1.

A Scenario-based Approach to Strategic

Planning
Tool Description – Scenario Matrix
Prof. Dr. Torsten Wulf, Christian Brands and Philip Meissner

Working Paper 4 /2010

Leipzig, August, 20th, 2010

 Center for Scenario Planning – Roland Berger Research Unit

2 1. Introduction

In this paper we describe the fourth-stage of our six-step scenario-based approach to

strategic planning, the scenario building. Having completed the trend and uncertainty

analysis using the ‘Impact/Uncertainty Grid’ tool in stage three, we will present in this paper

how plausible scenarios can be created based on the identified key uncertainties using the

‘Scenario Matrix’ tool (Figure 1). The overall goal of the scenario building step and its

‘Scenario Matrix’ tool is to generate and develop four distinct future scenarios. Of particular

interest in this context is the actual process of deriving the scenarios out of the previously

identified two critical influence factors using the so-called ‘scenario influence diagram’ and

‘scenario fact sheet’1. Before explaining the ‘Scenario Matrix’ tool itself we will examine what

scenarios actually are, what the basic idea behind them is and how they can help a company

to think ahead.

Figure 1: Six-Step Scenario-based Approach to Strategic Planning

1
See Wulf/Meissner/Stubner (2010), pp. 25-28.
 Center for Scenario Planning – Roland Berger Research Unit

3 2. Background Information Scenario Building

Building sound and plausible scenarios is a challenging task that needs to follow a structured

process. However, before describing the scenario building itself, one should first identify the

purpose that the creation of scenarios fulfills. Academics usually distinguish between three

purposes which scenarios can accomplish:

 First, scenarios are used as a onetime activity to predict and evaluate a specific, already

defined strategic plan of action.

 Second, scenarios are used as a onetime activity to support and enhance a specific

strategic planning process including related decisions.

 Third, scenarios are used from a onetime activity to an ongoing course of action within an

organization’s strategic planning process supporting the way in which an organization

learns2.

What all three purposes have in common, however, is that scenarios enable managers to be

better prepared for strategic decisions, especially in times of increased volatility and

environmental uncertainty.

The scenario building approach presented in this paper can be used for all three purposes

explained above. Nevertheless, it is mostly applicable to the third purpose as its holistic

approach aims at utilizing scenario planning for a company’s continuous strategic planning

activities. Establishing a common understanding, we see scenarios as a plausible

description of how the future may develop based on a coherent and internally consistent set

of assumptions about key relationships and driving forces3. Scenarios in the context of our

approach are not meant as a forecast or precise prediction nor do they state a desired

future4. Rather they produce a picture or a story describing a possible future which, as

explained in the previous paragraph, supports organizational learning and readiness for

unforeseen events. In the sense of this paper, scenarios provide different views on the

2
See Bradfield/Wright/Burt/Cairns/van der Heijden (2005), p. 807.
3
See Metz/Davidson/Bosch/Dave/Meyer (2007), p. 174.
4
See Lindgren/Bandhold (2009), p. 22.
 Center for Scenario Planning – Roland Berger Research Unit

4 nature of the future5. Put differently, scenarios try to answer so-called ‘What if…?’ questions,

evidently bringing risks as well as opportunities to an organization’s attention, rather than

concealing them6. In our six-step scenario-based approach to strategic planning, scenarios

go even one step further by answering ‘What if, then…!’ questions and hence giving

strategic recommendations for a specific course of action to be undertaken by organizations

in the four scenarios7. This aspect however, will not be examined in this paper, but in the

subsequent one focusing on the ‘Strategy Manual’ tool.

As explained in the previous steps (see ‘360° Stakeholder Feedback’ tool and

‘Impact/Uncertainty Grid’ tool), scenarios examine critical uncertainties and variations of

uncertainties in addition to important predetermined trends8. Different tools have been

developed in the past trying to fulfill the task of developing scenarios based on uncertainties

as well as trends. Again, three different approaches to scenario building can be

distinguished. The first approach uses extrapolative data analysis and trend models giving

each scenario a specific probability of occurrence and is subsequently forecasting oriented.

This quantitative approach is expert-led, that is the planner controls the process, completes

the narrow-focused scenario building task using proprietary tools, expert judgment and

historical time series data. The outcome usually is a brief document explaining the produced

quantitative data with a short storyline for three to six scenarios9.

The second approach uses both quantitative and qualitative analysis focusing on intuitively

run workshops as well as complex computer-based mathematical models to develop multiple

future scenarios. It is expert-led with some participation from senior managers within an

organization. The outcome is an extensive set of data-driven scenarios supported by an

extensive storyline as well as possible actions and their consequences10.

5
See van der Heijden/Bradfield/Burt/Cairns/Wright (2002), p. 63.
6
See Lindgren/Bandhold (2009), p. 22.
7
See Liebl (2002), p. 161.
8
See van der Heijden/Bradfield/Burt/Cairns/Wright (2002), p. 64.
9
See Bradfield/Wright/Burt/Cairns/Van Der Heijden (2005), p. 807.
10
See Bradfield/Wright/Burt/Cairns/Van Der Heijden (2005), p. 807.
 Center for Scenario Planning – Roland Berger Research Unit

5 The third approach is qualitatively and organizationally focused. The scenarios are

constructed within an organization using inductive or deductive processes monitored by an

experienced scenario practitioner. The outcome is a logically, qualitatively and discursively

described set of two to four scenarios all being equally probable11.

Examining the descriptions of the three approaches in more detail one can see that each

one has a different agenda and goal that it tries to achieve. The first approach follows the

classic idea of strategic planning, that is, it tries to find “the one best” strategy by giving

various scenarios different probabilities12. Based on these probabilities managers feel more

certain about the future, develop specific strategic actions for the most probable scenario

and execute them. The focus hence lies on obtaining better forecasts by perfecting

extrapolative data analysis or trend models13. This approach might work well in a stable

economic environment, but is very difficult to apply under volatile or uncertain conditions.

Additionally, the quantified scenarios are developed by experts who have hardly any

interaction with the outside world or the decision makers responsible for acting upon the

developed scenarios14. Of course, as for the other approaches the outcome of this method is

a document explaining the scenarios. However, the decision makers were not part in

developing the document meaning they seldom feel inspired, motivated or energized by the

scenarios usually describing a situation beyond their focal point of attention.

The agenda and goal of the second approach is slightly different and to a certain extent tries

to overcome the limitations of the forecasting oriented approach. By not only focusing on

computer-based models that try to attach a probability of occurrence to each scenario, this

approach involves workshops with senior managers discussing the results and scenarios

obtained from computer models. Senior managers might thus obtain strategic insights that

go beyond being presented with a specific figure on which they can develop a strategy.

11
See Bradfield/Wright/Burt/Cairns/van der Heijden (2005), p. 807.
12
Ansoff (1965)
13
See Wack (1985a), p. 73.
14
See Wack (1985b), p. 139.
 Center for Scenario Planning – Roland Berger Research Unit

6 However, they will still find it tremendously challenging to understand the uncertainty factors

and forces driving their value chain and subsequently the developed scenarios. Put briefly,

involving someone else’s model where only the results are discussed substitutes thorough

and concise thinking by senior managers, which is crucial when it comes to developing as

well as executing strategies15.

The third approach takes its agenda one step further by using company-internal inductive or

deductive processes to develop scenarios. At this stage all relevant stakeholders are

engaged in the scenario-development process. This also means that the probability of the

scenarios representing actual significance to a manager is higher. Nevertheless, scenarios

do not only represent information about a specific state of the world. Rather, they are also

about perceptions16. Thus, if one keeps the scenario building process only within the cosmos

of the organization, there is the danger of sticking to established mindsets and ignoring

previously identified uncertain factors that might have a high impact on the future

development of the business. Rather, the inclusion of internal and external stakeholders

seems necessary in order to incorporate the philosophy of expanding one’s mind and of

discussing what could happen into the scenario development process17.

This last approach is best suited to our understanding of scenario planning since it enables

all key stakeholders to advance organizational learning and readiness for unforeseen events

within a company. The other two approaches are either too complex or too expert-focused to

be fully integrated into an organization’s strategic planning process.

Having briefly explained the various scenario building approaches one can see that each

one has certain shortcomings beyond the ones discussed. In summary, they are very

resource intensive, lack a method for identifying extreme or unforeseen events and are often

15
See Wack (1985b), p. 140.
16
See Wack (1985b), p. 140.
17
See van der Heijden (2005), p. 240.
 Center for Scenario Planning – Roland Berger Research Unit

7 difficult to be introduced into an organization’s existing strategic planning activities18. We

believe that the ‘Scenario Matrix’ tool described in the next section overcomes these

shortcomings by giving extensive methodological support for developing and visualizing

scenarios. The tool known as the matrix approach was first described by Kees van der

Heijden (2005) and best fits our definition of how we perceive scenario planning.

18
See Lindgren/Bandhold (2009), p. 45-46.
 Center for Scenario Planning – Roland Berger Research Unit

8 3. Description of the ‘Scenario Matrix’ Tool

The ‘Scenario Matrix’ tool described in this paper follows an approach which was first

introduced by van der Heijden (2005). The scenario matrix is a deductive method useful for

constructing and describing scenarios in uncertain and volatile situations. Deductive

scenario methods are perceived as the most analytical and exhaustive ways of building

scenarios from an outside-in perspective19.

The scenario matrix builds and visualizes four scenarios based on two key uncertainty

factors. Four is regarded as the maximum number of scenarios that decision makers are still

able to manage20. The scenario matrix is complemented by two other tools that are important

for scenario building - the fact sheet and the influence diagram. Overall, four sub-steps are

necessary in order to design and describe scenarios on the basis of the scenario matrix tool.

3.1 Sub-step 1: Scenario identification

At the heart of scenario identification is the scenario matrix. The scenario matrix is based

upon the two key uncertainty factors that have been identified in step three, the trend and

uncertainty analysis, of our six-step process to scenario-based strategic planning using the

‘Impact/Uncertainty Grid’ tool. In order to construct the scenarios, it is necessary to project

each key future uncertainty with an extremely positive and negative outlook along the x and

y axes of the matrix. Subsequently, one can position the scenarios in the four quadrants of

the developed matrix, thus automatically generating four distinct scenarios (Figure 2). The

two key uncertainty dimensions are hence the basis for building as well as describing the

four scenarios. We commonly develop scenarios that look three to five years into the future.

This matches the typical time frame for strategic planning activities. Each scenario should be

given a concise and easily memorable name. When brainstorming for relevant names one

can for example consult historical events associated with the scenarios such as Greek

mythology. The scenario name should enable the reader to quickly capture the story of the

scenario and to understand the alternative worlds which the scenarios describe. It is
19
See van der Heijden (2005), p. 253.
20
See Wack (1985b), p. 146 and van der Heijden (2005), p. 245.
 Center for Scenario Planning – Roland Berger Research Unit

9 important to focus the name on the chain of causes and effects behind the scenario

description, the so-called influence diagram, rather than its end-state21.

Figure 2: Scenario Matrix

3.2 Sub-step 2: Creation of an influence diagram

In the second sub-step the stories behind the scenarios need to be built. These stories

describe the paths along which the world will arrive at the four alternative scenarios22 In

order to derive these stories, we generally build a chain of causes and effects leading to

these end-states. This chain of causes and effects is called the ‘influence diagram’ and

describes the strategic levers behind the scenarios23.

In order to develop the influence diagram it is necessary to establish a list of factors, forces,

trends and their interrelation. Here, the trends and uncertainties identified in step three of our

six-step process are a good starting point. It is important to select the most important factors,

to link them together, to look for interdependencies and to analyze how one development

impacts on another from the moment the scenarios are written up to five years into the

future.

21
See van der Heijden (2005), p. 260.
22
See van der Heijden (2005), p. 247.
23
See Wack (1985b), p. 140.
 Center for Scenario Planning – Roland Berger Research Unit

10 When visualizing different future developments using an influence diagram it is important to

ensure the authenticity and consistency amongst the various developments. Linking a trend

with a critical uncertainty needs to be done unambiguously using arrows displaying the

influence one development has on another. For example, describing a future scenario where

an increase in taxes for venture capital firms leads to high investments in biotech start-ups is

not plausible and consequently discredits the whole scenario building process. This can be

avoided by clearly distinguishing developments from specific events and testing whether the

developments are capable of going up or down, e.g. putting an ‘increase in’ in front of taxes

(Figure 3).24 It is thus the role of the scenario project leader to perform a sanity check testing

each linkage amongst trends and uncertainty factors for inconsistencies.

Figure 3: Influence Diagram

At this stage it should be noted that we advise the development of the scenario axis and

influence diagram to be conducted in a workshop setting. Contributors to the workshop

should be the participants identified in step one of the process using the ‘framing checklist’

tool. Usually, these contributors consist of senior executives, industry experts and specialists

involved in a company’s strategic planning activities. The workshop should be facilitated by a

moderator guiding participants through the steps described above. The key advantage of

24
See van der Heijden (2005), p. 233.
 Center for Scenario Planning – Roland Berger Research Unit

11 developing the scenarios in a workshop setting stems from the fact that all key participants

are actively involved in the process of ensuring consistent and plausible scenarios.

3.3 Sub-step 3: Scenario description

Once the influence diagram has been completed and all interdependencies have been

checked, one can start describing the four scenarios in continuous prose. Here, the

previously explained influence diagram should be used as a basis for describing the dynamic

nature of each development. By systematically describing why a certain development

happens and how this influences another development one creates the basis for writing the

story25. At this stage there are two writing techniques: First, one can write small text modules

for each trend and uncertainty on the influence diagram. Depending on the type of scenario,

these text modules have different forms, e.g. positive development of GDP for scenario A

versus stagnating GDP growth for scenario B. Upon completion of writing the text modules

one needs to place them in a logical order as done in the influence diagram. Subsequently,

the different text modules need to be connected. When writing the scenarios in this manner

one usually starts with the global or macro perspective breaking it down to an industry or

company level depending on the scope of the scenario project.

Second, one can apply a more creative technique by not focusing on each trend and

uncertainty on the influence diagram individually, but looking at the whole picture and taking

the influence diagram as an orientation. Here one should start with the final outcome of the

scenario and explain, using the various trends and critical uncertainties, what has to happen

in order to arrive at the final state. Given that this technique focuses on a free writing style

there exists the danger of giving strategic recommendations rather than describing the

environment of the scenarios.

25
See van der Heijden (2005), p. 259.
 Center for Scenario Planning – Roland Berger Research Unit

12 At this stage looking back at the ‘Impact/Uncertainty Grid’ one can see that our method for

describing the scenarios is not a random task of putting together unsystematic future

developments, but a precise and well structured process based on a thorough and validated

set of developments for the future. Next, one can complete the literary description of the

scenarios by giving each scenario a concise headline and sub-header as it is done in

newspaper articles. This step helps to capture the scenario reader’s attention, makes it

easier to communicate the essence of each scenario and above all stimulate creative

thinking about future developments.

3.4 Sub-step 4: Creation of a fact sheet

The last step of the ‘Scenario Matrix’ tool is to establish a brief fact sheet for each scenario.

A fact sheet should contain the relevant numbers, key indicators and a short description of

each scenario. When looking at a fact sheet, the reader should quickly understand the

current situation given the scope of the scenario, the relevant measures on which it is based

and what the scenario actually looks like (Figure 6).

Having completed the description of the scenarios and the fact sheet one should perform a

final check examining whether the scenarios fulfill the purpose they have been developed

for: Do the developed scenarios help to understand and anticipate uncertainties as well as

risks? Have the scenarios revealed strategic opportunities one was previously unaware of26?

If the answer to both questions is yes, then the scenarios should lead managers to perform

certain actions based on the scenarios. If the answer to the questions is no, than the

scenarios are simple guesswork and should hence be revised.

3.5 Application to the European Airline Industry

We recently applied the ‘Scenario Matrix’ tool to the European Airline Industry in a scenario

study written from the perspective of European Network carriers. The two key uncertainties

26
See Wack (1985b), p. 140.
 Center for Scenario Planning – Roland Berger Research Unit

13 identified are ‘degree of regulation of the industry in Europe’ and ‘price sensitivity of the

customer base’. Afterwards, we conducted a workshop with industry experts to develop four

industry scenarios. In this workshop each uncertainty was placed on one axis of the matrix

respectively and given a more positive and a more negative connotation – positive and

negative from the perspective of the companies for which the scenario analysis was done, in

this case European network carriers like Lufthansa, Air France-KLM or British Airways. In

terms of the ‘degree of regulation of the industry in Europe’, the positive development is a

‘protectionist regulation of the industry in Europe’ and the negative development an ‘open

regulation of the industry in Europe’. In terms of the ‘price sensitivity of the customer base’,

the positive development is a ‘decreasing price sensitivity of the customer base’ and the

negative development an ‘increasing price sensitivity of the customer base’. Afterwards,

each scenario was given a concise and easily memorable name (Figure 4).

Each name symbolizes the various developments within the scenario, e.g. ‘Europe under

Siege’ represents European Network Carriers having to deal with a very open regulation of

their industry while at the same time experiencing a decreasing price sensitivity of its

customer base. Hence, due to the open regulation their home market is under attack by

airlines from outside Europe while the whole market is benefiting from a consumer’s

willingness to pay more for flights.

Figure 4: Future Scenarios for the European Airline Industry

 Center for Scenario Planning – Roland Berger Research Unit

14 In the next step, an influence diagram displaying the developments necessarry to take place

until 2015 for the key uncertainties to develop was established. Developments included in

the influence diagram are e.g. the ‘level of innovation of European Airlines’ or the

‘service/comfort/price expectation of clients’ (Figure 5). As one can see, the authenticity and

consistency amongst the various developments is ensured. Based on these consistent

developments, the scenarios were extensively described in continuous prose.

Figure 5: Influence Diagram for the European Airline Industry until 2015

In a final step we produced a fact sheet for each of the scenarios (Figure 6). The fact sheet
for the ‘Network Fortress’ scenario contains relevant numbers, key indicators and a short
description of the scenario. The final outcome of having applied the ‘Scenario Matrix’ tool to
the European Airline industry is a precise set of four different industry scenarios until 2015
as well as an influence diagram and fact sheet for each scenario. Performing the check if the
developed scenarios fulfill their purpose provides a positive result. Feedback received from
industry experts having read the scenarios indicates that the scenarios anticipate
uncertainties and risks and show managers strategic opportunities that they were previously
unaware of.
 Center for Scenario Planning – Roland Berger Research Unit

15

Figure 6: Fact Sheet ‘Network Fortress’ Scenario for the European Airline Industry
 Center for Scenario Planning – Roland Berger Research Unit

16 4. Evaluation

The ‘Scenario Matrix’ tool is a well-structured, efficient and clear-cut method for developing

four scenarios supported by an influence diagram and a fact sheet. Its main advantages lie

in the logical and quick way in which the scenarios are developed. Thoroughly applying the

tool in an extensive form as it was done in the scenario study on the European Airline

industry takes about five to six man-days plus the resources required for the half to one day

long scenario workshop. Hence the manpower and resources required to apply the tool and

develop extensive scenarios are marginal compared to other scenario development

techniques.

Nevertheless, applying the tool to the European Airline industry produced some controllable

shortcomings. First, developing four scenarios based on two key uncertainty factors

supported by an extensive list of future developments stemming from the influence diagram

does not guarantee full completeness of the scenarios. Second, despite establishing an

influence diagram checking the authenticity of each influence factor, there remains a certain

danger that logical pitfalls do appear within the scenarios. Even so, when applying the

‘Scenario Matrix’ tool as part of the whole six-step process to scenario-based strategic

planning according to the specifications given in each tool description, the risk of the two

shortcomings appearing is only limited.

Concluding, the outcome of the ‘Scenario Matrix’ tool is a plausible set of four scenarios

indicating how an industry can develop in the future. These four scenarios enrich the

strategic planning process by leading to creative thinking and an active engagement with the

future and thus respond to a manager’s deepest worries. To further benefit from the various

scenarios one has to derive strategic implications for an industry or company for each

scenario. This process will be explained in our tool description focusing on the ‘strategy

manual’.
 Center for Scenario Planning – Roland Berger Research Unit

17 Bibliography

Ansoff, H.I. (1965): Corporate Strategy. New York: McGraw-Hill, 1965.

Bradfield, R., Wright, G., Burt, G., Cairns, G., van der Heijden, K. (2005): The Origins and

Evolution of Scenario Techniques in Long Range Business Planning, in: Futures 37, pp.

795-812.

Liebl, F. (2002): The Anatomy of Complex Societal Problems and its Implications for

Operational Research, in: Journal of the Operational Research Society 53, pp. 161-184.

Lindgren, M., Bandhold, H. (2009): Scenario Planning – The Link Between Future and

Strategy, Palgrave Macmillan: Houndmills, 2009.

Metz, B., Davidson, O.R., Bosch, P.R., Dave, R., Meyer, L.A. (2007): Climate Change 2007:

Mitigation, Cambridge University Press: Cambridge, 2007.

Van der Heijden, K., Bradfield, R., Burt, G., Cairns, G. Wright, G. (2002): The Sixth Sense –

Accelerating Organizational Learning with Scenarios, John Wiley & Sons: Chichester, 2002.

Van der Heijden, K. (2005): Scenarios – The Art of Strategic Conversation, 2nd Ed., John

Wiley & Sons: Chichester, 2005.

Wack, P. (1985a): Scnearios: Uncharted Waters Ahead, in: Harvard Business Review, Sep-

Oct 1985, pp. 73-90.

Wack, P. (1985b): Scenarios: Shooting the Rapids, in: Harvard Business Review, Nov-Dec

1985, pp. 139-150.

 Center for Scenario Planning – Roland Berger Research Unit

18

Wulf, T., Meissner, P., Stubner, S. (2010): A Scenario-Based Approach to Strategic Planning

– Integrating Planning and Process Perspectives of Strategy, HHL Working Paper: Leipzig

2010.
Contact:

HHL – Leipzig Graduate School of Management

Center for Scenario Planning – Roland Berger Research Unit
Address: Jahnallee 59, 04109 Leipzig, Germany
Phone: +49-341-9851675, Fax: +49-341-9851679
E-mail: [email protected]
Webpage: www.scenariomanagement.de