The default planning mode is a single best-estimate forecast: here is what we expect revenue to be, here is the headcount we’ll hire, here is the roadmap we’ll deliver. One plan, one future, all energy focused on executing against it.
The problem is that the future rarely looks like the plan.
The Illusion of Precision
Single-point forecasts create an illusion of precision. A budget that says “we will spend $4.2M in H1” sounds precise. It implies a level of confidence in the underlying assumptions — revenue trajectory, hiring timeline, project scope — that almost never exists in practice.
This illusion is dangerous because it discourages adaptive planning. If the plan is specific, deviating from it feels like failure. So organisations stick to the plan past the point where the plan is clearly wrong — because changing the plan requires admitting the original was flawed.
The Cost of Single-Scenario Betting
The cost of planning for only one future shows up differently depending on which direction reality diverges. If things go better than expected, the organisation is slow to accelerate — there’s no pre-built plan for upside. If things go worse, there’s no pre-built response — leadership has to improvise under pressure, in a compressed timeframe, with deteriorating options.
Both are expensive. The upside miss is often invisible — you never know how much faster you could have grown. The downside scramble is painful and visible — decisions made under pressure that a little prior thinking could have made better.
Building Adaptive Capacity
The alternative to single-scenario betting is not analysis paralysis. You don’t need to model every possible future. You need to model the futures that matter — the ones where you’d respond differently.
Start by identifying the two or three most uncertain variables in your operating environment. For each pair of extremes, ask: would we do anything differently? If the answer is yes, you need a scenario for it. If no, you don’t.
Most businesses need three to four scenarios, not fifty. The value is not in the modelling — it’s in having made the decisions in advance.
