Adaptive strategy.

How do you plan when you cannot predict the environment?

Conventional strategic planning assumes a forecast. You analyse the market, project where it goes, commit resources to a position, and execute. It works when the environment is stable enough that the projection survives contact with the year, and it has produced most of the strategy literature businesses still use.

Adaptive strategy starts from a different assumption: that the projection will be wrong in ways that cannot be known in advance, and that the useful work is therefore not producing a better forecast but building a position that stays viable across several of them.

The distinction turns on the kind of uncertainty involved. Some uncertainty is measurable: the range of outcomes is known and can be planned against with a contingency. The uncertainty that defeats strategic planning is the other kind, where the thing that changes is not a variable in the model but the model itself. No amount of scenario work anticipates a change in what customers consider the category to be, because scenario work varies the inputs and that change replaces them.

It is not the absence of a plan

This is the misreading that ruins it. "We are adaptive" is used, often sincerely, to describe having no strategy at all: reacting to whatever arrives, calling the reaction responsiveness, and never committing to anything long enough to find out if it works.

That is the opposite. Adaptive strategy requires more up front decision making, not less, because you have to decide in advance which commitments are permanent and which are provisional. Making that call under pressure, in the moment, is how businesses abandon the thing that was actually working.

The two layers

Every adaptive approach ends up with some version of the same structure: a small set of things that do not move, and a larger set that is expected to.

The stable core should be short enough to recite. Who this business is for. What it refuses to do. What standard it holds. What it wants to be known for. These are decisions that only pay off through repetition, and changing them resets the clock on every impression the business has made.

The adaptive layer is everything downstream: the language, the channels, the offers, the emphasis, the proof, the sequence. None of these deserve loyalty. Each is a current best answer to a question that will be asked differently next year.

Most strategic damage comes from getting these the wrong way round: businesses defend a tactic because it is familiar, and revisit their positioning because a quarter was bad. The familiar tactic is usually the thing that should go, and the positioning is usually the thing that has not been given long enough to work.

Deciding in advance what would change your mind

The most useful practice in adaptive strategy is also the least practised: writing down, at the point of commitment, what evidence would tell you the commitment was wrong.

It costs almost nothing and it changes the character of every later argument. Without it, evidence gets interpreted by whoever has the most invested in the current answer, and a plan is defended until it is embarrassing rather than until it is disproven. With it, the conversation is not about opinion, it is about whether a condition stated in advance has been met.

It also protects against the opposite error. A pre-agreed threshold gives a team a reason to hold a position through the noisy middle period, when a new approach has not yet worked and abandoning it feels prudent.

Where the advantage actually comes from

In a slow market, competitive advantage comes from having the better position. In a fast one it comes increasingly from the speed of the loop: how quickly a business notices something has changed, decides what it means, acts, and confirms the action worked. Two businesses with identical positions perform differently if one takes six weeks to notice what the other notices in three days.

This is a genuine shift in where durable advantage sits, and it is worth being careful about, because it is easy to overstate. Structural advantages, a cost base competitors cannot match, a distribution position they cannot buy, patents, switching costs, still beat responsiveness and always will. The claim is narrower: where no structural advantage exists, and for most businesses none does, the residual advantage available is the quality and speed of the loop, and that one can be built deliberately.

That is a capability, not a plan, and it has to be built and maintained like any other. It is what Digilu operates on behalf of clients, and it is the operating half of the category we work in: strategy expressed as an ongoing responsibility rather than as a document.

The costs, stated plainly

• It is harder to hold people to. A fixed plan makes accountability simple: did you do the thing. Adaptive work has to be judged on the quality of the decisions, which requires someone able to judge decisions.
• It resists reporting. "We changed course twice and both were right" is a genuine result and it does not fit on a slide next to a target.
• It is easy to counterfeit. Because adaptation is legitimate, it is available as an excuse for every missed commitment, and the difference between adapting and giving up is visible only if the thresholds were written down beforehand.
• It needs measurement to be worth anything. Without a reliable read on the environment, adaptive strategy degrades into reacting to the loudest voice in the room, which is where it started.

When conventional strategy is the better choice

When the environment genuinely is predictable, forecasting beats adapting, because commitment compounds and the overhead of continuous reassessment buys nothing. Long asset cycles, regulated markets and businesses whose advantage is structural rather than perceptual are all better served by picking a direction and holding it. The model here is for environments where the planning cycle is slower than the change.