What Makes a Decision Right

The right one does not exist yet

The question "how do you make the right decision" carries an assumption inside its own phrasing: that the right one already exists somewhere, waiting to be found by the right method. The assumption is comforting, because it hands responsibility to the method. But the rightness of a decision whose outcome hasn't happened yet isn't something you can go looking for, because rightness is measured against an outcome, and the outcome doesn't exist yet.

So the executive goes looking for the right method instead: one more data point, one more opinion, one more meeting. The method isn't wrong, but it answers the wrong question. The question worth asking isn't "which one is right." It's "what did I know, and what did I give up, in making this one."

Looking at the moment, not the method

Two executives can make the same decision, with the same data, in entirely different ways — and both can turn out to have been right. The difference between them isn't rightness. It's what the decision was built on: which information it used and which loss it accepted, and that is the one thing knowable before the outcome ever arrives. A right decision is not one whose outcome turned out well; it is one where you can say, afterward, exactly what you knew and what you gave up.

The distinction looks small but changes everything in practice. Take a pricing decision: one executive changes the price using three months of data and a loss accepted by name — "we will lose volume in this segment, and we accept that." Another changes the same price with no data at all, purely in reaction to a competitor's move. Both can land on the same outcome. But one leaves behind a method usable for the next pricing decision; the other leaves behind only a reaction.

The split is in the choosing, not the outcome

Labeling a decision right or wrong after the fact is easy, because the outcome is sitting right there. But that label is useless at the moment the decision gets made — at that moment there is no outcome yet, only options and their known costs.

Two decisions, one outcome

The same outcome can come from two different decisions. One is made by looking at the whole of the available information and naming the loss being accepted. The other arrives by luck, or by avoiding the decision long enough that time makes it instead. The outcomes may match, but the two are not the same thing — one is repeatable, the other is not.

What determines a decision's quality is not the outcome; it's how the decision was made — which information was knowingly set aside, which loss was knowingly accepted. If those two questions have answers, the decision was built well regardless of what happens next. If they don't, even a good outcome teaches the next decision nothing.

Naming the loss is the same move that unfreezes the leader who cannot decide; the difference here is a leader who isn't frozen but is still deciding by the wrong measure.

The same split shows up in hiring decisions. One executive hires after naming the candidate's weak point in writing — "this person decides slowly under pressure, and we accept that." Another hires the same candidate purely off a good feeling from the interview. Both can turn out well. But the first leaves behind a criterion the next hire can reuse; the second leaves behind only luck.

Missing isn't information, it's a boundary

Executives looking for the right decision usually reach for more information: one more report, one more opinion, one more scenario. But what's usually missing isn't information. It's a boundary on when enough information counts as enough.

An unbounded search never closes

• How much information is enough is never stated in advance.
• Which source gets the final word is never decided.
• There's no threshold for whether new information should change the decision at all.

A search run without those three boundaries never really ends in theory, because every new piece of information justifies the next one. What makes the search for the right decision unworkable isn't a shortage of information; it's that no boundary was ever set. The moment a boundary exists, the decision becomes makeable with whatever information is already on hand. Without one, the decision keeps sliding forward no matter how much information accumulates.

In practice the boundlessness looks like this: in a decision meeting, the phrase "let's get one more report" gets said for the third time, and each time it sounds reasonable, because none of them is excessive on its own. A team that never set the boundary in advance can never know which report is the last one; only time, or outside pressure, ends up deciding that for them.

Rightness is a label applied backwards

The phrase "the right decision" runs backwards by its own nature. What makes a decision right is the conditions it was made under, but the person who sees those conditions most clearly is the one judging the decision afterward, because that person now knows the outcome too. That doesn't make them correct. It just gives them an advantage the decision-maker never had.

Being right in the present tense

The only measure of being right in the present tense is this: was all the available information used, and is the accepted loss known by name. If both hold, the decision is the rightest one available at that moment, whatever the outcome turns out to be. The three-step method described in when a decision will not close is really a mechanical way of making both of those visible: writing the decision to one sentence surfaces the information; setting the threshold surfaces the loss.

What makes a decision right is not that it gets confirmed later; it's that it can be defended in the present. A defensible decision gets made the same way again even when the outcome turns out badly, because the method never depended on the outcome — only on the best use of what was known at the time.

This backward advantage shows up especially in board reviews. An executive defending a decision gets judged not against the information available at the time, but against what everyone now knows. The comparison isn't fair, but it happens everywhere anyway, because hindsight always sees more clearly.

From the field

A scene repeats across these notes: an executive describing a past decision says, "if only I'd known then." Most of the time it was knowable then too — it just hadn't been gathered yet, or it had been gathered and never said out loud.

Regret gets mistaken for information

Regret gets mixed up with information here. The executive feels as though they've known the thing they learned later all along, and rewrites the decision accordingly. The rewrite is comforting but not instructive, because it produces no rule for the next decision — only a regret.

This is a different question from the distinction in every hard decision is hard in its own way. That piece asks what kind of hard a decision is. This one asks what a closed decision should be judged against. The two work together: one names the kind of decision, the other says how it should be measured.

The observation from the field is this: there is no such category as "the right decision." There is only a decision made in a way that can be named. What separates it is whether you can say, at the moment it was made, which information was used and which loss was called by its name. If both can be said, the rest of the decision is arithmetic on the outcome — not a judgment on the decision itself.