Pomodoro didn't work for me either

I tried the twenty-five-minute thing for a semester and concluded I was bad at focusing.

That was the wrong conclusion. What was actually happening: it worked beautifully for problem sets and flashcards, and it consistently wrecked the work where I had to load a lot of context before anything happened.

Two kinds of work

Work made of small pieces. Ten calculus problems. Thirty flashcards. Reading a chapter and taking notes. Each piece finishes on its own, so when the timer rings, nothing is lost. The interval helps here — it makes starting easy and it stops you from grinding for three hours at declining returns.

Work you have to hold in your head. A proof. A hard problem you've been circling. The paragraph where an argument finally comes together. This work has a warm-up: five or ten minutes of loading the problem, the constraints, and what you already tried, before you're anywhere useful.

Ring a timer twenty-five minutes in and you're often interrupting at the exact moment it started working. Then you come back and pay the loading cost again.

I lost most of a semester to this before noticing it wasn't a discipline problem, it was a mismatch.

Match the block to the work

Short blocks, 25 to 30 minutes. Problem sets, flashcards, reading with notes, admin, editing. Anything you can pause cleanly.

Long blocks, 60 to 90 minutes. Writing, proofs, hard single problems, anything requiring a lot in working memory at once. Take a real break at the end rather than four small ones in the middle.

Very short blocks, 10 to 15 minutes. Not for productivity — for starting when you can't start. More on that below.

You can mix these in a day, and the mixing is the point. One system for all work is the thing that fails.

When starting is the problem

A different failure mode, and much more common than not being able to sustain focus.

You sit down. You mean to start. Forty minutes later you've read some things on your phone and you feel worse. This isn't a focus problem, it's an activation problem, and timers don't touch it.

Three things that help, in order of how much they've helped me:

Make the first action absurdly small. Not "study stats." Open the problem set and read problem one. That's the entire commitment. Almost always, once the file is open and you've read one problem, you keep going — the resistance was to starting, not to working.

Start in the middle of something. Leave a task half-finished at the end of a session, deliberately. Coming back to a half-written paragraph is much easier than coming back to a blank page, because the reentry decision is already made.

Change the room. This one is embarrassingly effective and I resisted believing it. Some places are where you work and some are where you don't, and your bed is definitively the second kind. If you can't start, move — library, a different floor, a coffee shop. The cost of walking somewhere is often less than the cost of the hour you were about to lose.

Phones

Everyone knows. Two things worth saying anyway.

Face-down on the desk is not enough. The presence of it costs something even unused. In a different room, or a bag on the far side of the table, is a different experience.

Blockers work until you learn to defeat them, which takes about a week. The durable version is physical distance, not software.

If you use your phone for the actual work — flashcards, readings — then the reverse trick: one app, and put the phone in a mode where nothing else can surface.

Breaks that restore anything

Scrolling is not a break. You end a scroll break more depleted than you started it, which is why the "five minute break" turns into forty.

What actually restores: standing up, walking, going outside, water, something with your eyes not focused on a rectangle 50 centimeters away. Boring, and the boring part is what makes it work.

The other thing about breaks: take them before you need them. A break at 90 minutes, taken deliberately, costs ten minutes. A break at 140 minutes, taken because you've stopped functioning, costs an hour and you spend part of it feeling bad about yourself.

The thing under all of this

Most focus problems I've had were not focus problems.

They were sleep. Or having no clear idea what the next action was, so I kept sitting down to "study for the midterm," which is not a task and cannot be started. Or working at the wrong time of day for the kind of work — I'm useless for anything requiring judgment after 10 p.m., and I spent years scheduling my hardest work exactly there.

So before you try a new system, three questions:

1. Am I sleeping? Not enough sleep looks exactly like an attention problem and no technique compensates.
2. Do I know what the specific next action is? "Study for the midterm" is not one. "Do problems 4 through 10" is.
3. Is this the right time of day for this kind of work?

More often than not, one of those was the answer, and no timer was going to fix it.

What I do now

Small work in 30-minute blocks, hard work in 90-minute blocks, phone in another room for the second kind, and one rule for when I can't start: open the file and read one problem.

That's not a system. It's the leftovers of four systems that didn't survive contact with an actual semester, which is roughly what a working method looks like.