When There’s No Playbook: Navigating Decisions in Unpredictable Situations

Some decisions cannot be made by reference to what worked before, because nothing comparable has happened. A regulation changes in a way no one anticipated, a market behaves against every historical pattern, a counterparty does something that has no precedent in the relationship. Experience, which is normally an organisation’s most valuable asset, becomes unreliable at exactly the moment it is most needed.

The instinctive responses are both wrong. One is to reach for the closest familiar situation and apply its logic, which imports assumptions that may not hold. The other is to wait for the picture to clarify — which works when information is merely delayed, and fails when the information does not yet exist for anyone.

What replaces precedent is structure. This article sets out five substitutes for a playbook, the three ways executives most often go wrong when a situation is genuinely new, and how to judge the speed a decision actually requires.

Key Takeaways

• Reason from the mechanism, not the case. Ask why something would work here, not whether it worked somewhere else.
• Name the analogy you are using. Unstated comparisons are the ones whose differences never get examined.
• Buy reversibility instead of certainty. When the chance of being wrong is elevated, the ability to withdraw is worth more than confidence.
• Set tripwires before committing, while judgement is still uncontaminated by sunk cost.
• Waiting is only useful if it produces information. In genuinely novel situations it frequently does not.

What This Article Covers

• How to tell if a situation is genuinely without precedent
• Five substitutes for precedent
• Three ways executives get this wrong
• How to decide how fast to decide
• Why outside perspective matters more, not less
• Frequently asked questions

How to Tell If a Situation Is Genuinely Without Precedent

Most situations that feel unprecedented are not. They are familiar problems arriving in unfamiliar packaging, and existing judgement applies. The distinction matters, because treating a routine problem as novel wastes time, and treating a novel problem as routine produces confident errors.

A situation is genuinely new when the mechanism is different, not merely the circumstances. A currency movement larger than any in your operating history is unusual but mechanically familiar — you know what currency risk does. A regulator applying an existing law in a way that inverts the logic of your structure is mechanically new, because the thing that previously made the structure work no longer applies.

The practical test: can you name a past situation where the same underlying force produced the outcome you are now trying to predict? If yes, you have precedent and should use it. If the honest answer is that you have a superficial resemblance and no mechanical match, you are in different territory.

Five Substitutes for Precedent

When there is no comparable case, five methods substitute for it. None requires historical data. All of them work by imposing structure on judgement rather than by supplying an answer.

1. Reason from the mechanism

Instead of asking what happened when others faced this, ask what would have to be physically, commercially, or legally true for each outcome to occur. This shifts the question from pattern-matching to causation, and it produces claims that can be checked. It is slower and considerably harder to fool yourself with.

2. Name the analogy, then list the differences

You will use an analogy whether or not you intend to. The discipline is to say it out loud — “we are treating this like the 2019 supplier failure” — and then enumerate every material difference. Half the time the analogy survives with adjustments. The other half, writing the differences down is what stops a confident mistake.

3. Invert the question: what would have to be true?

Rather than arguing whether a course of action is right, ask what conditions would make it right, then assess each condition separately. This converts an unresolvable debate about a conclusion into a set of tractable questions about premises, several of which can usually be tested cheaply.

4. Buy reversibility rather than certainty

Certainty is not available in a novel situation, so stop paying for it. Structure the commitment so that being wrong is survivable: a pilot rather than a rollout, a shorter initial term, a staged investment, an option rather than an obligation. Accepting a lower expected return in exchange for the ability to exit is usually rational when the variance is genuinely unknown.

5. Set tripwires before you commit

Define, in advance and in writing, the specific observable conditions that would trigger a reversal or a review — a customer volume below a stated level by a stated date, a regulatory determination going a particular way. Decided beforehand, these are objective. Decided afterwards, they move to accommodate what has already been spent.

When you cannot know whether a decision is right, make one you can afford to be wrong about — and agree in advance how you will know.

Three Ways Executives Get This Wrong

Three failure modes account for most poor decisions in novel situations. Each is a reasonable instinct applied where it does not belong, which is what makes them difficult to spot from inside the discussion.

Forcing a familiar frame

The most experienced person in the room recognises the situation as something they have seen before, and their confidence closes the discussion. Experience is usually the right guide, which is precisely why the exception is dangerous: the pattern-match happens automatically and is rarely stated as a claim that could be challenged.

Freezing while waiting for clarity

Deferring a decision feels prudent and costs nothing visible. But in a genuinely new situation, clarity may be months away or may never arrive, and in the meantime the choice is being made by default. Delay is a decision to accept whatever position you drift into, and it should be evaluated as one.

Deferring to the most confident voice

In the absence of evidence, groups gravitate toward whoever sounds most certain. Confidence in a novel situation is not a signal of accuracy, since nobody has grounds for it. This is a governance problem as much as an analytical one, and it is examined further in The Decisions Nobody Questioned: Warning Signs of Weak Internal Governance.

How to Decide How Fast to Decide

Speed should be set by two things: how reversible the decision is, and whether waiting will actually produce information. Most organisations set it by how uncomfortable the uncertainty feels, which is unrelated to either.

Where a decision is reversible and waiting produces nothing, move now and learn from the result — the information you want will come from acting, not from analysis. Where it is reversible but waiting produces something, take the smallest committing step that keeps you in the game while the picture develops.

Where a decision is irreversible and waiting produces information, wait, and be explicit about what you are waiting for and by when. Where it is irreversible and waiting produces nothing — the hardest quadrant — the objective shifts to converting it into something reversible: renegotiate for an exit right, stage the commitment, or reduce the scope until the downside is survivable.

Where the underlying difficulty is that the available information is contradictory rather than absent, the approach differs again — see When Your Advisors Disagree: Resolving Conflicting Information Before a Decision. Where the information is simply incomplete, see Why Strategy Fails: When Good Decisions Are Built on Incomplete Information.

Why Outside Perspective Matters More, Not Less

The usual argument for internal decision-making is that nobody understands the business better. In a situation without precedent, that advantage weakens: the organisation’s accumulated judgement was built on conditions that no longer apply, and deep familiarity makes it harder to notice that the mechanism has changed.

Independent input contributes three things here. It supplies analogies from outside your sector, where a mechanically similar situation may well have occurred even though nothing comparable exists in your industry. It provides someone whose standing does not depend on the decision, which is rarely true of anyone in the room. And it forces reasoning to be made explicit, because an outsider cannot follow assumptions that were never articulated.

This is the purpose of a standing advisory relationship rather than a project engagement: novel situations do not arrive on a schedule, and the value of an outside view depends on it being available when the situation appears rather than three weeks after. Our case study Ongoing Strategic Decision Support illustrates how this works in practice.