Scenarios for Success: Turning Insights into Action
Bill Sharpe and Kees van der Heijden (editors)
This book is a compilation of chapters written by people with extensive experience in scenario planning as a business tool. Many trace their experiences to Royal Dutch Shell, or Pierre Wack, and the methods developed there to use scenarios to prepare organizations for change. Because this is a compilation, there is no real storyline to the contents. Nevertheless, what comes through is the concept that “business as usual” planning prepares organizations poorly to cope with external changes which may be a departure from current circumstances. Very few business planning tools offers the possibilities that scenario planning does to identify and cope with the uncertain nature of these external changes.
This summary will condense a small portion of the book.
What conditions are needed for scenario planning to work? (1) For strategic scenarios, it must be led by the person with the decision rights (2) If leaders are not ready to change their mind; they are not ready for scenarios. Scenarios are a tool for change – usually significant change. Changing their mind requires engagement with the process as a participant. Just receiving a presentation of the output will not change minds. (3) Good scenario work takes time: time to immerse in the situation and time to think it through. This is a reflective practice and there can be a lot to absorb before reaching any decisions.
There are many links between systems thinking and scenario planning. Organizations comfortable with formal system thinking adapt quickly to scenario planning.
A very important feature of scenario planning is identifying the “predetermined elements”. These are things that have already happened, but the consequences of these events have not yet occurred. One example is the effect of winter snowfall on spring floods. Even if you know that the snow is out there, you can’t predict the floods because you don’t know about the warming rate and spring rains. So even if there are predetermined elements, there is also uncertainty. Some things are inevitable, but uncertain. This seems like a paradox that people struggle with. For this reason, scenarios need to combine rigorous and intuitive thinking, and develop a good set of facts to backstop the stories developed.
The future depends on a combination of predetermined and uncertain elements. Real value is created by determining which is which. Predetermined elements are ‘those events that have already occurred (or almost certainly will occur) but whose consequences have not yet unfolded”*. There are two types of predetermined elements of greater importance. The “obvious” elements need to be identified and agreement reached that these events are in progress. The “systemic” elements are more important and represent the effects of existing feedback systems on events. Because these systems are NOT obvious, they may require deep investigation. The uncertain elements reflect both uncertainty about whether certain events will ever happen and at what time they will happen.
The experience at Shell showed that converting scenarios into action depends on the persuasiveness of the scenarios and the mental models of the decision makers. For this reason, it is vital to develop multiple scenarios, one of which is a “do nothing” scenario. Without a scenario to represent the status quo, no decision maker can make a decision for substantial change – and maybe not even then.
Helping decision makers re-perceive the future is critical. Existing mental models are highly effective in a stable environment, but may be blind to important events and consequences in a volatile environment. Multiple scenarios also served to help decision makers learn about their mental models and re-perceive their usefulness and limitations. In this sense, scenarios were learning tools.
Good scenarios combine rational and creative elements. While the more creative elements may be entertaining, it is the combination with rational analysis that enables action to be considered. This is expressed in terms of “seeing” and “knowing”, where seeing is the creative and knowing the analytical aspects of the scenarios. Seeing is characterized by: “maps”, processes, fiction and openness. The contrasting characterizations for knowing are: forces, content, fact and closed.
The scenario planning practice of Wack was very deep, slow and intense. Current practice is shallower, quicker, and moderate. In particular, Wack’s work involved decision makers in developing the scenarios while modern consultants often do the work without the decision makers, who do not contribute to or engage in the development process or content. It is not obvious that these differences are important, as they may reflect difference in intent. Modern methods may be good for developing hints about the future more than deep understanding. They may be even more suitable for a more complex volatile world, but that is not obvious either.
Scenario planning recognizes two main environments surrounding the “client”. The client operates directly in the transactional environment. This could be a company and its suppliers, customers and competitors. Surrounding the transactional environment is the contextual environment. The contextual environment is resistant to influence by the client. Typical contextual actors or elements are governments, the economy, currency valuations and technological advances.
Changes in the contextual environment create opportunities for change in the transactional environment. The organizations that recognize and react properly to the changes in forces in the contextual environment gain competitive advantage.
In a strategic context, the authors advance 5 propositions.
- A key aspect of strategy involves the invention of original options that create new value constellations along the boundary between the contextual and transactional environments.
- The creation of new value constellations inevitably requires rethinking the roles and relationships of the stakeholders involved.
- To be fully effective, the strategic use of scenarios must involve the systematic consideration of options rather than leaving them to intuition or other unrelated or unstructured approaches.
- Change is most sustainable if pursued as a transitional activity.
- The concept of transitional change involves keeping aspects of the current state while replacing aspects of the current state with new features. Don’t change everything at once and don’t create the expectation of a fixed future state. Transition may be continuous.
- As a tool for transitional change, scenario planning is best pursued using a prototyping process called ‘staging’.
It is difficult to rationally describe future states based on today’s knowledge. This is especially the case when the systematic features of today’s processes are poorly known. Often, the best starting place for scenario development can be a comprehensive examination of the systems supporting the status quo. It is these systems that maintain the status quo and that would need to change for a different future to emerge.
A common approach to scenario building starts with identifying significant uncertainties about the future. These are reduced to two independent factors and displayed as a grid (see below); the four quadrants represent the scenarios in this system. In the example below (I made this up), the four scenarios represent four distinct futures with four different interactions between consumers and government approved claims.
Not all scenarios in such a scheme are equally believable, but all must be worked through before making decisions based on this approach. This can be a challenge; decision makers sometimes request elimination of all but one scenario, ask to develop that one, then choose between that scenario and the status quo. This defeats a good part of the reason to do scenarios and misses the value of the options created in the less obvious scenarios.
The next stage in development is identification of the systems that are in place (or might come into place) in each of the four scenarios. The authors offer what they call the “triangle of deep structure” which I approximate below.
The top two layers are usually directly observable, but the third layer is often driving the second layer in a discontinuous or non-linear way. Structures in the third layer are usually mapped as “casual loops” (see below) consisting if reinforcing and balancing loops. These systemic structures function to approximately maintain conditions, but when the reinforcing or balancing force becomes dominant, the system can break down or move to an entirely new condition. The reinforcing or balancing systems may respond slowly, creating the illusion of a more fundamental change and provoking reaction from stakeholders. These actions may cause the system to over-respond setting off a round of more volatile movements.
Next, the dominant systems in each of the potential scenarios are described and combined with the basic four scenario description. These systems are probably very different in detail in each scenario.
In fact, the evolution of these systems may create and maintain the distinctively different scenarios. Because of the implications for decision making, cooperative identification of the scenario-specific systems by scenario planners and decision makers is valuable in developing a belief in the potential validity of the analysis.
Time is the most difficult uncertainty to cope with. It matters almost as much when things happen, as whether they happen. The authors suggest thinking about time in three horizons. Horizon 1 is essentially the continuation of current forces. Horizon 2 is a mix of the status quo and truly new systems. Horizon 3 represents a shift to new systems and conditions. Horizon 1 is clear, horizon 2 is ambiguous (new or old? new and old?), and horizon 3 is a step into the unknown. The Greeks defined two kinds of time. Chronos is the time we usually think about; linear and regular. Kairos is time as an opportunity; this is the concept of seizing the day. As you progress from horizon 1 to horizon 3, the situation goes from being essentially chronos-driven to being kairos-driven. Scenarios that involve significant breaks between current and future dominant systems are less likely to proceed under chronos-time and more likely to proceed in kairos-time. A nice example of this is the historical spread of land line telecommunications in Africa compared to the spread of mobile phones in Africa. Because the new system had a different foundation than the old system, the transition from nearly no telecommunications to common mobile service was almost instantaneous.
The most disruptive points occur when the horizons cross. When the horizon 1 pattern “fails” and is replaced by a horizon 2 system, there is significant disruption while the horizon 2 system replaces horizon 1. The same things happen when horizon 3 overtakes horizons 1 and 2.
A central issue in scenario work is the effect of heuristics (habitual judgments) and bias on people’s perception. The original psychological studies were oriented towards conventional decision making, but the “traps” discovered there also apply in scenario development and selection. People:
- Over or under estimate the probability of an event based on their own experience, and they gain confidence in the estimates over time (representativeness heuristic).
- Estimate of probability of an event based on the frequency with which it occurs. More frequent events are more probable (availability heuristic).
- Anchor their analysis based on the first estimate they obtain. It can be demonstrated that random numbers can anchor estimation (anchoring and adjustment trap).
- Persist in their confidence in their estimates, even when presented contrary information (belief persistence).
- View information that confirms their existing opinion as reliable, but information that contradicts their existing opinion as unreliable (confirmation bias).
- Frame problems and notice information based on their existing mental models, so that information supports the mental model (experience bias).
- Express higher confidence in their judgments that objectively merited, even in the face of corrective information (overconfidence bias).
- Strive to dismiss options until only a single option and the status quo remain. This occurs essentially from the beginning of the process and leads to under-development of options (single outcome bias).
These constraints on thinking simplify potential decisions and can be very effective in dealing with truly commonplace, frequent events – but they are more dangerous when applied to infrequent, uncertain future events.
A similar investigation into the ease with which people reason inductively or deductively illustrates some of the problems people face when developing scenarios. Generally, people have a much easier time starting from the present situation and “explaining” different possible futures. It is much more difficult to assume some potential futures then imagine the series of events that led to the various futures. Apparently, deducing an outcome is familiar and comfortable, while inducing the past from the future is not. Groups tasked with inducing a past often skip back to the near present and deduce their way forward instead of working stepwise backwards. This is important because much of the value in scenario development comes from identification of turning points that are linked more to the future outcome than to the present state. The difficulty with induction suggests that extra effort is needed to help people become comfortable with induction and remain disciplined in its appropriate use.
Groups working on scenarios do not progress steadily. The more common pattern is periods of limited progress with occasional bursts of progress. Termed “punctuated equilibrium” by analogy to the evolutionary phenomena, this is a group phenomenon that is different from either groupthink or fragmentation. Groups struggling to make steady progress may make a sudden leap while facing a deadline. The leap may be the result of cognitive or creative breakthrough, or it may be the result of inappropriate application of a simplification. Given the tendency to “polish” things for presentations, it may be very difficult to distinguish which has happened.
The cognitive sciences tell us that humans are bounded in their rationality; they interpret and make sense of what they see going on around them through unique and interacting…lenses that comprise their mental models. And, as Pierre Wack has noted ‘in times of rapid change and increased complexity, the mental model becomes a dangerously mixed bag; enormously rich detail and deep understanding can coexist with dubious assumptions, selective inattention to alternative ways of interpreting evidence and projections that are a mere pretense.
If the main purpose of scenarios is to prepare for developing strategy, it is useful to consider the dimensions available for strategic action. Five dimensions are suggested together with the markers for each end of the dimension.
Scenarios can be successful…only when (1) they are based on a sound analysis of reality, and (2) they change the decision-makers’ assumptions about how the world works and compel him to change his image of reality.
Many strategies fail because they do not take a deep and realistic view of the world, including the context beyond the immediate current state of buyer-seller interactions. Finding the deeper structures at work can enable understanding of what is and is not important today and in the future. Because the look into the future involves so much uncertainty, it becomes very important to understand how our mental models limit our ability to view reality clearly. Our mental models and reality may differ significantly. In the end, the best we can do is to be aware of our limitations and work hard to overcome them.
My interpretation and comments
- Every individual begins their career with mental models based on their education and experience. The organizations they join soon graft on additional mental models. Over a long time, these mental models may dominate thinking. When people talk about organizational culture, these mental models play a huge role. I recently read a blog discussing the purchase of Instagram by Facebook. Why Instagram and not Kodak? Why Instagram and not Google or Facebook? Those organizations had mental models that did not let them imagine Instagram. Success and failure lock in mental models. It is very hard to change such mindset, which may be why most organizations express a desire to be learning organizations, but so few seem to learn much.
- Scenario work can be high risk, and this may be why there is resistance to it within organizations. Simplistically, all scenarios lead to three outcomes: (1) things will be about the same, so we don’t need to change much, (2) things will be much worse, and we will need to change a lot whether we want to or not, or (3) things will be much better and we will have to change, but at least that’s good for us. The real hope for most established companies is that things will be about the same in the future. That means that two out of three outcomes will call for significant change. Once people know that a significant change scenario is developed, it is very hard to ignore the information and carry on as usual. Everybody is much more comfortable with outcome (1).
- Most scenario work creates less value than possible because people avoid extreme outcomes. By confusing the scenario projection with prediction, they get hung up on the unlikeliness of the outcome rather than developing an appreciation for the possibilities that the unlikely outcome presents. I’m sure the leaders at Border’s considered the impact of Amazon to be quite minor in 2000, but the extreme idea that millions of dedicated book buyers would buy without touching the books first was too unlikely to be considered.
- Adaptive learning was mentioned in this book, but only a bit. Scenario work is a method of adaptive learning and I think its importance is that is in a way of learning by doing. There is plenty of scientific work that shows that adults learn more from doing than by hearing or seeing. Many sorts of experiments can’t be done in the real world, and certainly experimenting with the future is one of the least possible possibilities. Scenarios are one of the few ways to experiment with the future and simultaneously learn by doing. Conventional strategy development is not as experimental. The further importance of scenarios is that they often consider a much wider world than strategy. Many strategies make perfect sense, if the world around the competitors does not really change. Those strategies may make no sense, if the world surrounding the competitors is changing. Certainly every conventional entertainment business is in the middle of this sort of environmental change, and most are not doing well.
*text in italics are directly quoted from the book
Recent Comments