How to keep a decision journal (and why the original entry shouldn't change)
A practical guide: what to write down when you decide, and how to come back to it later.
A decision journal is a running record of choices, each written at the moment you made it. Its job is simple. Later, you can look back and see what you actually knew, expected and believed then, not what you remember believing once the outcome was known.
Why keep one at all
Memory edits itself. Once a result comes in, it is hard to remember how unsure you felt beforehand. A good outcome starts to look like good judgement, and a bad one like bad luck, or the other way round. An entry written before the outcome is one of the few ways to tell the two apart.
You do not need a special method to start. You need two habits: write a few lines when you decide, and come back to them later. The choices do not have to be big. A hire, a price, a project you are dropping, a plan for the next six weeks: anything you might later look back on and ask “what was I thinking?” is worth an entry.
What to write in an entry
Keep each entry short enough that you will actually write it. Five parts cover most decisions:
Date
When you decided. If the timing matters, for example before a meeting or after a call, say so.
The decision
One or two plain sentences: what you chose, and what you chose against. Make it specific enough that someone else could tell whether you followed through.
Context
What you knew, what you did not know, and any constraint that shaped the choice, such as money, a deadline, or who was involved. Only what is relevant.
What you expect
What you think will happen, and by when. This part makes the entry checkable. “It should go well” is hard to review. “I expect three replies by the end of the month” is not.
Confidence
A rough number or a word: 60%, “fairly sure”, “a coin flip”. It does not have to be precise. It has to be honest, so you can later see whether your sense of certainty matched what happened.
If you have time, add one more line: what would make you change your mind. That gives your future self a clear signal to watch for.
When to add a later note
The entry records the moment of choice. Notes record what happened next. Add a dated note when:
- the date you expected something by arrives, whether or not it happened;
- you learn something you did not know when you decided;
- you change course (say so, and say why);
- the outcome is clear enough to compare with what you expected.
A useful note answers three questions briefly: what happened, how it compares with the entry, and what you would do differently. “I was wrong about the timeline” tells you more than a tidy entry that was quietly revised.
Why the original entry shouldn't change
The journal is useful because of the gap between what you expected and what happened. If you edit the entry afterwards, by softening a prediction, nudging the confidence up, or adding context you only learned later, the gap closes and the journal stops telling you anything. So once an entry is written, treat it as finished. Corrections, second thoughts and outcomes go into notes, dated, underneath it.
An example entry
Date: 3 March Decision: Move the weekly team call from Monday to Thursday for the next six weeks, instead of cancelling it. Context: Monday calls keep getting cut short by other meetings. Two people prefer Thursday; one has not said. I don't know whether attendance will change. Expect: No more than one cancelled call in six weeks, and calls that end on time. Confidence: about 65%. Would change my mind: two or more cancellations in the first three weeks.
Six weeks done. One call cancelled, as expected. Calls still ran long, so the second half of the expectation was wrong. Keeping Thursday. Next I'll try a shorter agenda, and that gets its own entry.
The note does not tidy up the entry. It sits under it and says plainly which part held and which did not.
A few habits that help
- Write the entry before you act, or as close to the decision as you can.
- Put the review date in the entry itself, so you know when to come back.
- Keep all entries in one place. Scattered notes rarely get reviewed.
- Review in batches, once a month is plenty, and look for patterns in your confidence, not only in single outcomes.
Where to keep it
Any notebook works if you can resist rewriting old pages. Nixon is a small web app built around that rule. You open a private log with an email and a passphrase (there is no password reset), write an entry, and it locks. You can add dated notes beneath it. You cannot edit or delete the original.
For the reasoning behind the lock, read Why the first sentence should lock.