Process-Over-Outcome Decision Review
Judge investment decisions by process quality, not isolated results
- Difficulty
- Moderate
- Time to result
- ~ongoing to results
- Steps
- 5
- Confidence
- 96%
The Process-Over-Outcome Decision Review separates the quality of a choice from the result produced by chance, timing, or incomplete information. Before acting, the decision-maker records the available evidence, underlying assumptions, expected payoff, and conditions that would invalidate the thesis. After the outcome, the review asks whether the original reasoning was disciplined rather than merely whether money was made or lost. Repeated reviews create a feedback loop that strengthens useful decision rules while exposing dogmatism, hindsight bias, and luck. The method preserves a coherent investment philosophy but requires enough flexibility to recognize when evidence has genuinely changed. It is especially valuable in markets, where a sound decision can lose temporarily and a poorly reasoned bet can succeed by accident.
Origin
Marta Norton connected her transition from research to portfolio management with Annie Duke's distinction between decision quality and outcomes on Masters in Business.
Core principles
- 01A good decision can produce a bad short-term outcome
- 02A winning outcome does not validate a weak process
- 03Real-world feedback should refine rather than replace a philosophy
- 04Conviction must coexist with willingness to be wrong
How to run it
- 1
Capture the original thesis
Write down the evidence, assumptions, expected payoff, and relevant uncertainties before the outcome is known.
Pro tip Use only information that was genuinely available at decision time.
Watch out A thesis reconstructed afterward will be distorted by hindsight.
- 2
Define disconfirming evidence
Specify which observations would demonstrate that the thesis or its mechanism is wrong.
Pro tip Distinguish temporary price movement from evidence about fundamentals.
Watch out If nothing could change your mind, the position is ideological rather than analytical.
- 3
Observe the outcome
Record what happened without immediately treating success as skill or failure as error.
Pro tip Include the time horizon over which the original payoff was expected.
Watch out Short-term results may arrive before the thesis has had time to work.
- 4
Audit the process
Evaluate information quality, assumptions, sizing, alternatives considered, and adherence to the stated rule separately from performance.
Pro tip Ask whether you would make the same choice again with the same information.
Watch out Do not reward an undisciplined decision merely because it made money.
- 5
Update the decision rule
Use patterns across multiple reviews to refine how future decisions are made while retaining principles that remain supported.
Pro tip Change one identifiable rule at a time so its effect can be observed.
Watch out Overreacting to one result replaces learning with outcome chasing.
In the wild
An investor buys an attractively valued company after documenting its balance-sheet strength and three-year earnings thesis. The stock falls after a weak quarter, but the underlying evidence remains intact. The review classifies the process as sound and the short-term result as inconclusive rather than automatically selling.
→ The investor preserves discipline while continuing to monitor the evidence that could invalidate the thesis.
A manager makes a macro trade based on an unsupported prediction and earns a quick profit. The review identifies that no disconfirming conditions or sizing rule were established, so the gain is not treated as proof that the method should be repeated.
→ The team avoids institutionalizing a lucky but unreliable decision process.
Common mistakes
Equating profit with skill
A positive result may reflect favorable randomness rather than a repeatable process. Review the evidence and reasoning that existed before the gain.
Using price as the only verdict
Market movement does not always prove or disprove the underlying thesis. Identify whether fundamentals or merely sentiment changed.
Moving the goalposts
Redefining the thesis after every surprise makes it impossible to learn. Preserve the original decision record and compare it with what occurred.
Is it for you?
Best for
It is best for investors, executives, and teams making repeated decisions under uncertainty.
Not ideal for
It is not ideal for deterministic tasks where outcomes directly and reliably reveal execution quality.
From the transcript
“don't evaluate the outcome, evaluate how you're making your decision the process.”
“when you are an investor, you have to learn how to have a philosophy but remain flexible, be willing to be proven wrong in real…”
From the episode
Marta Norton on Direct Indexing and Investments
Marta Norton