Scenario Analysis and the Limits of Prediction

Abstract

The limits of prediction pose a critical risk for any organisation making long-range decisions about its strategy and investments. The accuracy of our forecasts degrades progressively due to the combined effects of complexity and uncertainty, compounded by the passage of time. Scenarios become useful in the zone beyond the limits of standard prediction. However, we question whether it is really the case that scenarios must only start where prediction ends. Do most organisations that invest in scenario planning have no expectation that the scenarios created will provide a credible picture of what the future might be like? Instead, we view scenarios as anticipatory systems with a distinctive type of predictive model, where prediction is not based on probability, but plausibility. The zone of plausibility is a space in which scenarios provide evidence-based, broad-brush, but targeted, prediction. We contend that other objectives for scenarios, (such as futures consciousness, or re-perception) are necessary but secondary to scenarios' predictive work in the zone of plausibility. Importantly, we argue that if we abandon hope that our scenarios will provide a constrained form of prediction, then we fail to harness the potential that scenarios can offer-when they are plausible, rigorous, and robust. KeywordsLimits of prediction • Complexity, uncertainty and prediction • Scenarios and prediction • Properties of scenarios • Plausible scenarios • Turbulence, Uncertainty, Novelty and Ambiguity (TUNA conditions) OverviewAll who claim to foretell or forecast the future are inevitably liars, for the future is not written anywhere-it has still to be built (Godet and Roubelat, 1996, p. 164).