A launch succeeds, and suddenly every meeting before it sounds visionary. A launch struggles, and the same meetings become a catalogue of obvious warnings. The ending can become the editor of the story.
There is a quieter way to celebrate good judgment: leave a record before the result arrives. A decision journal is useful because it gives your later self something specific to inspect. You can see what you believed, what you checked and what you were still guessing.
Farnam Street describes decision journaling as recording your thinking before acting, then reviewing it against subsequent events. The short working note below is our own format for everyday professional choices.
Choose a decision that can teach you something
Do not document every calendar invitation. Pick a choice whose reasoning you might need again: which offer to launch, whether to hire for a recurring bottleneck, or which audience deserves your next month of research.
A useful entry begins with a sentence containing a verb. “Our positioning” is a subject. “We will test a workshop for newly promoted managers before developing a longer programme” is a decision. It also contains a boundary. You are authorizing a test, not declaring a permanent business model.
Write the decision while a real alternative remains available. A retrospective justification written after the commitment has already become public has a different purpose.
Write a note your future self can challenge
Use these six prompts. A paragraph for each is usually enough.
| Prompt | What belongs in the note |
|---|---|
| What are we choosing? | A specific action and its boundary. |
| What did we observe? | Conversations, constraints or results you can locate again. |
| What are we assuming? | The part that still depends on a belief. |
| What else could we do? | A credible alternative, including waiting. |
| What would change our mind? | Evidence that would alter the choice. |
| When will we look again? | A date and the information expected by then. |
Keep observations and interpretations separate. Three customers requesting help with onboarding is an observation. A large underserved market is an interpretation. The second may eventually be supported, but the first does not establish it by itself.
Make uncertainty visible without inventing precision
Consider a fictional consultant deciding whether to create a group workshop. Two existing clients requested a similar session. Neither has approved a budget. The consultant has time for one pilot and cannot support a second audience this month.
An honest note might say: “We have evidence of interest in this problem, but no evidence that people will pay for this format. We will offer one clearly scoped pilot. We will revisit after the invitation period, using paid commitments and reasons for declining.”
That note is more useful than “This will probably work.” It identifies the uncertainty and the next observation. If you use numerical confidence estimates, explain what the estimate means. A percentage with no reasoning can merely give a guess a more authoritative costume.
Review the choice on two separate pages
On the first page, evaluate the reasoning using only information available at the time. Was the alternative fairly considered? Did anyone search for disconfirming evidence? Were the constraints real? Did the chosen test expose the important uncertainty?
On the second page, record the outcome and unexpected events. A cancellation caused by a venue closure is different from no customer wanting the workshop. Both affect the result; they teach different lessons.
Do not quietly edit the original entry to sound wiser. Add a dated update. Preserving the difference between earlier belief and later knowledge is the point of the exercise.
Turn the lesson into a change you can observe
End the review with one adjustment to the next comparable decision. Perhaps you will seek paid commitments before booking a room. Perhaps you will consult the person responsible for delivery earlier. Avoid conclusions so broad that they cannot guide action, such as “trust ourselves more.”
A journal is not a machine for making uncertainty disappear. A small collection of entries can reveal questions you repeatedly neglect, but it cannot isolate every cause of a business result. Use it alongside customer evidence and relevant expertise.
For your next consequential choice, leave a dated note and schedule its return. Good judgment deserves more than a flattering story after the event. Give it a record that can help you make the next decision better.
Continue with a useful after-action review when the work is complete.
Questions worth asking
Which decisions belong in a decision journal?
Record decisions that consume meaningful time, money or attention, especially when you expect to face similar choices again. Routine reversible choices rarely need a full entry.
How often should I review a decision journal?
Choose the review date when writing the entry. Match it to when useful evidence should appear, and leave the original entry unchanged.
