Pomodoro vs time blocking vs Flowtime: how to choose
It is Sunday evening and tomorrow is already on the screen. Four meetings: 09:30, 11:00, 14:00, 16:30. The longest clear stretch is eighty minutes, and it lands directly after lunch, which is the worst eighty minutes you own.
In another tab is an article about time blocking, because somebody in the team channel said it changed their life.
Choosing between these methods on vibes is why none of them lasted last time. They are not rivals — they fail in three different places, and the calendar you just looked at has already told you which failure is coming.
The rule underneath: match the method to your bottleneck
Ask what actually goes wrong on a bad day. There are three answers, and each method fixes exactly one of them.
You never start. The day dissolves, you look up at 16:00 and the work has not begun. That is Pomodoro’s problem to solve.
You start, and then the day gets taken. The intent was real; four other things ate the hours. That is time blocking.
You start, you get going, and something stops you — frequently the method itself, ringing at minute 25 exactly when the problem finally loaded. That is Flowtime.
One sentence to keep: pick for the thing that is broken — starting, protecting, or the bell.
Pomodoro, at its strongest
Twenty-five minutes on, five off, four rounds, then a real break. The case for it is narrow and very strong: it makes the first minute cheap.
Best day for it: shapeless. Nothing scheduled, several medium tasks, at least one you are avoiding. A Saturday with a tax return in it.
One worked afternoon: 13:30 first session on the task you like least, 13:55 stand up and leave the desk, 14:00 again. By 15:30 you have four facts about your afternoon instead of a feeling about it.
The mechanism is a promise to stop rather than a promise to work, which is the whole trick and is covered properly in the full guide.
Where it costs you: it is the highest-frequency method here. Sixteen small decisions a day about whether to keep going.
Time blocking, at its strongest
Every hour of tomorrow gets a job, decided in advance, written in a calendar — including lunch, including the gap you would otherwise donate to email.
Best day for it: contested. More obligations than hours, other people with access to your calendar, and a recurring sense that the important thing keeps losing to the loud thing.
` 08:40 Arrive. No email. 09:00 Report — section 2 only (blocked, declines meetings) 09:30 Standup 10:00 Report — section 2 continued 11:00 Client call 11:45 Spill block — unassigned 12:15 Lunch, away from the desk 13:15 Shallow: email, invoices, replies 14:00 Meeting 15:00 Read the paper you keep not reading 16:00 Spill block — unassigned 16:45 Lay out tomorrow. Ten minutes. `
Two things make that schedule survive contact with Tuesday.
Block about 60% of the day, not 100%. The two spill blocks are the difference between a plan and a fantasy. Everyone underestimates how long their own work takes; Daniel Kahneman and Amos Tversky named this the planning fallacy in 1979, and most demonstrations since have been students estimating their own coursework rather than adults estimating knowledge work — so treat it as a well-supported tendency, not a measured constant. Plan as though you were optimistic, because you were.
Name the block with the deliverable, not the project. “Report” becomes a four-hour smear. “Report — section 2 only” has an edge you can hit.
Where it costs you: ten to fifteen minutes of planning a day, and it is the only method here that fails publicly. A blown block sits in the calendar all week where you can see it.
Flowtime, at its strongest
No fixed interval. Pick one task, write down the time you started, work until you actually flag, write down the time you stopped, take a break scaled to the run, and log the interruptions.
Best day for it: one hard thing with a long load-in and an unpredictable length. Debugging. Drafting. Anything where minute 20 is when you become useful.
| Task | Start | Stop | Ran | Break | Interruptions |
|---|---|---|---|---|---|
| Fix the import bug | 09:12 | 10:34 | 82 min | 15 | 2 |
| Draft chapter 3 | 10:49 | 11:26 | 37 min | 10 | 0 |
| Draft chapter 3 | 11:36 | 12:31 | 55 min | lunch | 1 |
After a fortnight, that log tells you something no fixed interval could: your real unit. If the numbers cluster at 45 to 55, you were never a 25-minute person. The Flowtime write-up goes deeper on how to size the breaks.
Where it costs you: it applies no pressure to begin. Flowtime is for people whose problem is the bell, not the blank page. If your problem is the blank page, it will let you not start in a very organised way.
Side by side
| Pomodoro | Time blocking | Flowtime | |
|---|---|---|---|
| Fixes | starting | protecting | the bell |
| Unit | 25 min + 5 | the calendar hour | one task, variable |
| Decided | in the moment | the night before | while working |
| Best day | shapeless | contested | one hard thing |
| Overhead | none | 10–15 min daily | keeping a log |
| Breaks when | the work needs a run-up | the day is reactive | you cannot start |
| Other people | invisible to them | visible, defensible | invisible |
The decision rule
Open tomorrow and find the longest uninterrupted gap. That single number decides it.
Longest gap under 45 minutes. Do not start with a timer; you will only be measuring the wreckage. Block first — one defended 90-minute block, three days a week, beats any timer setting you could choose, and once the gap exists you run a timer inside it. Getting the gap at all is a scheduling problem before it is a discipline problem, which is where deep work for beginners starts.
A wide-open day, and nothing has started by 10:40. Pomodoro. Fifteen minutes, now. Blocking an empty day is a way of spending the morning planning the morning.
One hard thing and a deadline this week. Put the blocks in first and work backwards from Friday, then run Flowtime inside them. Blocking protects the hours; Flowtime stops you cutting a good run short at minute 25.
None of the above, and you want a default. Block the week on Sunday in ten minutes, run Pomodoro inside the blocks, and check in a fortnight whether your sessions keep wanting to be longer. If they do, move up — either to Flowtime or to the 90-minute ultradian block — and stop fighting it.
Can I use more than one in the same week?
They run at different resolutions, so yes, and the combination is the normal end state rather than a compromise. Blocking is a planning layer you touch once a day; the timer runs inside the block.
They only conflict if you make them agree to the minute, which nothing requires.
The pairing that needs care is a Flowtime run inside a tightly packed calendar. A variable-length method in a fixed-length container only works if the block behind it is empty — put the spill block there first, or do not run Flowtime that day.
Where each one breaks
Time blocking breaks on reactive days. Support rotas, on-call, small children, any job where other people set your next ten minutes. Every block becomes a lie by 10:30, and a calendar full of lies is worse than no calendar, because it converts an unavoidable day into a personal failure. If three days a week are reactive, block the other two and leave the reactive ones honest.
Pomodoro breaks on long load-in work. When it takes twenty minutes to get a problem into your head, a bell at 25 arrives roughly when you became useful. Lengthen the interval or change method. Do not keep the 25 out of loyalty.
Flowtime breaks when you will not keep the log. The record is the entire value. Two weeks of half-filled rows tell you nothing, and at that point you are working without a timer — allowed, but not a method.
And all three break identically when used to avoid the real problem. If the work is unclear, or you dread it, or it is the wrong job, no interval and no block and no log will touch that. A method is scaffolding around a decision you have already made.
Fomodo covers the timer half of this and carries all three shapes: Pomodoro, Flowtime under the name Deep Work, and two ultradian lengths for people whose real unit turned out to be ninety minutes. The limitation is the other half — it is not a calendar. It cannot see your 11:00 meeting, it will not defend a block, and it has no opinion about yesterday afternoon. The blocking happens in your calendar app and nothing here changes that.
What it does handle is the ugly case. The phone dies, iOS kills the app, you come back at 15:20 — it works out how much real time passed and puts the session back where it should be, which matters most at exactly the moment you would otherwise have written off the afternoon.
Find the longest gap in tomorrow. That number picks the method. You only have to run it.
