I Made the Decision Two Weeks Before I Had to, and It Finally Stuck

For years I treated decisions as something that happen at the moment they happen, and then spent the moment being someone I did not recognise. The fix that finally worked was absurdly simple: I moved the decision back two weeks, to a version of me who was calm and had the facts, and let him decide. The in-the-moment version turned out to be perfectly good at following instructions. He was just terrible at making choices.

There is a name for the gap between those two versions, and it is the gap I keep writing about: the reason a person gives for a decision and the process that actually produced it are rarely the same thing. In the moment, I produce reasons after the fact — “it is a special occasion”, “one more won’t matter” — that I would never have produced two weeks earlier. The two-week-ago version of me is not wiser. He is just not in the middle of wanting the thing.

The concrete version, the one I actually used: I wrote down the decision in full sentences, dated it, and put it somewhere I would see it before the moment arrived. “On Friday, when the work dinner happens, I am having two drinks and leaving at nine-thirty.” That is the whole technique. No app, no accountability partner. A dated sentence, written in advance, that the later version of me could either follow or explicitly break — and having to explicitly break it, instead of silently deciding, turned out to be most of the work.

Why it worked is that it removed the rationalization in advance. The in-the-moment reason “it is a special occasion” cannot survive contact with a sentence I wrote two weeks ago on an ordinary Tuesday, because the Tuesday version already knew Friday would call itself special. He had met this Friday before. The decision was never about willpower at the moment; it was about whether the moment got to write its own reasons, and it no longer did.

The one place it backfired taught me the limit. I pre-committed to something for the wrong reason — I wanted to look disciplined, not to change the outcome — and when the moment came, breaking the pre-commitment felt like a moral failure rather than a bad plan. A pre-commitment is only as good as the reason you wrote it. If you cannot state the reason in a sentence that survives being read back later, the advance decision is just a trap you set for yourself.

The research side calls this pre-commitment, or an implementation intention when it is the “if-then” form — “if it is Friday night, then two drinks and home by nine-thirty.” The “if-then” form works better for me than a bare rule, because it gives the later version an explicit trigger instead of a vague standard to judge himself against. I am not a clinician and this is not therapy; it is just the thing I caught myself doing that the literature happens to have a name for.

What I pre-decide now, and what I deliberately leave open, is a short list. I pre-decide the things where the in-the-moment version reliably lies to me — food after ten, the second round, the “quick check” of a feed. I leave open the things where I genuinely need new information, because pre-deciding those is just guessing in advance and calling it discipline. The skill is telling those two categories apart, and I get it wrong less now.

The transferable version is smaller than the jargon: the person deciding and the person following the decision are two different people, and you are allowed to pick which one gets the pen. Hand it to the calm one, in advance, in writing. The one in the moment will be fine. He is very good at following instructions.