Building an investment decision journal

01Save the decision-time snapshot

Record the thesis, key sources, market reference, scenarios, risk budget, and unresolved questions as they existed when you decided. Later information should be added as an update, not allowed to overwrite the original state.

The immutable snapshot is what makes a review honest. Without it, memory tends to make the old thesis resemble whatever eventually happened.

02Record the action and the alternative

Save the chosen action—paper shares, a defined-risk paper structure, watch, or no position—and the strongest alternative considered. State why the chosen path fit the thesis and risk budget better.

This captures the opportunity cost of the decision instead of judging it against cash in the abstract.

03Use event-driven review triggers

Schedule review around the evidence that can change the thesis: a filing, product milestone, regulatory decision, financing event, price condition, or falsifier. A calendar reminder can support the process but should not be the only reason to reopen the record.

Each update should say what changed, which original claim it affects, and whether the decision response changed.

04Separate outcome from process

At the end of the review window, score source quality, calibration, risk discipline, execution accuracy, and update behavior before looking at the simulated return. This ordering reduces the temptation to call every winner a good decision.

The purpose of the journal is not to eliminate uncertainty. It is to make learning cumulative rather than anecdotal.

Use a consistent review template across winners, losers, and declined ideas. Comparable fields reveal recurring calibration errors that a collection of free-form post-mortems can hide.