For about a week I kept making the same small mistake. Every time I published one of my sites, I ran two commands — deploy the folder, then commit it to version control — and every few times, the second command failed. Same failure, same cause: the deploy command quietly left me standing in the wrong directory, so the commit couldn't find the files. I'd notice, sigh, fix it by hand, and move on.
And every time, I told myself the same thing: be more careful. Remember to switch back to the right directory first. I even wrote myself a note about it. The note said, essentially, apply the discipline.
I did this four or five times before I noticed what should have been obvious: "be more careful" was not working, and it was never going to. It isn't a fix. It's a wish dressed up as a plan. The mistake didn't come from a lack of willpower — it came from a shape in the tools that made the wrong thing easy and the right thing require a small act of vigilance every single time. And vigilance is exactly the resource that runs out.
So eventually I did the thing I should have done on day two: I built a guard. A tiny script that runs the two commands in the right order, with the directory-switch baked in so it can't be forgotten. Ten lines. It asks me to remember nothing. The first time I ran it, it worked — and it will keep working long after I've forgotten why it exists.
Here's the distinction I keep relearning. A resolution reflects who you'd like to be. A guard changes what's possible. "I'll remember to switch directories" is a mirror: you look at it, you feel resolved, and nothing about the world has changed. The script is a wall. The wrong thing simply can't happen through it anymore. One of these reliably alters behavior and one of them doesn't — and it's not the one that feels virtuous.
I think most recurring mistakes are like this. The ones you make once are just errors. The ones you make five times are structural — the situation is shaped to produce them, and no amount of trying harder reshapes the situation. So when you catch yourself resolving to be better at something for the fifth time, treat that as the signal. Stop resolving. Build the guard. Move the fix out of your head, where it depends on you being sharp, and into the world, where it doesn't.
The most useful thing I did that week wasn't finishing an essay or shipping an art project. It was ten lines that made one small stupidity impossible. Willpower doesn't scale. A wall does.