Contents
General Thinking & Meta-Models
Section 1
The Core Idea
Outcome-blind evaluation is judging a decision by the quality of the process and the information available at the time, not by the outcome that followed. A good decision can lead to a bad result; a bad decision can lead to a good result. Luck and uncertainty ensure that outcomes are an imperfect signal of decision quality. The discipline is to separate "did we decide well?" from "did it work out?" and to reward and learn from the former while using outcomes to update beliefs and improve process.
In deciding and judging, the model corrects two errors. The first is outcome bias: judging decisions by results. We tend to call a decision good when it works and bad when it fails, even when the same process could have produced the opposite result. That distorts learning and incentives: people are rewarded or punished for luck, and we draw the wrong lessons from history. The second error is process worship: ignoring outcomes entirely and only caring about procedure. That is also wrong — outcomes contain information. The right stance is outcome-blind for evaluation of the decision (was the process sound? were the right inputs considered?) and outcome-aware for learning (what did we learn from what happened? how do we update?).
The model is central to probabilistic thinking and to building a decision-making culture that does not over-reward or over-punish luck. In investing, a single outcome tells you little about whether the thesis was right; you need many trials or an explicit outcome-blind review of the thesis and process. In strategy, a failed initiative may have been the right bet; a successful one may have been a bad bet that got lucky. Outcome-blind evaluation asks: conditional on what we knew and the process we used, was this a good decision?
Section 2
How to See It
You see the need for outcome-blind evaluation when good process leads to bad results or bad process leads to good results — and people still judge the decision by the result. The diagnostic is the mismatch: we are praising or blaming based on outcome when we should be evaluating based on process and information at the time.
Business
You're seeing Outcome Blind when a team makes a well-researched bet that fails and is treated as a failure of judgment, while another team makes a reckless bet that pays off and is celebrated. The first decision may have been good (right process, bad luck); the second may have been bad (wrong process, good luck). Outcome-blind evaluation would judge both by the quality of the decision, not the result.
Technology
You're seeing Outcome Blind when a post-incident review concludes "we should have done X" purely because X would have prevented this incident — without asking whether X was reasonable given what was known before the incident. The outcome is driving the lesson; outcome-blind evaluation would ask what the right decision was given prior information and whether the process was sound.
Investing
You're seeing Outcome Blind when a manager is lauded for a big winner or fired for a big loser without examining whether the thesis and process were sound. A single outcome is a noisy signal. Outcome-blind evaluation would review the thesis, the sizing, the process — and use the outcome to update beliefs, not to judge the decision in isolation.
Markets
You're seeing Outcome Blind when a policy is deemed a success or failure by the result (e.g. inflation went down) without asking whether the policy was the right call given what was known and what alternatives existed. Outcomes are informative, but they should not be the only basis for judging the decision. Separate "was the decision good?" from "did it work out?"
Section 3
How to Use It
Decision filter
"When evaluating a decision, ask first: given what was known at the time and the process used, was this a good decision? Then ask: what did the outcome teach us? Use outcome-blind evaluation for accountability and incentives; use outcomes for learning and updating. Do not judge decisions by results alone."
As a founder
Evaluate people and decisions by process and information at the time, not only by outcomes. Reward good process even when results are bad; correct bad process even when results are good. In post-mortems, separate "what did we decide and why?" from "what happened?" and "what do we learn?" Use outcomes to update beliefs and improve process, not to assign blame or credit for luck.
As an investor
Judge investment decisions and managers by thesis quality, process, and sizing — not by a single outcome. A good thesis can lose; a bad thesis can win. Over time, good process should outperform, but in the short run outcomes are noisy. Use outcome-blind review in portfolio and manager evaluation; use outcomes to refine theses and process.
As a decision-maker
When you or others are judged by results, push for outcome-blind evaluation: "Given what we knew and how we decided, was this the right call?" When learning from history, use outcomes to update — but do not let a single outcome override the lesson that good decisions can fail and bad decisions can succeed. Build a culture that rewards good process and learns from outcomes without conflating the two.
Common misapplication: Ignoring outcomes completely. Outcome-blind evaluation is for evaluating the decision — was it a good call at the time? Outcomes still matter for learning: they provide evidence to update beliefs and improve process. Do not use "outcome blind" to avoid accountability when results are bad; use it to separate luck from skill and to learn correctly.
Second misapplication: Applying it only in hindsight. The discipline is to set the standard in advance: we will evaluate decisions by process and information at the time. If you only invoke outcome-blind when a decision fails, it looks like an excuse. Establish the norm before the outcome is known.
Section 4
The Mechanism
Section 5
Founders & Leaders in Action
Jeff BezosFounder & CEO, Amazon, 1994–2021
Bezos distinguished type 1 (irreversible) from type 2 (reversible) decisions and urged speed on type 2 so that outcome does not paralyse. His "disagree and commit" emphasises process — get the right input, then commit — rather than waiting for certainty. Outcome-blind evaluation fits: evaluate the decision by the process and information at the time; do not let a single outcome override the quality of the decision.
Warren BuffettChairman, Berkshire Hathaway
Buffett has long said that he evaluates decisions by the process and the logic at the time of the decision, not by the short-term outcome. He has also said that the stock market is a voting machine in the short run and a weighing machine in the long run — implying that outcomes eventually reflect quality but that short-run outcomes are noisy. His discipline is to focus on the quality of the decision and to let outcomes inform but not define judgment.
Section 6
Visual Explanation
Outcome Blind — Evaluate the decision by process and information at the time; use outcomes for learning. Good decision + bad luck = still a good decision. Bad decision + good luck = still a bad decision.
Section 7
Connected Models
Outcome blind sits with models about bias, process, and incentives. The connections below either describe the bias it corrects (outcome bias, hindsight bias), ground it in theory (expected utility, probabilistic thinking), or touch on incentives and process (skin in the game, process overhead).
Reinforces
Outcome Bias
Outcome bias is the tendency to judge decisions by their results. Outcome-blind evaluation is the corrective: judge by process and information at the time. The reinforcement: outcome bias is the error; outcome-blind evaluation is the discipline that corrects it. Name the bias when you see it; apply the evaluation standard to reduce it.
Reinforces
Expected Utility Theory
Expected utility theory says a good decision maximises expected utility given beliefs — the outcome is one draw from a distribution. Outcome-blind evaluation is the behavioural application: evaluate whether the decision was correct ex ante (did it maximise expected utility given what was known?), not ex post (did the outcome turn out well?). The reinforcement: both separate decision quality from realised outcome.
Tension
Hindsight Bias
Hindsight bias is the tendency to see the outcome as having been predictable after it occurs. Outcome-blind evaluation tries to judge as of the time of the decision, without the benefit of hindsight. The tension: we have to work to reconstruct "what was known then" and to ignore "what we know now." The discipline is to make the evaluation outcome-blind despite our hindsight-biased instincts.
Tension
Process Overhead
Process overhead is the cost of adding steps and checks to decision-making. Outcome-blind evaluation can add process (e.g. documenting rationale, pre-registering criteria). The tension: too much process can slow decisions and add bureaucracy; too little and we fall back to outcome-based judgment. Balance: enough process to evaluate decisions fairly, not so much that good decisions are delayed or avoided.
Leads-to
Probabilistic Thinking
Probabilistic thinking is reasoning in terms of probabilities and distributions rather than single outcomes. Outcome-blind evaluation follows: if outcomes are draws from a distribution, then a single outcome is a noisy signal of decision quality. The lead: the more you think probabilistically, the more natural it is to evaluate decisions by process and to use outcomes for learning rather than for judgment.
Leads-to
Skin in the Game
Skin in the game means decision-makers bear the consequences of their decisions. Outcome-blind evaluation does not ignore consequences — it separates "did they decide well?" from "did it work out?" The lead: you can have skin in the game (you bear the outcome) and still evaluate decisions outcome-blind (we judge the decision by process). Incentives can be outcome-based while evaluation is process-based; the key is to not conflate "they lost money" with "they made a bad decision."
Section 8
One Key Quote
— Annie Duke, Thinking in Bets"Resulting is the tendency to equate the quality of a decision with the quality of its outcome."
"Resulting" is outcome bias: we call a decision good when it works and bad when it fails. The quote names the error. Outcome-blind evaluation is the corrective: the quality of the decision depends on the process and information at the time, not on the outcome. Good decisions can fail; bad decisions can succeed. Separate the two so that incentives and learning align with decision quality.
Section 9
Analyst's Take
Faster Than Normal — Editorial View
Set the standard in advance. Outcome-blind evaluation only works if it is the norm before the outcome is known. If you invoke it only when a decision fails, it looks like an excuse. Establish upfront: we will evaluate decisions by process and information at the time. Then apply it consistently to successes and failures.
Separate evaluation from learning. Use outcome-blind evaluation for accountability and incentives: was this a good decision? Use outcomes for learning: what did we learn? What should we update? The same outcome can lead to "good decision, bad luck" (evaluate the decision positively; learn from the outcome) or "bad decision, good luck" (evaluate the decision negatively; learn from the outcome). Do not use outcome-blind to avoid learning when results are bad.
Document the process. To evaluate outcome-blind, you need a record of what was known and how the decision was made. Encourage or require a brief decision memo: what did we know? what were the options? what did we choose and why? That creates the basis for outcome-blind review later.
Single outcomes are noisy. In investing, strategy, and hiring, a single outcome is a weak signal of decision quality. Use outcome-blind evaluation for the decision; use many outcomes (or explicit process review) to assess the decision-maker or the process. Over time, good process should produce better outcomes — but in the short run, separate the two.
Section 10
Test Yourself
Is this mental model at work here?
Scenario 1
A team made a well-researched bet on a new market. The market collapsed for reasons no one predicted. The team lead is demoted.
Scenario 2
A founder took a reckless bet that happened to pay off. The board celebrates her judgment. She repeats the bet and loses badly.
Scenario 3
A post-mortem says 'we should have done X' because X would have prevented the incident. No one asks whether X was reasonable given what was known before the incident.
Section 11
Summary & Further Reading
Summary: Outcome-blind evaluation is judging a decision by the quality of the process and information at the time, not by the outcome. Use it in deciding and judging to correct outcome bias, align incentives with decision quality, and learn correctly from history. Evaluate outcome-blind; use outcomes for learning. Set the standard in advance and document process so outcome-blind review is possible. Pair with outcome bias, hindsight bias, expected utility, probabilistic thinking, process overhead, and skin in the game.
Further Reading
01
Book
Kahneman on outcome bias, hindsight bias, and the separation of decision quality from result. Foundational for outcome-blind evaluation.
02
Book
Mauboussin on luck vs skill and how to evaluate decisions when outcomes are noisy. Practical application of outcome-blind thinking in investing and strategy.
03
Paper
Classic paper demonstrating that people judge decisions by outcomes even when the outcome was due to chance. The empirical basis for outcome-blind evaluation as a corrective.
Why this matters next
mental modelsIncentives
Outcome Blind applied the Incentives mental model
mental modelsProxy
Outcome Blind applied the Proxy mental model
mental modelsUtility
Outcome Blind applied the Utility mental model
mental modelsQuality
Outcome Blind applied the Quality mental model
mental modelsInflation
Outcome Blind applied the Inflation mental model
mental modelsOutcome Bias
Outcome Blind applied the Outcome Bias mental model
Frequently asked questions
What is Outcome Blind?
Outcome Blind is a mental model used for better thinking and decision-making.
How do you apply Outcome Blind?
To apply Outcome Blind, identify situations where this framework is relevant, then use it as a lens to evaluate your options and decisions. The model is most useful when combined with other complementary mental models.
What category does Outcome Blind fall under?
Outcome Blind falls under the General Thinking & Meta-Models category of mental models. Other models in this category can be found on the General Thinking & Meta-Models hub page.
Why is Outcome Blind important?
Outcome Blind is important because it provides a structured way to think about problems that would otherwise be approached with intuition alone. Understanding this model helps you avoid common reasoning errors and make better decisions.
Continue exploring
5WAWALEMFTGL
Mental model
5 Whys
A root-cause analysis technique that drills past symptoms by asking why five suc
Mental model
All Models Are Wrong
George Box s insight that every model is a simplification of reality — the quest
Mental model
Amara's Law
We tend to overestimate the effect of a technology in the short run and underest
Mental model
Eisenhower Decision Matrix
A prioritization framework that sorts tasks into four quadrants based on urgency
Mental model
First Principles Thinking
Breaking down complex problems to their most fundamental truths, then reasoning
Mental model
Goodhart's Law
When a measure becomes a target, it ceases to be a good measure — incentives shi
More like this, in your inbox
I send a newsletter every week — free, no spam, unsubscribe anytime.