Decision-making
Decision Review
A Decision Review examines one important decision. Facts, possible explanations and risks are brought together in a written Decision Brief. The client remains responsible for the decision.
There is no obligation to commission further work after the review.
Illustration · the decision fork
A decision with its reversal point written down
What might explain a difficult staffing decision
The operations director of a fictional equipment-servicing firm has a problem and a proposal. Repeat visits are rising: engineers return to the same customer within two weeks, and customers complain. Her proposal is simple: hire eight more engineers. The board meets in three weeks. Each of her managers offers a different cause with complete confidence.
That is the decision owner’s real problem. It is not a lack of ideas. It is that she cannot tell which idea is right, and a hire is expensive to reverse.
Rival hypotheses, and the cue that separates each
- H1 · Real capacity shortageEngineers are rushed, so jobs are left incomplete. Cue: overtime and backlog are high across all job types, and repeat visits are spread evenly.
- H2 · The wrong engineer for the jobDispatch assigns by availability, not skill. Cue: repeat visits cluster on a few complex job types, whoever attends.
- H3 · The measure rewards closing, not fixingA visit counts as done when the ticket closes, and the bonus follows. Cue: repeat visits concentrate among the fastest closers.
Her existing data can separate these in about two weeks of analysis and a handful of conversations. If H1 holds, hiring is right. If H2 or H3 holds, eight new engineers would inherit the same dispatch rule and the same bonus, and the repeat rate would barely move. The action follows from the cue: hire only if capacity is the cause; otherwise change dispatch or the closing measure first, which is cheap and reversible. The Brief also asks about people: engineers should not become the scapegoat for a measure that rewarded them, and the customers who complained are part of the picture. All figures here are invented to show the logic.
When is it useful?
Decision Review is narrow on purpose. It works well when the question is sharp enough to be wrong, and it works badly when the decision has already been taken and only needs a signature.
A good fit
- A decision is on the table now, with conflicting signals or divided views among the people who must carry it out.
- A friction keeps returning despite good intentions and earlier measures.
- A change works on paper and does not take hold in behaviour.
- A choice about people or organisation with much at stake and little reversibility.
- A sponsor who can name the decision and is willing to read an inconvenient conclusion.
A poor fit
- The decision is already made and you need support for it.
- You want a standard training programme or an off-the-shelf model.
- The question is a general strategy exercise with no decision attached.
- You want someone to take the decision off your hands.
- Access to the people affected is ruled out in advance.
Working approach
Five stages over three to four weeks
The duration is indicative and depends on how specific the case is and how quickly people are available.
- 1. IntakeOne conversation to sharpen the question: what must be decided, by whom, by when, and what would count as a good outcome. We also settle who is affected and what may be shared. You get: a one-page statement of the decision, agreed in writing.
- 2. Decision ecologyThree to five short interviews with the people who will live with the decision, plus documents and figures that already exist. We map interests, constraints, history, and who is not in the room. You get: a list of observations, each marked as observed or inferred.
- 3. Working sessionWith the decision owner, rival explanations side by side. We pick the few signals that discriminate between them, weigh downside, reversibility and timing, and set a decision threshold. You get: a shortlist of cues and a first view of the threshold.
- 4. Decision BriefA concise written document, usable at the leadership table, with an explicit prediction and the conditions under which you would change course. You get: the twelve-component Brief, with the strongest rival kept in.
- 5. Outcome reviewLater, prediction is compared with what happened, so the next decision starts smarter. Where it makes sense, a review at about 3 months and, for slower decisions, at 12 months, agreed in writing. We promise a look at the outcome, not a result. You get: a dated review plan.
What the work involves
Indicative scope
- ScopeWe examine one clearly defined decision. We agree the approach, conversations and timing in advance.
- Your timeThe decision owner: an intake, a working session and a review of the draft. A handful of colleagues: short conversations. Someone who can supply existing documents and figures.
- You receiveThe written Decision Brief, a short presentation of its logic if wanted, and a dated plan for the outcome review.
- You do not receiveA decision made for you, a training programme, a restructuring plan, a software tool, or a guarantee about the result. Large data analysis or more than five interviews are a different scope.
- BoundariesIf the question changes after our first conversation, we discuss it before changing the engagement. Nothing is expanded without agreement.
The organisation decides what to do with the Brief. We show the information supporting each recommendation and what remains uncertain. A later review of outcomes can be agreed separately; no outcome is guaranteed.
Ethics and confidentiality
Before starting, we agree what information is needed, who can see it and how it will be stored. Confidential details should not be sent by ordinary email. We also consider the consequences for people who are not in the room. See Ethics for more on the approach.