How to Review a Decision Without Judging Only the Outcome
To review a decision fairly, separate the information available when you chose from the outcome you learned later. Ask whether you checked the facts that mattered, considered realistic options, and acted on your stated reasons. Then examine the result and decide what, specifically, you would change next time. A favorable result does not automatically make the process sound, and an unfavorable result does not automatically make it careless.
What is outcome bias?
Outcome bias means evaluating a decision partly by how it turned out, even when the outcome does not tell you whether the original reasoning was good. In their 1988 study, Jonathan Baron and John Hershey examined how outcome information affected evaluations of decisions.
Hindsight bias is related but different: once you know the ending, it can seem more predictable than it was beforehand. Baruch Fischhoff's research examined this effect of outcome knowledge on judgments under uncertainty.
These distinctions matter because “It went badly” and “I ignored information I already had” call for different lessons. The first describes an outcome. The second identifies something you may be able to change.
Review the decision in three passes
- Then: What did I know, what was uncertain, and which options were realistically available?
- Action: What did I plan, what did I actually do, and what changed between those two moments?
- Now: What happened, what information is genuinely new, and what would improve my next decision?
Begin with your original note if you have one. Read it before composing the lesson. If you did not write anything down, distinguish what you remember from what you can verify. A dated email may confirm a deadline; your current impression cannot establish exactly how confident you felt.
You can still learn without an old entry. Just avoid presenting a reconstruction as a perfect record. For future choices, use this decision journal template.
A worked example: the offer you declined
- Then
- I wanted the role, but the stated travel requirement would conflict with commitments I already had.
- Plan
- Ask whether travel could be reduced. Decline if the requirement was fixed.
- Action
- I asked. The company confirmed the requirement. I declined.
- Later
- A friend joined the company and loved the work.
- First reaction
- Maybe I made a terrible mistake.
Your friend's experience is new information about their experience. It does not establish that the travel requirement would have suited you, or that the company would have offered you different terms.
A fair review could be: “I checked the condition that mattered and chose consistently with it. I still feel disappointed about what I gave up.” A choice can make sense and still contain a loss.
Now change one detail: suppose you declined because you assumed the travel requirement was fixed, without asking. You may still have made the same choice after asking, but the process has an identifiable gap. A useful lesson is “Check negotiable conditions before treating them as facts.”
Does a fair review mean ignoring bad results?
No. Results can expose a missed risk, an inaccurate estimate, or an ineffective action. Review them carefully. The question is whether you can connect the result to a specific feature of the decision rather than using the ending as a verdict on everything.
Ask: “Was this risk reasonably knowable? Did I dismiss evidence? Did my circumstances change? Did I carry out the plan?” More than one answer can be true. Responsibility and uncertainty can coexist.
Write a lesson you can actually use
“Trust myself more” may express a wish, but it leaves the next moment vague. “Before declining an offer over one condition, ask whether that condition can change” names a situation and an action.
Keep the lesson proportional to the evidence. One disappointing job does not prove that you always choose badly. One successful delay does not prove that waiting is always wise. If you suspect repetition, compare several records and look for exceptions as well as similarities.
“I asked the question I needed to ask. I chose based on the answer. I miss the possibility, but I would not change that part of the process. Next time, I will also ask to speak with a current team member before deciding.”
When should you stop reviewing?
A useful review produces a correction, a question worth checking, or an honest recognition that the outcome was uncertain. When you keep reopening the same material without new evidence or a different next action, pause the review.
You can return when something changes. If the decision is already made and the same argument keeps restarting, read overthinking a decision you already made.
Shadow OS is developing expanded decision-record and look-back features for an upcoming app update. You can use the review above in a notebook today. The Decision Archetype Quiz offers a different starting point: a description of how you think you choose, which you can compare with your actual records.
The research supports the distinctions noted above. The templates and illustrative examples are editorial exercises, not validated assessments or evidence of an app's effectiveness.