The Role of Uncertainties in Scenario Planning: A Philosophy

If we knew how the future would turn out, we would not need to study or explore scenarios. But nearly every headline suggests uncertainty about the future, and many of them state it outright.

Creating myths about the past and the future

Scenario planning arose as a very human response to uncertainty. As storytelling animals, we encapsulate uncertainty in our narratives. Many times, humans created great myths about the origins of the universe and the roles humans play in it. Uncertain about our origins, we created stories to fill in the blanks.

Scenario planning creates myths about the future. As much as we think of ourselves as rational and data-driven, only events very close to us can be evaluated using data.

History does offer us some data in the form of written and verbal narratives. The fossil record, and the chemicals in rocks, offer some insight into the past, as do structures in the universe, such as the cosmic microwave background radiation. But there is a lot we don’t know about the deep past though we understand much of the context—many of the details remain vague or missing. 

Questions, such as why there are so few medium-sized dinosaurs in the fossil record, persist in areas like paleontology. Going back much further, how life arose on the planet, and how it diversified—and why the basic plans of plants and animals are locked in all remain a mystery. DNA may eventually reveal some of those answers—computer modeling some—but not all. 

The past will always exist through a veil of incomplete data. From the burning of the Library at Alexandria to metamorphic rock churning and twisting the sedimentary record, to deep blankets of igneous flows and the torrents of water unleashed as ice caps melted, data was erased or displaced. Regardless of the power of our instruments, we can’t recreate the past where no data exists.

The role of uncertainties in scenario planning: Bad forecasts and the lack of data from the future

When it comes to the future, data simply does not exist. Despite physics leaving the door open, so to speak, for time travel, there is no evidence that sentient beings will ever be able to travel into the future or the past.

That means that all speculation about the future proves essentially data free. No assertion, no matter how grounded in current data, will yield a 100% accurate forecast about the future. The variables are too many, and the relationships too complex.

I keep a slide in my presentations demonstrating that even in industries with very deep pockets, their analytical skills fail them. Most forecasts don’t show the previous year’s forecast, only the current forecast. If you look at almost any forecast and track it over the years, two things will be clear quickly.

First, if the domain is large enough, there may be many forecasts, and those forecasts do not align, and it is not just because the forecasters chose to count different things. All forecasters use different models with different assumptions that yield different answers. Those assumptions attempt to replace uncertainty with intelligent, rational guesses based on historical data, most of the time statistically extrapolated forward—sometimes directly manipulated to reflect the forecaster’s view of the future. All of it with built-in assumptions that may or may not be shared.

Second, historical forecasts, if they can be found, will illustrate adjustments year-over-year for the inaccuracy of the previous forecast. That is why you don’t often see forecasts that track forecast performance—most forecasts don’t perform well. 

In 2016, for instance, Zion Market Research suggested the global VR market would reach nearly $27B. IDC estimates 2022 won’t break $410M in VR headset sales. To be fair, Statista.com’s aggregate analysis estimates 2022 at $25.2B when counting all the components of VR and AR. The Zion Research was VR alone.

As another example, a 2018 paper, Oil projections in retrospect: Revisions, accuracy and current uncertainty (Wachtmeister, Henke and Höök), explores the inaccuracy of pricing and supply forecasting. It also points to areas of “stability” where forecasts are more accurate. The global aggregate forecasts are not accurate, and I would argue that the “stability” of some components derives from the shortness of the forecast horizon. Over long periods, governments, business models, technology, and other factors will also disrupt the underlying foundations associated with near-term stability. 

In November of 2022, when this post was released, the world was in a precarious position around oil, with OPEC openly defying requests from Western leaders not to curtail production as global energy prices spiked from the Russian war against Ukraine. The confederation known as OPEC+ agreed to steep oil production cuts in early October 2022, removing two million barrels per day from the world’s supplies. Saudi representatives listed economic drivers, such as rising interest rates, and social drivers, such as the arrogance of wealth, as reasons for reducing supplies. They did not explicitly list the vulnerability of Western governments and economies stretched thin by the pandemic and support for Ukraine.

I add this very contemporary example as a contrast to the exuberant forecasts associated just a few years earlier with the fracking of oil shale in the United States, a technology that was to reduce the dependence of the U.S. on foreign oil. In the conclusion to a working paper on fracking, The U.S. fracking boom: Impact on oil prices (Frondel, Manuel; Horvath, Marco) saw fracking’s impact as minor. OPEC, it concludes, still matters.

Using monthly data spanning from January 2000 to December 2016 and a supply-side model similar to that proposed by Kaufmann et al. (2004), this study investigated the effect of fracking on the world crude oil price. Among the key results was a correlation analysis that found a statistically significant negative long-run relationship between the oil price and OPEC supply volumes that exceed the announced OPEC quota, indicating that OPEC still matters. 

The research also found a negative influence of the increased U.S. oil production due to fracking on the oil price, a result that is in line with the studies of Borenstein and Kellogg (2014) and Kilian (2017b). The simulation exercise demonstrated that oil prices would have been around $40 to $50 per barrel higher if the U.S. fracking boom had not occurred, an effect that is quite high because the study specifies the demand side is taken as given. This exercise indicated the importance of modeling both the supply and demand side to avoid overestimating the impact of a single factor, such as the additional oil supply due to fracking. 

Despite hopeful forecasts of a future increase in U.S. oil independence, fracking did not mitigate market reactions to the recent OPEC+ move nor to the energy aspects of Russia’s Ukrainian war. Fracking may well have made some difference, but it was not a game changer that significantly reduced the volatility of the oil market.

“Game-changing” technology, like fracking, may not be as “game-changing” as its advocates assert. Other technology, such as Apple’s iPhone, often dismissed by pundits and competitors early on in its lifecycle, change multiple aspects of life in positive and negative ways. Those competitors created reactionary myths about the iPhone that did not adequately explore its potential impact. 

Keep in mind that technology is a moving target. The first iPhone is a far different device than the one Steve Jobs first unveiled. But even then, those who explored the future robustly could have imagined the iPhone role. Because there is no data about the future, organizations and individuals must rely on stories about the future to offer a set of possibilities—they must go out of the way to ensure that those stories don’t reinforce bias and that they expand the possibilities by purposefully avoiding unrealistic constraints.

In the pre-industrial world, oil was not counted, at least not crude oil from the ground. Pre-industrial economics was much more interested in olive oil than crude oil. When considering uncertainty, this is not a disingenuous comment because our reliance on fossil fuels has only spanned a fraction of human existence. 

There is no reason to assume that the pace of technological change will render fossil fuels to a historical footnote—but also no reason to imagine that we won’t overcome our dependence on them. 

In scenario planning, experience limits imagination. It is unlikely that any Romans ever imagined a post-agricultural world, even though they did use technology, including roads, aqueducts and other plumbing. But they did imagine an empire and succeeded in using their technology to spread their culture around the Mediterranean and Europe.

The continued use of fossil fuels should be thoroughly explored not to create better forecasts but to expose more alternatives. Stories about a dystopian future often accompany the changes necessary to detach humanity from its dependence on fossil fuels. That is, again, a demonstration of bias constraining imagination. Stories are used as often to craft propaganda as they are to inspire courage or demonstrate positive aspects of change. Scenario narratives, in their best light, often form the basis for social and economic change driven by the hope of making the positive aspects of the stories real and mitigating the risks exposed in them.

Exposing bias, arrogance and ignorance

Humans will continue to speculate about the future if they adopt scenario thinking or not. Scenario planning offers a systematic way to explore multiple alternative futures. 

Other approaches to storytelling often employ uncertainty to describe a future that instills fear or hope based on applying a limited set of values to the uncertainties to craft the narrative to purpose. 

Political movements often paint opponent policies in terms of light or dark. There is no room for subtlety in political advertising—a form of civic discourse perhaps that proves the antithesis of the principles of civic discourse. Advertising offers no room for response. It simply appears to the eyes or the ears. Any “dialog” occurs when the opposition drops an equally non-interactive set of assertions about the future.

Bias, consciously or unconsciously, excludes even the possibility that certain futures could exist or excludes a specific set of values from uncertainties. The bias of some conservative politicians, for instance, against the principles of a “liberal democracy,” removes certain possibilities from their vision of potential futures. Bias, especially practicing confirmation bias, seeks only evidence and positions that reinforce a given belief system. Because bias can be a social structure, it often manifests passively, meaning that those who practice it live in a world that does not recognize its bias.

Arrogance, unlike bias, is never unconscious. Arrogance assumes that an individual or a group knows better than others and seeks to craft a future that reinforces their beliefs. Like bias, arrogance constrains possibilities, but it does so purposefully, not passively.

Ignorance derives from the inability to imagine the future, often imposed by those in power or through isolation and lack of access to knowledge, which may also be a mechanism those in power use to reinforce ignorance. Ignorance results in not understanding uncertainty, or if it is recognized, being able to imagine the possibilities because of a limited worldview.

Bias, arrogance, and ignorance all constrain the expansive nature of scenarios. They prune the edges of the search tree so that the possible uncertainty values assume becomes a subset, which, if employing scenario planning, abuses the scenario planning technique so that it does not challenge underlying assumptions but rather reinforces them.

If I can’t know, how do I manage uncertainty?

The first and most important action when thinking about the future is to put a name on those things that are uncertain. It may be a long list. Name the things meaningful to you, the things you can’t know how they will turn out but that you would like to yearn to—concepts and ideas that, if you knew their future, might change your life.

Not all of the uncertainties listed by an individual will be important to a strategic planning question. The focus question in scenario planning creates a boundary—a test for the applicability of uncertainty to an area of exploration. 

When thinking about the future of work, for instance, an uncertainty like regional access to water will likely not change the way people work in general. It may play a critical role in the future of work in that region—even where locations for hospitable work exist, but will likely not affect the technologies or practices of work globally.

On one hand, the focal question constrains uncertainties to those meaningful to the question—and on the other, it acts as a prompt to seek an exhaustive list of uncertainties related to the solution space. The focal question triggers the search for uncertainties beyond an individual’s immediate domain.

No individual will generate an exhaustive list of uncertainties. Teams of people typically build the list, complementing their knowledge with interviews with stakeholders and experts. If scenarios don’t start with sufficient uncertainties, they will not result in stories about the future that are as rich as they could be.

Uncertainties act as characters in scenarios. They have behaviors and attributes that describe how they are represented and how they behave in the scenarios. The interactions of the uncertain characters drive and inform scenario plots.

Regardless of the core uncertainty’s strength in driving the matrix, scenarios require many other players for the story to present as plausible, interesting, and complex.

Refresher: What are scenario planning uncertainties?

Uncertainties represent concepts critical to a scenario planning domain, the value of which cannot be determined. Uncertainties exist at various abstractions, from concepts and plans controlled by individuals or the organization to those controlled by industry or other broader group, to those over which businesses and people have very little control throughout the scenario planning horizon.

When a company creates, a new product falls into the most controllable uncertainty category. Although the company may not know when it will invent a new product or when a product under development will ship, they own almost all the resources required to make that decision and execute it.

At the next layer, group uncertainties, organizations may play a significant role in setting things like industry standards or the nature of supply chains.

At the highest level, individuals have little or no control. They may lobby for political goals, and they may even offer illegal incentives for politicians to bend to their will, but the breadth and complexity—and the timeframes involved, make most uncertainties very unclear, directionally and temporally.

Concepts like climate change fall into this category, as do political movements. The recent overturn of Roe v. Wade was uncertainty at both ends of the case’s lifecycle. With so many decades of abortion-related issues in the U.S., it was not clear that 1973 would be the year of change, any more so than 2022 would be a year when that decision was overturned. Pundits may have forecasted a directionality, but until Dobbs v. Jackson Women’s Health Organization hit the Supreme Court docket, they couldn’t see a likely moment.

Elected leaders may have appointed people to sway the courts, but the ultimate decision came down to the Supreme Court Justices in 1973 and 2022 voting on a question on their own. Yes, those justices may represent a moment as much as their legal perspectives and knowledge, but neither case could be forecasted as “the” case a decade before, and no one could forecast a decade out, either outcome.

Uncertainty is defined not just by a lack of data but by differences of opinion. Scenario planners migrate toward issues of controversy where no single opinion holds sway, and if it does, that opinion is often formed by bias.

As for data, a similar uncertainty profile emerges: disagreements about the interpretation of data.

Perspective: Driving forces and uncertainty

Scenario planners often talk about driving forces and uncertainty. Driving forces describe those epics like climate change or demographics, areas with momentum toward viable forecasts that may lead to a sense of certainty in its broadest sense.

I avoid seeing driving forces as anything but uncertainty. While climate change permeates many aspects of the future, its exact nature remains uncertain. We do not know precisely when certain features of climate change will arrive in a systemic way that permanently changes the weather. We do not know, for instance, the tipping point of sea level rise that will disrupt a major metropolitan area. We do not know if nations will heed warnings or exacerbate existing problems.

In demographics, we can know the count for how many people of each generation will exist, though some argue even about those numbers. We do not know how the generations will relate—or how they will behave or influence each other.

In my practice, I see uncertainty and change as the only driving factors—nothing else is certain enough that it cannot be swept aside or modified significantly over the planning horizon. History is filled with social and technological moments that surprise those living through them. A single human life may not see any of these moments, some may experience many of them.

At the end of World War II, the future Queen Elizabeth could not have imagined much of the world that existed at the time of her death. In England, the Monarchy may be seen as a driving force, but the Monarchy of today differs significantly from the one Queen Elizabeth inherited, and the same is true of the British Commonwealth she led.

Many leaders have appeared to be on the side of a lasting right in the past, leading great movements toward a predestined future only to see their movements become paragraphs edited down for space due to the human proclivity to create great, sweeping movements. Others arrived that, against the odds, politics and human intransigence, succeeded changing the world forever.

The human plot reverberates. Lasting change does occur, often along the lines that people seek—but it is rarely enduring, seldom linear—with ideas often tugging along the trajectory of a sine wave in which the amplitude, the frequency and the phase are forever opaque beyond the current moment, and because the slope changes over time, historical data is not useful in informing the future.

Scenarios offer a technique to explore the possible way uncertainties will play out. Scenarios will define all of the possible paths, but they strive to create highly divergent futures that force people to question their assumptions and imagine new possibilities. If they succeed in this goal, they can act as a counterpoint to propaganda and other forms of storytelling that seek not to expand the possible, but to constrain it, to paint change in the shadow of fear or to offer deliverance without acknowledging the path forward.

Scenarios start with the humble admission of not knowing. By documenting uncertainties in scenario planning, we refocus on exploring what might be to unleash the possible, and hopefully, navigate away from the dreadful.

Did you enjoy The Role of Uncertainties in Scenario Planning? Leave a comment.

For more serious insights on strategy, click here.