RPIK Pestleweb - Thesis PDF

University of Reading
Is there a better way to

analyse the business environment?
By
RJ Collins
Student number: 83000716
Management Challenge submitted in partial fulfilment of the requirements for the

degree of Master of Business Administration
Executive summary
The focus of this thesis is an investigation of a new method for modelling the business
environment. This thesis recognises the key importance of deep and timely knowledge
of the business environment for strategy development and effective continuous
execution. Further, it identifies issues relating to the effectiveness of the business
environmental modelling task both in professional practice and in business education.
Reports in the literature of a failure of businesses to engage in effective environmental
scanning are corroborated in this thesis by evidence gathered in a workshop session
with MBA students.
First, this thesis introduces a new method for modelling the business environment,
called PESTLEWeb. In the first instance, it is intended that this method be used in an
academic setting, particularly in support of teaching the analysis of context for strategy
development. Consequently, this thesis adopts a rigorous academic approach to the
methods evaluation.
It is argued that the PESTLEWeb method has several properties that distinguish its
effectiveness compared to traditional keyword-driven methods of identifying, analysing
and synthesising business environmental knowledge. Properties of the new method are
investigated experimentally and then separately by an observational study in a practical
workshop setting. Results of a longer-term follow-up review of workshop attendees
application of the method are also reported
Results from the experimental portion of the research are presented. These include an
investigation both of the cognitive aspects of the tool and subjects affective responses
to it. In regard to cognitive performance, no strong evidence is reported to indicate that
the detail contained within the PESTLEWeb model is significantly more memorable
than a text-based analysis. However, a measure of affective variables for subjects shows
a strong statistically significant preference for the new method in terms of subjects
judgments of how logical, rational and convincing the models are.
Results from the observational study show that the method may be introduced to an
MBA class in a relatively short and straightforward workshop setting. It is shown that
(i)
learners are able to understand and use the method to some extent within the workshop
itself. A longer-term follow up of a self-selecting group of workshop students
demonstrates that some, at least, are able to develop sophisticated and rich models of the
business environment using the PESTLEWeb method.
Recommendations from this thesis include an incremental process of further

investigation and academic engagement with the PESTLEWeb method with the ultimate
aim of introducing it into business education. Such a process would involve continuing
with the current research through experimentation and further workshop sessions and
engaging in academic research seminars in order to refine the method and expose it to
further critique.
( ii )
Acknowledgments
I would like to thank Professor Francis J. Aguilar, Professor Emeritus of Business

Administration at the Harvard Business School, for providing me a copy of his book
Scanning the Business Environment. His book is widely acknowledged as being the
original work in the field of business environmental scanning. I would also like to thank
Dr Harold E. Klein, Associate Professor Strategic Management at the Fox School of
Business, Temple University, Philadelphia. Dr Klein kindly provided copies of his
papers which were otherwise unavailable to me. Dr Kleins work is the closest analogue
to the PESTLEWeb method that I have discovered in the literature.
I would like to thank the MBA students of Henley Business School who contributed to
this work by taking part in both the experimental study and a workshop on the
PESTLEWeb method. In particular I would like to thank the members of the Picasso
study team who provided helpful feedback about the method during our exam
preparation.
The staff of Henley Business School have been crucial in enabling me to complete this
work. It was particularly generous and helpful to allow me the time and space to present
a PESTLEWeb workshop to an MBA class.
Finally, I would like to thank my partner, Monika Tesa ov, who never complained
once during all of the long-nights, early mornings and studious weekends of this MBA,
when I was unable to spend time with her.
( iii )
Contents
1 Introduction................................................................................................... 1
1.1 Problem statement ......................................................................................... 1
1.2 Relationship of the project to the terms of reference and to personal objectives
...................................................................................................................... 2
1.3 Structure of the thesis .................................................................................... 4
2 Introduction to PESTLEWeb ......................................................................... 5
2.1 Drawing conventions..................................................................................... 5
2.2 An example PESTLEWeb model for the automotive industry........................ 7
3 Review of current thinking .......................................................................... 11
3.1 Understanding the business environment ..................................................... 11
3.2 Criticisms of current methods and practices of environmental scanning....... 16
3.3 Visual argumentation as an appropriate tool in this domain.......................... 19
3.4 Conclusion .................................................................................................. 28
4 Investigation................................................................................................ 29
4.1 Conceptual framework................................................................................. 29
4.2 Ethical considerations.................................................................................. 31
4.3 Procedure 1: An experimental investigation of cognitive and affective
responses to PESTLEWeb models............................................................... 31
4.4 Procedure 2: Workshop-based observation and longer-term follow up......... 42
4.5 Conclusion and recommendations................................................................ 60
5 Reflection.................................................................................................... 67
5.1 Evaluation of findings in relation to current thinking ................................... 67
5.2 Experience of the research process .............................................................. 68
5.3 Achievement of personal objectives............................................................. 70
6 References................................................................................................... 72
Appendix A : Response paper used for the experiment in Study 1............................... 77

Appendix B : PESTLEWeb evaluation form from workshop....................................... 81
Appendix C : Examples of PESTLEWeb models developed by other people............... 83
Appendix D : Evidence from the management challenge log....................................... 88
Appendix E : Workshop presentation on PESTLEWeb ............................................... 91
Tables
Table 1: Constitution of Subjects ................................................................................ 36

Table 2: Subject's Responses to Categorisation Question on Previous 'Educational
Focus'............................................................................................................ 37
Table 3: Results of 2 Sample t-test for Procedure 1, Comparison of PESTLEWeb vs.
Text Presentation (Results significant at the 5% level highlighted in red) ...... 39
Table 4: Cronbach's Alpha for Results of Procedure 1................................................. 40
Table 5: Kirkpatrick Levels of Training Evaluation..................................................... 44
( iv )
Figures
Figure 1: Core Elements of the PESTLEWeb Drawing Convention ............................ 5

Figure 2: Reference, Comment and Consequence Elements of the PESTLEWeb
Drawing Convention.................................................................................... 6
Figure 3: Illustration of the use of a Generic Association between Elements in a
PESTLEWeb Diagram................................................................................. 7
Figure 4: Part 1 of a PESTLEWeb Diagram for the Global Automotive Industry ........ 7
Figure 5: Part 2 of a PESTLEWeb Diagram for the Global Automotive Industry ........ 8
Figure 6: Part 3 of a PESTLEWeb Diagram for the Global Automotive Industry ...... 10
Figure 7: Development of Economic Strategy (Andrews, 1980)................................ 11
Figure 8: Aguilar's Causal Relationships between External Industry Conditions and
Areas of Management Attention ................................................................ 12
Figure 9: Example of Toulmin-Style Legal Argument, redrawn from
http://repgrid.com ...................................................................................... 25
Figure 10: Conceptual Model of an Order-Processing System, Redrawn from
Checkland 1993......................................................................................... 27
Figure 11: Conceptual Model of Factors Leading to Effective Analysis of the Business
Environment .............................................................................................. 29
Figure 12: PESTLEWeb Version of the Business Environment Information ............... 35
Figure 13: Text Version of the Business Environmental Information .......................... 35
Figure 14: Companies Have a System in Place for Scanning the Business Environment
.................................................................................................................. 50
Figure 15: Business Environmental Analysis for Previous MBA Assignments............ 50
Figure 16: Subjects Ease of Analysing Business Environment for MBA Exams ........ 51
Figure 17: Traditional PESTLE Analysis Provides Deep Insights ............................... 51
Figure 18: Subjects Ease of Explaining the Context of their Business to Others......... 52
Figure 19: Subjects Ease of Understanding the PESTLEWeb Method ....................... 52
Figure 20: PESTLEWeb Provides a Better Explanation than Traditional PESTLE
Analysis .................................................................................................... 53
Figure 21: Subjects Belief that use of PESTLEWeb Would Improve the Quality of
their Analysis ............................................................................................ 53
Figure 22: Subjects Intention to Adopt the PESTLEWeb Method.............................. 54
Figure 23: Subjects Enthusiasm to Learn More about the PESTLEWeb Method........ 54
Figure 24: Subjects Enthusiasm to Access PESTLEWeb Models Produced by Other
People ....................................................................................................... 55
Figure 25: Subjects Willingness to Share PESTLEWeb Models with Other People ... 55
Figure 26: Subjects Views Concerning the Usefulness of PESTLEWeb to their
Business .................................................................................................... 56
Figure 27: PESTLEWeb CoP Webpage ...................................................................... 57
Figure 28: Two PESTLEWebs for 'Globalisation and Nigerian Content Development'
Ladeji, D. (2011) ....................................................................................... 83
Figure 29: PESTLEWeb for the Oil Industry by Kemi Otaru, Henley MBA Student... 84
Figure 30: PESTLEWeb Model for the Car Industry (1), By Hermann Bauer, Henley
MBA Student ............................................................................................ 85
MBA Student ............................................................................................ 86
MBA Student ............................................................................................ 87
(v)
1 Introduction
1.1 Problem statement
The focus of this thesis is an investigation of a new method for modelling the business
environment. From an academic standpoint, the need for such a method arises from two
threads in the literature. Firstly, that understanding the business environment plays an
important role in strategy development, an idea that has recurred over an extended
period in the literature of business strategy (Aguilar, 1967; Andrews, 1980; Barney,
1991; Drucker, 1994; Horton, 2010). Secondly, the observation that this task is
frequently not done well (Fuld, 2003; Schoemaker, 2008; Schoemaker and Day, 2009).
At a practical level, the motivation for this thesis arose from the difficulties experienced
by the author and his MBA study team when attempting to analyse various business
case studies. In particular, whilst it was simple enough to apply the common PEST or
PESTLE (Political, Economic, Social, Technological, Legal and Environmental)
acronyms and generate a list of discrete issues relating to the business environment, it
was nevertheless much more difficult to synthesise these atomic facts into a coherent
argument. My team were not alone in that difficulty and indeed the course tutors had
already identified this as a general problem:
In carrying out a PEST or PESTLE analysis it is all too easy to

produce lists of factors, many of which may be of little or no
importance in developing strategy Developing a clear
understanding of the cause-and-effect relationships between the
factors in the PESTLE model is more challenging.
Tovstiga and Aylward (2008)
It was thus apparent that whilst a common issue in business strategy development (at
least in education) had been identified, no solution was readily available. The method
described in this thesis was created by the author during the course of his MBA study to
address this issue: how to model the business environment in a manner that captures
rich, cause-and-effect relationships. The issue was how to build a model that captures
complexity and multi-dimensional relationships and enables rationale inferences and
deductions to be made. In particular, a method was required that enabled both business
students and practicing business strategists to elicit, externalise, capture and disseminate
(1)
coherent arguments about the business environment and to act as both a tool for
thinking and as a repository for Knowledge Management.
In the first instance, it is intended that the proposed method be used in an academic
setting, particularly in support of teaching the analysis of context for strategy
development. Consequently, this thesis adopts a rigorous academic approach to the
methods evaluation.
1.2 Relationship of the project to the terms of reference and

to personal objectives
The solution proposed to the issue identified in Section 1.1 is a graphical model that has
been termed PESTLEWeb. This name is intended to signify both its main focus
(PESTLE issues in the business environment) and the idea that this is expressed in a
web or network of causatively related threads. Such a model specifically addresses
the issue of 1.1 since it effectively encodes not only the discrete items or issues of
the domain of interest but also the rich web of relationships between them.
The PESTLEWeb model is a direct development of earlier graphical notations

developed by the author in the field of complex systems engineering (Collins, 1997a,b).
These models were themselves derived from earlier work in visual argumentation
(Toulmin, 1958; Schmidt 1986), cognitive maps (Eden, 1989, 1992; Eden et al, 1992;
Bougon, 1992), soft-systems methodology (Checkland, 1993) and the representation of
complex technology systems (Shlaer and Mellor, 1992; Selic et al, 1994; Hoare, 1985).
As such, the PESTLEWeb model and this thesis relate directly to the authors personal
objectives as stated in his original application form for the Henley MBA:
I have a great interest in teaching and research. I believe that some

of my research background in complex systems engineering and
psychology has applications within strategy development and
corporate finance. I could see that doing an MBA would open
opportunities for research and teaching.
(2)
The PESTLEWeb model is just such an application of tools and methods from complex
systems engineering and psychology to the domain of business strategy development.
This objective was further refined in the proposal for the project on which this thesis is
based:
The proposed project is only the first step of a larger programme of

research I have in mind in the domain of Visual Analytics for
analysis and synthesis of business strategy.
It is clear from the research presented within the following thesis that many interesting
avenues of enquiry remain unexplored. Some of the core questions, for example relating
to usability in a classroom setting, have been addressed but other questions have
arisen, such as in the area of individual difference, the nuances of the notation and its
long-term effectiveness in professional practice. It is thus reasonable to say that the
current work has provided a useful basis for a larger programme of research.
Another objective of the project is summed up in the following terms in the original
research proposal:
this [research] could provide a powerful and effective curriculum

resource for Henley Business School (the research will show if this
view is correct or not). If successful this project will result in a
Henley Branded model/method for analysing the business
environment and produce a valuable tool for teaching.
The experimental and observational studies documented within this thesis directly
address the efficacy of the PESTLEWeb model within a business school teaching
environment.
A final objective is that the publication of this thesis will represent a next step towards
the eventual goal of adoption of the PESTLEWeb as a teaching resource within Henley
Business School and beyond.
(3)
1.3 Structure of the thesis
A prologue to this thesis (Section 2) presents a brief introduction to the PESTLEWeb
modelling method. This section is introduced to provide a context and motivation for
the sections that follow. In practical terms, both the literature review and the
investigation make more sense following an explanation of the PESTLEWeb modelling
convention itself.
The thesis proper begins with a review of current thinking (Section 3). The argument
presented within that review may be summarised thus:
1. Successful business strategy crucially depends on developing a rich, accurate

and timely understanding of the business environment.
2. Scanning of the business environment is frequently not done well.
3. Visual argumentation is a powerful and effective tool to support analysis,

understanding, communication and reasoning from complex multi-dimensional
datasets and hence is appropriate in this problem domain.
Major sections within the review address each element of this argument in turn.
The tutorial and review sections are followed by an investigation of issues relating to
the efficacy of the PESTLEWeb method (Section 4). The investigation is separated into
two major sections:
1. An experimental procedure intended to rigorously test the hypothesis

relating to the cognitive and affective properties of the method; and
2. A workshop-based observational procedure intended to test the method in a

more realistic practical setting.
Each of these sections follows the pattern of presenting objectives, followed by the
design of the study and finally findings and analysis.
Both of the studies are brought together and discussed within the Conclusion and
Recommendations (Section 4.5).
(4)
The final section of this thesis presents a critical reflection on the content of the thesis,
conduct of the project and the learning that has taken place as a result (Section 5).
2 Introduction to PESTLEWeb
2.1 Drawing conventions
At its core, PESTLEWeb is based on the ubiquitous PESTLE tool that is used to support
the identification of Political, Economic, Social, Technological, Legal and Ecological
issues in the business environment (Thompson and Martin, 2005). However,
PESTLEWeb is a drawing convention rather than a tabulated, textual analysis.
The base building-blocks of PESTLEWeb are a series of boxes labelled to represent

each element of the PESTLE analysis. These boxes are coloured in order to aid
identification and differentiation. Those base elements are then joined with directed
lines representing the causal relationships between elements.
Figure 1: Core Elements of the PESTLEWeb Drawing Convention
In order to be useful for analysis, as well as representing PESTLE issues and their
causal relationships, it is also important to be able to identify the logical consequences
of those causal relationships. This is particularly true because, to be productive, we need
the analyst to move from an abstract analysis of the business environment to the specific
consequences that impinge on their own business.
(5)
Additionally, it is a common convention in the more formal diagrams of software
engineering to include drawing elements that enable annotation and comments. Such
elements may sit apart from the formal modelling (representational) elements of the
diagram, enabling a reader to better understand the diagrams. Within the PESTLEWeb
notation, we include a generic comment symbol and a specific variant of a comment
used to indicate source references for further information. The latter symbol enables the
diagram to be embedded within a richer network of source information.
Figure 2: Reference, Comment and Consequence Elements of the PESTLEWeb Drawing

Convention
Finally, the convention includes an element to annotate relationships between elements

that are not causal in nature. These elements are intended to be less powerful in the
PESTLEWeb model than in the causal relationship element identified above. Rather
than expressing a consequential logic, they represent a more generic association
between elements. These two types of connectivity between elements (causal and
relational) have previously been shown to be useful in a similar context in a
diagrammatic notation introduced by the author (Collins 1997a; 1997b)
Consider, for example, Figure 3, which shows two elements from a PESTLEWeb
analysis of the automotive industry. The two issues in the diagram are not causally
related, but there is an important relationship or association that the analyst wants to
emphasise in the model. This association may be implicit or the joining line may be
labelled explicitly with the nature of the association. Such associations may also be
annotated with comment elements.
(6)
Economic
Growth in China: most cars <
1.6lt because 60% of car
buyers are from middle or
low-income households
PESTLE
Association
Technological
Japanese car industry
has focused primarily
on small fuel efficient
cars
PESTLE
Figure 3: Illustration of the use of a Generic Association between Elements in a PESTLEWeb

Diagram
2.2 An example PESTLEWeb model for the automotive

industry
2.2.1 Starting the model
This section illustrates the development of a PESTLEWeb model for the automotive
industry, showing how the model is built section by section. Figure 4 shows an extract
from the full model.
Environmental Economic Political

Pollution from Externalities - Introduction of
operation and Pollution cost does Carbon Tax and
production of cars not fall to the user. Carbon Trading
PESTLEWebTM PESTLEWeb TM PESTLEWeb TM
Environmental Social Political Technological

Growing public Government R&D
Increasing political
spend on
Congestion awareness of green will to address environment friendly
issues. environmental issues technologies
PESTLEWebTM PESTLEWeb TM PESTLEWeb TM PESTLEWeb TM
Consequence
Economic Economic
Long term trend Shift towards Opportunity!
(increase) in oil purchase of low- Increased demand for
prices emission vehicles small, efficient cars
PESTLEWeb TM PESTLEWeb TM
Figure 4: Part 1 of a PESTLEWeb Diagram for the Global Automotive Industry
(7)
This shows the consequential chain that leads from car pollution and congestion through
the social and economic consequences of that pollution and on to the political response
to those social and economic issues. Note that some of these consequences are
economic and some are technological.
At the bottom right-hand side of Figure 4, we see the final consequence for a specific
business that emerges from this causal logic chain. This consequence, of course, would
be different depending on the circumstances of the particular business; for an
automobile manufacturer producing large gas-guzzling cars, the consequence might be
a significant threat. In this example, we have shown the opportunity that may arise for a
manufacturer of smaller fuel-efficient engines.
2.2.2 Considering consequences
In Figure 5, we see the development a section of the automotive PESTLEWeb model

that brings together the following issues:
The recent credit crisis and its political and hence economic ramifications for the
car industry;
Social, Economic and Technological factors leading to the US car industry being
strongly weighted towards large-car manufacture; and
Issues of unintended consequence of government support of existing
manufacturers.
(8)
One aspect of this diagram should be highlighted to illustrate a general point about
PESTLEWeb models. The Consequence element labelled Reduction (delay) in
purchase of new vehicles could be considered a generic Economic issue rather than a
Consequence; so, why is it so labelled?
The answer is that the Consequence elements are the link between the general
analysis of the business environment and the immediate consequences for the specific
business for which the analysis is being performed. In a pure Industry Analysis this
element might well be labelled a generic Economic node. However, in this case we
are developing the model from the position of a business that can foresee immediate
consequences and implications for their business.
In general, the Consequence elements will feed most directly into later stages of
strategic analysis for the business. They will often appear as the Opportunities and
Threats in a later SWOT (Strengths, Weaknesses, Opportunities and Threats) analysis.
2.2.3 Threads
Threads are series of causally related issues that run through the model. These can be
considered the stories that run through the analysis. In other words, each of the
PESTLEWeb elements should be regarded neither as atomic nor as simply connected
to its precursors and consequences. Rather, analysts need to consider threads of
causally related issues in order to develop a deep understanding of the business
environment. Such PESTLEWeb threads are a close analogue to those introduced by
Collins (2000) and are used to explain and analyse behaviour in large-scale complex
systems.
The section of the automotive industry PESTLEWeb model in Figure 6 introduces

issues associated with the emerging economies and bias in certain countries towards
specific types of vehicle. In the full integrated model, elements from this thread cross-
linked to elements in the previous threads. The concept of threads is useful in this
context because they enable coherent themes to be drawn from the interconnected
network of model elements.
(9)
Technological Economic
Eighty-eight per cent Shift towards
of new vehicles sold
purchase of low-
in Brazil are flex-fuel
models emission vehicles
PESTLEWebTM PESTLEWeb TM
Consequence
Economic Economic Consequence Economic Political
US car industry favours Loss of tax revenues and
large car production The main growth markets Japan, India and Brazil Government
Low affluence in lowering of GDP as car
are the developing world and could move into a position manufacturing moves to
intervention in
BRIC countries the leading models are small of significant comparative support of domestic
far-east and emerging
Economic fuel efficient cars
advantage in car economies car companies
China: most cars < 1.6lt PESTLEWeb TM PESTLEWebTM production PESTLEWebTM PESTLEWebTM
because 60% of car buyers
are from middle or low-
income households
Technological
Japanese car industry PESTLEWebTM
has focused primarily
on small fuel efficient Consequence
cars Japan, India and Brazil in
PESTLEWebTM a better position than the
Technological US to exploit shift towards
small fuel / efficient cars
Indian car market has
always focused on
small cheap cars
PESTLEWebTM
Economic Consequence
Changing
Recent spike in oil consumer habits:
prices switching to smaller,
fuel-efficient vehicles
PESTLEWebTM
Technological
Car development
and production is a
long-term process
PESTLEWebTM
( 10 )
3 Review of current thinking
3.1 Understanding the business environment
Successful business strategy depends crucially on developing a rich, accurate and

timely understanding of the business environment.
3.1.1 Environmental scanning as a key component of strategy

development
In his seminal work on strategy, Andrews posited an ecological view of strategy as the
process of matching what a company can do within the universe of what it might do to a
formulation of behaviour representing what it should best do (Andrews, 1980).
Andrews dedicated a whole chapter of the work to The company and its environment:
Relating opportunities to resources and his argument is best summarised in Figure 7,
which is drawn from that chapter.
Figure 7: Development of Economic Strategy (Andrews, 1980)
( 11 )
In the original work in which PEST analysis was introduced, Aguilar made explicit the
structured nature of the Business Environment Scanning Task (Aguilar, 1967); that is,
that there is logic of causal relationships as shown in Figure 8.
Figure 8: Aguilar's Causal Relationships between External Industry Conditions and Areas of
Management Attention
The general logic of this process has stood the test of time and, for example, is reflected
in more recent work focused on the Resource-Based View (RBV) of the firm. In
particular, Barney (1991) has commented on the complementarity between
environmental models of competitive advantage and the resource-based model. Indeed,
it has been argued that the RBV specifically builds on two previous broad approaches to
strategy (the internally and externally focused) by combining internal and external
perspectives (Collis and Montgomery 2008).
Drucker has identified the failure to match a companys behaviour with its changed
environment as a principle cause of stagnation (Drucker 1994). Drucker is more
concerned with issues of the micro-environment pertaining to markets, customers and
competitors than the issues of the macro-environment that this thesis is focused on.
However, the point is still well made; a business is embedded in a dynamic context and
the leaders of any business would do well to take account of that context in an
intelligent and purposeful manner. When Drucker refers to a theory of the firm he is
referring to a theory of the firm in the context of an environment. Therefore, the firm
must engage in environmental scanning on an ongoing basis in order to monitor forces
that may have an impact on the business (Albright, 2004).
All of this is not to say, of course, that the firm must be regarded as a passive element
within its environment. This view is somewhat in contrast to the process suggested by
Aguilar in Figure 8 and entails a more dynamic and co-evolutionary relationship
between firm and environment. As has been cogently argued by Astley and Fombrun, a
( 12 )
firm is an active component within an ecology and as such may both respond
dynamically to and affect the evolution of that environment (Astley and Fombrun, 1983;
Astley 1984):
However, the argument here is that for this co-evolution to be strategically valuable to
the firm it must be purposeful. A rich model of the external environment that enables
and promotes sense-making of the environment by business leaders is a valuable
component of being effectively purposeful.
It has been argued that in an increasingly dynamic, global environment strategy has
much to do with dynamic capabilities such as environmental sense-making. Dynamic
capabilities are the routines, activities and micro-processes by which an organisation
achieves a new configuration, key amongst which is an early warning system of
effective environmental scanning (Horton, 2010). The argument for this is that:
When there are high levels of ambiguity, complexity or unpredictability

in the business environment, strategy must be discovery driven, involve
exploration and learning using qualitative methods.
Horton (2010)
For a firm to maintain competitive advantage it must purposefully create, codify and
leverage information about an evolving business environment:
[a] crucial component of managerial behavior in rapidly changing

environments is problem sensing, the cognitive processes of
noticing and constructing meaning about environmental change so
that organizations can take action.
Kiesler and Sproull (1982, cited by Barr et al, 1992)
Given this argument, logic would dictate that, in order to develop a good theory of the
firm in an environmental context, one must first develop rich, relevant and useful
theories or models of the business environment itself. The current work is focused
principally on the elicitation, externalisation, development and formal representation of
theories of the business environment to support such strategic sense-making.
( 13 )
The ability to manage such high-value and relevant information relating to the business
environment is an example of what Pandza and Thorpe term a high-order capability:
The concept puts forward the view that firms in possession of such a
high-order capability are deemed able to purposefully create, extend or
modify their knowledge base to respond to the shifts in a business
environment.
Pandza and Thorpe (2009)
The idea here is that effective environmental sense-making may be considered a meta-
cognitive skill for a business that has a particular role in gaining competitive
advantage. This is because such sense-making is one element of an adaptive capability
that enables effective and appropriate responses in a dynamic environment.
It may be for this reason that effective environmental scanning has been recognised as
an important component of entrepreneurial behaviour. McEwen (2008) has argued the
link between environmental scanning, organisational learning and entrepreneurial
success:
Despite the importance of the entrepreneurs' knowledge to new venture

success, many are faced with a capability gap because of the discrepancy
between their current knowledge and the information that is relevant to
the current business environment. One way of acquiring and using
outside information is through environmental scanning.
McEwen (2008)
Thus, it can be seen that effective business environmental scanning has been recognised
as of key importance to business strategy development and execution for several
decades. The arguments have remained relevant in one form or another. It is a basic
skill for business management, but performed well it may also be a higher-order
business capability that enables effective responses in a dynamic environment.
( 14 )
3.1.2 Management of environmental knowledge
An organisations ability to create and share knowledge is important for establishing
and sustaining competitive advantage (Kanawattanachai and Yoo, 2007). This, together
with the argument in the previous sections (that knowledge of the business environment
is a valuable resource for the business) leads naturally to the conclusion that a
mechanism to codify and share such knowledge would be of significant benefit to
organisations.
In general, methods for representing knowledge (codes) provide a mechanism for

capture and re-use which is of economic benefit to firms:
In todays rapidly changing business environment, an organizations

ability to create and share knowledge is important for establishing and
sustaining competitive advantage.
Kanawattanachai and Yoo (2007)
However, also importantly, such codes provide a mechanism and catalyst for the
development of new theories:
Once they are available, new codes often make possible the development
of new models and new theory.
Hkanson (2007)
If businesses are to be managed effectively in complex environments, then their internal

knowledge structures must in some ways reflect that complexity. This follows logically
from the law of requisite variety proposed by Ashby (1958):
Control can be obtained only if the variety of the controller is at least as

great as the variety of the situation to be controlled.
Richardson (2004)
If such internal complexity is to be useful for intentional adaptation (as opposed to a

post-hoc, reactive adaptation) then knowledge of the external environment must be
( 15 )
either implicitly or explicitly coded within the system itself. Whilst such information
could be implicitly coded, for example through a highly agile, adaptive and devolved
organisational structure, the argument here is that an explicit codification is a valuable
resource. This is because codification of important and relevant information for the firm
(in this context strategically important knowledge about the business environment)
enables shared cognitive sense-making. That is, it allows company executives and
leaders to develop and articulate a shared understanding of the environmental features
that are important to their business (Oliver and Montgomery, 2008).
The issue is that, in practice, there can be significant inconsistency amongst leaders in a
firm in their views about the business environment for that firm. It is difficult to see
how there could be effective and intelligent strategic alignment in the face of diverse
or even contradictory views about the environment. To address the issue of diverse
mental models amongst leaders within a business, Bougon has described the creation, in
a workshop setting, of congregate, cognitive maps (Bougon, 1992). Such maps
represent the collective, consensus view of a group externalised in a graphic network.
The argument being that we should be searching for a method that not only allows
individuals to perform analysis of the environment, but also for those models to be
externalised and communicated so as to develop consensus in a group of leaders.
The paper by Bougon (1992) thus provides a cogent argument for methods like
PESTLEWeb to be used by leaders within a firm to formulate a shared, consensus view
of the environment within which they operate.
3.2 Criticisms of current methods and practices of

environmental scanning
Scanning of the business environment is frequently not done well.
3.2.1 Lack of quality and quantity of environmental scanning
As pointed out in the previous section, the value of analysis of the business environment
has been repeatedly argued in the academic business literature. Nevertheless, there is
significant evidence that, in practice, this activity is frequently not done well if it is done
( 16 )
at all. For example, a survey of 140 corporate strategists revealed that over 65% of them
admitted that their organisations had been surprised by high-impact external events in
the proceeding five years (Fuld, 2003).
Fewer than 20% of global firms have sufficient capacity to spot, interpret and act on the
weak signals of forthcoming threats and opportunities (Schoemaker and Day, 2009).
97% of respondents to a survey by the same authors reported that their firms lacked any
formal process or systems to prevent similar surprises in the future. A further survey of
100 global senior managers found that 80% of respondents perceived their future need
for peripheral vision to be greater than their current capacity (Schoemaker, 2008).
A failure to perform this important environmental scanning task can have significant
damaging effects on a business. Pearce et al (1982) have pointed out that failure to
effectively absorb information about the business environment can be a major factor in
the failure of small and growing businesses. Indeed, they argue that inattention to the
world beyond the office door is an important cause of many small business failures.
3.2.2 Failures in codifying and communicating environmental

information
There is evidence to suggest that even when business leaders do engage in more or less
formal scanning of the business environment they may fail to draw correct insights and
inferences from the data. This typically arises as a result of the difficulty business
leaders have in navigating derived environmental information as a consequence of its
volume and lack of organisation (Albright, 2004).
It is not enough to simply collect information about the environment. In order for the
receipt of information to result in action it must be analysed and the results of that
analysis must be communicated within the organisation (Albright, 2004).
Communication is important since it is rarely the case that business leaders have
complete autonomy in strategic decision making. In order for the firm to be strategically
aligned, there must be a degree of consensus in the world view (weltanschauung) of
its leaders (Kaplan and Norton 1992, 1993, 1996, 2004).
( 17 )
However, it has been argued that leaders mental models of the competitive
environment can be rather diverse within a given organisation (Daniels et al, 1994).
This has important implications for strategy development; if there is no clear, consensus
understanding of the external environment, then how can a rational strategy (other than
a highly reactive, emergent strategy) be developed? Business leaders may fail to
integrate and communicate new information in a useful and productive manner that
enables the business to gain consensus and hence react in a rational manner. Thus, Barr
and Hough (1997) report an analysis of six firms strategic responses to environmental
changes and consider the organisations ability to recognise and successfully react to
those changes. An important observation in Barr and Houghs work is that there are
significant differences in the mental models that leaders with a given firm have about
the changing business environment.
Such differences have important consequences in terms of the ability of the organisation
to recognise and react to environmental dynamics. In particular, they argue that before
an organisation can adapt it must first understand. That is, leaders must formulate a
shared model of the environment and the implications of changes:
Different beliefs about cause and effect have been established as a

plausible explanation for differential responses to environmental
change.

What our study suggests is that firms need a map in which they have
confidence, before they undertake a major journey.
Barr and Hough (1997)
This work effectively makes the argument that, in addition to engaging in a process of
environmental scanning or sensing, organisations need to be able to codify and
communicate information so that they can engage in well-informed, rational and
consensual decision making.
( 18 )
A clear and rational organisation of environmental information is important, since it
enables effective navigation to salient features in the data. However, perhaps more
important, the codification of information can make explicit the cause-and-effect
relationships that lead from the global business environment, through implications for
industries and markets to end results for specific businesses. It is this chain of logic
which is the crux of environmental scanning, and which so often proves difficult for
analysts (see the quotation from Tovstiga and Aylward on page 1).
3.3 Visual argumentation as an appropriate tool in this domain
Visual Argumentation is a powerful and effective tool to support analysis,

understanding, communication and reasoning from complex multi-dimensional data-
sets and hence is appropriate in this problem domain.
Previous sections have presented arguments for effective scanning of the business
environment as a key component of strategy development. They have also shown that
environmental scanning is frequently not done well. These arguments effectively define
the issue this thesis addresses.
The following sections move towards the solution domain for this issue. In particular, a
theoretical underpinning for the application of visual argumentation is presented. Visual
argumentation refers to graphical models with a more or less formal syntax and
semantics that enable the expression and codification of complex arguments.
One of the most coherent and convincing arguments for the use of graphics over written
language has been made by Tversky (2005):
Graphics have an advantage over language in expressiveness; graphics

use elements and relations in graphic space to convey elements and
relations in real or metaphoric space. As such, they allow inference
based on the visuospatial processing that people have become expert in
( 19 )
as part of their everyday interactions with space. As cognitive tools,
graphics facilitate reasoning, both by externalising, thus by offloading
memory and processing, and by mapping abstract reasoning into spatial
comparisons and transformations. Graphics organise and schematize
spatial and abstract information to highlight and focus the essential
information. Like language, graphics serve to convey spatial and
abstract concepts to others. They make private thoughts public to a
community that can then use and revise those concepts collaboratively.
Tversky (2005)
The following sections present arguments in support of this case, first from the basis of
the underlying mechanisms of reasoning and memory (broadly the domain of cognitive
psychology) and, second, through reference to various forms of visual argumentation
and other graphic models that have been successfully applied elsewhere.
3.3.1 Evidence from cognitive psychology and education

Any method that seeks to address the issues associated with business environmental
analysis should take into account various aspects of human cognitive psychology. Thus,
for a method to be effective, it can be presumed that it needs to encourage and support
remembering and thinking (key elements of the general framework for cognitive
activities identified by Craik and Lockhart (2008)). Thus:
Remembering: To enable analysts to absorb and recall business

environmental information quickly and efficiently in order to allow them
to cognitively process larger datasets and to enable the intellectual use of
that information after analysis; and
Thinking: To be able to reason, logically and rationally across datasets to

enable deep insights and useful deductions to be made.
With regards to remembering, a key feature of effective memory and subsequent recall
is that learners need to engage in deep cognitive processing (Craik and Tulving, 1975;
Craik and Lockhart, 1975, 2008). This typically occurs when a learner transforms
information from one form to another (for example from words into a picture) or when
the learner consciously makes associations between items of information.
( 20 )
As can be seen from the introduction to PESTLEWeb in Section 2, the method requires
the analyst to both transform information (from the linear text of source material into
graphical elements) and to create multi-dimensional associations between items. It
should be noted that, since PESTLEWeb has a pre-defined semiotic system and syntax,
the degree of transformation is not as great as it might be with a completely
unconstrained diagramming technique, such as a mindmap (Buzan, 1993). The rationale
for the design decision of having a defined semiotic system in this case is to promote
ease of learning, exchange of information and development of collaborative models.
In regard to thinking, Buckingham Shum et al (1997) have presented the case for
graphical argumentation:
the intellectual demands of visualising the structure of an

argument can provide insights into its strengths and weaknesses, thus
facilitating its more rigorous construction. Furthermore, exposing
an arguments structure in this way facilitates in subsequent
communication because important relations can be more easily
perceived and analyzed by others.
Buckingham Shum et al (1997)
The differences between text-based, and graphical-based argumentation have been

tested experimentally by Bauer and Johnson-Laird (1993). Their work tested the ability
of subjects to perform more or less complex reasoning tasks based on information
presented in text form or using a graphical model of the argument. They report that
subjects responded faster and drew many more valid conclusions from diagrams than
from the verbal premises. Although the reasoning tasks presented in these experiments
were of a rather specific kind (double disjunctions), they nevertheless add weight to the
idea that arguments presented graphically can support reasoning and logical deduction.
However, it is not just the low-level aspect of cognition relating to complex deductions
which is of relevance here. Rather it is the higher-level aspects associated with expert
performance over that of the novice. After all, in the context of a business school or in
professional practice, we are concerned not only with the minutia of correct, logical
( 21 )
deduction but rather with a holistic expertise in analysis and theory-building. As Glaser
and Bassok (1989) have pointed out, in the context of reviewing key trends in
instructional theory:
Experts and novices may be equally competent at recalling specific

items of information, but experts chunk these items in memory in
cause and effect sequences that relate to the goals and subgoals of
problem solution and use this information for further action. The
progression from declarative knowledge to well-tuned functional
knowledge is a significant dimension of developing competence.
Glaser and Bassok (1989)
This has a notable resonance with the quotation from Aylward and Tovstiga above,
although it is more general in scope. This paper contributes to the case that, in teaching
analysis of the business environment, the emphasis must surely be on a progression
from identification of facts (declarative knowledge) towards providing intellectual tools
and experience in chunking information into cause-and-effect sequences.
One of the authors of a key work on scanning the business environment, Schoemaker,
has written on the subject of MBAs and business school education and calls for a new
approach to business education (Schoemaker, 2008). Schoemaker calls for a greater
focus on uncertainty and paradox and identifies scanning the periphery as one of the
five key challenges facing business leaders. He points out that companies can discover
many opportunities by attending to and acting on weak signals in the periphery (the
business environment). Schoemaker is generally critical of reductionist models having
such prominence in business education. In this context, however, the PESTLEWeb
method is intended to be analytical whilst not being principally reductionist. A
PESTLEWeb model is intended to embrace the complexity of the business environment
with its many dimensional relationships, to make it explicit, to lay it bare and hence
promote understanding and reasoning within the context of that complexity, not in
denial of it.
( 22 )
In a later paper, Schoemaker and Day make an explicit link between scanning the
business environment and common psychological predispositions towards error,
including filtering, distorted inference and bolstering pre-formed ideas through selective
search (Schoemaker and Day 2009). Both of these arguments (that well understood
psychological mechanisms can lead to disastrous failures and that purely reductionist
approaches to problem solving can fail to identify significant threats) have been
presented by the author in previous works (Collins and Leathley, 1995; Collins and
Thompson, 1997a, b).
Schoemaker and Days proposal for dealing with psychological error is to employ a
series of strategies aimed towards sense-making and include testing multiple
hypotheses, canvassing the wisdom of the crowd and developing diverse scenarios. A
PESTLEWeb model is intended to support these activities since, as argued above, it
aims to capture some of the rich complexity of the business environment. A
PESTLEWeb model is not principally reductionist, since it does not reduce observations
of the environment into discrete atomic facts. Rather, it enables diversity in sense-
making because its core component of discourse is a thread or causatively related
structure of potentially contradictory or uncertain elements.
As described in Section 3.1.2 above, the Congregate Cognitive Maps described by

Bougon (1992) provide an excellent model for the way in which PESTLEWeb models
might be used in practice to canvas the wisdom of the crowd, i.e. for models to be
constructed collaboratively by analysts and/or leadership teams and then exposed for
wider review.
Of course, such crowds need not be confined to a single firm. Considering the view of
Astley described in Section 3.1.1 (that firms may operate in a co-evolutionary manner
with their environments), PESTLEWeb models may provide a mechanism to both
identify and work with exogenous partners.
( 23 )
3.3.2 Evidence based on other examples of visual argumentation
PESTLEWeb models are but one example of many semi-formal, graphical tools for
thinking that have been employed from disciplines including business and strategy,
philosophy, law, government policy development, computer software development,
capture of design rationale, analysis of systems safety and analysis of soft, often
social, systems.
For example, the SPIRE method (which shares a number of features in common with
PESTLEWeb) has been described by Klein and Newman (1980). The SPIRE approach
is described as a Systematic Procedure for Identifying Relevant Environments
(SPIRE). Whilst SPIRE ultimately results in a graphical representation of the
relationships between issues that resembles PESTLEWeb models, it is not, at its core, a
graphical method. Rather, issues in the environment are coded with their relationships
into text, being formatted in a computer-readable form. These issues and relationships
are intended to be generated by analysts or business leaders based on inspection of the
firm and its environment. Computer software is then used to identify environmental
issues that have common causes and consequences and to group these into clusters for
graphical display.
The paper by Klein and Newman is motivated by similar concerns to this thesis and
makes similar arguments in a number of regards. However, it does appear to place a
greater emphasis on software for the identification of common groups rather than
considering the construction of the graphical argument as part of the intellectual
discovery process, as argued here. A final distinguishing feature of the SPIRE method is
that the groups of issues are not typed by categories such as Political, Social etc.
Rather, the clusters are grounded in an analysis of the problem domain. The analogue
here is the Grounded Theory of Glaser and Strauss (1999), in which categories are
obtained from the domain of study rather than imposed upon it.
Eden has been influential in the development and use of cognitive maps within
organisations (Eden, 1989, 1992; Eden et al, 1992):
( 24 )
Cognitive maps can be seen as a picture or visual aid in
comprehending the mappers' understanding of particular, and
selective, elements of the thoughts (rather than thinking) of an
individual, group or organization. They may also be seen as a
representation that is amenable to analysis by both the mapper and
others.
Eden (1992)
Cognitive maps share many similarities with PESTLEWeb models but are generally less
constrained. That is, although they serve to map a series of ideas or issues and show
linkages between concepts, those issues are not typically typed (categorised).
PESTLEWeb models might thus be considered as a domain-specific example of
cognitive maps. The relationships encoded in cognitive maps have a variety of
meanings and may be simply bi-relational, directed (uni-relational) or causative.
The notion of visual argumentation seems to have arisen first with the work of Steven
Toulmin (Toulmin, 1958; Schmidt 1986). Toulmin proposed a system of six interrelated
components which could be applied to the analysis of arguments. These ideas may be
employed both to analyse arguments, to represent them and to construct them. An
example of a legal argument constructed to the Toulmin model is shown in Figure 9.
Figure 9: Example of Toulmin-Style Legal Argument, redrawn from http://repgrid.com
( 25 )
Notice that the Toulmin model is typed by the various components of the argument
(Claim, Modality etc.). This is similar to the PESTLEWeb model, but it is even more
constrained since there is defined syntax of relationships between typed elements, e.g. a
Backing always applies to a Warrant and is not syntactically correct in other
positions.
The Toulmin model of argumentation raises another question in relation to

PESTLEWeb: does PESTLEWeb in fact constitute a language for argumentation?
PESTLEWeb provides a syntax for expressing implication/causation, and for
relating issues. It does not, however, provide the equivalent rich syntax for the
elements of argument that are present in the Toulmin model. The rationale for
PESTLEWeb in its present form is that it is intended to be relatively easy to learn.
Introduction of Toulmin-style structural elements might improve the richness and
formality of the argument, but may also make the language more difficult to learn and
use. This question is beyond the scope of the research presented in this thesis, which is
limited to an investigation of the core PESTLEWeb method. However, it does provide
an opportunity for further comparative research with more complex notations.
The use of graphical legal arguments in the Toulmin style to support jury decision
making have been investigated experimentally by Tesa ov (2007), with interesting
results. Whilst mean performance levels did not vary between graphical and textual
arguments, there was a statistically significant difference in the spread of performance.
That is, there was a significant degree of individual difference. Some individuals did
very much better using graphics opposed to text, whereas some did rather worse. The
implication would be that providing a choice of medium for argumentation would be
more appropriate than an absolute adoption of one or the other.
Within the domain of large-scale systems engineering, the author has developed a
graphical notation intended to capture the rationale for complex design decisions the
Essential Logic Model (ELM) (Collins 1997a, b). Perhaps surprisingly, such decisions
( 26 )
share a number of features in common with an analysis of the business environment. In
particular, items in the model may be:
1. Mutually contradictory;
2. Dynamic (subject to more or less rapid change); and/or
3. Uncertain.
Indeed, such features of the model and of the underlying reality it represents are a key
motivator for constructing such a model; the model enables the complexity to be
externalised and managed. PESTLEWeb adopts some of the notational conventions of
the ELM in that the graphical elements representing relationships in the model
distinguish between those that are purely relational and those that have a causative
component.
In contrast to the previous diagrammatic models, Checklands Soft System

Methodology introduces conceptual models of human activity systems (Checkland,
1993) that have a less formal syntax (see Figure 10). This example represents the flow
of information or material, but Checkland uses essentially the same syntax and
notational conventions to represent influence between elements in the analysis and
sometimes the sequencing of activities. This looseness in syntax is in sharp contrast to
the notations used by business analysts and computer scientists to represent technology
systems (Hoare, 1985; Shlaer and Mellor, 1992; Selic et al, 1994; Friedenthal et al
2009).
Figure 10: Conceptual Model of an Order-Processing System, Redrawn from Checkland 1993
( 27 )
PESTLEWeb sits somewhere between these extremes; elements of the model are
typed, and there are different notational conventions for two distinct classes of
relationship. However, no constraints are imposed on the order of elements and their
relationships or their cardinality.
3.4 Conclusion
This thesis addresses the question Is there a better way to analyse the business
environment? This question arises from the duel observations that (1) modelling the
business environment is an important prerequisite for strategy formation (Section 3.1.1)
and (2) that the process is frequently flawed and often entirely absent (Section 3.2.1).
This review provides some strong indicators of where such an improved method might
be found. In particular, a case is made for the use of visual argumentation as a tool for
conducting and representing analysis of the business environment (Section 3.3). This
argument is based in the literature of cognitive psychology, business strategy, law and
education, with reference to other disciplines that have effectively employed such
techniques to address similar issues.
Having made a theoretical argument for the use of visual argumentation to analyse and
represent the business environment, the remainder of this thesis seeks to investigate a
specific example of a visual language designed specifically for that purpose, namely
PESTLEWeb. The investigation is structured to reflect the literature that has been
presented. Thus, aspects of the method in relation to cognition are first investigated,
followed by a study that is more focused on the practical and pragmatic aspects of such
a method; teaching, learning and application.
( 28 )
4 Investigation
4.1 Conceptual framework
A conceptual model for this investigation, as recommended by Hair et al (2007), is
presented in Figure 11. This diagram identifies and categorises the key factors leading
to effective analysis of the business environment, based on Section 3.
Figure 11: Conceptual Model of Factors Leading to Effective Analysis of the Business Environment
( 29 )
The structure of the investigation reflected the conceptual model and included:
1. An experimental procedure intended to rigorously test the hypothesis

relating to the cognitive and affective properties of the method; and
2. An observational procedure to test the pragmatic application of the tool,

conducted in two parts:
a. A workshop-based observational procedure intended to test the method

in a more realistic, practical setting.
b. A longer-term follow up to the workshop session to investigate adoption

and use of the method by students.
To summarise the research design according to the categories of Hair et al (2007):

Procedure (1) was specifically a causal research design intended to test the cognitive
and affective properties of the PESTLEWeb method in a rigorous, quantitative fashion.
Procedure (2a) was a descriptive research design intended to develop frequencies of
variables associated with learning. Procedure (2b) adopted an exploratory design which
is essentially discovery oriented; Students were taught how to use the method and the
results were observed.
Considering the fixed vs. flexible taxonomy of Robson (2002), Procedure (1) was a
Fixed Experimental design and Procedure (2a) was principally a Fixed Non-
experimental design. However, as is explained below, it also included elements of a
flexible, ethnographic study. Procedure (2b) employed a flexible design of non-
structured observation.
Procedure 1 is documented separately from procedures 2(a) and 2(b) in the following
sections. 2(a) and 2(b) are discussed together since they are essentially short-term, and
long-term aspects of the same activity. A final section brings together the results in a
final conclusion. Figure 11 indicates the relationship between the key factors driving
effective analysis of the business environment and the elements of the study intended to
shed light on those factors.
( 30 )
4.2 Ethical considerations
It was recognised that both experimental and observational studies may raise ethical
issues, such as that of the absence of consent for covert observation (Gray, 2005; Hair et
al 2007). Consequently, a review of the designs for both procedures was conducted with
regards to such issues. It was concluded that both procedures were free of such issues on
the following grounds:
1. All participation was on the basis of informed consent.

2. Participants were all competent adults, free to opt out of the procedures at
any time.
3. The designs were completely open and honest there was no covert
observation and participants were clearly informed about the intention of the
research and the extent of their involvement.
4. All results were made anonymous. Participants were given the option of
joining a mailing list to receive the results of the research, but identities were
not correlated with specific results from questionnaires or response papers.
5. In relation to responsibilities towards sponsors, staff of Henley Business
School were completely cognisant of the intended procedure and there was
judged to be negligible risk to the schools reputation.
4.3 Procedure 1: An experimental investigation of cognitive

and affective responses to PESTLEWeb models
4.3.1 Objectives
The primary objective of Procedure 1 is to test, under experimentally rigorous
conditions, whether there are differences in cognitive and affective responses between
PESTLEWeb models compared to purely textual descriptions of industry contexts. The
procedure is intended to test if there are measurably different properties of subjects
responses to the proposed graphical method compared with traditional text-based
descriptions. Such a test is important because we would like to understand if the
PESTLEWeb method has desirable properties that would indicate its efficacy in use. In
particular, it is useful to understand its use in a business school teaching context.
( 31 )
Considering the alternate outcome, it should be noted that, if there were no measurable
differences between the proposed graphical method and traditional text-based
presentations, this would call into question the notion that there were any advantages in
using the graphical method.
The hypotheses tested in this procedure fall into two categories:
1. The Cognitive: That is, is there a difference between the two forms of
presentations in peoples ability to remember the information?
2. The Affective: That is, people have different opinions about the two modes
of presentation. Do they judge one to be better than the other? In particular,
do people regard one as being more persuasive, engaging (interesting) and
rational than the other.
Two key hypotheses to be tested may thus be postulated:
H1 : There is a significant difference in subjects ability to remember information

relating to the business environment presented as a PESTLEWeb model, as
compared with presentation in a text-only format
H2 : There is a significant difference in subjects opinions about the suitability of

presenting information relating to the business environment as a PESTLEWeb
model, as compared with presentation in a text-only format
The corresponding null hypotheses are:
H1o: There is no significant difference in subjects ability to remember information

relating to the business environment presented as a PESTLEWeb model, as
compared with presentation in a text-only format
H2o: There is no significant difference in subjects opinions about the suitability of

presenting information relating to the business environment as a PESTLEWeb
model, as compared with presentation in a text-only format
A secondary objective of Procedure 1 is to investigate if various sub-groups (based on

gender and educational background) demonstrate different responses to the
( 32 )
PESTLEWeb as opposed to a textual mode of presentation, as compared with the total
population of participants.
4.3.2 Design
The experiment was designed to replicate a traditional lecture in a highly controlled
manner. That is, for information on a business environment to be presented using
PowerPoint slides with a recorded, descriptive voice-over. The experimental set-up was
similar in many regards to that employed by Tesa ov (2007) to investigate the use of
legal argumentation. However, difficulties associated with ensuring good control
between runs for that procedure were overcome by the use of a recorded and strictly
timed voice-over.
The hypotheses were tested in a classic, randomised, between-groups experimental

design (Miller, 1986). The intention was to enable the results to be analysed using a
students t-test to determine if there were any statistically significant differences
between the two experimental conditions (Rowntree, 1981).
Subjects were volunteers from two different intakes of postgraduate level business
students studying for Masters of Business Administration (MBA) at a UK business
school. Subjects were randomised between two experimental conditions by distributing
randomised (shuffled) invitations to experimental sessions, hence mitigating against
assignment bias. Subjects were provided with no prior information about the content or
subject matter of the experimental sessions other than that they were to attend an
experiment in support of an MBA Management Challenge.
Subjects were each given a response booklet comprising both categorisation questions
(gender, educational background) and questions relating to the content of the procedure
(see Appendix A). Subjects were given sufficient time to complete the initial
classification before the procedure proper was started. Subjects were asked to provide
responses on a ten-point Likert-type, interval scale (Hair et al, 2007).
( 33 )
Groups were assigned to two different experimental conditions. Both were presented
with an identical, pre-recorded voice-over. The majority of slides were identical, but
the key information slide differed in each case. In the first condition, the information
was presented as PESTLEWeb model (see Figure 13 ). In the second case, the
information was presented in text form (see Figure 12).
The technique of using a completely automated presentation resulted in a high degree of

control between the two experimental conditions. Thus:
1. All spoken description was identical between the two conditions;
2. The timing of information presentation was automated, and identical

between the two presentations; and
3. All hesitations, errors of speech and mispronunciation were edited out of the
pre-recorded audio.
The content of the text in condition (2) was closely based on the PESTLEWeb model in
condition (1).
The response questionnaire was divided into two primary sections:
1. Questions intended to test H1 (five questions); and
2. Questions intended to test H2 (three questions).
A final question was related to H1and required the subject to replicate as much of the
respective information slide as possible. This was intended to discover if subjects could
replicate the general form of the argument, whilst not depending on any specific
content.
Questions in group 1 (testing H1) consisted of required recall of the content of the
information slides with a range of cognitive difficulties, i.e. varying from those that
required simple recall of the contents of one item to those that depended on both recall
of facts and the relationship between elements.
( 34 )
Environmental
Introduction of
Road Congestion Carbon Taxing
PESTLEWebTM
Environmental
Opportunity!
Growing public Increasing political Increased
Pollution from awareness of will to address government funding Increased demand
operation of cars green issues environmental for green for efficient, low
issues technologies pollution vehicles
PESTLEWebTM
Environmental
Pollution from
Production of cars
PESTLEWebTM
Environmental
Easy to each oil has Shift towards
already been purchase of fuel
extracted efficient vehicles
PESTLEWebTM
Long-term upward
trend in oil prices
Increasing demand
for oil
Figure 12: PESTLEWeb Version of the Business Environment Information
1. There is increasing political will to address environmental issues

2. This political issue is a consequence of the growing social issue
of public awareness of green issues
3. The issue of growing public awareness of green issues, in turn
arises from the environmental issues of road congestion, pollution
from operation and pollution from production of cars
4. The consequences of increasing political will to address
environmental issues is the introduction of carbon taxing (an
economic issue) and increased government funding for green
technologies (a technological issue)
5. The issue of increased public awareness of green issues also
leads to the consequence of an economic shift towards purchase
of low-emission vehicles
6. This shift is, a consequence of the economic long-term upward
trend in oil prices
7. The upward trend in oil prices is a consequence of both the
economic issue of increasing demand for oil and the
environmental issue that the easy to reach oil has already been
extracted
8. The consequence of the introduction of carbon taxing, increased
government funding for green technologies and the shift towards
purchase of fuel efficient vehicles creates the opportunity of
Increased demand for efficient, low pollution vehicles
Figure 13: Text Version of the Business Environmental Information
( 35 )
4.3.3 Pilot study for Procedure 1
The principle aims of the pilot study were to:
1. Ensure that the technology for the experiment (the automated presentation)
functioned correctly and was presented at a practical speed; and
2. The response paper (Appendix A) was clear and unambiguous.
The pilot was conducted first informally with group of four adults who had little or no
experience with PEST analysis. A second pilot of the presentation and questionnaire
was conducted with the authors MBA study team; a group of five adults with
significant familiarity with PEST analysis.
In both cases, the technology of the experiment worked well and there were no issues in
terms of comprehending the questionnaire.
4.3.4 Findings and analysis

4.3.4.1 Description of subjects
A total of 47 subjects took part in the experiment (n=47), broken down as shown in
Table 1.
Condition Male Female Total
Group 1 PESTLEWeb 14 6 20
Group 2 Textual 21 6 27
47
Table 1: Constitution of Subjects
( 36 )
Educational Focus Number
Business 15
Engineering 8
Engineering, Science 4
Science, Business 4
Science 3
Business, Engineering 2
Social Sciences, Business 2
Business, Arts 1
Business, Law 1
Social Sciences 1
Law 1
Law, Arts 1
Law, Business 1
Law, Social Sciences, Arts, Business 1
Science, Business, Engineering 1
Science, Social Science 1
47
Table 2: Subject's Responses to Categorisation Question on Previous 'Educational Focus'
4.3.4.2 Differences between the two groups: Results of Testing H1o and H2o
An analysis of the responses of the two groups is presented in Table 3. In this table,
groups have been compared using a 2-sample t-test. Results significant at the 5% level
are highlighted in red.
It can be seen that for the Cognitive (memory) group, whilst there is some difference
between the two groups, this is not statistically significant in most regards. Only Q2
shows a p value less than 5% (3.76%). Thus, these results do not provide sufficient
evidence to reject the null hypothesis H1o. This result holds true for each item treated
( 37 )
individually, as well as for the set of items Q1 to Q5 being treated together (Sum
Cognitive).
The results from the Affective group, however, show a strongly statistically significant
result. Each of the items Q6 to Q8 shows a significant difference well below the 1%
level. Thus, there is sufficient evidence here to reject the null hypothesis H2o.
( 38 )
Cognitive Affective
Sum Sum
Item Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Sum all
Congnitive Affective
Group 2 Mean 3.00 1.74 2.22 2.04 1.89 4.85 3.26 6.33 25.3 10.9 14.4
Group 2 SD 1.24 0.90 1.50 0.76 1.22 2.63 2.18 2.25 12.7 5.6 7.1
Group 1 Mean 3.15 2.55 2.95 2.05 2.15 6.80 6.95 8.35 35.0 12.9 22.1
Group 1 SD 1.35 1.47 1.76 1.28 1.27 1.96 1.50 1.46 12.0 7.1 4.9
Difference between SDs -0.11 -0.57 -0.26 -0.52 -0.05 0.66 0.67 0.79 0.6 -1.5 2.1
Mean Difference 0.15 0.81 0.73 0.01 0.26 1.95 3.69 2.02 9.6 2.0 7.7
Std Error Difference 0.38 0.37 0.49 0.32 0.37 0.67 0.54 0.54 3.6 1.9 1.7
df 39.06 29.44 37.11 28.80 40.17 44.99 44.83 44.34 42.2 35.1 44.9
t-statistic 0.39 2.18 1.49 0.04 0.71 2.91 6.87 3.71 2.6 1.0 4.4
p-value (2-sided, Ha: 1 2 ) 0.70 0.0376 0.14 0.97 0.48 0.0057 0.0000 0.0006 0.0114 0.3154 0.0001
p-value (1-sided, Ha: 1 < 2 ) 0.65 0.9812 0.93 0.52 0.76 0.9972 1.0000 0.9997 0.9943 0.8423 1.0000
p-value (1-sided, Ha: 1 > 2 ) 0.35 0.0188 0.07 0.48 0.24 0.0028 0.0000 0.0003 0.0057 0.1577 0.0000
Table 3: Results of 2 Sample t-test for Procedure 1, Comparison of PESTLEWeb vs. Text Presentation (Results significant at the 5% level highlighted in red)
( 39 )
In summary, the evidence suggests that, whilst the PESTLEWeb diagram did not
enhance memory and recall of information relating to the business environment,
subjects nevertheless found it more persuasive, engaging (interesting) and rational.
4.3.4.3 Are the cognitive and affective categories reliable?

The complete set of subject responses was tested for internal consistency using
Cronbach's alpha measure. The results are shown in Table 4. Given the familiar criteria
of reliability of a Cronbachs Alpha score greater than 0.7, it can be seen that only the
Affective group of items are internally consistent.
Question set Cronbachs Alpha
All questions 0.64
Cognitive group 0.33
Affective group 0.78
Table 4: Cronbach's Alpha for Results of Procedure 1
It is thus reasonable to conclude that the Affective measures in this experiment are
internally consistent, whereas the Cognitive measures are not. Subjects tend to have a
consistent (positive) view about the persuasiveness, interest and rationality of the
PESTLEWeb model as compared with text presentation. However, Q1 to Q5 do not
consistently measure participants ability to remember information from one type of
presentation compared to the other.
4.3.4.4 Exploration of further hypothesis: Gender and educational background

No significant statistical difference was found between sub-groups of participants. That
is, there were no statistically significant differences based on either gender or
educational background.
4.3.4.5 Discussion of results and analysis of Procedure 1

This section focuses on a detailed discussion of the experimental results and their
interpretation rather than a more general discussion of the implication of these results
for business education and practice. This latter subject is covered in sections 4.5 and 5 .
( 40 )
The most important finding from this procedure is that subjects did indeed see a
PESTLEWeb model as being more persuasive, engaging (interesting) and rational than
a textual mode of presentation. These subjects were drawn from a group considered to
be representative of the intended users of a PESTLEWeb model; adults with significant
management experience and a postgraduate level of knowledge and skill in business.
This result suggests that PESTLEWeb would be a valuable tool when attempting to
present an analysis of the business environment to such an audience.
It should be emphasised that there was a strongly statistically significant difference

between the two groups. For all of the affective variables, there was less than 1%
probability that this results could have been obtained by chance alone. For Q7, it was
less than a 1 in 1,000 chance that such a result was as a result of chance.
Of course, this is not to say that these results could be generalised without care. This
was a self-selecting group drawn from one business school class and could therefore be
biased in a number of respects. However, for this group the result was statistically
significant and this at least provides a positive indicator for a more general difference.
The second result is perhaps counter-intuitive, but is consistent with previous research
findings. Subjects did not find the pictorial PESTLEWeb model any more or less easy
to remember than a text presentation. This result is consistent with the work of Craik
and Lockhart (1975, 2008), who argued that it is depth of cognitive processing that
leads to enhanced memory rather than mode of presentation per se. This suggests an
opportunity for further investigations in which memory effects are tested not for
presentation of PESTLEWeb models, but for their production. That is, to test if subjects
are better able to remember such analysis if they are actively involved in the production
of the PESTLEWeb model.
It should perhaps be considered if these two are results inconsistent. Are subjects
(affective) judgments about the method consistent with their performance in purely
cognitive (memory) terms? The argument is that these results are not inconsistent. That
a person is persuaded by an argument is not necessarily contingent on them being able
to remember the details of that argument. It only depends on their judgment that they
were persuaded by the argument as it was presented. The same is true of an arguments
( 41 )
logicality and rationality. Moreover, the memory test is a rather low-level test of
subjects ability to recall content, whereas the affective measures are indicative of a
higher-level intellectual process representing a change in mental state.
Without descending into abstract philosophy, a final point should be appreciated;

subjects were not tested on the extent to which they were actually persuaded by the
argument, but rather they were asked their opinion on the persuasiveness of the
presentation. These are subtly but crucially different things. Indeed, one could design an
experimental procedure that directly addresses the question of the extent to which
subjects were persuaded by the argument. Thus, subjects might be asked: To what
extent do you agree with the conclusion that there is an opportunity for fuel-efficient
vehicles rather than a question such as Do you think the presentation was
persuasive? This suggestion provides another opportunity for further research in this
domain.
Considering the test of sub-groups of subjects, there were probably too few individuals
in each sub-group to determine a statistically significant difference even if such a
difference had indeed been present. It is thus suggested that further runs of Procedure 1
be conducted in order to increase the sub-group size. It would be valuable to know if
there was a difference in performance between textual and visual argumentation of this
type between sub-groups based on gender, language, educational background and stated
preference in learning styles. Such information would enable learners and analysts
alike to select a method of presentation more appropriately for a given audience.
4.4 Procedure 2: Workshop-based observation and longer-

term follow up
4.4.1 Objectives
This study was designed to determine whether PESTLEWeb could be used productively
within an educational setting. That is, rather than just being a theoretical tool of
academic interest, could it actually be used to enhance student learning and analysis of
the business environment? A further objective was to determine if there was any
application of learning subsequent to the initial learning activity.
( 42 )
The questions to be addressed in this observational study were:
Would students
1. Be interested in the method?

2. Be able to understand the method in the time-frame of a one-hour tutorial
workshop?
3. Engage with the method to the extent that they could apply it to a short case
study?
4. In the longer term (post workshop), adopt the method productively in their MBA
assignments and/or workplace?
Additionally, through the use of pre- and post- workshop questionnaires, there was
an objective to gather information relating to:
1. Their companys current ability to scan the business environment;

2. Their use and opinions of other, traditional methods of analysing the business
environment;
3. Their opinions about the PESTLEWeb method as presented in the workshop;
and
4. Their views on the likelihood of their future use of the method.
As recognised above, information gathered in an observational study is invariably less

definitive than that gathered through experimentation. However, the questions
addressed in Procedure 2 strive to get to the crux of the matter: Is PESTLEWeb suitable
for use by MBA students? Can it be taught in a sufficiently short timescale? Can
students understand it? Can they use it? Will they use it? Do they believe it helps them
perform deeper and richer analysis? Having gained a working knowledge of the
method, will they choose to apply it in practice?
It is these latter questions on which the adoption and use of the PESTLEWeb method
rests.
4.4.2 Design
4.4.2.1 Method of evaluation
The design intent of this procedure was to replicate, as far as possible, a realistic
teaching session within an MBA context and then to monitor the longer-term outcomes
( 43 )
of the training. This is appropriate since it is exactly this context in which the
PESTLEWeb method is intended to be applied in the first instance.
A common model for evaluating learning of this type is the Kirkpatrick Levels of
Training Evaluation, as outlined in Table 5 (Kirkpatrick and Kirkpatrick, 2009).
Level 1:
To what degree participants react favourably to the learning event
Reaction
Level 2: To what degree participants acquire the intended knowledge,

Learning skills and attitude based on their participation in the learning event
Level 3: To what degree participants apply what they learned during

Behaviour training when they are back on the job
Level 4: To what degree targeted outcomes occur as a result of the learning

Results event and subsequent reinforcement
Table 5: Kirkpatrick Levels of Training Evaluation
The procedure described in this section was focused on levels 1 to 3 of the Kirkpatrick
model. Levels 1 and 2 were evaluated observationally during the workshop session.
Level 3 was evaluated through a longer-term follow up with students.
A design decision was made to avoid the use of traditional happy sheets measuring the
Level 1 (Reaction) responses of participants. This decision was made for several
reasons:
1. Given the skewed, self-selecting nature of the group and the ambiguous
nature of the presenter being a fellow student, it was considered unlikely that
the results would be free of compliance effects.
2. Level 1 reactions were available observationally in the behaviour of the

subjects (e.g. level of engagement and activity).
3. The developers of the Kirkpatrick model themselves argue that, in studies of

the value of training, there is often an unfortunate bias towards Level 1
measures, where in fact it is levels 2, 3 and 4 that provide the most valuable
( 44 )
indicators of training value (Kirkpatrick and Kirkpatrick, 2009; Kirkpatrick
and Kirkpatrick, 2010).
4.4.2.2 Workshop design

The workshop involved the presentation of the slides included as Appendix E. At slide
14, subjects were asked to trial the PESTLEWeb method as part of a structured activity.
Subjects were given a short case study which they were asked to analyse using the
PESTLEWeb method. Specifically, they were asked to identify PESTLE issues
through silent brain-storming, copy these issues to post-it notes and then arrange these
into a PESTLEWeb model on a white-board. Subjects were then asked to critically
review their results to consider dynamics and additional causes and consequences in
their model.
The observational procedure used in this part of the study cannot be easily categorised
using the familiar classes of qualitative research. However, it has elements that are
familiar from ethnography (Hair et al, 2007), focus groups and action research
(Eriksson and Kovalainen, 2008). Note that there was no intention for the observer to
judge the quality of the analysis produced by subjects. It was felt that this was not
compatible with the type of activity taking place in the workshop, which was principally
pedagogic. Rather, the intention was for subjects to judge the usefulness of the method
and their ease of learning it.
As well as observations conducted during the session, a significant part of the data
gathering for this investigation was achieved through pre- and post- workshop
questionnaires. These provided the participants the opportunity to provide feedback on
the workshop and to address specific questions relating to the study objectives.
It was recognised at the outset that the likely number of subjects for this study would be
insufficiently large to draw significant, generalisable conclusions. Additionally, subjects
represented a skewed population since they were self selecting. Thus, it was recognised
that any results would be indicative rather than conclusive.
( 45 )
4.4.2.3 Design of the pre- and post- workshop questionnaire
The intention for both pre- and post-workshop questionnaires was not to test specific
hypotheses, but rather to gather descriptive statistics concerning the responses. The
intention was to shed light on the learning experience itself and the pre-workshop
context for that learning.
The pre-workshop questionnaire was designed to provide a baseline of subjects

current attitudes and use of business environmental scanning methods as taught on their
MBA course. The core intent here was to discover if there was a need for an improved
method in the minds of the participants.
Subjects were asked to respond on a five-point, Likert-type interval scale (Hair et al,
2007). The intention was to report findings using straightforward descriptive statistics
(mean, standard deviation etc.) in both numerical and graphical form. A five-point scale
was selected following the recommendation of Cox (1980). It was noted that there has
been some criticism of odd-number Likert scales including a neutral mid-point, but that
the literature is generally equivocal on this matter (Garland, 1991). Note however the
comments on this subject in Section 4.4.4.6.
Question 1 focused on the subjects current business and their systems for scanning the
business environment. This question was driven by Day and Schoemaker (2006), who
identified that there is a common business issue associated with a failure to scan the
business environment. Thus, even if subjects had not identified an issue in their current
skills and behaviours in scanning the business environment, a lack of a formal system
within their business would signal a latent need.
Questions 2, 3 and 4 were intended to probe the subjects pre-workshop use of the
methods for scanning the business environment as taught on their existing MBA course.
In particular, they focused on analysis for assignments, exam case studies and the
context of their business.
Question 5 was not intended to elicit information about the subjects analysis of the
business environment, but rather their views on the ease with which the results of that
analysis can be communicated with others.
( 46 )
The post-workshop questionnaire was designed to provide information focused at level
2 (Learning) and level 3 (Behaviour) in the Kirkpatrick model. As a set, the post-
workshop questions are intended to probe a single construct in the minds of the
participants: is the PESTLEWeb method better than the methods than subjects have
previously been taught? That is, do participants believe:
1. It is relatively easy to learn? (Q1)

2. It provides better explanations than other methods they have experienced?
(Q2)
3. Using it would improve their analysis? (Q3)
4. That they in fact intend to use it for MBA assignments? (Q4)
5. That they would like to learn more about the method? (Q5)
6. That they would like to collaborate in the use of the method? (Q6 and Q7);
and
7. That the method would be useful for their business? (Q8)
4.4.2.4 Investigating Kirkpatrick Level 3 results for the workshop

Given the Kirkpatrick model, it was considered important to determine if there was any
evidence of longer-term transfer of learning from the workshop into study or
professional practice. Thus, subsequent to the workshop, attendees were given the
opportunity to join a Community of Practise (CoP) for PESTLEWeb and to share the
results of their analysis (Wenger and Snyder, 2000). The decision to create a CoP arose
directly from the authors engagement in the Knowledge Management module of the
Henley MBA. People who joined the workshop were subsequently requested to share
their own PESTLEWeb analysis and asked to comment on their own use of the method.
4.4.2.5 Observational method

Two approaches were adopted for observation. Firstly, the author acting as workshop
leader noted behaviour during the workshop and subsequently recorded this in a journal.
Since it was recognised that there was a difficulty associated with both leading a
workshop and making detailed observational notes, a second observer (a member of the
authors MBA study team) also attended and observed the workshop. Information from
the second observer was collated in a post-workshop debrief.
( 47 )
4.4.3 Pilot study
A pilot workshop was held with the authors MBA study team. This involved an
informal presentation of the workshop slides (Appendix E) and the completion of the
pre- and post-workshop questionnaires (Appendix B).
Whilst the pilot group were already familiar with the PESTLEWeb method, they had
not previously seen the workshop slides, nor had they seen the case study that was to be
used within the workshop.
Some concern was expressed that the case study was too easy in that it made the
various PESTLE issues very explicit in the text. This, however, was tempered by the
view that the workshop group would be unfamiliar with the PESTLEWeb method and
that there was a need to make the information relatively easy to analyse in a short
workshop session.
One significant error was found in the pre- and post-workshop questionnaires; the rating
scale was ranked from Strongly disagree to Strongly disagree. This was, of course,
subsequently corrected.
4.4.4 Findings and analysis

Results for pre- and post-workshop questionnaires are presented in sections 4.4.4.1 and
4.4.4.3. As well as providing a histogram of results for each question, summary
statistics are also provided. In this case, each of the available response options has been
mapped onto a numerical value (Strong disagree = 1 Strong Agree = 5) to obtain
measures of mean, standard deviation, kurtosis and skewness.
4.4.4.1 Observational results for the workshop

It was apparent at the outset of the workshop session that participants were fatigued
after a full day of workshop presentations and activities. Nevertheless, 20 attendees
elected to remain after their study day and begin a second, hour-long workshop at 17:30.
Workshop attendees were attentive during the presentation but asked only two
questions, as follows:
Q. Would you include the PESTLEWeb pictures in you MBA assignments, or just
use a text version of the results?
( 48 )
A. I included the diagrams in my IMP assignment, followed by a section of text
that identified the main threads of the argument and key results
Q. Which tool did you use to draw the PESTLEWeb diagrams?

A. I have used both Microsoft PowerPoint and Visio. I have produced templates for
both of these tools and I will show you later how to access those
Workshop attendees appeared to start the activity at slide 14 rather hesitantly. From the
observers point of view, this appeared to be due to lack of energy after a full day of
study rather than any serious confusion about the method. In particular, attendees
appeared reluctant to move from the stage of identifying an initial list of issues towards
organising them as a graphic model. This is obviously a key step which differentiates
the PESTLEWeb model from traditional PESTLE analysis. The workshop presenter
(the author) moved from group to group, encouraging them to start arranging the issues
they identified into PESTLEWeb models on the flip-charts provided.
Over a period of some minutes, each of the groups did indeed begin to structure their
issues into coherent models with, as might be expected, more or less success. The
workshop presenter moved from group to group and challenged them to review their
models and look for further potential causes and consequences in order to enrich their
models.
After 15 minutes, the workshop leader called a halt to the activity and presented the
remainder of the slides. Attendees were referred to a website containing supporting
information for the presentation. They were finally requested to complete questionnaires
and hand them in as the session ended.
( 49 )
4.4.4.2 Results of pre-workshop questionnaire
Question 1) My company has a well-developed system in place for scanning the

business environment
7
Question 1
6
Mean 2.74
5 Standard Error 0.23
Median 3.00
Frequency
4
Mode 3.00
3 Standard Deviation 0.99
Sample Variance 0.98
2
Kurtosis -0.94
1 Skewness -0.17
Range 3
0
Strongly Disagree Neutral Agree Strongly Minimum 1
Disagree Agree Maximum 4
Bin
Figure 14: Companies Have a System in Place for Scanning the Business Environment
Question 2) I have found it easy to analyse the business environment as part of my

MBA assignments
12
Question 2
10
Mean 3.00
8
Standard Error 0.20
Median 3.00
Frequency
6 Mode 3.00
Standard Deviation 0.88
4 Sample Variance 0.78
Kurtosis 1.30
2 Skewness -1.09
Range 3
0
Minimum 1
Strongly Disagree Neutral Agree Strongly
Bin
Figure 15: Business Environmental Analysis for Previous MBA Assignments
( 50 )
Question 3) I have found it easy to analyse the business environment for my MBA
exam case studies
8
Question 3
7
Mean 2.74
6
Standard Error 0.20
5 Median 3.00
Frequency
4 Mode 3.00
3
2 Kurtosis -0.70
1
Skewness 0.01
Range 3
0
Minimum 1
Bin
Figure 16: Subjects Ease of Analysing Business Environment for MBA Exams
Question 4) I think that PESTLE analysis gives me deep insights into the context for my
business
7
Question 4
6
Mean 3.05
Median 3.00
Frequency
4
Mode 3.00
2
Kurtosis -0.55
1 Skewness -0.11
Range 4
0
Disagree Agree
Maximum 5
Bin
Figure 17: Traditional PESTLE Analysis Provides Deep Insights
( 51 )
Question 5) I find it easy to explain the context of my business to others
12
Question 5
10
Mean 3.21
8
Standard Error 0.24
Median 4.00
Frequency
6 Mode 4.00
Kurtosis 0.26
2 Skewness -1.15
Range 3
0 Minimum 1
Bin
Figure 18: Subjects Ease of Explaining the Context of their Business to Others
4.4.4.3 Results of post-workshop questionnaire
Question 1) I found the PESTLEWeb method straightforward to understand
8
Question 1
7
Mean 3.84
6
Standard Error 0.26
5 Median 4.00
Frequency
4 Mode 4.00
3
2
Kurtosis 0.86
1 Skewness -0.99
0
Range 4
Disagree Agree
Maximum 5
Bin
Figure 19: Subjects Ease of Understanding the PESTLEWeb Method
( 52 )
Question 2) A PESTLEWeb model provides a better explanation of the business
environment than traditional PESTLE
10
9 Question 2
8 Mean 4.11
6 Median 4.00
Frequency
5 Mode 4.00
3
Kurtosis -1.00
2
Skewness -0.17
1
Range 2
0
Bin
Figure 20: PESTLEWeb Provides a Better Explanation than Traditional PESTLE Analysis
Question 3) Using PESTLEWeb would improve the quality of my analysis of the

business environment
12
Question 3
10
Mean 3.84
8
Standard Error 0.23
Median 4.00
Frequency
6 Mode 4.00
Kurtosis 2.65
2 Skewness -1.44
Range 4
0
Minimum 1
Bin
Figure 21: Subjects Belief that use of PESTLEWeb Would Improve the Quality of their Analysis
( 53 )
Question 4) I will use the PESTLEWeb method for my future MBA assignments
7 Question 4
6 Mean 4.00
Standard Error 0.19
5
Median 4.00
Frequency
4 Mode 4.00
3
2 Kurtosis -1.48
Skewness 0.00
1
Range 2
0 Minimum 3
Maximum 5
Disagree Agree
Bin
Figure 22: Subjects Intention to Adopt the PESTLEWeb Method
Question 5) I would like to learn more about the PESTLEWeb method
10
9
Question 5
8
Mean 4.21
7
Standard Error 0.16
6 Median 4.00
Frequency
5 Mode 4.00

3
Kurtosis -0.82
2
Skewness -0.34
1 Range 2
0 Minimum 3
Strongly Disagree Neutral Agree Strongly Maximum 5
Disagree Agree
Bin
Figure 23: Subjects Enthusiasm to Learn More about the PESTLEWeb Method
( 54 )
Question 6) I would like to see PESTLEWeb models for my industry produced by
others
8 Question 6
7 Mean 3.84
6
Standard Error 0.23
Median 4.00
Frequency
5
Mode 3.00
2 Kurtosis -1.51
Skewness -0.01
1
Range 3
0
Minimum 2
Bin
Figure 24: Subjects Enthusiasm to Access PESTLEWeb Models Produced by Other People
Question 7) I would be willing to share my PESTLEWeb models with other people
7
Question 7
6 Mean 3.53
Standard Error 0.23
5
Median 3.00
Frequency
4 Mode 3.00
3
2 Kurtosis -1.00
Skewness 0.10
1
Range 3
0 Minimum 2
Disagree Agree
Bin
Figure 25: Subjects Willingness to Share PESTLEWeb Models with Other People
( 55 )
Question 8) PESTLEWeb would be useful to my business
7 Question 8
6 Mean 3.84
Standard Error 0.21
5
Median 4.00
Frequency
4 Mode 4.00
3
2 Kurtosis -0.77
Skewness -0.18
1
Range 3
0 Minimum 2
Disagree Agree
Bin
Figure 26: Subjects Views Concerning the Usefulness of PESTLEWeb to their Business
4.4.4.4 Sample size for Procedure 2a

In all, 20 questionnaires were submitted with one of these being only partially complete
and hence un-usable. All of the statistics in the above section are descriptive and based
on the population as measured, as opposed to being a sample for inferential purposes.
4.4.4.5 Results relating to Kirkpatrick Level 4 transfer of learning

As stated in Section 4.4.2.4, workshop attendees were offered the chance of joining a
CoP focused on the PESTLEWeb method. This CoP was convened via a LinkedIn
Group, as shown in Figure 27. At the time of writing, the CoP had attracted 35
members, 31 of whom are Henley MBA students and the remainder are colleagues of
the author.
After the workshop, a number of PESTLEWeb CoP members developed their own
PESTLEWeb models, either for personal interest or with the intention of submitting
these as part of an assignment towards the Henley MBA. Examples of these models are
included in Appendix C.
Each of these models has a significant level of detail both in terms of the number of
issues identified and the logic that ties them together. It is somewhat difficult to judge
( 56 )
the quality of the analysis from these diagrams alone, since presumably that judgment
would depend on the various authors final conclusions and inferences based on the use
of the model. However, it can be said that the authors of these models have been able to
execute them successfully; that is, they are syntactically and semantically correct
according to the PESTLEWeb drawing convention. Additionally, the authors have
demonstrated their preference for using the PESTLEWeb method over traditional
methods.
That the PESTLEWeb method has been adopted in this way, in a relatively short
timescale and with little promotion, may be the most significant indicator of its
usefulness.
Figure 27: PESTLEWeb CoP Webpage
( 57 )
4.4.4.6 Discussion of results and analysis of Procedure 2
The mean response to each question in the pre-workshop questionnaire is close to 3.
This suggests that an average subject in this study group is not currently experiencing
great difficulties in the area of business environmental analysis. However, that a
significant number of subjects scored at the mid-point of the scale suggests that, in
follow-up studies, it would be useful to increase the number of options available or to
remove the neutral mid-point (Cox, 1980; Garland, 1991).
However, a more important observation may be that a significant fraction of subjects

responded in a negative manner. Thus, 42% of subjects either disagreed or strongly
disagreed with the statement My company has a well-developed system in place for
scanning the business environment (Question 1). This result would be broadly in line
with the industry survey reported by Day and Schoemaker (2006). 32% of subjects
either disagreed or strongly disagreed with the statement I think that PESTLE analysis
gives me deep insights into the context for my business (Question 4). 21% of subjects
either disagreed or strongly disagreed with the statement I find it easy to explain the
context of my business to others (Question 5). It is a value judgment as to what a
desirable distribution to these questions might be. However, it does not seem
unreasonable to suggest that a business school would consider it desirable for business
students on an advanced course of study in this domain to respond more positively, i.e.
with a mean well above the neutral position and significantly fewer responses in the
negative region. Thus, particularly in respect to Question 4, it would be desirable to
equip students with a method of analysis which they judged to provide them deep
insights into the context of their business.
It is recognised, of course, that such a self-assessment questionnaire is clouded by

opinion over fact. Thus, although a respondent might lack confidence in their ability to
perform deep analysis, their performance in reality may exceed their expectations.
Again, however, presumably a business school would desire an autoepistemic outcome
in which the student both knew how to perform an accurate analysis and was confident
in their own abilities. This state is not reflected in a significant proportion of the
respondents to the pre-workshop questionnaire.
( 58 )
Considering the post-workshop questionnaire, we can note that the mean level of
response for each question is above the neutral position. Whilst there are some negative
responses (e.g. two respondents either disagreed or strongly disagreed with the
statement I found the PESTLEWeb method straightforward to understand), a review
of the graphs in Figure 19 to Figure 26 indicates a marked preference for responses in
the positive.
Concerning the direct comparison of the PESTLEWeb method to the traditional

PESTLE method, Question 2 showed 79% of subjects either agreeing or strongly
agreeing with the statement A PESTLEWeb model provides a better explanation of the
business environment than traditional PESTLE. Similarly, 79% of subjects either
agreed or strongly agreed with the statement Using PESTLEWeb would improve the
quality of my analysis of the business environment.
In relation to level 3 of Kirkpatricks model (Table 5), 68% of subjects indicate an

intention to adopt the use of the PESTLEWeb method in their academic studies
(Question 4: I will use the PESTLEWeb method for my future MBA assignments).
Cronbachs Alpha for the complete set of post-workshop questions is 0.82, showing that
they have a high degree of internal consistency and suggesting that they are measuring a
single construct. As described in Section 4.4.2.3, this construct is intended to address
Level 2 and Level 3 of the Kirkpatrick hierarchy and be an indication of both the value
of the acquired knowledge and the likelihood of future application
In summary then, the pre-workshop results show that, whilst on average students are
not experiencing particular issues in the domain of business environmental analysis,
nevertheless a significant proportion are. The pre-workshop results are probably not
aligned with a desirable position from the point of view of a business school. The post-
workshop results provide a consistent indicator that the proposed PESTLEWeb method
has advantages over existing methods measured in regard of a nexus of items relating to
ease of learning, depth of analysis and likelihood of future use.
( 59 )
4.5 Conclusion and recommendations
The two studies described above together address the efficacy of the PESTLEWeb tool
from two directions. Firstly, a classical between-groups style experiment was
conducted to determine if there were any statistically significant differences between the
presentation of business environmental information in the form of a PESTLEWeb
model as compared with a more traditional text format. This investigation aimed to
determine whether there were differences both in either cognitive or affective responses
between two groups. Secondly, an observational study, supported by pre- and post-
workshop questionnaires, was conducted to provide supporting evidence (or otherwise)
that the PESTLEWeb model could be usefully applied in a business school teaching
setting
As described in previous sections, the first of these procedures had the advantage of the
potential to provide statistically significant measures of difference, under well-
controlled experimental conditions. The disadvantage of this procedure was, however,
that it was only able to investigate a relatively narrow set of conditions and responses
given the available time and resource limitations. The second procedure provided
qualitative evidence in a more realistic teaching setting. It can be regarded as a
legitimate trial of the use of PESTLEWeb but, of course, it would be invalid to
generalise these results to other settings, other student groups and other tutors.
4.5.1 Conclusion and recommendations arising from Procedure 1

The results of the experimental study showed no significant difference in recall of
models presented in PESTLEWeb form vs. traditional text. There was, however, a
strongly statistically significant difference between the two groups in terms of affective
variables. Simply put, subjects believed that the PESTLEWeb models were more
persuasive, engaging and rational than a traditional text presentation.
It is useful to deconstruct these two results. Firstly, why might we expect subjects to
remember more from the PESTLEWeb model, and why is it that in fact they do not?
Perhaps biased by the familiar aphorism A picture is worth a thousand words we are
led to the intuitive, and yet incorrect, view that pictures are intrinsically more
memorable than the written word. However, this view is not born out in the literature of
( 60 )
cognitive psychology (Craik and Tulving, 1975; Craik and Lockhart, 2008), nor by the
experimental investigation presented in this thesis. The theory and evidence instead
points towards the process of remembering rather than intrinsic characteristics of the
thing remembered. As discussed in Section 3.3.1, an essential element of improved
memory and subsequent recall is active transformation by the learner.
Thus, an obvious suggestion for subsequent experimentation would be to investigate if

there are significant differences in recall when subjects construct their own
PESTLEWeb models rather than have them presented in their fully constructed state.
However, such an experiment would itself suffer from some difficulties. Even
establishing a significant difference in these circumstances would not distinguish
PESTLEWeb over any other form of transformation (for example, constructing a rhyme
or drawing a mindmap) and without further work it would do no more than add weight
to the arguments of Craik and Lockhart (1975, 2008). Thus, whilst this is an attractive
direction for further investigation, it is a non-trivial one.
There is another complicating factor in the experimental tests of memory and recall
using PESTLEWeb. Subjects engaged in the experiment were unfamiliar with the
notation; they had not learnt it, nor had they practiced using it. Thus, there was a
significant level of unfamiliarity with the task. Subjects in the experiments were both
learning to interpret a PESTLEWeb model with no prompting or pre-training and also
attempting to remember the content of the presented model. It could easily be that the
cognitive difficulty associated with the interpretation and learning task effectively offset
any potential advantages in memory.
This condition again encourages further experimentation, for example by exposing

subjects to tutorial material to enable them to gain familiarity with the notation. After
the model of de Groot, we might expect that experts would have better memory of the
models than novices as a result of familiarity and ability to chunk elements of the
model into familiar sub-components and patterns (de Groot, 1978). This would be an
attractive notion of course; if business students could be taught a generalisable,
extensible modelling tool in their early years of study it could then be used as a medium
for presenting more complex arguments in economics, business environment and
strategy later on.
( 61 )
Considering the cognitive aspects of PESTLEWeb, there was no attempt made here to
replicate the work of Bauer and Johnson-Laird (1993). That is, it would be interesting to
test specifically if subjects are more or less able to make logical deductions and
inferences from a PESTLEWeb model than from a text-only presentation. Nonetheless,
whilst a direct replication of Bauer and Johnson-Laird using PESTLEWeb models
would be relatively straightforward, the results would again be limited; the original
experiment considered rather specific and difficult to understand logical statements
involving double disjunctions. It is not obvious that a direct replication of the study
would shed light on more common reasoning tasks within the domain of business
environmental analysis. A more general investigation of students ability to reason using
the PESTLEWeb tool would, however, be an attractive line of investigation for future
work.
We turn now to the measure of affective variables, which are important in their own
right (to be of practical use, users would need to find them convincing, rational etc.), but
also important to the extent they contribute to level 3 and 4 outcomes in the Kirkpatrick
Learning Evaluation model shown in Table 5 (Kirkpatrick and Kirkpatrick, 2009).
Whilst there was a statistically significant result, we might ask what exactly that result
was. Subjects judged the PESTLEWeb models to be more convincing than those people
exposed to the text stimulus. This is not to say, of course, that the models were actually
more convincing. There are a number of reasons why subjects might have assessed the
PESTLEWeb model as being more convincing, not least of all compliance effects that
might lead students to conclude that they should judge these models as more
convincing. Another useful avenue of experimentation should be to test if subjects were
actually more convinced by the argument presented in a PESTLEWeb rather than a text-
based argument. Such an experiment would be a natural and straightforward extension
to the experiment described in this thesis, in which subjects would be asked to indicate
their level of agreement with conclusions under the two experimental conditions. Again,
following on from the discussion above, such an investigation would have to deal with
any learning effect associated with an initial use of the tool.
Thus, a key recommendation arising from the experimental section of this study is to
continue with further experiments. Such experiments would serve to tease out the exact
( 62 )
nature of any advantages or disadvantages in PESTLEWeb in particular, and the use of
visual argumentation in general in the context of business in education and in practice.
4.5.2 Conclusion and recommendations arising from Procedure 2

The pre-workshop questionnaire for Procedure 2 revealed that significant numbers of
subjects had a negative experience of previous methods. For example, around half of
subjects considered that their company did not have a well-developed system for
scanning the business environment (confirming the results of Day and Schoemaker,
2006). Views as to whether traditional keyword-driven PESTLE provided deep insights
were evenly split, with as many subjects strongly agreeing as strongly disagreeing.
These results certainly suggest that this is a worthwhile area to investigate.
Procedure 2 demonstrated that PESTLEWeb could indeed be taught, in a reasonably

short time, in a classroom setting at a business school. The post-workshop questionnaire
results were strongly weighted towards the positive end of the response ranges for all
questions. The follow-up study showed that some students were able to use
PESTLEWeb productively to construct relatively sophisticated models of the business
environment.
These results, however, are also open to criticism. That the tool can be taught in one
setting, by one tutor, to one class does not guarantee that the tool could be introduced
universally in a business school setting. To substantiate this argument would require the
tool to be piloted by other tutors and in other settings.
Moreover, that some students adopt the tool and appear to use it effectively is no
guarantee of universal efficacy. For example, this was a self-selecting group. There is
every possibility that those students who were most highly motivated and most attracted
to systematic, analytical tasks elected to use the tool. This evidence therefore provides
little information about the likely effects of the tool being introduced to a general
audience of business school students. For example, for less able students , the added
overhead of learning the methodology might actually work against their ability to
perform useful analysis.
( 63 )
We have also not seen the grade results for students adopting the PESTLEWeb method.
That students choose to use the tool, and produced complex models, is no guarantee
that an independent review by an assignment assessor will give a favourable result. For
this to be achieved, several further conditions need to be in place. Students need to be
able to re-synthesise their analysis into coherent conclusions that find favour with
assessors. Assessors themselves would need to understand the PESTLEWeb notation
and be able to follow the arguments presented (bearing in mind however that students
were advised to translate their conclusions from the analysis into a normal discursive
form rather than leaving them as a PESTLEWeb graphic).
Finally, the opportunity cost associated with constructing a PESTLEWeb model needs
to be considered. For example, it may well be that the insights provided by a light-
weight analysis of the environment may be sufficient to proceed with effective strategy
development, where effort might be more liberally applied. The introduction to this
thesis presents the arguments for why this should not be so, arguing that deep
environmental analysis is indeed warranted.
4.5.3 Final recommendations

The final recommendations arising from this thesis are intended to address the two
primary issues identified within the review of current thinking in Section 3, that:
1. Scanning of the business environment is of critical importance to business

(Barney, 1991, Albright, 2004); and
2. The task is frequently not done well or not done at all (Schoemaker and Day,
2009).
The review of current thinking and the investigations within this thesis have provided a
strong candidate to address these issues. However, more work needs to be done to
further develop understanding in this domain and to promote the practical application of
solutions.
( 64 )
In particular, it is recognised that there will always be limitations when somebody sets
out to evaluate a tool or method that they have created themselves. Whether
intentionally or purposefully, there is an opportunity for skewed results. Considering
this, a key recommendation is that the work of this thesis be subject to replication by
independent researchers. Indeed, there is an opportunity here for other Henley MBA
students to replicate, adapt and test both PESTLEWeb and the whole class of graphical
techniques as part of their own work.
Thus, it is recommended that step-wise piloting and introduction of the PESTLEWeb

method should be continued under conditions of careful monitoring, i.e. a process of
incremental expansion of the use of the tool, coupled with further empirical and
observational research. An incremental programme of research and application would
present a low-risk route towards adoption and use of the method.
Thus, an incremental process is proposed, as follows:
1. Gather feedback from assignment markers for the MBA Strategic Direction
(SD) and Global Business Environment (GBE) sections of the course to
discover their responses to the use by students of the PESTLEWeb method.
2. Run a series of further pilot workshops for the method as addenda to the
existing SD and GBE courses at Henley.
3. Present a research seminar to tutors and other academics at Henley to

introduce the method and gather further critical feedback.
4. Provide tutorial information on the PESTLEWeb method via the Academic

Resource Centre at Henley.
5. Encourage independent replication and validation of the method.
6. Formally move to introduce the method as part of core teaching of SD, GBE
and during introductory modules for the MBA.
7. Complete further observational and empirical studies, including:
a. PESTLEWebs efficacy for example, its effect on reasoning rather

than memory and long-term follow up on student performance; and
( 65 )
b. Comparative studies with more or less complex notations.( See the
comments comparing PESTLEWeb with the Toulmin model of
argumentation in Section 3.3.2)
8. Seek to promulgate the results of the PESTLEWeb research at conferences

and in academic journals with the intended aim of promoting the acceptance
and widespread use of the method beyond Henley Business School.
4.5.4 Final conclusion

This thesis has recognised the key importance of deep and timely knowledge of the
business environment for strategy development and effective, continuing execution.
Importantly, issues relating to effectiveness of the environmental modelling task have
been identified both in professional practice and in business education. Reports in the
literature of a failure of businesses to engage in effective environmental scanning have
been corroborated in this thesis by evidence gathered in a workshop session with MBA
students.
This thesis has introduced a new method for modelling the business environment, called
PESTLEWeb. It has been argued that this method has several properties that increase its
effectiveness as compared with traditional keyword-driven methods of identifying,
analysing and synthesising business environmental knowledge. These positive
properties of the new method have been investigated both experimentally and
observationally in a workshop session. Whilst no evidence was found to indicate that a
PESTLEWeb model was significantly more memorable than a text-based analysis, a
measure of affective variables showed a strong, statistically significant preference for
the new method. The observational study showed that the method could be introduced
to an MBA class in a relatively short and straightforward workshop setting. It was
shown that learners were able to understand and use the method to some extent within
the workshop itself. Additionally, a longer-term follow up of a self-selecting group of
students has demonstrated that some, at least, are able to develop sophisticated and rich
models of the business environment using the PESTLEWeb method.
Recommendations from this thesis have proposed an incremental process of further

investigation and academic engagement with the method with the ultimate aim of
introducing it into business education.
( 66 )
5 Reflection
5.1 Evaluation of findings in relation to current thinking
I consider the most relevant published current thinking in this domain to be the work by
Schoemaker and Day (2009). Their book Peripheral Vision makes excellent cases both
for the role of environmental scanning in business strategy development and execution
and also that this task is frequently performed poorly or not at all. Schoemaker and Day
present some discussion about the interpretation of environmental information and make
suggestions such as Formulate multiple hypotheses, Encourage constructive
conflict and Use dialogue to share the big picture. They do not, however, provide a
concrete methodology for enacting any of these.
For me, the most significant aspect of PESTLEWeb is its contribution to interpretation
and understanding. PESTLEWeb is concerned with sense-making and theory building.
The development and use of tools to enable these processes has been a significant
interest for me for more than two decades; a significant part of my PhD thesis
concerned the development and use of extended metaphors to enable teachers to
contextualise teaching experiences. Students and business leaders do not lack access to
information about the business environment; they do, however, lack a methodology to
first analyse the data in a rich manner and then re-synthesise this into a coherent theory.
A key theme of my own education and research work is that it has always been multi-
disciplinary. The PESTLEWeb work builds on this and draws on literature from a
variety of fields. For example, the process of theory building within PESTLEWeb has
resonances with the processes of qualitative research and in particular the Grounded
Theory of Glaser and Strauss (1999). The use of PESTLEWeb as a tool for thinking
draws on the work of Craik and Lockhart (1975) and de Groot (1978), which I first
came across whilst completing research in a university psychology department. The use
of graphical notations to build models representing complex realities is embedded in my
own teaching in complex systems engineering and in the many graphical notations of
modern computing, IT business analysis and software engineering (Shlaer and Mellor,
1992; Selic et al, 1994; Friedenthal et al 2009).
( 67 )
The PESTLEWeb work, of course, rests centrally on its application in the domain of
business strategy development. The work of Klein and Newman (1980) on the SPIRE
method provides a natural precursor to PESTLEWeb, as do the authors working in the
area of cognitive maps (Eden, 1989, 1992; Eden et al, 1992; Bougon, 1992)
I believe it is important to note, however, that despite its relatively formal appearance,
PESTLEWeb is not primarily intended to be a rigorous and logically correct
modelling language in the sense those terms might be used by a computer scientist or
mathematician. In contrast to the work of Klein and Neman, the PESTLEWeb method is
not intended to be a predominantly computational or algorithmic method. Instead, it
should be considered a humanistic tool that has at its heart the human capabilities of
grouping, arranging, structuring, story-telling, theory-building and communicating.
PESTLEWeb models are stories through pictures, albeit that those pictures adopt a
regular semiotic and syntax. The diagrams allude to formalism as an enabler to
comprehension and communication, but they are not bound by it; for example, no great
attention has been paid in this thesis to the exact semantics of the two classes of
relationships depicted in the notation. That has not been necessary; PESTLEWeb rests
at the balancing point between the recklessly casual and the restrictively causal.
I feel very enthusiastic in terms of the relationship of the research outcomes to the
original management problem. There has been significant interest shown in the
PESTLEWeb method during the course of this research project. This has been most
clearly evinced by the workshop follow-up study described in Section 4.4.4.5. MBA
students at Henley Business School are starting to use the method and there seems no
reason why growth in the use of the method should not continue.
5.2 Experience of the research process

Early on in this project I made a decision to complete both an experimental and an
observational study. I recognised that the experimental work in particular was somewhat
controversial in a business school environment. Whilst not explicitly excluded in the
rules for the Management Challenge, neither was it highlighted as a likely option. I had
not come across a single, relevant example of an experimental investigation in my
( 68 )
literature research (see the entry in my research diary for Thursday 16 September
included at Appendix D).
Why then, select such a potentially difficult option? Firstly, in many respects,
experimentation is not as difficult as it might appear. It afforded me the opportunity to
collect a significant amount of rigorous data in a short time. I predicted that two, two-
hour sessions at Henley would enable me to collect rigorous, comparative data from
sufficient numbers of subjects to produce statistically significant results (or to indicate
that no such result was likely). Of course, the design and preparation took some
significant effort, particularly given that the experiment was to be presented in a
completely controlled and automated fashion. However, my skills were well developed
in this area and it appeared to be a good option.
Secondly, in purely academic terms, experimentation provides a power for explanation

and generalisation that far exceeds qualitative and/or observational research. I have a
firm belief in the power of experiment and its ability to determine significant truths. The
observation that little research in business appears to have been based on experiment is
not sufficient cause to give up the method. It would be my contention (along with
Pfeffer and Sutton (2006) and Schaffer and Thompson (1992) ) that business research
would be all the better for a greater emphasis on rigorous evidence.
Before starting the research process, I was very confident that I had the technical
knowledge to complete the research. I was, however, very conscious that there were
some significant differences between this research and the work I had previously done.
For example, whilst I had performed experiments in a physical sciences laboratory and
attended multiple courses on experimental psychology, I had never actually completed
any sort of experimental work involving people. Additionally, although I had completed
a significant amount of observational work as part of my PhD research, this was
predominantly with individuals and focused on the use of computer-based learning. In
that work, much of the data was generated from recording tools embedded in the
software which was the subject of the research. The observational study was based
around a training workshop presented to a group of postgraduates. This was a format I
was very familiar with and was confident about when planning the work. However, in
practice, when the time came to do both the experimental work and the observational
( 69 )
workshop session, I was quite anxious (see the entries in my research diary for Monday
4 October and Thursday 31 March included at Appendix D).
In retrospect, both the experimental and observational studies were more successful than
I would have expected. Volunteers from the Henley MBA were extremely generous in
taking part, the technology for the experiment worked well and the observational study
had a lot of positive responses (see the entry in my research diary for Friday 1 April
included at Appendix D).
My most significant regret was that I did not test actual reasoning ability using
PESTLEWeb during the experimental portion of the work. This would have been a very
natural extension of the work by Bauer and Johnson-Laird (1993). This does, however,
afford the opportunity for further research.
5.3 Achievement of personal objectives

This Management Challenge has been of significant benefit to achieving my personal
objectives. The quotation from my original application form for the Henley MBA,
included in Section 1.2, is worth repeating in this context:
I have a great interest in teaching and research. I believe that some of

my research background in complex systems engineering has
applications within strategy development and corporate finance. I
could see that doing an MBA would open opportunities for research
and teaching.
I see this research project as a significant stepping stone towards a more significant
programme of work in this area. I have managed to import an area of knowledge which
I am very familiar with and apply it in a new domain. I believe that the literature review
in Section 3 substantiates the view that this application of previous work is both valid
and useful. The work has also thrown up a range of further research questions and
provided a foundation of information about the method itself.
( 70 )
A key objective of this research project was to produce a re-usable curriculum resource
that could be used by Henley Business School. Appendix E provides a set of workshop
material which has been trialled in a realistic setting; a Henley MBA class. The
PESTLEWeb CoP website includes a variety of other resources, including drawing
templates and extensive examples that are already being used by MBA students. The
results of the observational study demonstrate the practicality of teaching and using this
method.
In my original objectives for this research project, I had dismissed the idea that I would
develop significantly in my ability to conduct research. This has not been entirely the
case. In fact, I have learnt a significant amount about the practical application of
experimentation in this area and the difficulty of performing such experiments.
I had, however, set as a core objective that this research project would create
opportunities for further research at Henley Business School. At this point, I have every
reason to be optimistic in this regard. I have had significant and positive support from
both students and staff at Henley for the current work and there is a growing interest in
the PESTLEWeb model. The submission of this thesis represents a significant
milestone.
Word Count = 17,678
( 71 )
6 References
Ashby, W.R. (1958) Requisite variety and its implications for the control of complex
systems, Cybernetica, 1, 83-99
Astley, W.G. (1984) Toward an Appreciation of Collective Strategy, Academy of

Management Review, 9 (3), 526-535
Astley, W.G. and Fombrun, C.J. (1983) Collective Strategy: Social Ecology of
Organisational Environments, Academy of Management Review, 8 (4), 576-587
Aguilar, F.J. (1967) Scanning the Business Environment, The Macmillan Company:
New York
Andrews, K.R. (1980) The Concept of Corporate Strategy: Revised Edition, Richard
D. Irwin, Inc: Homewood, Illinois
Albright, K.S. (2004) Environmental Scanning: Radar for Success, Information

Management Journal, 38 (3), 38-45
Barney, J, (1991) Firm Resources and Sustained Competitive Advantage, Journal of

Management, 17 (1), 99-120
Barr, P.S.; Huff, A.S (1997) Seeing Isn't Believing: Understanding Diversity in the
Timing of Strategic Response, Journal of Management Studies, 34 (3), 337-370
Barr, P.S.; Stimpert, J. L.; Huff, A.S (1992) Cognitive Change, Strategic Action and
Organizational Renewal, Strategic Management Journal, 13, 15-36,
Bauer, M. and Johnson-Laird, P.N. (1993) How Diagrams Can Improve Reasoning,
Psychological Science, 4 (6), 372-378
Bougon, M.G (1992) Congregate Cognitive Maps: A Unified Dynamic Theory of

Organization and Strategy. Journal of Management Studies, 29 (3), 369-389
Buckingham Shum, S. J., MacLean, A., Bellotti, V.M.E., Hammond, N.V., (1997)
Graphical Argumentation and Design Cognition, Human-Computer Interaction, 12(3),
267-300
Buzan, T. (1993) The Mindmap Book BBC Books: London
Checkland, P. (1993) Systems Thinking, Systems Practice, Chichester: John Wiley

and Sons
Collins, R.J. and Leathley, B. (1995) Psychological Predispositions to Errors in Safety,

Reliability and Failure Analysis, Safety and Reliability, 14, 642
Collins, R.J. and Thompson, R. (1997a) Systemic Failure Modes: A Model for
Perrow's Normal Accidents in Complex, Safety Critical Systems in Advances in Safety
( 72 )
and Reliability. Proceedings of the ESREL 97 International Conference on Safety and
Reliability, 17-20 June 1997 , C. Guedes Soares (Ed), Elsevier Science: Oxford
Collins, R.J. and Thompson, R. (1997b) Searching for Systemic Failure Modes in
Advances in Safety and Reliability. Proceedings of the ESREL 97 International
Conference on Safety and Reliability, 17-20 June 1997 , C. Guedes Soares (Ed),
Elsevier Science: Oxford
Collins, R.J. (1997a) The Essential Logic Model: A Method for Documenting Design
Rationale in Safety Critical Systems, ESREL 97
Collins, R.J. (1997b) A Tutorial on the Essential Logic Model, International

Committee on Systems Engineering, UK Conference 1997
Collins, R.J. (2000) Improving Systems Integrity by Using Thread Analysis for Design
Validation In A Decade of Progress A New Century of Opportunity: 10th Annual
International INCOSE Symposium Proceedings, 16-20 July 2000
Collis, D.J. and Montgomery, C.A. (2008) Competing on Resources, Harvard

Business Review, 86 (7/8), 140-150
Cox, E.P. III (1980) The Optimal number of Response Alternatives for a Scale: A
Review, Journal of Marketing Research, Vol. XVII (November 1980), 407-422
Craik, F.L.M., and Lockhart, R.S. (1975) Levels of processing: A framework for
memory research, Journal of Verbal learning and Verbal Behaviour. 11, 671-684
Craik, F.L.M., and Lockhart, R.S. (2008) Levels of Processing and Zinchenkos
Approach to Memory Research, Journal of Russian and East European Psychology,
46(6), 5260
Craik, F.L.M., and Tulving, E. (1975) Depth of processing and the retention of words
in episodic memory, Journal of Experimental Psychology: General, 104, 268-294
Daniels, K., Johnson, G. and de Chernatony, L. (1994) Differences in Managerial

Cognitions of Competition, British Journal of Management, 5(2), 21-29
Day, G.S. and Schoemaker, P.J.H. (2006) Peripheral Vision: Detecting weak signals
that will make or break your company, Harvard Business School Press: Boston,
Massachusetts
de Groot, A.D. (1978) Thought and Choice in Chess, Mouton Publishers: The Hague,
Netherlands
Drucker P. F, (1994) The theory of business Harvard Business Review, Sep/Oct, 95-
104
Eden, C. (1989) Cognitive maps as a visionary tool: strategy embedded in issues

management, Long Range Planning
Eden, C., Ackermann, F. and Cropper, S. (1992) The Analysis of Cause Maps.
Journal of Management Studies, 29, (3), 309-324
( 73 )
Eden, C. (1992) On the Nature of Cognitive Maps, Journal of Management Studies,
29 (3), 261-265
Eriksson, P. and Kovalainen, A. (2008) Qualitative Methods in Business Research,

Sage: Los Angeles
Friedenthal, S., Moore, A., Steiner, R. (2009) A Practical Guide to SysML: The
Systems Modelling Language, Elsevier: Amsterdam
Fuld, L. (2003) Be Prepared, Harvard Business Review, 81(11), 20-21
Garland, R. (1991) The Mid-Point on a Rating Scale: Is it Desirable?, Marketing

Bulletin, 2, 66-70, Research Note 3
Glaser, B.G. and Strauss, A.L. (1999) The Discovery of Grounded Theory: Strategies
for Qualitative Research, Aldine de Gruyter: Hawthorne, New York
Glaser, R., & Bassok, M. (1989) Learning theory and the study of instruction, Annual
Review of Psychology, 40: 631-666, Palo Alto, CA: Annual Reviews Inc.
Gray, D.E. (2005) Doing Research in the Real World, Sage Publications: London
Hair, J.F., Jr, Money, A.H., Samouel, P. and Page, M. (2007) Research Methods for
Business, John Wiley & Sons
Hkanson, L. (2007) Creating knowledge: The power and logic of articulation,

Industrial and Corporate Change, 16, (1), 5188
Hoare, C.A.R. (1985) Communicating Sequential Processes, Prentice Hall: New York
Horton, J., (2010) Strategy 2.0 Innovating: How We Develop and Execute Strategy
MHD Supply Chain Solutions, 56-59
Kanawattanachai, P. and Yoo, Y. (2007) The Impact of Knowledge Coordination on

Virtual Team Performance Over Time, MIS Quarterly, 31(4), 783-808
Kaplan R S and Norton D P (1992) The balanced scorecard: Measures that drive
performance, Harvard Business Review, Jan-Feb, 71-80
Kaplan R S and Norton D P (1993) Putting the Balanced Scorecard to Work, Harvard
Business Review Sep-Oct, 2-16
Kaplan R S and Norton D P (1996) Using the balanced scorecard as a strategic

management system, Harvard Business Review, Jan-Feb, 75-85
Kaplan, R. S., & Norton, D. P. (2004). Strategy maps: Converting intangible assets into
tangible outcomes. Boston, Harvard Business School Press: Boston: Massachusetts
Kiesler, S. and L. Sproull. (1982) Managerial response to changing environments:

Perspectives on problem sensing from social cognition Administrative Science
Quarterly. 27, 548-570
( 74 )
Kirkpatrick, J, Kirkpatrick, J. (2009) The Kirkpatrick Model: Past, Present and Future,
Chief Learning Officer, Nov 2009, 20-55
Kirkpatrick, J.D. and Kirkpatrick, W.K. (2010) Roe's Rising Star: Why Return on
Expectations is Getting so Much Attention, T+D, Aug 2010, 64(8), 34-38
Klein, H.E. and Newman, W.H. (1980) How to integrate new environmental forces into
strategic planning, Management Review, July 1980, 40-48
Ladeji, D. (2011) Globalisation & Nigerian Content Development: A Study of IT

Practices in the Upstream Petroleum Industry, Henley Business School: Henley on
Thames
McEwen, T. (2008) Environmental Scanning and Organizational Learning in

Entrepreneurial Ventures, Entrepreneurial Executive, 13, 1-16
Miller, S. (1986) Experimental Design and Statistics, 2nd Ed, Methuen and Co. Ltd:
London
Oliver, A.L., Montgomery, K. (2008) Using Field-Configuring Events for Sense-

Making: A Cognitive Network Approach, Journal of Management Studies, 45(6),
1147-1167
Pandza, K., Thorpe, R. (2009), Creative Search and Strategic Sense-making: Missing
Dimensions in the Concept of Dynamic Capabilities, British Journal of Management,
20, S118-S131
Pearce II, J., Chapman, B.L. and David, F.R. (1982) Environmental Scanning For
Small And Growing Firms, Journal of Small Business Management , 20(3), 27-34
Pferrer, J. and Sutton, R.I. (2006) Evidence Based Management, Harvard Business
Review, Jan 2006, 62-74.
Richardson, K.A. (2004) Systems theory and complexity: Part 2, E:CO, 6(4), 77-82
Robson, C (2002) Real World Research: a Resource for Social Scientists and
Practitioner-researchers, 2nd ed. Blackwell: Malden, MA
Rowntree, D. (1981) Statistics without Tears: A primer for non-mathematicians,

Penguin: Harmondsworth, Middlesex
Schaffer, R.H., and Thompson, H.A. (1992) Successful Change Programs Begin with
Results, Harvard Business Review, Jan-Feb 1992, 80-89
Schmidt, D. P. (1986) Patterns of Argument in Business Ethics, Journal of Business

Ethics, 5 (6), 501-509
Schoemaker, P.J.H. (2008) The Future Challenges of Business: Rethinking

Management Education, California Management Review, 50, (3), 119-139
Schoemaker, P.J.H. and Day, G.S. (2009) Why We Miss the Signs, MIT Sloan
Management Review, 50(2), 43-44
( 75 )
Selic, B, Gullekson, G., Ward, P.T. (1994) Real-Time, Object-Oriented Modeling,
John Wiley and Sons: New York
Shlaer, S. and Mellor, S.J. (1992) Object Lifecycles: Modeling the World in States,
Yourdon Press, Prentice Hall: Englewood Cliffs, New Jersey.
Tesa ov, M. (2007) Can Diagrams Help in Jurors Decision-Making?, Oxford and
Cherwell College: Oxford
Thompson, J.L. and Martin, F. (2005) Strategic Management: Awareness and Change.
5th ed., Thompson Learning: London
Toulmin, S.E. (1958) The Uses of Argument, Cambridge University Press: Cambridge
Tovstiga, G. and Aylward, M. (2008) Strategic Direction. Study Guide, Henley

Business School
Tversky, B. (2005) Visuospatial Reasoning in The Cambridge Handbook of Thinking

and Reasoning, K.J. Holyoak and R.G. Morrison (Eds), Cambridge: Cambridge
University Press
Wenger, E.C and Snyder, W.H. (2000) Communities of Practice: The Organizational
Frontier, Harvard Business Review, Jan-Feb 2000, 139-145
Yunggar, M., (2005) Environment Scanning for Strategic Information: Content

Analysis from Malaysia, Journal of American Academy of Business, Cambridge, 6 (2),
324-331
( 76 )
Appendix A: Response paper used for the experiment
in Study 1
Please do not turn over the

page until asked to do so
Please provide some background information about yourself
I am. Female Male (Tick one)
English is my first language:

Yes No (Tick one)
My education has been focused on: (Tick as many as required)
Science Languages Social Sciences Arts
Business Medicine Engineering
When I learn new things, I prefer to get new information in the form of: (Tick one)
Written Words Pictures Spoken Words
I have a visual impairment that is not corrected by glasses (including colour

blindness):
Yes No (Tick one)
( 77 )
Based only on your understanding and memory of the
information presented to you:
1. What issue or issues were presented as causing a shift towards the purchase of fuel-
efficient vehicles?
2. What issue or issues were presented as a consequence of a political issue?
3. What social issues were presented in the analysis?
4. What issues were presented that led directly to the final market opportunity?
( 78 )
5. What issues were presented as being the cause of the growing public awareness of
green issues?
6. Please rate how persuasive you found the presentation by ticking one of the boxes
below:
Not at all persuasive Not very Somewhat Very persuasive

persuasive persuasive
7. Please rate how engaging (interesting) you found the presentation by ticking one
of the boxes below:
Not at all engaging Not very Somewhat Very engaging

engaging engaging
8. Please rate how logical/rational you found the presentation by ticking one of the
boxes below:
Not at all logical Not very Somewhat Very logical

logical logical
( 79 )
9. In the space below, please try to replicate as much of the information as you can, in
the same form in which it was presented.
- End of Response Paper -
( 80 )
Appendix B: PESTLEWeb evaluation form from
workshop
This evaluation is intended to get your feedback on the PESTLEWeb TM method and to
find out if you think it will be useful for you.
Before the Workshop

Please tick one box in response to each question
1) My company has a well- Strongly disagree Strongly agree

developed system in place for
scanning the business
environment
2) I have found it easy to Strongly disagree Strongly agree

analyse the business
environment as part of my
MBA assignments
3) I have found it easy to Strongly disagree Strongly agree

analyse the business
environment for my MBA exam
case studies
Strongly disagree Strongly agree

4) I think that PESTLE analysis
gives me deep insights into the
context for my business

5) I find it easy to explain the
context of my business to
others
( 81 )
At the end of the workshop

1) I found the PESTLEWebTM
method straightforward to
understand
2) A PESTLEWebTM model Strongly disagree Strongly agree

provides a better explanation
of the business environment
than traditional PESTLE
3) Using PESTLEWebTM would Strongly disagree Strongly agree

improve the quality of my
analysis of the business
environment

4) I will use the PESTLEWebTM
method for my future MBA
assignments

5) I would like to learn more
about the PESTLEWebTM
method

6) I would like see
PESTLEWebTM models for my
industry produced by others

7) I would be willing to share
my PESTLEWebTM models with
other people

TM
8) PESTLEWeb would be
useful to my business
If you would like to learn more about this research and the PESTLEWebTM method,
please put your name and e-mail address below.
Name: ____________________________________
e-mail: ____________________________________
( 82 )
Appendix C: Examples of PESTLEWeb models
developed by other people
Analysis of Nigerias Socio-Economic Journey: 1960-Mid 2000s
Economic
Political Instability No long term Niger-Delta Militancy
Unprepared military &
economic blueprint Crisis
civilian rulers
PESTLEWebTM
Overdependence on High Government Inadequate High Unemployment

Oil & Gas taxation of petroleum liberalisation of the rate
industry economy
Significant levels of State failure in Neglect of Oil

public corruption providing adequate producing
regulation & basic communities
social infrastructure
Social
High Population
Transparency International
ranks Nigeria as 130 out of
180 countries on the
Corruption Perceptions
Index (CPI) PESTLEWebTM
Analysis of Nigeria Socio-Economic Journey: Mid 2000s to date
Advent of continuing
More technocrats in Reforms of current
stable democratic
public service petroleum laws &
experience
structures
Nigerian Content
Bill
Ongoing deregulation
of all sectors of the
economy
Economic
Govt intervention to
emphasize job Petroleum Industry
creation Bill
Technological
PESTLEWeb TM
Globalized business
models of IOC & other
companies aided by
technological
developments
PESTLEWebTM
Social
Political
Amnesty declared for Less volatile
Niger-Delta militants environment for E&_ Increased focus on
activities oil producing
communities
PESTLEWeb TM
PESTLEWeb TM
Figure 28: Two PESTLEWebs for 'Globalisation and Nigerian Content Development' Ladeji, D.
(2011)
( 83 )
Figure 29: PESTLEWeb for the Oil Industry by Kemi Otaru, Henley MBA Student
( 84 )
(Threats) BMW has to be
(Opportunity) highly prepared for raw material
Fuel efficient engines Increased investment and
engineering
have
driven,
beenculture
and agile developed
innovative
in
of producer
research in light weight and
price instabilities especially
in case of price increase
Developed Markets and green trend
Japan (Hybridcompetitive
could create cars) alternative materials for
Toyota Prius
advantage car s
(Threat) to pay CO2 penalty, Reference

if emission targets wont be Emerging economies
achieved with every model in Oil price shock in 2008 Lack of resource Shortage raw Steinberger T.; PwC Growing Mega-City's in HighIntroduction
fuel tax and ofCO2
Desire to avoid taking over developed
the fleet $147 a barrel Peak Oil 2015 materials price such markets in car production
Automotive, (2010); Beyond developed and environmental
legislations /incongestion
Japan
Shortage starts 2013 dependence on oil Crisis current situation and
=> negative effect on as steel and rare soils global total 55 mio 2000 developing economies Zones in big city's
max. 125g/km 2015of
outlook for Europeans
profit margin => increased price global total 90 mio 2016 developed economies
automotive industry;
Pricewaterhouse Coopers
High fuel tax and CO2 Fuel efficient engines (Opportunity) highly
legislations
legislations ininEurope
Japan Fuel efficient
engineering driven,engines
innovative
have been developed in
max. 130g/km
125g/km 2015 Reference andhave
agilebeen developed
culture in
of Producer
EU (highly efficient Japan (Hybrid cars)
max. 95g/km 2020 diesel engines) EU 710
High fueltax
penalty per
and CO2 Moritz (2010); Future could create competitive
10g/km above
legislations CO2
in Japan Development in the Toyota Prius
advantage
legislation per sold
max. 125g/km 2015car Automotive Industry and its
from 2015 onwards Markets; Management
Engineers
(Threats) Has to redefine its (Opportunity) revolutionise

value proposition Ultimate the market with individual
driving machine", no core premium mobility from A - B
competency yea and low e.g. for business people
economy of scale for new (blue ocean strategy)
Development of revolutionary developments
High
High fuel tax
tax and CO2
CO2 Fuel efficient engines People in developed
propulsion system for customer Development of new
legislations
legislations in Japan have been developed in economies shifted
cars: Hybrid, Electrification, personalised environmental
max. 125g/km
125g/km 2015
2015 Japan (Hybrid cars) towards smaller fuel
efficient cars Hydrogen etc? friendly mobility concepts
long term advantage Toyota Prius biggest change since Otto's (Threat) Low profit margin
invention of petrol engine Reference
Winterhoff M (2008); Future
business for premium car
of Mobility 2020, Athur D. producers
Little
Stronger impact in US car New stakeholders for

Low fuel tax in order to General American car Agreement of CAFE Increase of Growing trend toward Growing global trend
industry due to non fuel efficient infrastructure such as car rental
protect American has high fuel Standard 2011 in the environmental individualisation towards mobility
products and higher oil price (mobility service) necessary to
automotive industry in consumption US awareness in
impact on fuel price collaborate
the short term developed economies
=> customers shift towards small
and efficient Japan/EU cars
(Opportunity) partner may

become a target for upwards (Threat) producer becomes
New stakeholders/partners for integration of BMW in order to dependent to a strategy
(Threat BMW) Japanese (Threat) Could also affect (Opportunity) Producer could increase profits and secure partner
electric components, batteries,
competitors have sales of SUVs in the US move into the gap with small core competence (e.g..
engines and infrastructure such
technological head start with efficient models Battery, Engine)
as electric energy suppliers
hybrid cars
necessary
Figure 30: PESTLEWeb Model for the Car Industry (1), By Hermann Bauer, Henley MBA Student
( 85 )
Risk of Euro zone Non Euro EU countries such
Financial Situation in developed breakup due to possible as UK not recovered yet
Government dept % of GDO
transfer payments from 2010: 71,57 %
recovered economies 2015: 83,94%
such as Germany
Financial Crisis caused by Customer car purchases Huge impact onto EU Eurozone member such as
Lehman Brothers bankruptcy automotive industry. Greece, Irland go bankrupt
stopped suddenly and and need huge loans from
end of 2008 force capital Governments need to
flows within the whole world
were delayed in to the support with short work and EZB. Italy, Portugal and
economy future which effected the stimulation packages Spain suffer as well
car industry (Threat) destruction of
margin by exchange rate of
export market devalues
towards Euro (if cars from
Germany)
Japan Economy 2011 still

Oil price shock in mid of Huge impact onto Japan
not recovered after financial High Risk of global
2008 $147 a barrel automotive industry. Global automotive Crises Government net dept
Governments need to production overcapacity % of GDP 2010: 121.73 %
Exchange rate fluctuations
support with loans and
stimulation packages about 37% becomes Prognosis 2015: 153.95 %
visible
(Opportunity) increase of
margin by exchange rate of
export market up values
towards Euro (if cars from
People in developed Destructive impact onto USA Economy 2011 still not Germany)
economies shifted American automotive recovered after financial
towards smaller fuel industry. Governments need Crises Government net dept
to support with loans and % of GDP 2010: 66,16%
efficient cars
stimulation packages Prognosis 2015: 85,52 %
Risk of US Government
Bankruptcy
(break down of global
economy)
GM and Chrysler go (Opportunity) to use crisis
bankrupt and need to be as a chance to make
reincarnated by the necessary changes for future
government growth when weaker
competitors shacked out
=> huge image impact
( 86 )
Brazil:
Brazilian Population Estimated Brazilian GDP

increase from
191,1 mio 2008
per capita (PPP)
US$ 15.200,- in 2020
Developed Markets and green trend
estimated 220 mio 2020
Brazil GDP Growth Rate Brazil Real per capita Koreas car sales urged by Cars per 1000 people Brazil was the worlds 5th - EU tries to reduce car (Opportunity) Huge for
(YoY) income US$ 4,480 (US$ 30% annually from 1980 US 765, Germany 500, largest car market in 2009, import tariff reduction from
10.304 purchase power China 40, India 12, European car producers
0,5% onwards (similar to the US with car sales rising 11.4% current level 35%
parity terms similar to 1960th and Japan 1970th ) Russia 124, Brazil 158 to a record 3.14 million downwards, but struggles
especially luxury cars
www.tradingeconomics.com
Korea 1980) www.tradingeconomics.com Analyst predict 4 th in 2010 in Brazil
Reference Reference Reference (Threat) European car

McManners P & 88% of new cars are flex Osborn A (2010); EU Auto
Kiss P. (2009); Think BRIC!, fuel running on ethanol Major imports are German producers have
Herbolzheimer E (2009); Makers Anticipate Export Luxury cars
Comparative Study, KPMG, Global Automotive Industry; and petrol Boost From Mercosur Trade
competitive disadvantage
Hungary Exam 2 GBE Case Study; 2008 sales increase Talks in Brazil without selling
Henley Business School, of 24% fuel flex technology
Greenlands-Henley
Russia:
Russia Russia Real per capita Estimated Russian GDP Cars per 1000 people
US 765, Germany 500,
Car sales forecasted to Russians look for latest (Opportunity) Russia
GDP Growth Rate (YoY) income US$ 3.330 (US$ per capita (PPP) reach 6 million by 2012 technology and design
16.000 PPP terms similar US$ 31.000,- in 2020 China 40, India 12, offers huge for European
2,7% therefore high amount of
to Korea 1980) Russia 124, Brazil 158
imported cars
car producers especially
www.tradingeconomics.com www.tradingeconomics.com luxury cars if tariffs fall
Russian Population slight Koreas car sales urged by Reference

McManners P &
Reference
decrease from 30% annually from 1980 Russian Prime Minister
141,8 mio 2008 onwards (similar to the US Herbolzheimer E (2009); Spiegel (2010); 'From Putin would like a free trade
estimated135 mio 2020 1960th and Japan 1970th ) Global Automotive Industry; Lisbon to Vladivostok' Putin agreement between the EU
Exam 2 GBE Case Study; Envisions a Russia-EU Free & Russia could eliminate
Henley Business School, Trade Zone current import tariff (25%)
Greenlands-Henley
India:
India Cars per 1000 people Rising trend towards small Indian import tariffs of
India Real per capita Estimated Indian GDP per (threat) very difficult to
GDP Growth Rate (YoY) 8,9% capita (PPP) US 765, Germany 500, premium hatchback cars about 100% cars and 13,5
income US$ 718 (US$ China 40, India 12, compete as an importer
challenging Thailand & Korea US$ 6.700 in 2020 parts
3.000 purchase power Russia 124, Brazil 158
parity terms)
more likely to assembly in
a CKD facility on the local
Indian market
Strong focus on small and

Indian Population increase Reference cheap cars such as Reference (opportunity) If Indian
from 1.148 mio 2008 Tata Nano 2500,-
India to become TOP 5 India's Export focus may
estimated 1.403 mio 2020 Kiss P. (2009); Think BRIC!, BBC News (2010); Indias producer and exporter for be driven by WTO tariffs are reduced in the
Comparative Study, KPMG, but also upgrading to small car market booming small car's in 2020
premium cars is happening towards reduction of their future, market would
Hungary
protectionism policy become very attractive for
small premium car import
China:
Reference
McManners P &
Chinese Governmental Tremendous increase in Herbolzheimer E (2009);
stimulation in 2009 with sale by 48% Global Automotive Industry;
tax cuts 5-10% for cars (compared with June Exam 2 GBE Case Study;
with engines smaller 1,6 2008) Henley Business School,
liters Greenlands-Henley
China China Real per capita income Koreas car sales urged by
Cars per 1000 people Expected parallels in (Threat) Opportunity just
US 765, Germany 500, China with tremendous form small engine car
GDP Growth Rate (YoY) approaching US$ 4,000,- 30% annually from 1980
China 40, India 12, producers in China; Market
9,6%
(US$ 12.000,- purchase onwards (similar to the US
Russia 124, Brazil 158 growth period ahead
power parity terms similar to 1960th and Japan 1970th) demand depends highly on
www.tradingeconomics.com www.tradingeconomics.com
Korea 1980) national macroeconomic
policies
Deregulation of global
Chinese Population Estimated Chinies GDP Reference markets such as China (Opportunity) Huge
increase from per capita (PPP) import tariff reduction from
1.328 mio 2008 Harwitt E. (2001); The potentials for foreign
US$ 20.000,- in 2020 Impact of WTO Membership 220% to 25% due to WTO
estimated 1.424 mio 2020 producers of
on the Automobile Industry Membership 2006
in China, The China
differentiated products in
Quarterly China
( 87 )
Appendix D: Evidence from the management challenge
log
The following quotes are extracted from my Management Challenge Log and provide a
representative view of the personal and process issues recorded there.
Wednesday 1 September
I feel really excited as I kick-off this part of the MBA. I believe that there is an
opportunity here both to tick the box for the MBA but more importantly to do
something meaningful for me. Having used the PESTLEWeb notation with the study
team and for my own IMP I am confident in my own mind that I have something useful
here. But I also know from experience from previous research that what is intuitively
obvious is frequently not always the case. Indeed, some of the most interesting science
comes when the results are unexpected. Still, I see here a convergence between my
academic and teaching work in complex systems, my background in psychology and
computing (particularly IT-type business analysis) and some issues in the field of
strategic analysis.
Thursday 16 September
I really think that a series of experiments for PESTLEWeb is the right way to go. I have
the experience, and I have neat method in mind to perform it but will Henley accept
this? The Management Challenge Guide does not exclude experiment but they are
only mentioned once in the context of plagiarism (Falsifying data, evidence, or
experimental results). The Manager as Investigator study guide mentions
experimentation in a number of places, and I think that the case would be arguable
Wednesday 27 September
In my proposal, I planned to do an observational study as well as the experimental
work. The issue is that I want the observational study to be as realistic as possible. I
need to find out if somebody could actually teach an MBA class how to use this method
in a relatively short space of time and have them use it effectively. I have the experience
of teaching other, similar postgraduate students quite similar methods. For example,
( 88 )
teaching people how to use SysML to analyse complex systems. But MBA students are
a different group. It is a stereotype but I imagine my technical students to be more
familiar with graphical modelling and MBA students to perhaps be more experienced at
making arguments verbally and in the form of essays. So, will the graphical methods
work with this diverse group? Will there be significant individual differences? Maybe
this is part of the interest for the MC what should work, may not work in practice.
Even such results would be interesting if rather disappointing.
Monday 4 October
I am really starting to regret setting up this experiment. If the technology does not work
then it will a complete waste of time. I only have a very limited time-window to get
these results and if it does not work I am in trouble. Additionally, what happens if
insufficient people volunteer? The whole effort rests on this working and there are some
serious risks.
Friday 29 October
First run of the experiment today. I was nervous as hell, but so relieved to see that the
technology all worked. As it was, it was simply a matter of getting people into the room,
giving the warm up script and setting the experiment to run itself. As a general method
this is a great way to do classroom experimentation. It allows absolute control between
two experimental conditions and (assuming the technology works), really reducing the
stress of trying to give a presentation under experimental conditions.
Friday 10 December
Completed the second run of the experiment with HB40 today. Great response! I am
pretty sure I will have enough subjects now to show a statistical result, if any exists. If
not then I better start rethinking! I can process the number tomorrow.
Monday 14 March
I think that I am all set for the workshop. Tass has agreed to help do some observation
work for me, which makes me more relaxed about the process. I think that the
presentation slides are looking OK and the case study is reasonably approachable. I
would really like to be running a half-day workshop, which would give people more
( 89 )
time to get into the material. Also, is the case study realistic? It is one thing to be given
this sort of material in a nice neat condensed form but what about the real-life
situation of having to hunt down relevant information? I guess that is always the issue
with case studies and probably why Henley favours real-world data rather than case
studies. But that is a problem for later I cant let this workshop get too large to be
practical.
Thursday 31 March
I am doing the workshop at Henley tomorrow. Its a fantastic opportunity, but I stand
every chance of falling on my face. Why should any of the students hang around after a
long day studying? What happens if they dont get it? Or worse, they get it but they
dont like it?
Friday 1 April
I did the first real PESTLEWeb workshop today. It was a fantastic experience it felt
like coming full-circle since it was held in the same room as I attended my first kick-off
workshop for the MBA. I really love doing this sort of teaching; with a really
enthusiastic, smart bunch of people. Its not really teaching at all; more like facilitated
learning. Give the people the ideas and they pick them up and run with them. I was
amazed and delighted at how many people stayed on at the end of the day for my
workshop. Would I have been that generous? I remember being pretty tired at the end of
my MBA workshops. But then I think I would have stayed myself. If nothing else I
would have been intrigued by another MBA student trying to do this.
As expected I got mixed results from the workshop. Clearly some people were very
switched on to it quite quickly. But then other people were less impressed. That is not
so bad. If even a few people find this useful then it might get used more frequently.
After all, this is a self-selecting group nobody has to use the method.
( 90 )
Appendix E: Workshop presentation on PESTLEWeb
( 91 )
( 92 )
( 93 )

RPIK Pestleweb - Thesis PDF

Uploaded by

Copyright:

Available Formats

RPIK Pestleweb - Thesis PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

RPIK Pestleweb - Thesis PDF

Uploaded by

Copyright:

Available Formats

University of Reading

Is there a better way to

Student number: 83000716

Management Challenge submitted in partial fulfilment of the requirements for the

Recommendations from this thesis include an incremental process of further

I would like to thank Professor Francis J. Aguilar, Professor Emeritus of Business

Appendix A : Response paper used for the experiment in Study 1............................... 77

Table 1: Constitution of Subjects ................................................................................ 36

Figure 1: Core Elements of the PESTLEWeb Drawing Convention ............................ 5

In carrying out a PEST or PESTLE analysis it is all too easy to

Tovstiga and Aylward (2008)

1.2 Relationship of the project to the terms of reference and

The PESTLEWeb model is a direct development of earlier graphical notations

I have a great interest in teaching and research. I believe that some

The proposed project is only the first step of a larger programme of

this [research] could provide a powerful and effective curriculum

1. Successful business strategy crucially depends on developing a rich, accurate

2. Scanning of the business environment is frequently not done well.

3. Visual argumentation is a powerful and effective tool to support analysis,

1. An experimental procedure intended to rigorously test the hypothesis

2. A workshop-based observational procedure intended to test the method in a

The base building-blocks of PESTLEWeb are a series of boxes labelled to represent

Figure 1: Core Elements of the PESTLEWeb Drawing Convention

Figure 2: Reference, Comment and Consequence Elements of the PESTLEWeb Drawing

Finally, the convention includes an element to annotate relationships between elements

Figure 3: Illustration of the use of a Generic Association between Elements in a PESTLEWeb

2.2 An example PESTLEWeb model for the automotive

Environmental Economic Political

Environmental Social Political Technological

Figure 4: Part 1 of a PESTLEWeb Diagram for the Global Automotive Industry

2.2.2 Considering consequences

Figure 5: Part 2 of a PESTLEWeb Diagram for the Global Automotive Industry

In Figure 5, we see the development a section of the automotive PESTLEWeb model

The section of the automotive industry PESTLEWeb model in Figure 6 introduces

Figure 6: Part 3 of a PESTLEWeb Diagram for the Global Automotive Industry

3.1 Understanding the business environment

Successful business strategy depends crucially on developing a rich, accurate and

3.1.1 Environmental scanning as a key component of strategy

Figure 7: Development of Economic Strategy (Andrews, 1980)

When there are high levels of ambiguity, complexity or unpredictability

[a] crucial component of managerial behavior in rapidly changing

Despite the importance of the entrepreneurs' knowledge to new venture

In general, methods for representing knowledge (codes) provide a mechanism for

In todays rapidly changing business environment, an organizations

If businesses are to be managed effectively in complex environments, then their internal

Control can be obtained only if the variety of the controller is at least as

If such internal complexity is to be useful for intentional adaptation (as opposed to a

3.2 Criticisms of current methods and practices of

Scanning of the business environment is frequently not done well.

3.2.1 Lack of quality and quantity of environmental scanning

3.2.2 Failures in codifying and communicating environmental

Different beliefs about cause and effect have been established as a

3.3 Visual argumentation as an appropriate tool in this domain

Visual Argumentation is a powerful and effective tool to support analysis,

Graphics have an advantage over language in expressiveness; graphics

3.3.1 Evidence from cognitive psychology and education

Remembering: To enable analysts to absorb and recall business

Thinking: To be able to reason, logically and rationally across datasets to

the intellectual demands of visualising the structure of an