Decision Quality vs Outcome Quality: Stop Judging Yourself by Results
A good decision can have a bad outcome, and a bad decision can have a good outcome. The teams that improve fastest separate the two and rate themselves on the…
On this page▾
- Decision quality and outcome quality are different. Most reviews conflate them.
- Judge decisions on the information and process available at the time.
- Outcome reviews answer 'what happened'; decision reviews answer 'should we do it the same way again'.
- Most learning lives in the gap between a good decision and a bad outcome.
Poker professionals figured this out a generation ago and Wall Street learned it the hard way. The lesson has barely reached corporate leadership: results are noisy, and judging decisions by results turns smart teams into superstitious ones. The fix is small and structural — separate the review of the decision from the review of the outcome.
The four-cell frame
- Good decision → learned, repeat
- Bad decision → lucky, do not normalise
- Good decision → learn, repeat anyway
- Bad decision → root-cause, change process
Three of the four cells are mis-handled by default. Lucky wins get celebrated and normalised. Unlucky losses get punished. Real learning lives in the 'good decision, bad outcome' cell, which is exactly the cell most companies bury because nobody wants to defend a bad result.
How to run a decision review
- 1What did we know at the time?Reconstruct the information set. Resist filling in what you learned afterwards.
- 2What process did we use to decide?Who was consulted, what alternatives considered, what risks named.
- 3Given what we knew, was the decision reasonable?If a thoughtful peer would have made the same call with the same information, the decision was sound.
- 4What would we change about the process next time?Different inputs, different stakeholders, different timing, different rubric. Not the outcome — the process.
Applied to people decisions
A senior hire that flames out is the canonical case. The temptation is to blame the recruiter, the interview loop, or the hiring manager. The decision-quality question is different: with the references, the work sample, the panel feedback and the comp competition at the time, was the decision reasonable? Sometimes the answer is yes — and the right response is to change the process (deeper references, longer trial work) rather than punish the people who decided.
Teams that run decision reviews — explicitly separating process from result — make significantly bolder decisions over time. They have learned that a bad outcome will be examined fairly.
The hindsight trap
Hindsight bias makes past decisions look obvious. Resist it. The discipline is to assess decisions as they were made, with the information they had — not as they look now with the knowledge of what happened. The simplest counter-measure: write the decision down in advance, including what you expected to happen. That document is the only honest reference point later.
Read next
All playbooksJeff Bezos's most-cited framework: irreversible decisions deserve deliberation, reversible ones don't. Most organizations confuse the two — and pay for it in…
Post-mortems happen after the failure. Pre-mortems happen before — and Gary Klein's research shows they boost problem identification by 30%.
Inversion. Second-order thinking. Opportunity cost. The handful of mental tools that turn ordinary judgment into strategic thinking — without requiring an MBA.