Decision Quality and Outcome Favorability: Why a Good Result Does Not Confirm a Sound Decision
A favorable outcome and a defensible decision are different claims, resting on different evidence, and an organization that only examines the first loses the ability to govern the second.
The Question That Gets Asked
When a consequential decision is reviewed after the fact — by a board, an audit committee, an executive retrospective — the question that tends to get asked first is simple: did it work? A result is examined. If the result is favorable, the decision is treated as having been sound. If the result is unfavorable, the decision is treated as having been flawed.
This is a natural instinct, and in many contexts a harmless one. But as a governance practice, it collapses two claims that are not the same claim, and the collapse has consequences for how an organization learns from its own decisions.
Two Different Properties
Decision defensibility is a property of the decision basis at the moment of commitment: what alternatives were visible and considered, what evidence was available relative to what the decision required, whether uncertainty was acknowledged rather than suppressed, whether contrary information was surfaced rather than filtered out, whether the scale of commitment was proportionate to what could reasonably be known at the time.
Outcome favorability is a property of what happened afterward — shaped by the decision, but also by factors the decision-maker did not control and in many cases could not have known: competitor behavior, market timing, regulatory shifts, chance events that had nothing to do with the reasoning behind the decision.
These two properties can align. They frequently do not. A well-reasoned decision, built on the best evidence reasonably available, can still produce a poor outcome because conditions moved in a direction that could not have been anticipated. A poorly reasoned decision — one that ignored available contrary evidence or failed to consider a materially relevant alternative — can still produce a good outcome, because the world happened to cooperate.
Why the Conflation Matters
If governance review only examines the outcome, it produces two systematic distortions over time.
The first: decisions that were genuinely defensible get treated as failures when circumstances turn unfavorable, teaching the organization that risk itself — not poor reasoning — is what gets punished. Decision-makers respond rationally to this signal by avoiding any commitment whose outcome cannot be guaranteed, even when the underlying reasoning was sound. The organization becomes more conservative without becoming more disciplined.
The second, less visible distortion: decisions that were poorly reasoned get treated as successes when circumstances happen to turn favorable. The weak process that produced them is never examined, because the scorecard shows a win. The organization reinforces a decision-making habit that will eventually fail, simply because it has not yet been tested by unfavorable conditions.
In both cases, the governance body loses the ability to distinguish a decision problem from a world problem. It is examining the wrong object.
What the Distinction Requires
Recognizing this does not mean outcomes are irrelevant to governance — a pattern of poor outcomes is legitimate evidence that something in the decision process deserves scrutiny. But the outcome by itself does not tell a governance body which situation it is looking at: a sound decision undone by an unforeseeable event, or a weak decision that has not yet met an unfavorable one.
Establishing which is the case requires a separate inquiry, conducted independently of how things turned out: what was known, what was available, what was considered, and whether the commitment was proportionate to that basis at the time it was made. This is not a question about whether a record exists or can be reconstructed — it is a question about the standard applied to whatever basis is available, and whether that standard is being applied at all, or substituted for by the simpler, more available signal of how things turned out.
The Limit
This distinction does not excuse poor decisions simply because "the process looked fine" — some failures genuinely do expose that known risks were never surfaced, or that available contrary evidence was disregarded. The point is narrower: the outcome alone cannot tell you which case you are in. Determining that requires looking at what was actually visible and considered at the time, independent of what happened next.
Institutional Implication
A governance body that wants to actually improve decision quality over time needs a practice for assessing defensibility that does not depend on outcome — applied consistently, including to decisions that turned out well. Otherwise the organization is not learning from its decisions. It is only learning from its luck.