Write a Letter to Yourself Before a Big Life Change

2026-10-02 · Yesterself

Six months after a big change, you will have one clean sentence about why you did it. Something like: the city had stopped fitting. Or: the job was going nowhere. Or: we had been over for a year and we both knew it.

That sentence will feel like the truth, and it will sound reasonable to whoever hears it. It is also a summary written after the results came in, by someone who already knows how it turned out.

This letter is the one you write before that happens. Not a pep talk, not a prediction, not a plan. A record of the state you are in while you still hold your reasons in your own hands.

What the letter is actually for

The letter is aimed at something narrow: preserving a testimony you will not be able to reconstruct later.

You are the only witness to your own reasoning, and you will be interviewed months from now, after being told the ending. That is the whole problem.

Baruch Fischhoff studied this in the 1970s. He and Ruth Beyth asked students to rate how likely various outcomes of Nixon's trips to China and the Soviet Union were, then later asked the same students to recall the ratings they had given. Many recollections had drifted toward what the students by then believed had happened. The uncertainty they genuinely felt was gone from the record in their heads.

Michael Ross made a related argument about personal histories: to recall what we used to think, we start from what we think now and adjust, according to whether we believe we are the sort of person who changes. Efficient, and not accurate.

So the change rewrites the story, and it rewrites it in both directions:

• If it goes well, you get I always knew. The doubt disappears, and with it the evidence that you can choose well under real uncertainty.
• If it goes badly, you get I ignored every sign. This one is more expensive. You end up prosecuting a version of yourself who did not exist, someone who had information you did not have.

A letter written beforehand does not settle which reading is right. It just means one of them can be checked.

The window

The window is narrow. It opens when you know the change is coming and closes when the first consequences arrive, because consequences immediately edit everything upstream of them.

• A move. Write while the apartment is still full. Once it is boxed, the place stops being where you live and becomes something you are leaving, which is a different building entirely.
• A job change. The night you accept, or the night before you resign. Not the first week, when relief or panic is doing the talking.
• A diagnosis, yours or someone else's. The gap between a scan and a result is unglamorous and useful: what you feared, what you had already decided, what you wanted to ask.
• A breakup. During the ambivalence, not after. Once someone has done something unforgivable in week three, the previous four years get resorted around it.

Nothing here is medical, clinical, or therapeutic advice, and writing is not treatment. If you are facing a diagnosis, or the weeks around one are more than you can carry, that belongs with a doctor or a licensed therapist. This letter is at best something you bring into the room with them.

What goes in it

One page is enough. The instinct to wait until you can write it well is the main reason these letters never get written.

1. The decision, in one dated sentence. What you are doing, or deciding between, and today's date. Write it for someone who has not been in your head.
2. Your reasons, in the order you actually feel them. Not the presentable order. If the first reason is that you are tired of being the person who stays, put that first.
3. The version you tell other people. Write it out, then write how it differs from number two. That gap is the most valuable line in the letter and the first one memory loses.
4. What you are afraid of, named. Not "I'm nervous." The specific outcome, described concretely enough to recognize if it shows up.
5. What you expect, and how sure you are, in words. "I think it is more likely than not that I will regret the money." Rough is fine; a number pretends to precision you do not have.
6. What would count as this being right, and what would count as it being wrong. Written now, before the goalposts can move without your noticing.
7. What you are giving up. Name it plainly. This is the item most likely to be denied later, when saying you never wanted it is easier.
8. The ordinary details. Where you are sitting, the noise outside, what the commute costs, what you ate. These make the letter readable as a place rather than an argument.
9. The conditions under which you would reverse it. Some changes cannot be reversed. Write the conditions anyway; they describe what you were optimizing for.

If you would rather the letter arrive on its own than sit in a folder you never open, Yesterself is built around that mechanic: any sentence can be postmarked to a future date and delivered on that day. It is an end-to-end encrypted journal with no account and no email, and it is not released yet; the site takes a waiting list.

Writing it so it survives

Assume the reader has forgotten the context entirely, because you will have. Use names instead of initials. Say what the meeting was about rather than "after Tuesday." Private shorthand that feels obvious today is opaque in a year.

Then leave the ugly parts in. The temptation is to write a letter you would not mind being found, and a letter with the pettiness edited out is a letter with the reasons edited out. Petty reasons are load-bearing more often than we admit.

Do not perform confidence you do not have. A letter that says "I don't know, and I am doing it anyway" is worth more later than a manifesto, because it is the only kind of letter that can ever surprise you.

One sitting is enough. On a third draft you are no longer recording the state you were in; you are composing a defense of it, and the draft you threw away was the honest one.

Opening it

Pick the date now and put it where the letter lives: on the envelope, in the filename, on the capsule. A letter with no delivery date surfaces by accident, years late, on a day you were not braced for. When it does arrive, read it under two rules.

You are not grading the outcome. You are grading the reasoning against what you knew at the time. A good decision can produce a bad result and remain a good decision; the letter exists precisely to keep those two things separate.

Read the whole thing before reacting to any line of it. The first paragraph of an old letter tends to provoke a verdict that the fourth paragraph would have complicated.

Most letters come back one of three ways. The reasons held, and you now have evidence that you can choose under uncertainty. Or they were quietly replaced by better ones and you did not notice the substitution. Or the stated reasons were never the real reasons, and the letter shows what you were actually solving for, which is usually useful and unwelcome at once.

None of those require you to do anything. You can read it once, note what it says, and put it away.

If you want one more step, write a short reply the same evening: what the letter got right, what it could not have known, and a line addressed to the version of you a year further on. That reply becomes the next letter, and the practice compounds instead of sitting in a drawer as a souvenir.

The letter has already done its job by existing, which is to make sure that when the story gets rewritten, somebody kept the original.

Common questions

When should I write this letter?

After you know the change is coming, but before the outcome starts arriving. The last quiet week before a move, the evening you accept the offer, the days before an appointment or a conversation you already know you are going to have.

What if I have not actually decided yet?

Write it anyway, and say so on the page. An honest record of being torn is more useful later than a confident record written after the fact, because it shows you what the choice actually cost at the time.

Should anyone else read it?

Not unless you want them to. This letter works best when nobody is being persuaded, including you. If you find yourself writing a defense, you have started a different document.

When should I open it?

Set a date now, far enough out that the outcome has landed. Six months and one year are both reasonable. You can also open it early, on the day you notice yourself telling a very tidy version of the story.