What does the future hold?

This is an important question for governments, businesses, organisations and individuals. Because when we make decisions today, we are often making bets on an unknown future e.g. about retirement or energy needs and provision.

Some are sceptical, rightly so, about the limitations of prediction in a complex and uncertain world. [See new blog posts from Ralph Stacey and Chris Rogers].

Yet we often need to act amid uncertainty: be that in planning large infrastructure projects; developing health and social care policies; or making choices about our own retirement provision.

So, how can we address the question: what does the future hold?

 

Mystics, statistics and learning

There are statistical models and mystical models. Although polar opposites in their reliance on data, both statistical and mystical models assume that they can make predictions about the future with some significant degree of certainty. Learning models, on the other hand, offer ways of working with uncertainty.

Statistical models assume the world is predictable, the future knowable, that ‘tomorrow’ will be largely like ‘today’ in all important aspects. They assume regularity and ignore uncertainty. They downplay the likelihood of unknowns and extreme events, although extreme events are more common in nature than you might expect (see long tails and power laws). They also downplay free-will, assuming that people will largely respond in ways that are similar to how they’ve responded in the past. And they assume that small differences in those responses won’t add up to anything much, even though there’s increasing understanding of butterfly effects (remember Gerald Ratner’s ill-advised comment?), ‘viral’ change and social movements. If data is king, then Big Data is emperor. Why do we like statistical models in business? Because they give the illusion of certainty and can reduce the anxiety of leadership and decision-making in an inherently uncertain world.

Mystical models assume the world is preordained, that the future can be intuitively seen now, by a select few with special gifts or skills e.g. successful business leaders, tech entrepreneurs, or fortune tellers. In assuming an unfolding future, mystical models downplay human agency and free-will; the ability of human beings to act and react in ways that are imaginative, rebellious, constructive and destructive. Hindsight about successful predictions or interventions becomes the measure of those who have the gift. Data is used selectively, in hindsight, if at all. I’m sceptical.

Learning models. So, how can we consider what the future holds in an uncertain world? If we cannot know the future in advance, by using mystics or statistics, then the alternative is to learn more quickly about the future that is emerging in the present. Learning models consider patterns and how they are changing. They help us choose our responses, rather than responding more automatically, and to discover how the interplay of all our responses is playing out.

The Dynamic Patterning Framework is a learning model which uses a systematic, data-driven and people-focused process to help formal leaders and decision-makers discover how their organisation[1] is changing. It assumes that the seeds of the future are in the present, but accepts that no-one can predict how they will play out as people interact in the course of their day-to-day work. Instead it actively and systemically explores multiple, diverse perspectives about what is changing. It brings together small, human-scale data about what’s going on in an organisation – e.g. how people are working together, what they’re talking about and how they feel – to help leaders discover weak signals of potential opportunities or problems, and to explore what is emerging. As well as being data-driven, this view takes human agency and free-will seriously. It understands that we are simultaneously co-creating the organisation that we are responding to; that what we say and do matters. Yet particular things that we say and do may sometimes matter rather more, or rather less, than we might expect.

STATISTICAL MODELSPredictable Based on yesterday Perfect (past) data Single expert perspective Answers Certainty

LEARNING MODELSPatterned Explores today Small data Multiple perspectives Questions Uncertainty

MYSTICAL MODELSPreordained Sees into tomorrow No clear link to data Single expert perspective Answers Certainty

In an uncertain world, learning more quickly about the future that is emerging is the only way to discover what the future holds.