We all use a variety of models to anticipate the future. From mental models to simple formal models to large, quantified systems, we rely on these to create forecasts about possible future outcomes. While it is common for us to talk about questioning our mental models or questioning the limits of large quantitative models, we don’t talk as much about shifting our models when working on the future.
Many of us have a tendency to apply a single model to a topic, let’s say, the 10-year future of an industry. Relying on a single model (mental or otherwise) to forecast what is possible, we run the real risk of creating an unrealistic set of forecasts. Why? It would assume that all the major variables in the model, and their relationships with one another, would remain constant and produce the same sort of patterns and outcomes over the entire 10 year period. Change to a key piece of regulation, the emergence of a new business model, or a transition in global supply chains each would require a new model of the industry in order for us to produce more realistic forecasts.
To take a more specific example, everyone is currently trying to anticipate the future of the US economy. COVID-19 rendered our pre-2020 models about the workforce, individual industries, and our particular organizations outdated. It’s not just that the pandemic introduced events to be accommodated – bumps in our timelines or time series – it’s that it required us to work out new models in order to more realistically forecast what might happen next with regard to the workforce, an industry, or our organization.
To return to our 10-year future example, the models you use to think through the first three years might be different than the ones you need to think through the subsequent seven years. Thinking critically about the future is not simply identifying events that are possible. It is also about shifting between the best models for forecasting the ever-changing emerging landscape.