This is highly ironic, since I'm all about commitment devices as a tool to not procrastinate. But fun fact: you can use procrastination itself strategically as a commitment device to cap the amount of time you spend on something.

This is really just a corollary of Parkinson's law, that "work expands so as to fill the time available for its completion". Here's a cute way to put it, apparently original to Wikipedia^[1]:

If you wait until the last minute, it only takes a minute to do.

So, that's all. If you don't want to let yourself spend more than X hours on something, don't start working on it till X hours before the (hard) deadline.

Is this like Structured Procrastination?

No, that's quite different. Structured Procrastination is an elaborate psychological game you can play with yourself in which you get yourself to do important work by framing it as procrastination on some other supposedly more urgent task — a decoy task. It reminds me a little of setting your clocks 10 minutes fast to trick yourself into being more punctual. I personally hate things like this. Some people think of Beeminder as being in this category — or that it's even worse than self-trickery, namely, self-blackmail. I don't think either are fair characterizations. I see it as merely incentive alignment. Like strategically removing your ability to go down fractal rabbit holes all day while writing a blog post because you started it at 9pm and if you don't hit publish by midnight you'll be kicked out of Inkhaven.

What about "Strategic Procrastination"?

Sigh, if you google that, you get a lot of platitudes about how it's good to take breaks and let your subconscious mind process information. I'm sure that's true, but not at all what I mean. But since "strategic procrastination" and "structured procrastination" are both taken, we could make up a concept handle like Parkinson capping or t-minus timeboxing or just-in-timeboxing. Maybe one of those will stick?

The Principle of Delayed Commitment

This one should maybe be its own post but since it's related, let me quote myself from a footnote of an old Beeminder blog post:

The principle of delayed commitment, also known as the principle of least commitment in [old-school] AI circles, is the simple decision-theoretic fact that you shouldn't commit to a course of action until you have to. Flexibility is valuable; laziness is a virtue. It's also one of the Lean Software Development principles. And I'm sure it comes up in other fields as well.

In short, deliberately waiting till a strategically chosen amount of time before a deadline to start working can be perfectly rational, by preserving optionality, and meta-rational, by leveraging the panic of a looming deadline in order to focus.

PS: Remember that if you're nodding along too hard to this advice, you might need to reverse it.

1. ^{^}

   I hear you, Wikipedia doesn't allow original research. Unless you do this:

   

   I believe that what Wikipedia calls the Stock-Sanford corollary of Parkinson's law is a bonafide example of citogenesis. Wikipedia cites a 2013 book which states the so-called Stock-Sanford corollary with no citation of its own. But checking the history of the Wikipedia page, the cute corollary was added in 2010! (Side note that GPT-5-Thinking figured this out but I confirmed it myself.)