How to keep a decision journal for a decision you make over and over

Most decision-journal advice pictures a single decision: you write your reasoning once, the outcome resolves once, and you compare the two. A decision you make over and over — take the extra shift, have one more drink, answer the message tonight instead of tomorrow — does not work that way. It predicts and resolves on the same weekly loop, sometimes a dozen times before you have even noticed it is the same decision wearing a new date. That changes what an entry is for, how many you need, and when to stop writing fresh ones.

Why the usual format assumes a decision that only happens once

The standard entry — what I expect, why, what would change my mind — is built for something like a job offer or a move: rare, high-stakes, resolved once, and worth a careful paragraph because you will not get many chances to learn its lesson. A recurring decision inverts every part of that. It is low-stakes any single time, resolved within days rather than months, and you get another attempt next week whether you learned anything or not. Writing the full careful version each time is not wrong, exactly, but it is the wrong amount of effort for what the entry needs to do, and an entry that takes too long to write is an entry you stop writing by the third occurrence.

What the first few entries are actually testing

Treat the first three or four occurrences as genuinely uncertain, because they are — you do not yet know if this is a pattern or a coincidence. Keep each entry short: the trigger that made the question come up, what you decided, and one line on how you expected to feel about it a day later. That last line is the one that matters here and is usually missing from a rushed entry. It is not asking whether the decision was right; it is asking whether you can predict, in advance, how this particular version of the decision tends to go. If you cannot yet, that is the honest answer for entry two, and it is exactly what the next entry is for.

Read them side by side, not one at a time

A single entry about a recurring decision teaches you almost nothing on its own — of course you can tell a story about why last Tuesday's extra shift made sense. The information is in the set, not the instance, and it only shows up when you put four or five entries next to each other and read across them instead of down through them one at a time. Look for what repeats: the same trigger, the same time of day, the same "just this once" reasoning attached to a different fact each time. A single entry has no way to catch that it is not, in fact, a one-off — five entries with the same shape make it hard to keep pretending each one is exceptional.

The moment you already have your answer

There is a specific tell that you have enough entries: you can predict what the next one will say before you write it. If you already know that "I'll just answer tonight" tends to be followed, a day later, by the same flat feeling the last four times produced, writing a fifth full entry is not gathering new evidence — you are just watching yourself confirm something you already know. That is the signal to stop journaling the decision and start writing the rule instead. Continuing to journal past this point is not neutral; it quietly launders an already-answered question as if it were still open, which is its own kind of avoidance.

Write the rule, not the fortieth entry

Once the pattern is clear, the recurring decision is not a decision anymore — it is a standing question with a known answer, and that is what a principle is for. Take what the entries actually showed you, not what you wish they showed, and write it as a rule with a real edge: something like "I don't take an extra shift on a night I have plans, even a small one" rather than a vague intention to be more careful about it. This is the same move covered in more depth elsewhere — telling whether something belongs in a decision journal or has graduated into a principle worth writing down — but the specific version here is that the journal itself is what surfaces the evidence. You are not guessing whether a pattern exists; you read it out of entries you already wrote.

Keep a lighter entry for when you break the rule

Writing the rule does not mean the recurring decision never needs a journal entry again — it means the entries that follow have a narrower, more useful job. Once you have a standing answer, save a fresh entry for the times you override it, and be honest in it about why: a genuine exception, or the same old pressure wearing a better excuse. Those exception entries are worth more than the original dozen combined, because they are the ones that will eventually tell you whether the rule needs an edge case added or whether it is holding up exactly as written.

CreedOS's daily check-in has a short note field next to whatever you are tracking, which is enough room for a few lines on a given day, and a creed's version history keeps the earlier wording around once a recurring decision graduates into a written behavior standard — so you can still see the reasoning that produced the rule, not just its final sentence. It's free, with no in-app purchases.