I made a call back in March. It took me two weeks to sit with it, look at it from enough angles, and finally commit.
Then I spent the next eight weeks slowly remaking it.
Not because anything changed. Because the discomfort never fully went away.
The pattern I didn’t want to name
There’s a version of due diligence that’s actually second-guessing dressed up as rigor.
I’m good at spotting it in other people. A team stalls on a decision they already made. A founder revisits the pricing model they settled last quarter — not because the market shifted, but because there’s a meeting on Tuesday and it feels like the responsible thing to stress-test.
I’m less good at spotting it in myself.
What I’ve started noticing is that the symptom isn’t confusion. The symptom is specific: the question keeps appearing in my head even though I already answered it. Not a new version of the question. The same one.
That repetition is diagnostic.
If a question keeps coming back without new information attached to it, I’m not thinking. I’m running an old anxiety loop and calling it reflection.
Why it’s not the same as checking your assumptions
There’s a real version of revisiting a call that’s useful — the assumption you accepted without testing, that runs quietly in the background until friction surfaces it. That should get questioned.
But a decision I made consciously, with the information I had, is different.
- An assumption is something I accepted without actually testing.
- A settled decision is something I did test. I looked at it. I committed.
Re-opening a settled decision isn’t due diligence. It’s the carrying cost of not fully trusting the call I already made.
That cost is real. It’s just invisible in the way slow leaks are invisible — you don’t see the water, but you notice you’re always cold.
What I found when I started tracking it
I kept a rough log for about three weeks. Every time I found myself re-litigating a closed question, I wrote it down.
Two things came out of it.
First: most of the reopenings happened in low-stakes moments. The drive somewhere. The dead space between tasks. Late afternoon, when I had enough energy to worry but not enough to do anything useful with it.
That’s not reflection. That’s the brain filling dead air with something that feels like productivity.
Second: there were a small number of genuine re-opens. Cases where new information had actually arrived — a conversation, a changed constraint, a number that didn’t match what I expected. Those felt different. They came with a reason.
The distinction was much easier to see in the log than it was inside the experience.
What I changed
I didn’t try to stop the loop. You can’t reliably stop a thought by deciding to stop it. That just adds effort and usually makes things worse.
What I did instead: every time the question came back, I asked one thing:
What new information arrived since I last answered this?
If the answer was nothing, I wrote “nothing new” — literally, in a note — and moved on.
It sounds trivial. It isn’t. Naming the absence of new information makes the revisit feel like what it is: not work. And once it stops feeling like work, the loop has less pull.
Over a few weeks, the frequency dropped.
Not to zero. Some decisions are wired into identity in ways that make them stickier. Those are worth a different kind of attention. But for most recurring re-litigations, naming the absence is enough.
The honest part
The deeper reason I keep reopening certain calls isn’t intellectual.
It’s that the decision made something real. It closed off other options. And some part of me doesn’t want it to have been final — wants to believe the question is still fluid, still open, still something I can change if I decide I need to.
Keeping a decision “open” in your head is a way of preserving optionality that doesn’t actually exist anymore.
The call was made. Its consequences are already in motion. Re-running it doesn’t offer a second chance. It’s just overhead with no upside.
That’s the thing I have to keep coming back to.
Not “trust the process” — too abstract. Something more concrete:
This is a closed file. Opening it again costs time I don’t have and changes nothing that matters now.
Most of the time, that’s true.
Most of the time, the file should stay closed.