Decision hygiene: building a review your team will actually use

The decision that worked for the wrong reasons

A marketing director in a retail group made a bold call to pull the entire seasonal budget behind a single campaign. It worked. Sales beat target, the board was delighted, and the decision entered company folklore as a masterstroke. What nobody examined was that a competitor had stumbled into a supply crisis the same fortnight, clearing the shelf space the campaign happened to fill. The director had been bold, and the director had been lucky. The team learnt the wrong lesson: bet big, trust your gut. The next year they bet big again, the luck did not repeat, and the campaign failed in ways the first success had hidden in plain sight.

This is the quiet problem with how most teams learn from their decisions. They do not, really. They learn from outcomes, which is a different and far less reliable thing.

Resulting: judging the decision by the result

There is a habit of mind so common it has a name in poker, where the gap between decision and outcome is brutally visible. It is called resulting: judging the quality of a decision purely by how it turned out. A good outcome means a good decision. A bad outcome means a bad decision. It feels like common sense. It is, in fact, one of the most expensive errors a leadership team can make, because it severs the link between what a team can control and what it cannot.

A decision is a bet placed under uncertainty with the information available at the time. The outcome is the bet plus everything outside your control: timing, a competitor's move, a currency swing, the weather, sheer chance. Strong decisions sometimes produce poor outcomes. Weak decisions sometimes produce good ones. Over a single instance, the outcome tells you almost nothing about the quality of the thinking that produced it.

When a team judges by outcome alone, two things happen, and both corrode capability. Decisions that worked are immune from scrutiny, so their flaws survive and get repeated. Decisions that failed are condemned wholesale, so the sound judgement inside them gets thrown out with the bad luck. The team optimises for outcomes it cannot control and stops examining the process it can. This is not a knowledge gap. It is a hygiene gap, and it compounds.

The deeper cost is cultural. In a team that judges by outcome, the safest move is to avoid being associated with a bad result, which is not at all the same as making good decisions. People hedge, defer, and seek cover. The bold, well-reasoned bet that happened to miss becomes a career liability, while the reckless punt that happened to land becomes a reputation. Over time the team learns to manage its luck exposure rather than improve its judgement.

What a decision review is actually for

The corrective is structured review of the decision process, separate from the outcome. This is not a post-mortem in the usual sense, which tends to fire only after something has gone visibly wrong and quickly becomes a search for whom to blame. A decision review examines the quality of the decision regardless of how it turned out, including the ones that worked.

The shift is from "did it work?" to "given what we knew and could have known, was that a good bet, and would we make it again?" Those are different questions, and only the second one builds judgement that holds.

This connects directly to how we think about building capability that endures. Judgement is not transferred by being told the right answer. It grows through owned, consequential practice, examined honestly. We make the same argument about delegating reversible decisions to the people who will grow by making them: capability comes from repetition with real stakes. A decision review is the examination step that turns those repetitions into actual learning rather than just accumulated outcomes. Without it, a team can make a thousand decisions a year and learn almost nothing, because experience without reflection is mostly just elapsed time.

Separating the bet from the bounce

The single most useful move in a decision review is to score the two things separately. For any decision under examination, ask two distinct questions.

First, process quality: given the information available and reasonably obtainable at the time, how sound was the reasoning? Did we frame the decision well, consider the real alternatives, weigh the cost of being wrong, and decide at the right speed for the stakes?

Second, outcome: how did it actually turn out?

Plot any decision against those two axes and you get four cells, and three of them are where the learning lives. A good process with a good outcome is the easy case, though it is still worth confirming the process held rather than the luck did. A good process with a bad outcome is the cell teams handle worst: this is a well-made bet that lost, and the instinct to punish it teaches everyone to stop making good bets. A poor process with a good outcome is the most dangerous of all, because the result hides the flaw and the team is about to repeat it with the confidence of the recently rewarded. A poor process with a bad outcome is the only cell where outcome-based judgement happens to land on the right answer, and even then for the wrong reason.

The retail marketing director's campaign sat squarely in the dangerous cell: poor process, good outcome. A review that scored the bet separately from the bounce would have caught it. Outcome-based judgement guaranteed it would not.

What it looks like in practice

A managing director in a logistics business instituted what she called a monthly "decisions hour". Not a project review, not a performance meeting, a standing sixty minutes where the leadership team picked two or three decisions from the prior quarter and examined the process, not the result. The rule she set was deliberate and counter-intuitive: at least one of the decisions examined each month had to be one that had worked. Her reasoning was sharp. If you only review the failures, the team learns that review is punishment, and they brace for it. If you also review the wins, the team learns that review is hygiene, and they relax into it.

The first session that examined a success was the one that changed the culture. The team pulled apart a logistics-routing decision everyone had filed as a triumph and discovered the outcome had ridden on a temporary fuel-price dip nobody had factored in. The decision had been fine; the celebration had been premature. Nobody was punished, because the outcome was good. But the team learnt not to bank a process they had not actually tested. Within two quarters, "would we make that bet again?" had become a phrase people used unprompted, which is the real marker that a practice has taken.

A head of product in a software company ran the opposite experiment first, and it failed instructively. He introduced decision reviews, but he ran them only after things broke, and he ran them looking for the cause, which in practice meant the person. Attendance held because it was mandatory, but the content hollowed out. People arrived with defences pre-built. The reviews produced tidy narratives in which the decision had been reasonable and the problem had been someone else's execution. He was getting performances, not examinations, and he was getting them because he had built a forum where the cost of honesty was higher than the cost of spin.

What he changed was the frame and the safety underneath it. He moved the review off the back of failures and onto a regular cadence that caught wins too. He separated, out loud, the question of decision quality from the question of outcome, so that owning a process flaw on a decision that had worked carried no penalty. And he made his own decisions the first ones reviewed, including one of his that had a good outcome and, on inspection, a thin process. A leader who reviews their own bets honestly, in front of the team, does more for psychological safety in ten minutes than a values poster does in a year. The reviews started producing real findings within a month, because the people in the room had finally seen that honesty was survivable.

The safety problem underneath

Both of those stories turn on the same hinge, and it is worth naming directly: a decision review only works if it is safe to be honest in it. The mechanics are simple. Two questions, two axes, a regular slot in the calendar. The hard part is never the mechanics. The hard part is that examining a decision means examining the judgement of the person who made it, and in most teams that feels like exposure.

This is the same territory we examine in decision-making under pressure: the constraint is rarely the model, it is the operator. A leader who needs to be seen as right will quietly resist any review that might surface a miss, and the team reads that resistance instantly and adjusts. A team that has watched a colleague punished for a well-reasoned bet that lost will never again bring a real decision to the table, only a defensible one.

The fix is not a technique. It is a stance the leader has to hold and model. It draws on the same inside-out capability that runs through the DUAL model, specifically the willingness to accept an uncomfortable reality about yourself without deflecting or performing. A leader who can say "that was my call, the outcome was fine, and the reasoning was thinner than it should have been" gives the whole team permission to do the same. A leader who cannot say it has, in effect, banned learning, however many reviews they schedule.

The practical signals that the safety is real are observable. People volunteer their own misses before being asked. Decisions that worked get examined as readily as decisions that failed. The phrase "good decision, bad outcome" gets used and meant. When those appear, the review is doing its work. When the room goes quiet and the narratives get tidy, it is not, and no amount of process will rescue it until the safety is fixed first.

The reader's next step

If you want to test whether your team is learning from its decisions or just accumulating outcomes, the diagnosis is uncomfortable and quick.

Three questions to sit with

• When did we last examine a decision that worked? If the honest answer is "we don't", your team is reviewing only failures, which means review reads as punishment and your wins are carrying unexamined flaws forward.
• What happened to the last good bet that lost? Trace one well-reasoned decision in the past year that turned out badly. Was the reasoning credited and the luck named, or was the person quietly marked down? The answer tells you what your team has learnt about the cost of honesty.
• Could someone safely tell me my reasoning was thin on a decision that worked out? If you cannot picture it happening, the constraint is not your team's candour. It is the safety you are or are not modelling, and that is where the work starts.

The pattern these questions reveal is diagnostic. A team that cannot examine its wins, or punishes its honest losses, does not have a process problem. It has a hygiene problem and a safety problem, and both are developable. Our use cases show what this kind of capability work looks like in applied settings.

A discipline, not another framework

The leaders we work with rarely need a more sophisticated way to make decisions. What they tend to lack is the discipline to examine the ones they have already made, honestly, separating the quality of the bet from the bounce of the result, on a cadence regular enough that it becomes hygiene rather than an inquest.

It is a small practice with a large return. Score the process and the outcome separately. Review the wins, not just the losses. Make it safe by going first. Do that consistently and a team stops mistaking luck for skill, stops punishing good bets that lost, and starts compounding judgement instead of just outcomes. If you want to think through how to build this across a leadership team rather than in one person, that is the conversation our services are built for, and you can start it with us.

The leaders described here are representative composites drawn from patterns we observe in practice, not identifiable individuals.