Decision Log. The Memory of Why
How to record, revisit, and learn from decisions over time
By the end of this module
Purpose
Decision Log is Foundry's most underused feature, and the one that pays back the most over time. It's a structured record of decisions made with enough context to revisit them later, learn from them, and build a personal library of good and bad bets.
What to Log
Log any decision where:
- The reasoning is non-obvious and you might forget it
- The outcome is uncertain and you want to track whether it proved correct
- You're making assumptions that could be tested
- Multiple alternatives existed and you chose one
Examples: technology choices, pricing strategy, hiring/not-hiring, positioning decisions, product feature decisions, partnership choices.
The Decision Form
- Title. Short label for the decision
- Decision Made. Exactly what you decided
High-value optional fields:
- Context. What was happening that led to this decision?
- Reasoning. Why this choice over the alternatives?
- Alternatives Considered. What else could you have done?
- Assumptions. What do you have to believe for this to be right?
- Risks. What could go wrong?
- What Would Prove This Wrong. The falsification condition
Revisit At. Set a date to review this decision. Appears on Today when overdue.
Revisit Flow
When a revisit date arrives, it appears on Today under "Overdue Revisits." Navigate to the decision, review the original context, then:
1. Record a verdict (VALIDATED / PROVED_WRONG / SUPERSEDED / STILL_ACTIVE)
2. Add an outcome note explaining what happened
3. Reschedule if you want to check again later, or dismiss to mark complete
This is where the compound learning happens. Over time, your decision log becomes a personal track record.
Decision Status
| Status | Meaning |
|---|---|
| -------- | --------- |
| ACTIVE | Current, operative decision |
| REVISIT_PENDING | Revisit date set and approaching/passed |
| SUPERSEDED | Replaced by a newer decision |
| ARCHIVED | No longer relevant |
Common Mistakes
Not setting revisit dates. Without a revisit date, decisions accumulate but never close the learning loop.
Writing too vaguely. "We decided to focus on B2B" is weak. "We decided to target SMB operations teams in the $10K-50K ARR range because they have budget authority and low coordination overhead" is strong.
Decision Lifecycle
Current operative decision. Revisit date set.
Revisit date has passed. Appears on Today.
Proved correct on revisit. Closed with learning.
Falsified on revisit. Outcome note captures why.
Replaced by a newer decision via REPLACES link.
Too early to call. Revisit date extended.
Transitions
A decision is a loop, not an event. Every state above is a moment in that loop, the value is in closing it, not in logging it.
Practice Checkpoints
Sign in to track your progress and mark checkpoints complete.
Next in Foundations Path
Focus. The Sustainability Dashboard
Strategic Priority lives here. So does the burnout early-warning system.