California Management Review
California Management Review is a premier academic management journal published at UC Berkeley
by Trudi Lang and Rafael Ramirez
Image Credit | Bogdan Magenta
Many entities produce scenario sets and make them publicly available with the intent that they be used widely by leaders. These include scenarios for finance, the energy transition, climate change, international relations, and specific industries. (Unfortunately, however some of this work is more accurately described as runs of models but presented as scenarios, which research has found to be confusing.1)
For example, the Network for Greening the Financial System (NGFS), the Taskforce on Climate-related Financial Disclosures (TCFD), and the U.S. Commodity Futures Trading Commission produce or advise on the use of climate-related finance scenarios. In the same way, the International Energy Agency, the World Energy Council, IHS Cambridge Energy Research Associates, Shell and BP have all published scenario sets on the energy transition. For climate change, the now famous Intergovernmental Panel on Climate Change (IPCC) scenarios have been issued in several iterations with the expectation they will guide widespread global business action and government policy. In international relations, the Center for Strategic and International Studies published scenarios on the future of the US - China relationship and Chatham House has done so on trans-Atlantic relations. In addition, industry groups such as the International Air Transport Association (IATA), professional associations such as United European Gastroenterology, international organizations including the World Economic Forum, and business networks such as BSR have developed scenarios for use by members and stakeholders.
Unsurprisingly, the uncertainty concerning what future course the COVID-19 pandemic might take and what the nature of the ‘post-pandemic’ world may be, has led companies such as Salesforce, think tanks such as the International Institute for Strategic Studies and member associations such as the World Energy Council to publish COVID-19 related scenarios for others to use.
But – is the purported usefulness of these scenarios translated into actual useful application? Our experience as teachers of one of the best scenario planning programs in the world suggests that in many cases this is not so.
Scenarios, or plausible descriptions of how the future context might unfold, when done well are designed to help executives and their organizations better navigate deep uncertainty.2 Scenarios do this by enabling executives to frame and reframe their understanding of what could happen in their specific context – thereby supporting the generation of new perspectives and options to guide priority-setting (in the here and now) in useful ways.3
The Oxford Scenario Planning Approach (OSPA), which we have developed with colleagues over the last two decades, urges executives to develop bespoke scenarios tailored to their organization’s unique circumstances.4 We have found that this ensures the scenarios that are produced are highly useful. However, as we saw in the opening paragraphs, many sets of public scenarios are released by companies, associations, inter-governmental organizations, think-tanks and consulting firms with the purported intent of having executives in other settings - indeed in ‘any’ setting - use them. Governments are even starting to mandate the use of some of these types of ‘scenarios’ to meet regulatory requirements.5
While there is plenty of guidance about how leaders can develop their own bespoke scenarios for their own specific uses,6 to our knowledge there is no detailed guidance about how to effectively use public scenarios developed by others, particularly those produced ‘for anyone’ (as opposed to those created ‘for someone’ but then made publicly available for everyone and anyone). This matters because scenario sets are developed in a specific context for a certain purpose and cannot simply be transferred into a different setting and be useful without first being re-purposed.
Utilizing an existing set of scenarios produced by someone else is not at all straightforward.
Scenario sets made ‘for anyone’ about a given system (the future of this or that) tend to have a positivist or cartesian stance, where the scenarists have stood outside the system looking at it objectively. In contrast, those scenario sets which have been designed for one particular user take a phenomenological stance, where the future context is assessed from their unique perspective.7 To make the cartesian scenario sets useful, they need to be helped to become phenomenological – that is, specific to the user’s own situation.
For example, we had a superb student several years ago who was a senior planner in a large international corporation which had taken up a set of well-researched scenarios produced by a world-class team working with the World Economic Forum (WEF). These scenarios had been developed over many months through inclusive stakeholder workshops in all continents and discussed extensively by professors, industry experts, consultants, CEOs of major NGO’s and firms, as well as government ministers. The planners in our student’s firm set out to use these generic scenarios to test their company’s strategy.
Problems arose when the corporation’s senior executives requested views on a new issue they wanted to assess and discuss – an issue which had not been covered by the original WEF scenarios. Preparing the materials for this senior management conversation, the planners responded by creating an additional scenario that did help the senior team to have the conversation they wanted to hold, but the planners found it was impossible to incorporate this newly added scenario into the original WEF set without substantially changing how the scenarios in the set related to each other. In other words, the new scenario was now informing a major strategic concern but did not fit into the existing set whose purpose was to assess the corporation’s strategies more generally. This is problematic because much of the strategic value derived from a scenario set arises from being able to compare the differences among them.
Meanwhile, elsewhere in the corporation a shorter-term set of scenarios was created to address more urgent matters. These ‘ad-hoc’ scenarios could be related as a set with the new scenario the planners had developed for the senior executives. However, the new home-made scenarios only covered some of the possibilities the WEF scenarios had considered. So, the firm ended up with two incompatible sets - one fully developed, but not fitting their own immediate use requirements, which they had tried to amend with mixed results; and another set made for more short-term and narrower strategic concerns which was not helpful for the wider intent that had led the company to consider using scenarios in the first place. In the end, the firm dropped the WEF set and created its own new bespoke set of scenarios that was better able to serve the purpose of testing their own strategy, as it covered the specific issues this company alone needed to address.
This example illustrates our view that utilizing an existing set of scenarios produced by someone else is not at all straightforward and can lead to confusion and problems which can be avoided.
So, if executives do choose to go down this route of using a set of scenarios produced elsewhere such as those highlighted above, how do they maximize the upsides of doing so while mitigating the downsides?
The purported advantages of using an existing, generic set of scenarios are said to be savings in cost and time – which are especially helpful for organizations with limited resources such as small and medium sized enterprises. Proponents also suggest that a set of scenarios developed by someone else can provide a potentially fresh outside perspective, or a place from which to start one’s own scenario planning. Yet there are plenty of potential disadvantages to taking this route as the experience of our student illustrated.
Combining our six decades of practicing, researching, and teaching scenario planning that has contributed to the development of the OSPA,8 we offer five tips to avoid the pitfalls of deploying public scenarios for one’s own use.
1. Start by clarifying the user, the intended use or purpose of using the scenarios, and the specific applications for your organization.
For scenarios to be useful and used, the intended user and their purpose for deploying a set of scenarios - as well as determining as precisely as possible how they are going to apply the outcomes - must first be determined. Our research finds a high correlation between securing these matters and perceived satisfaction from using scenario planning.9 For example, in your organization is it the head of strategy, or the CEO, or the director of a business unit, or head of the risk committee, or the director of HR, or the chair of your Board that will be the primary user of the scenarios? And what is their purpose for using them? Is it to reframe the organization’s identity, or to test strategy, or to make sense of changing conditions, or to generate new options, etc.? And finally, how exactly are the outcomes of the work to be applied and when? For example, will they inform a capital allocation decision your organization has to take over the next month, or inform a big negotiation, or be part of a strategic conversation at your next Board off-site meeting? Being crystal clear about these three elements of user, use and application will ensure that the most appropriate set of scenarios is chosen and effectively tailored to the specific context in which they will be used. It will also guide you as to what you will need to adapt from the original set and what you will need to add as well as drop. When people do not do this, they sometimes instead seek a ‘focal question’ which we find does not work as well to anchor the scenarios so that they are useful. Without a user and a use, even well focused scenarios risk being useless – and thus seen to be expensive and a waste of resources, risking scenario planning getting a bad reputation that is unwarranted and avoidable.
2. Make sure you choose true scenarios, rather than sensitivity analyses of forecasts or model runs of extant models that misuse the term.
As we stated above, the term ‘scenarios’ is often confusingly used to depict model runs and, sometimes also sensitivity analyses (upside and downside projections). This is done especially in fields such as finance and climate science but is not what is meant by the term ‘scenarios’ here. Scenarios are contrasting descriptions of how the future context could unfold, based on the various outcomes of a wide range of inter-connected factors in areas such as technology, ideology, economics, social arrangements, regulations and the environment. A sensitivity analysis on the other hand is more closely related to forecasting in which individual elements of the forecast are given different values to see what the impact would be on the forecast, while model runs simulate outcomes of different values of the underlying parameters of the model. Both models and forecasts essentially work with one frame of the future whereas scenarios come in sets providing contrasting and divergent frames of the future to help address deep uncertainty.10
3. Identify the most appropriate set of scenarios.
When using an existing set of scenarios, the focus shifts from ‘manufacturing’ one’s own bespoke scenarios to ‘choosing’ from what is available. An excellent set of scenarios is one that is useful, relevant, plausible and challenging in relation to the user’s intended use. Choosing a set of scenarios should be guided by these criteria (as should manufacturing) and involve comparing different sets as a first iteration and then, after considering this carefully, deciding on the actual set to use in a second iteration. For instance, different scenarios on energy transitions give coal a very different future role, so a decision has to be made about which set is most helpful for the intended purpose; one where coal is kept as a viable energy source for a long time or one where it is taxed out of existence earlier? Or if the purpose is to use scenarios to test the robustness of your organization’s strategy in light of the impact of COVID-19, how well will the given set of scenarios help you do this? Do they challenge the very assumptions you want questioned – or other assumptions which are less important for your own intended use? Other more general questions include understanding whether the scenarios provide a set of perspectives that are relevant and plausible with regards to what was previously assumed by key people in your organization about their future context, and which scenario set is going to deliver the most insights and value? And, as we see in the next tip, how much extra work is required to render the original set sufficiently helpful?
1. Have you defined who the primary user of the scenarios is in your organization; what purpose they have for using the scenarios; and how the outcomes will be used?
2. Are you using actual scenarios, rather than sensitivity analyses or model runs?
3. Does the scenario set you have chosen provide the basis for generating relevant, plausible, and challenging perspectives for the user?
4. Have you tailored the scenarios to make them fit-for-purpose?
5. Are the scenarios now sufficiently developed to generate useful insights for the user and the purpose of the scenario planning?
4. Tailor the scenarios and render them fit-for-purpose.
Once you have chosen a relevant set, the next important step is to tailor the scenarios to your organization, specifically to the particular user in it and their reason for wanting to use them. This includes considering what factors are missing in each scenario that would make them more pertinent. You may also want to revise the time horizon from the original set – do you need to shorten it or lengthen it? Does a scenario from the set need to be dropped or replaced? In addition, each scenario should be fleshed out to include developments in the organization’s own immediate business or transactional environment, which will almost certainly be missing from the original generic set. For example, what new entrants could emerge in your business environment in each scenario? Who would win? Who would lose? What changes in roles would be seen in each scenario? What new collaborations could be envisaged? What would be the new competitive dynamics? These questions enable the scenarios to become sufficiently tailored and detailed to be productively useful for your organization.
5. Test the usability of the scenarios.
Once these tailoring activities have been undertaken, the usability and value of the scenarios can then be evaluated against the purpose that was identified at the beginning of the process, with at least two test runs recommended, so as to debug the problems and add any elements you find are missing. Relevant questions to ask at this stage include what developments are currently occurring or beginning to unfold that validate each of the scenarios? What signals do the scenarios suggest should already be tracked to understand their unfolding? Are the signals different enough from one scenario to another to indicate you have a sufficiently diverse set of scenarios or are they too similar and you need further work to differentiate them? Do the events in the early years of the scenarios suggest interesting options? If not, do the scenario storylines need to be further developed? Do the scenarios lend themselves to be represented in contrasting system maps? Do the scenarios generate useful strategic insights? If not, is further work on them needed? Once you are satisfied that the scenarios are now highly useful and insightful you can progress to using them for the purpose you set out.
While we believe the benefits of executives developing bespoke scenarios for their own organizations is without question preferable, the reality is that there are many scenario sets being published with the intent of having others use them. Our aim in outlining these tips is to provide executives with very practical guidance about how to maximize the value of using these scenarios while avoiding making mistakes which are expensive and can be avoided.
1. R. Ramirez, T. Lang, and G. Peterson, Seven sure-fire ways to do scenario planning badly: A guide to poor practice, Long Range Planning. (1 April 2021).
2. J. Kay and M. King, “Radical Uncertainty: Decision-making beyond the numbers”. (New York: W. W. Norton & Company, 2020).
3. R. Ramirez and A. Wilkinson, “Strategic Reframing: The Oxford Scenario Planning Approach”. (Oxford: Oxford University Press, 2016).
4. See for example R. Ramirez, S. Churchhouse, A. Palermo, and J. Hoffmann, Using scenario planning to reshape strategy, MIT Sloan Management Review. (Summer 2017): 31-37.
5. See for example how the UK Government is requiring pension schemes to use a form of scenarios to report on financial risks.
6. See for example P. Scoblic, Learning from the future: How to make robust strategy in times of deep uncertainty, Harvard Business Review (July-August 2020): 39-47.
7. R. Ramirez and J. Selsky. Strategic planning in turbulent environments: A social ecology approach to scenarios, Long Range Planning, 49 (2016):90-102.
8. R. Ramirez and A. Wilkinson, “Strategic Reframing: The Oxford Scenario Planning Approach”. (Oxford: Oxford University Press, 2016).
9. R. Ramirez, Y. Bhatti, and S. Tapinos, Exploring how experience and learning curves decrease the time invested in scenario planning. Technological Forecasting and Social Change, 151 (2019):1-12.
10. J. Kay and M. King, “Radical Uncertainty: Decision-making beyond the numbers”. (New York: W. W. Norton & Company, 2020).