C organization.
Let us develop the step-by-step approach for the multi-organization,
building directly upon the step-by-step approach for the single organization.
Then we will apply this approach to each of the multi-organizational types:
the joint venture, the merger, and the alliance/partnership. We analyze C
as an organization even if it is not a separate legal entity, which is the case
in some alliances and partnerships. Even if C is not a separate, independent
organization, the partnership (or whatever the form of multi-organization)
has goals, people responsible for leading it, systems for coordination and
control, etc. It has to be created with organizational design components in
mind so that it can be managed to meet its goals. So think of C as an organi-
zation in its own right, no matter how big or small, or how independent or
intertwined it may be with its parents.
To begin, you want to first examine the focal organization, C. Next,
examine the parents, A and B. (If there are more than two parent organi-
zations, then examine all of them, especially those that are most influential
in shaping the design of the new, focal organization.) Finally, examine
the relationship between C and its parent organizations, A and B. Our
thesis is that C is better understood when we examine it together with
A and B.

The multi-organization step-by-step approach

To analyze a multi-organization, here are the steps you should follow:

1. Analyze C as a separate organization applying our single organization
step-by-step approach. Remember that C is an organization with its own
goals, strategies, environment, and so on (which includes A and B in the
environment; this is very important). Here, you want to examine the
218 Burton, DeSanctis, and Obel

misfits and possible changes for bringing C’s design components into
alignment.
2. Analyze A separately using our step-by-step approach. Again, the
emphasis is on the misfits within A and possible changes.
3. Analyze B separately using our step-by-step approach. Again, the
emphasis is on the misfits within B and possible changes.
In the above, it is important to examine A, B, and C as separate organiza-
tions, even if they share some personnel, goals, coordination and control
systems, or other design components. For purposes of analysis and design,
each one is its own unit of analysis and has its own organizational boundary.
Analyzing each as a separate unit of analysis will allow you to see points of fit
and misfit and identify possibilities for change in organizational design,
especially for C.
The next step is to examine the relationships between A and C, and then
B and C.
1. Examine the relations between A and C. Here, you want to analyze and
understand:
 What goals does A have for C? What goals does C have for A?
 What results, outcomes, and outputs does A expect from C? And,
vice versa?
 What resources – leadership, personnel, and financing – does A give
to C?
 What policies and limitations does A place upon C?
 What agreements and contracts does C have with A?
2. Repeat step 1 to examine the relations between B and C as above.
The next step is to do a higher level of misfit analysis and possible change.
1. Take the analyses above for A and C.
 Make two lists side by side – one for A and one for C. Then compare
A and C on each of the dimensions from your analysis: goals,
strategy, environment, etc. Many of these entries will be different,
perhaps almost all. But focus on the different ones that will impede C
in realizing its goals; these are significant misfits.
 Develop a list of changes in the relationship between A and C that
will ameliorate these misfits and facilitate C in the accomplishment
of its goals.
2. Repeat step 1 for B and C.
3. Do a comparison of A and C with B and C. If these relationships are
quite different, then there is likely to be conflict between A and B, which
will make it difficult for C as it pursues its work. C then needs to address
219 Design dynamics: managing change and multi-organizations

these issues and make a determination of whether to live with them or

work with A and B for a solution.
Next we want to review how the step-by-step approach just described
can be applied to three types of multi-organizations: the joint venture, the
merger, and the alliance/partnership.

Joint venture

The joint venture C is an organization that is given resources and supported

by its parents, A and B, but C is largely independent of A and B. The first
step is to do separate analyses of A, B, and C; and then turn to their
relations.
As an example, consider a bio-tech start-up that is a joint venture of two
larger pharmaceutical companies. You should begin with a single organiza-
tion step-by-step approach for the joint venture C; then do the same for the
parents, A and B. Next analyze the relationship between C and its parent
organizations. Suppose the goal for C is to develop new drug-assessment
algorithms which are to be used by both of its parents, A and B, but in
different ways. Suppose that A has assigned six professional algorithm devel-
opers to the joint venture and also the CEO. In addition, A has made available
substantial funding. In return, it wants access to any of the assessment
algorithms developed by C for its application in its entity drug development.
This agreement is formalized in a very detailed contract, particularly the
intellectual property (IP) rights. Suppose B, on the other hand, has a different
relationship and goals with C. B has supplied a large number of highly trained
but less experienced scientists, and it has provided the space, equipment, and
support personnel for the joint venture. B hopes to learn how to develop
assessment algorithms for its own entities but will not use the specific
algorithms developed by C. That is, it does not have IP rights in the joint
venture. In sum, A and B have differing relationships with C, and these are
formalized in a detailed contract at the time of creation of the venture.
This is a quick sketch of the relations between the joint venture C and its
parents, A and B. The joint venture C should plan its organizational design
and assess its misfits as the venture evolves. C should be assessed as an
independent unit of analysis. Then it should look at the goals that A and B
have for the joint venture and the resources they are supplying. Can C realize
its goals? Are there sufficient resources? If there are misfits here, then C, the
220 Burton, DeSanctis, and Obel

new development laboratory, must decide to live with them, or approach the
parents to obtain a resolution.

Merger

The merger is the result of A and B becoming a new organization, C. At first

glance, you might think that A and B will go away as C is formed and can be
ignored. But the shadows of A and B can be long. C cannot be understood
(alone) without an assessment of where it came from. The people, climate,
coordination, and control systems, and many other organizational design
components of A and B will be imported into C and will influence C’s
organizational design in significant ways. For this reason, an analysis of the
C organization prior to its formation, and an assessment of the parents’
influence on C, can be very helpful.
GlaxoSmithKline (GSK) is a recent merger of Glaxo Wellcome and
SmithKline Beecham. The new GSK is made up of people, facilities, drugs,
brands, etc., from the parents. In the step-by-step approach, you would
examine GSK – the C organization. You also would assess Glaxo Wellcome
and SmithKline Beecham – the parent organizations. As the merger occurs,
the parent organizations are becoming shadow or legacy organizations, but
they greatly affect the new GSK.
To assess the design of the new, merged organization, revealing possible
misfits and ways to change them, you should first do separate step-by-step
analyses of each of the three organizations, GSK, Glaxo Wellcome, as it was
prior to the merger, and SmithKline Beecham, as it was prior to the merger.
Then turn to differences that will create misfits after the merger. Our
suggested approach is as follows:
1. Make two lists side by side – one for GSK and one for Glaxo Wellcome.
Then compare GSK and Glaxo Wellcome on each of the organizational
design components: goals, strategy, environment, etc. Some of these
entries will be similar, and some will be different.
2. Focus on the differences, as these will create variation in the particular
dimension for the merged organization. Assess if these differences are
important and how they should be handled. Develop a list of changes
in the relationship between GSK and Glaxo Wellcome that will
facilitate reduction of misfits, thus enhancing alignment in the new
organization.
221 Design dynamics: managing change and multi-organizations

3. Look at the misfits in the new potential organization. Consider how they
should be addressed. Here you should consider how to yield the ‘‘best’’ of
the two parent organizations.
The normal financial approach to merger analysis is the opposite: antici-
pate the benefits of a merged company which has inferred synergies or cost
savings. But organizational misfits between the parent organizations in a
merger are very likely to create higher-level misfits in the new, merged
organization, C, which can linger for some time, and not yield the hoped-
for results. Failed mergers which do not realize the promised synergies are
usually those which spawn misfits when incompatible organizations, A and
B, become merged organizations without full organization analysis, identi-
fication of misfits, and managerial steps to fix misfits. Any one of the factors
of climate, leadership style, people, incentives, and IT systems is sufficient to
derail the benefits of a merger.

Strategic alliance or partnership

A strategic alliance, or partnership, is planned collaboration among multiple

firms for conducting business activity together for purposes of mutual
benefit. Alliances are contract-driven and arise in many different forms:
suppliers with manufacturers, sharing of IP in research and development,
joint product development, and integrated marketing efforts, among others.
Although the C organization formed by A and B is typically not a stand-
alone organization, a strategic alliance, once formed, can be considered as an
organization in its own right in that it has its own goals, strategies, leader-
ship style, etc. Usually, the goals for C are spelled out in the contract
agreements between the parents. Many have a joint oversight and policy
committee made up of members of the parent organizations. In addition,
the leadership (e.g., top management team) of the alliance is typically shared
by representatives from each parent company. C is, in a sense, an implicit or
fathom organization with very few resources of its own and little autonomy
from A and B. The assignment of people, leaders, infrastructure, and other
organizational components of C are drawn from the parent organizations
and typically remain there as the partnership operates. In this sense, an
alliance is the opposite of a merger.
As in the case of a merger, planning for the organizational design of an
alliance is vital. In addition to stating the goals in the formal alliance
222 Burton, DeSanctis, and Obel

agreement, the parent organizations, A and B, should also plan on defining

the strategy, environment, configuration, and all other components of
organizational design. A complete design is important to alliance success.
One way to do this is to follow our step-by-step approach, identifying the
organizational design components for C and their relationship to the parent
organizations. If you are creating an alliance, you need to have a very good
understanding of the strategic alliance from an information-processing
view: who does what, how, and based upon what information.
As an example, consider a strategic alliance between an automobile
manufacturer and one of its suppliers. Saab and others have many such
alliances. They begin with a reasonably well-understood agreement and/or
formal contract between the two parents. The goal of the strategic alliance as
its own organization is to develop and manufacture parts, modules, or
subassemblies which perform the automotive function, but more import-
antly are integral to the automobile. GKN, a British conglomerate, is another
example of such an alliance. GKN makes drive shafts for many companies
and each one must be engineered for that specific application. The strategic
alliance is the development process in which the drive shaft is adapted to the
automobile, but the automobile including modules from other suppliers will
adapt to the drive shaft – all of which are yet to be built. The information-
processing demands are very high with many reciprocal adjustments where
decisions are made jointly by the parent organizations.
When forming a strategic alliance, we suggest that you begin with an
analysis of the anticipated C organization, following the step-by-step ap-
proach as outlined in this book. Next do the same for A and B, and any other
parents that are a part of the alliance. Then, move on to the higher-level
misfits between the strategic alliance and the parent organizations. The
analysis and resolution of the higher-level misfits must be addressed for a
successful strategic alliance or partnership.
To summarize, we have presented a step-by-step approach for the multi-
organization. We then examined three kinds of multi-organizations: the
joint venture, the merger, and the strategic alliance/partnership. Each in-
volves at least three organizations: two or more parent organizations and the
focal organization. The joint venture, the merger, and the alliance can be
quite different from each other in their legal and contractual nature; how-
ever, from an organizational design perspective, their components are the
same as any organization. Thus, they should be analyzed using a single
organization step-by-step approach, following the approach outlined in
Chapters 1–10 of this book. Relationship analysis, as described above, can
223 Design dynamics: managing change and multi-organizations

then be used to identify higher-level misfits that will reveal incompatibilities

between the C organization and its parents. From there you plan for needed
changes to increase alignment between C and its parents, A and B, thus
enhancing the information capacity of C and improving its potential to
meet its goals.

Summary

In this final chapter, we have reviewed the step-by-step approach and then
extended it in terms of the difficulty of change for your organization.
Examining and fixing the misfits within a step and then moving on to the
misfits between steps is the recommended approach to achieving a holistic
organizational design. The step-by-step approach brings a disciplined and
comprehensive approach to the complex process of organizational design.
We showed how you can think through the costs and benefits of implement-
ing design changes in your organization. Finally, we demonstrated how
multi-organizations, such as joint ventures, mergers, and strategic alliances,
can be understood within the step-by-step approach.
Now you have completed the step-by-step approach for an organizational
diagnosis and design. You have completed a guided tour of your own
organization and achieved a better understanding of the design challenges
for better goal attainment.

Glossary

Complementarities: different (yet compatible) design components, rather

than similarities in design components between two parent organizations;
often used as a motivation for two organizations forming a multi-
organization.
Joint venture: the forming of a new organization by two or more
independent organizations; the organization typically remains independ-
ent of the parents, except for the resources that they bring together in the
joint venture.
Merger: the forming of a new organization by combining two or more
independent organizations together.
224 Burton, DeSanctis, and Obel

Multi-organization: the formation of a new organization, C, by multiple

parent organizations, A and B; the multi-organizational forms are joint
ventures, mergers, and alliances or partnerships.
Strategic alliance: planned collaboration among multiple firms for
conducting business activity together for purposes of mutual benefit;
typically takes the form of a contract among the collaborating firms;
sometimes referred to as a partnership.
Strategic partnership: planned collaboration among multiple firms for
conducting business activity together for purposes of mutual benefit;
typically takes the form of a contract among the collaborating firms;
sometimes referred to as an alliance.
 References

Arrow, Kenneth (1974): The Limits of Organization, New York: W. W. Norton.

Ashby, W. R. (1956): Introduction to Cybernetics, London: Methuen.
Bartlett, C. A., & Ghoshal, S. (1998): Managing Across Borders: The Transnational Solution, 2nd edn,
Harvard Business School Press.
Bartol Kathryn M. & Locke, Edwin (2000): ‘‘Incentives and Motivation,’’ in Sara L. Rynes and Barry
Gerhart (eds.), Compensation in Organizations: Current Research and Practice, San Francisco: Jossey-
Bass, pp. 104–50.
Boudreau, M.-C., Loch, K. D., Robey, D., & Straub, D. (1998): ‘‘Going Global: Using Information
Technology to Advance the Competitiveness of the Virtual Transnational Organization,’’ Academy of
Management Executive 12(4), pp. 120–8.
Bourgeois, L. J., III (1980): ‘‘Strategy and Environment: A Conceptual Integration,’’ Academy of
Management Review, 5, pp. 25–39.
Bourgeois, L. J., & Eisenhardt K. (1988): ‘‘Strategic Decision Processes in High Velocity Environments:
Four Cases in the Minicomputer Industry,’’ Management Science, 34, pp. 816–35.
Bower, J. L. (2001): Merloni Elettrodomestici: Building for a New Century (HBA 9–301–112), Cambridge,
MA: Trustees of Harvard College.
Burns, T., & Stalker, G. M. (1961): The Management of Innovation, London: Tavistock.
Burton, R. M., & Obel, B. (1984): Designing Efficient Organizations: Modelling and Experimentation,
Amsterdam: North Holland.
(1988): ‘‘Opportunism, Incentives, and the M-form Hypothesis: A Laboratory Study,’’ Journal of
Economic Behavior and Organization, 10, pp. 99–119.
(2004): Strategic Organizational Diagnosis and Design: The Dynamics of Fit, Dordrecht: Kluwer
Academic Publishers.
Burton, R. M., Lauridsen, J., & Obel, B. (2002): ‘‘Return on Assets Loss from Situational and Contingency
Misfits,’’ Management Science, 48 (11), pp. 1461–85.
(2004): ‘‘The Impact of Organizational Climate and Strategic Fit on Firm Performance,’’ Human
Resource Management, 43(1), pp. 67–82.
Carey, D. C., Ogden, D., & Roland J. A. (2004): The Human Side of M & A: Leveraging the Most Important
Factor in Deal Making, Oxford: Oxford University Press.
Carroll, T. N., Burton, R. M., Levitt, R. E., & Kiviniemi, A. (2005): ‘‘Fallacies of Fast Track Heuristics:
Implications for Organization Theory and Project Management,’’ submitted to Management Science.
Chandler, A. D. (1962): Strategy and Structure: Chapters in the History of the Industrial Enterprises,
Cambridge, MA: MIT Press.
Choo, C. (1998): The Knowing Organization, New York: Oxford University Press.

225
226 References

Cullen, J. B. (2002): Multinational Management: A Strategic Approach, New York: Southwestern-Thompson

Learning.
Cyert, R. M., & March, J. G. (1963): A Behavioral Theory of the Firm, Englewood Cliffs, NJ: Prentice Hall.
Davidow, W. H., & Malone, M. S. (1992): The Virtual Corporation, New York: Edward Burlingame Books/
Harper Business.
Denison, D. R. (1996): ‘‘What is the Difference between Organizational Culture and Organizational
Climate? A Native’s Point of View on a Decade of Paradigm Wars,’’ Academy of Management Review,
21(3), pp. 619–54.
DeSanctis, G., & Fulk, J. (eds.) (1999): Shaping Organization Form: Communication, Connection, and
Community, Newbury Park, CA: Sage.
Donaldson, L. (2001): The Contingency Theory of Organizations, Thousand Oaks, CA: Sage.
Doty, D., Glick, H. W. H., & Huber, G. P. (1993): ‘‘Fit, Equifinality, and Organizational Effectiveness: A
Test of Two Configurational Theories,’’ Academy of Management Journal, 38(6), pp. 1196–250.
Duncan, R. B. (1972): ‘‘Characteristics of Organizational Environments and Perceived Environmental
Uncertainty,’’ Administrative Science Quarterly, 17(3), pp. 313–27.
Eisenhardt, K. M., & Brown, S. L. (1998): ‘‘Time Pacing: Competing in Markets that Won’t Stand Still,’’
Harvard Business Review, 76(2), pp. 59–69.
Fenton, E. M., & Pettigrew A. M. (2000): ‘‘Theoretical Perspectives on New Forms of Organizing,’’ in
Andrew M. Pettigrew and Evelyn M. Fenton (eds.), The Innovating Organization, London: Sage.
Galbraith, Jay R. (1973): Designing Complex Organizations, Reading, MA: Addison-Wesley.
(1974): ‘‘Organization Design: An Information Processing View,’’ Interfaces, 4(3), pp. 28–36.
Gibbons, R. (1998): ‘‘Incentives in Organizations’’, Journal of Economic Perspectives, 12(4), pp. 115–32.
Gibson, C. B., & Birkinshaw, J. (2004): ‘‘The Antecedents, Consequences, and Mediating Role of
Organizational Ambidexterity,’’ Academy of Management Journal, 47(2), pp. 2009–226.
Gittell, J. H. (2003): The Southwest Airlines Way, New York: McGraw-Hill.
Gresov, C. (1989): ‘‘Exploring Fit and Misfit with Multiple Contingencies,’’ Administrative Science
Quarterly, 34(3), pp. 431–54.
Gresov, C., & Drazin, R. (1997): ‘‘Equifinality: Functional Equivalence in Organization Design,’’ Academy
of Management Review, 22(2), pp. 403–28.
Hahn, G. J. (1999): ‘‘The Impact of Six Sigma Improvement – A Glimpse into the Future of Statistics,’’ The
American Statistician, 53(3), pp. 208–16.
Håkonsson, D. D., Lauridsen, J., Obel, B., & Burton, R. (2005): ‘‘How Misfits between Strategy and
Leadership affect Organizational Performance,’’ presented at the AOM Conference, Hawaii
(submitted for publication).
Hambrick, D. C. (2003): ‘‘On the Staying Power of Defenders, Analyzers, and Prospectors,’’ Academy of
Management Executive, 17(4), pp. 115–19.
Hansen, M. T., & Nohria, N. Fall (2004): ‘‘How to Build Collaborative Advantage,’’ MIT Sloan
Management Review, 46(1,) pp. 22–30.
Heneman, R. L., Fisher, M. M., & Dixon, K. E. (2001): ‘‘Reward and Organizational Systems Alignment:
An Expert System,’’ Compensation and Benefits Review, 33(6), pp. 18–29.
Hennart, J. F. (1993): ‘‘Explaining the Swollen Middle: Why Most Transactions Are a Mix of Market and
Hierarchy,’’ Organization Science, 4, pp. 529–47.
Heydebrand, W. V. (1989): ‘‘New Organizational Forms,’’ Work and Occupations, 16(3), pp. 323–57.
INSEAD (1994): ‘‘The 3M Company: Integrating Europe (A) and (B),’’ ECCH case collection 494–023–1
and 494–023–2.
227 References

Keen, Peter G. W., & McDonald, M. (2000): The Eprocess Edge, Berkeley, CA: McGraw-Hill.
Kerr, S. (1975): ‘‘On the Folly of Rewarding A while Hoping B,’’ Academy of Management Journal, 18(4),
pp. 769–83.
Kotter, J. P. (1988): The Leadership Factor, New York: Free Press.
Kottler, P. (2000): Marketing Management: The Millennium Edition, Upper Saddle River, NJ: Prentice Hall.
Kowtha, N. R. (1997): ‘‘Skills, Incentives, and Control: An Integration of Agency and Transaction
Cost Approaches,’’ Group and Organization Management, 22(1), pp. 53–86.
Lawrence, P. R. (1981): ‘‘Organization and Environment Perspective,’’ in A. H. Van de Ven and W. F. Joyce
(eds.), Perspectives on Organization Design and Behavior, New York: Wiley, pp. 311–27.
Lawrence, P. R., & Lorsch, J. W. (1967): Organization and Environment, Boston: Harvard Business Press.
Likert, R. (1967): The Human Organizations, New York: McGraw-Hill.
Makadok, R. (2001): ‘‘Toward a Synthesis of the Resource-based and Dynamic-capability View of Rent
Creation,’’ Strategic Management Journal, 22, pp. 387–401.
March, J. G. (1991): ‘‘Exploration and Exploitation in Organizational Learning,’’ Organization Science, 2,
pp. 71–87.
March, J. G., & Simon, H. A. (1991): Organizations, New York: John Wiley & Sons.
McCall, M. W. Jr., & Hollenbeck, G. P. (2002): Developing Global Executives, Cambridge, MA: Harvard
Business School Press.
McGregor, D. (1969): The Human Side of Enterprise, New York: McGraw-Hill.
Milakovich, M. E., & Gordon, G. J. (2001): Public Administration in America, Thorofare, NJ: Thompson
Learning Incorporated.
Miles, R. E., & Snow, C. C. (1978): Organizational Strategy, Structure, and Process, New York: McGraw-
Hill.
(1986): ‘‘Network Organizations: New Concepts for New Forms,’’ California Management Review, 28,
pp. 62–75.
Miles, R., Snow, C. C., Mathews, J. A., Miles, G., & Coleman, Jr., H. H. (1997): ‘‘Organizing in
the Knowledge Age: Anticipating the Cellular Form,’’ Academy of Management Executive, 11(4),
pp. 7–20.
Miller, D. (1992): ‘‘Environmental Fit Versus Internal Fit,’’ Organizational Science, 3(2), pp. 159–78.
Mintzberg, H. (1983): Structures in Fives, Englewood Cliffs, NJ: Prentice Hall.
Nagar, Venky (2002): ‘‘Delegation and Incentive Compensation,’’ The Accounting Review, 77(2),
pp. 379–95.
Naman, J. L., & Slevin, D. P. (1993): ‘‘Entrepreneurship and the Concept of Fit: A Model and Empirical
Tests,’’ Strategic Management Journal, 14(2), pp. 137–53.
Nokia (2003): ‘‘ Fostering Innovation, ’’ ECCH case collection 403–040–1.
Nonaka, I., & Takeuchi, H. (1995): The Knowledge-Creating Company: How Japanese Companies Create the
Dynamics of Innovation, New York: Oxford University Press.
Obel, B. (1993): ‘‘Strategi og Ledelse – Er der en sammenhæng?’’ in Steen Hildebrandt (ed.), Strategi og
Ledelse, Veje og Visioner mod år 2000, Herning: Systime, pp. 396–408.
Ouchi, W. G. (1980): ‘‘Markets, Bureaucracies, and Clans,’’ Administrative Science Quarterly, 25,
pp. 129–41.
PHP Institute (1994): Matsushita Konosuke: His Life and His Legacy, Tokyo: PHP Institute.
Polanyi, M. (1996): The Tacit Dimension, London: Routledge & Kegan Paul.
Porter, M. E. (1985): Competitive Advantage: Creating and Sustaining Superior Performance, New York: The
Free Press.
228 References

Powley, E. H., Fry, R. E., Barrett, F. J., & Bright, D. S. (2004): ‘‘Dialogic Democracy Meets Command and
Control: Transformation through the Appreciative Inquiry Summit,’’ Academy of Management
Executive, 18(3), pp. 67–81.
Quinn, R. E., & Kimberly, J. R. (1984): ‘‘Paradox, Planning, and Perseverance: Guidelines for Managerial
Practice,’’ in J. R. Kimberly, & R. E. Quinn, (eds.), Managing Organizational Transitions, New York:
Dow-Jones-Irwin, pp. 295–314.
Richard, M. (2001): ‘‘Toward a Synthesis of the Resource-based and Dynamic-capability Views of Rent
Creation,’’ Strategic Management Journal, 22(5), pp. 387–401.
Rivkin, J., & Fleming, L. (2004): ‘‘Complexity, Networks, and Knowledge Flow,’’ Research Policy, 33(10),
pp. 1615–34.
Robbins, S. P. (1990): Organization Theory: Structure, Design and Applications, Englewood Cliffs, NJ:
Prentice-Hall.
Roberts, J. (2004): The Modern Firm: Organizational Design for Performance and Growth, New York:
Oxford University Press.
Rynes, S. L., & Gerhart, B. (2000): Compensation in Organizations: Current Research and Practice, San
Francisco: Jossey-Bass.
Scott, W. R. (1998): Organizations, Rational, Natural and Open Systems, Englewood Cliffs, NJ: Prentice-
Hall.
Siggelkow, N., & Rivkin, J. (2005): ‘‘Speed and Search: Designing Organizations for Turbulence and
Complexity,’’ Organization Science, 16(2), pp. 101–22.
Sinha K. K., & Van de Ven, A. (2005): ‘‘Design of Work Within and Between Organizations,’’ Organization
Science, 16(4), pp. 389–408.
Smith, A. (1904): An Inquiry into the Nature and Causes of the Wealth of Nations, London: Oxford
University Press.
Sorenson, O., Rivkin, J., & Fleming, L. (2004): ‘‘Complexity, Networks, and Knowledge Flow,’’ Research
Policy, 33(10), pp. 1615–34.
Stinchcome, A. L. (1965): ‘‘Social Structure and Organizations,’’ in James D. March (ed.), Handbook of
Organizations, Chicago: Rand McNally, pp. 142–93.
Tagiuri, R., & Litwin, G. H. (1968): Organizational Climate, Cambridge, MA: Harvard University Press.
Thompson, J. D. (1967): Organizations in Action, Oxford: Oxford University Press.
Tushman, M. L., & Romanelli, E. (1985): ‘‘Organizational Evolution: A Metamorphosis Model of
Convergence and Reorientation,’’ in L. L. Cunnings, & B. M. Stacer, (eds.), Research in Organizational
Behavior, San Francisco: SAI Press, pp. 171–222.
Velstring, T., Rouse, T., & Rovit, S. (2004): ‘‘Integrate Where It Matters,’’ MIT Sloan Management Review,
46(1), pp. 15–18.
Venkatraman, N. (1989): ‘‘The Concept of Fit in Strategy Research: Towards a Verbal and Statistical
Correspondence,’’ Academy of Management Review, 14(3), pp. 423–44.
Williamson, O. E. (1975): Markets and Hierarchies: Analysis and Antitrust Implications, New York: Free
Press.
Woodward, J. (1965): Industrial Organization, Theory and Practice, Oxford: Oxford University Press.
Zammuto, R. F., & Krakower, J. Y. (1991): ‘‘Quantitative and Qualitative Studies in Organizational
Culture,’’ Research in Organizational Change and Development, 5, pp. 83–114.
Zuboff, S. (1988): In the Age of the Smart Machine: The Future of Work and Power, Oxford: Oxford
University Press.
 Index

ABB 68 calm environment 44–5, 49, 76–7

adaptiveness 159 Carey, D. C. 215
agency theory 184–5 Carlsberg 214
airlines 11, 41, 47, 164 Carroll, T. N. 110
analyzer strategy 24, 25, 79 cellular organization 95, 132
with innovation 30–1 centers of excellence 87
without innovation 29–30, 31 centralization 160–1
Arrow, K. 6 Chandler, A. D. 23, 64, 207
Ashby, W. R. 37, 46 change 14, 15, 27–8, 29, 166
assembly line 126 benefits of 208
autonomy 127, 128, 177, 178 difficulties of 209–12
punctuated equilibrium 214
Bartlett, C. A. 85, 90 readiness to change 141
Bartol, K. 197 time-paced innovative action 214–15
biotechnology industry 28 Choo, C. 167
Birkinshaw, J. 31 Cisco 190, 215
blob complexity 70, 76–7 clan model of coordination and control
Bon Gout 44–5 164–5, 171
bonus-based incentive systems 189–91, climate: see organizational climate
195–6 Coca-Cola Inc. 27
Boudreau, M.-C. 96 communication 123, 128; see also knowledge
Bourgeois, L. J. 38, 40 exchange
Bower, J. L. 83 communications 71, 84, 93–4, 170
Brown, S. L. 214 Compaq 215
Burns, T. 3 compensation 186; see also reward, incentive
Burton, R. M. 3, 15, 123, 204, 209 complementarities 216–17
configuration 57 complexity, environmental 40–3, 47, 48, 49
coordination 80 complexity, organizational 69–73, 74
environment 37–8 blob 70
incentive systems 196 fits and misfits 75–80
leadership style 136 flat 71–2
misfits 206, 207, 208, 211, 213 symmetric 72–3
organizational climate 140 tall 70–1

229
230 Index

complicated task design 113–14, 119 divisibility 111, 114–15, 117

compromise 145 divisional configuration 63–6, 83, 89, 196
configuration 57–69, 86, 123, 204 Donaldson, L. 208
divisional 63–6 Doty, D. 10
fits and misfits 75–80, 196 Duncan, R. B. 38
functional 61–3
matrix 66–9 education 124, 185, 188
simple 59–61 effectiveness 11
conflict 140, 143, 144, 145 configuration and 64, 66, 68
contingency theory 38 efficiency 11, 143
contractual relationships 190, 191, 221–2 configuration and 61, 62, 66, 68
coordination and control systems 158–66 efficiency and effectiveness, balance between
clan 164–5 12–13, 14, 15, 16, 208; see also goals
family 161–2 Eisenhardt, K. 38, 214
fits and misfits 175–9 Eli Lilly 116
machine 162–3 employees 184, 189; see also workforce
market 163–4 education, training and experience 124,
mosaic 165–6 185, 188
coordination of tasks 4, 9, 57, 60, 82–4, 109 number of 122, 123, 124
across geographical boundaries environment 14, 18, 37–51, 204, 209, 214
configuration and 62–3, 64, 67, 68, 80 calm 44–5, 49
and task design 114 fits and misfits 49, 50
and people management 125, 127 locally stormy 45–6
creativity 163 perception of 38, 41, 44
Cullen, J. B. 91 turbulent 46–7, 50, 208
customer relationship management systems 171 varied 45, 50
Cyert, R. M. 136 environmental imperative 37, 38
equifinality 10
data-driven information systems 169, 177 equity 197; see also fairness
Davidow, W. H. 92 event-driven information systems 168–9
decentralization 160–1, 172, 173 executive
decision-making, hierarchical 7 choice of principles for systems design 158
defender strategy 24, 26–7, 77 and task design 113, 116
delayering 69 role in configurations 60–1, 63, 64, 65–6
delegation, preference for 136–7, 146, 185 experience 124, 185, 188
Denison, D. R. 135 exploration and exploitation 24, 32, 47, 85, 89
DeSanctis, G. 82
developmental climate 144–5, 150 factory approach to managing people 126–7,
differentiation 38, 69, 74–5, 88 131–2
Digital Corporation 26 fairness 142, 143, 182, 183, 197
Disney, Walt 162 family model of coordination and control 161–2,
distributed organizations 82–103 175–6
fits and misfits 99–101 fashion industry 30
geographic: see geographic distribution feedback 42, 159–60, 162
knowledge exchange: see knowledge exchange Fenton, E. M. 3
231 Index

firm Gerstner, Lou 141

information-processing view of 5, 6–8, Ghoshal, S. 85, 90
16, 123 Gibbons, R. 185
internal or external focus 58, 92, 95 Gibson, C. B. 31
fits and misfits 14–16, 204–6, 208, 218 Gillette 214
climate 148–51 Gittell, J. H. 165
complexity 75–80 GlaxoSmithKline 220, 223
configuration 75–80, 196 global organizational design 86–7
coordination and control 175–9 goals 3–17, 175, 203, 206, 208
environment 49–50 and complexity 69
geographic distribution 98–101 and incentives 183
incentives 193–7 and strategy 34–5
information systems 157–9 Google 29
knowledge exchange 99–101 Gordon, G. J. 37
leadership 148–51 Gresov, C. 208
people management 130–3 group climate 142–3, 150
performance and 5, 207, 208, 213 Grove, Andy 44
strategy 33–5
task design 118–20 Haakonson, D. D. 24, 136, 208
flat complexity 71–2, 78 Hahn, G. J. 143
flexibility Hambrick, D. C. 24
and configuration 60, 67 Hansen, M. T. 96
and coordination and control systems 159–61, Hennart, J. F. 186, 190
163, 178 heterogeneity 165–6
and decisions to locate 85 Hewlett-Packard 26, 215
and environment 46, 47 hospitals 163
and incentive systems 190 human resources: see people management
and knowledge exchange 95
forecasting 42, 47, 139 IBM 31, 141, 170, 215
formalization 159–60, 172, 173, 178, 209 IKEA 165
fragmented task design 114–15, 119–20 incentives 182–98
free rider problem 192 acceptance of the system 196–7
Fulk, J. 82 behavior-based 184–5, 186, 192
functional configuration 61–3, 196 bonus-based systems 189–91
functional specialization 58–9, 60, 73, 74, 87 design of 183–7
effectiveness and efficiency of 188
Galbraith, J. R. 3, 6, 7, 57, 78, 132 fits and misfits 193–7
General Electric 11, 42, 46, 78, 190 free rider problem 192
General Motors 123, 215 group-based 193
geographic distribution 82, 83, 84–91 personal pay system 187–8
fits and misfits 98–101 profit sharing systems 191–2
global structure 86–7 results-based 184, 186, 189,
international structure 87–8 192, 193
multi-domestic structure 88–90 skill pay system 188–9
transnational structure 90–1 informated organization 94–5
232 Index

information 8 knotty task design 116, 120

amount of 167–8, 173, 174–5 knowledge 92
flow of 61–2, 72, 123, 142 explicit 124, 167
sharing 145, 163, 177, 178 tacit 124, 167, 173, 174, 177
tacit nature of 167–8, 173, 174 knowledge exchange 82, 91–6
information-processing capacity 6–8, 57, ad hoc communications 93–4
92, 123–4 fits and misfits 99–101
climate and 144, 145 informated 94
complexity and 71, 73 cellular 95
configuration and 61, 66 network 95–6
environment and 46 Konosuke, Matsushita 122
goals and 12–13 Kotter, J. P. 136
people management and 125 Kottler, P. 24
information-processing demands 43, 116, 132 Kowtha, N. R. 184, 185, 186
information systems 166–72 Krakower, J. Y. 140
data-driven 169
event-driven 168–9 laboratory approach to managing people 127, 132
fits and misfits 175–9 Law of Requisite Variety 46
people-driven 169, 171 Lawrence, P. R. 37, 38
relationship-driven 171–2 leader style of leadership 138, 150
information technology 8, 94, 175 leadership 135–52, 165
infrastructure 157, 166, 175, 176, 177, 178 theory X and theory Y 136
innovation 28, 163–4, 177, 214–15 leadership style 136–40, 146
integration 38 credibility of the leader 141, 143, 144, 145
Intel 44 fits and misfits 148–51
interdependency 166 leader 138
configuration and 63, 66, 68 maestro 136–7
environment and 41, 48 manager 137–8
people management and 128 producer 138–9
and task design 110, 111, 114, 117 theory X and theory Y 136
internal process climate 143–4, 150 learning 12, 24
international organizational design 87–8 LEGO 27, 45
inventory management 7, 162 Lenovo Group 170
IT infusion 93, 98 Likert, R. 136
line functions 61
jello effect 67, 68 Litwin, G. H. 140
Jet Blue 11 local responsiveness 85, 96, 97
joint venture 215, 216, 219–20, 222 locally stormy environment 45–6
Jordan, Michael 191 location decisions 84–5
just-in-time inventory 7, 162 Locke, E. 197
Lorsch, J. W. 37, 38
Keen, P. G. W. 94
Kellerher, Herb 164 machine model of coordination and control
Kerr, S. 183 162–3, 171, 177
Kimberley, J. R. 145 maestro leadership style 136–7, 150
233 Index

Makadok, R. 24 Nokia 163–4

Malone, M. S. 92 Nonaka, I. 167
management, top 68, 76, 135, 185
manager leadership style 137–8, 150 Obel, B. 3, 15, 50, 123, 209
March, J. G. 12, 24, 57, 124 configuration 57
market model of coordination and control 163–4 coordination 80
mass production 110, 114, 126 environment 37–8
matrix configuration 66–9, 78–80, 91, 96, 133 incentives 196
McDonald, M. 94 leadership 136
McDonalds 114 misfits 208, 211, 213
McGregor, D. 136 office approach to managing people 128, 132–3
merger 215, 216, 220–1, 222 open systems theory 37
Merloni Elettrodomestici 83 optimal sourcing 84–5, 97
middle management 70, 71, 159 orderly task design 112–13, 118–19
Milakovich, M. E. 37 organization
Miles, R. 24, 58, 95, 96 definition of 9, 122
Miller, D. 208 goals and scope of 3–17
Mintzberg, H. 57 organizational climate 140–5, 147
misfits 5, 204–5, 207–8, 212–15; see also fits developmental 144–5
and misfits fits and misfits 148–51
and balancing design dimensions 14–16, group 142–3
210, 211 internal process 143–4
fixing 209 rational goal 145
identifying 203, 212, 217 organizational design 3, 4, 10, 18–19, 101
and mergers 220–1 changing 206–12
opportunity losses for 206–7, 207–8 dynamics of 203–24
sources of 212–13, 214 holistic approach 204, 208, 211
monitoring 159–60, 162, 183 information-processing view of 6–8, 42,
morale 141, 142, 144 123, 222
mosaic model of coordination and control misfits 14–16, 212–15
165–6, 171 multi-contingency view 3, 16
Movex 159 multi-organization 215–17
multi-domestic organizational design 88–90 open systems theory 37
multi-organization 215–17, 217–23 role of environment in 38
joint venture 219–20 organizations, distributed 82–103
merger 220–1 fits and misfits 99–101
strategic alliance 221–3 global 86–7
international 87–8
Naman, J. L. 208 multi-domestic 88–90
Navision 159 transnational 90–1
NEC 90 OrgCon software 5
network organization 95–6 Ouchi, W. G. 164
new product development 116
NK complexity theory 63, 65, 68 parallel processing 72
Nohria, N. F. 96 partitioning a task 4, 9, 57, 58
234 Index

partnership 215, 216, 221–2 Rivkin, J. 38, 50, 63, 208

people-driven information systems 169–70, 171 Robbins, S. P. 9
people, management of 122–34, 136, 175 Roberts, J. 23
factory model 126–7 Romanelli, E. 12, 214
fits and misfits 130–3 rules 159–60, 187–8
laboratory model 127
office model 128 Saab 222
shop model 125–6 Samsonite 44
People Soft 159, 169 SAP 159, 169, 175
performance 15–16, 37, 151, 191, 204 scapegoating 141, 143
configuration and 68–9 scope, defining 9–10
coordination and control and 159 Scott, W. R. 37, 110
environment and 39, 40, 41 shop approach to managing people 125–6, 131
incentives and 197 Siggelkow, N. 38, 50, 63, 208
and management of people 130 Simon, H. A. 57, 124
and misfits 5, 207, 208, 213 simple configuration 59–61, 87
personal pay incentive system 187–8, 195 Singapore Airlines 11
philosophy 158, 189 Sinha, K. K. 110
planning 139, 221 Six Sigma programs 143, 162
policy 160, 188 skill pay incentive system 188–9, 195
Polyani, M. 167 skills 124, 130, 190
Porter, M. E. 23 Slevin, D. P. 208
Powley, E. 166 Smith, Adam 62
preference for delegation 136, 146 Snow, C. C. 24, 58, 96
Proctor and Gamble 90 Sorenson, O. 167
producer leadership style 138–9, 151, 178 Southwest Airlines 11, 41, 164
product/service/customer orientation 58, 59, 73 span of control 69
professionalization 123–4, 125, 127, 129 specialization 62, 69
profit-sharing incentive systems 191–2, 196 sports, team-based 191
prospector strategy 24, 28–9, 78 Stalker, G. M. 3
punctuated equilibrium 214 standardization 111, 112
step-by-step approach 18–19, 203–6, 216
Quinn, R. E. 145 configuration and complexity 73–5
coordination, control and information systems
rational goal climate 145, 151 172–5
rationality, bounded 124 distributed organization 96–9
reactor strategy 24, 25–6, 76 environment 47–9
readiness to change 141, 147 incentives 192–3
regulation 47 leadership 146–8
relationship-driven information systems 171–2 multi-organization 217–23
Renault 116 people management 128–30
repetitiveness 111, 117 scope and goals of organizations 13–14
resources, slack 7, 8 strategy 31–3
rewards 141, 182, 186; see also incentives task design 117–18
risk 28, 44, 136, 164, 184–5, 195 Stinchcome, A. L. 183
235 Index

strategic alliance 215, 216, 221–2 training 124; see also education
strategy 23–36, 110, 175, 186, 206, 207 transnational organizational design 90–1
analyzer with innovation 30–1 trust 140, 143, 144, 145
analyzer without innovation 29–30 turbulent environment 46–7, 50, 208
defender 25–6 Tushman, M. L. 12, 214
fits and misfits 33–5
prospector 28–9 uncertainty 6, 37, 38, 40, 42, 110, 111, 184
reactor 25–6 environmental 37, 38, 40, 42, 184
strategy space 25 task design 7, 110, 111
structure 23, 110, 175, 207; see also uncertainty avoidance 136, 146, 147
configuration Unilever 90, 116, 166
symmetric complexity 72–3 unit of analysis 9, 13, 23, 215–17
unpredictability 41–2, 43, 47, 48–9, 110
Tagiuri, R. 140 UPS 94
Takeuchi, H. 167 utility companies 44
tall complexity 70–1, 77–8
task design 109–21 values 140, 165
complicated 113–14 Van den Ven, A. 110
fits and misfits 118–20 varied environment 45, 50
fragmented 114–15 Velstring, T. 96
knotty 116 virtualization 58, 92, 98
orderly 112–13, 118 vision and mission statements 18
orderly task design
task design space 118 Wal-Mart 163
tasks Welch, Jack 190
assignment of 60, 61 Whole Foods 192
coordination of 4, 9, 57, 112 Williamson, O. E. 63
design of 109–21 Woodward, J. 110
divisibility 111, 114–15, 117 work practices 87, 89, 141
partitioning of 4, 9, 57 workforce
repetitiveness 111, 117 education, training and experience 124,
technology design 109 185, 188
telecommunications 170; see also number of people 122, 123, 124
communications professionalization 124
tension 140–1, 148, 178
Thompson, J. D. 7, 110, 111 Xerox 13, 31
3M Corporation 11, 28, 89, 127
time-paced innovative action 214–15 Zammuto, R. F. 140
Toyota 116 Zuboff, S. 94