What is the Pomodoro technique?
The Pomodoro technique is a way to plan a day, protect a stretch of attention, and take a break on purpose. Francesco Cirillo started it in the late 1980s with a tomato-shaped kitchen timer. A traditional Pomodoro is 25 minutes of work and about 5 minutes of rest, with a longer break after four rounds. It is a focus timer you can run with a clock, a phone, or an agent that starts the timer for you. The 25-minute length is the familiar form, not the whole method.
What is the Pomodoro technique?
The Pomodoro technique is a complete system for dealing with time, not only a 25-minute countdown. Francesco Cirillo created it in the 1980s while he was a student who could not focus, using a tomato-shaped kitchen timer. He has written that the iconic 25-minute sessions are one expression of a broader way of working with time.
Cirillo’s own site describes the method as more than a timer: a timer, a pencil, and paper are enough to plan work, see how much effort a task takes, and recover from interruptions.[1] On his personal site he writes that in the 1980s he created the Pomodoro Technique to understand his relationship with time, and that it grew into a method for planning, handling interruptions, and learning.[3] The book page puts the same origin in the late 1980s, when he was a university student experimenting with that tomato-shaped kitchen timer.[4]
The page about Cirillo adds a detail that matters if you only know the timer. In the 1990s he worked as a software-development mentor. Clients were struggling to stay focused, he suggested the method, one team tried it, and other teams followed.[2] He also says the 25-minute sessions are not the technique itself. If you treat a pomodoro as a magic length, you miss the planning and the record of what you finished.
A useful way to hold the idea: you choose a small piece of work, you protect it from other tasks for one interval, you stop when the timer ends, and you write down that you did it. The next section is the sequence Cirillo printed in his 2006 paper.
How do you run one Pomodoro?
Pick one activity, set a timer for 25 minutes, and work only on that activity until it rings. Mark the round, rest for 3 to 5 minutes, and start again. Every four Pomodoros, take a longer break of 15 to 30 minutes. If the interval is interrupted for good, void it and start a fresh one.
Cirillo’s 2006 paper, revised in 2007, is the operational write-up. A traditional Pomodoro is 30 minutes: 25 minutes of work plus a 5-minute break. At the start of the day you choose tasks from an Activity Inventory and put them on a To Do Today sheet. You need a kitchen timer, that daily sheet, and the inventory of work that shows up later.[5]
A Pomodoro can’t be interrupted; it marks 25 minutes of pure work.
[5]
The paper treats the interval as indivisible. There is no half Pomodoro. If someone or something breaks it, you mark that Pomodoro void, as if you had never set it, and you start another. When the timer rings you mark an X beside the activity and you stop, even if a few more minutes would finish the sentence. The short break is 3 to 5 minutes. He writes that those minutes let you disconnect so the last 25 minutes can settle, and that you should not keep working “just for a few more minutes.”[5]
- Choose today’s tasks and write them in priority order.
- Set the timer for 25 minutes and work on the first task only.
- When it rings, mark an X and step away for 3 to 5 minutes.
- Repeat. After four marks, take 15 to 30 minutes.
- If you are pulled away, void that Pomodoro and start a new one.
How long is a Pomodoro?
In Cirillo’s paper the work interval is 25 minutes and the following break is about 5 minutes, so one cycle is half an hour. The longer break after four cycles is 15 to 30 minutes, depending on how hard the set was. Those numbers are the traditional form. They are a rule of the method, not a measurement of the ideal human attention span.
In the paper, Cirillo works the length out from two needs: a unit of effort you can count, and a span your attention can hold. He calls roughly 20 to 35 minutes ideal, and one Pomodoro became one countable unit. Estimating a task as “two Pomodoros” is the point of the record. If you stretch the interval whenever the work feels close to done, the X’s stop meaning the same thing and the day’s plan gets slippery.[5]
A pomodoro timer can be a physical timer, a phone, or software. The object does not matter. What matters is that it rings and that you obey the ring. ntimer includes the traditional lengths as a preset so you do not have to rebuild the arithmetic each morning.
ntimer’s cycle presets include Pomodoro 25/5, with a 15-minute long break every 4 rounds, plus 52/17 and Ultradian 90/20. Flowtime sets the break to 1/5 of the focus time you just spent, and Third Time sets it to 1/3. You can run several timers at once, one per work: 2 on the Free plan and up to 10 on Pro.
What does research say about short breaks?
A 2022 meta-analysis of micro-breaks, breaks of no more than 10 minutes, found small gains in vigor and small drops in fatigue. The overall effect on performance was not statistically significant. Effects on performance showed up for less demanding cognitive tasks, and longer breaks were associated with a larger performance boost. That is evidence about short breaks, not a trial of the Pomodoro technique.
Albulescu and colleagues pooled 19 records and 22 samples, 2,335 people in all. Micro-breaks raised vigor (d = 0.36) and reduced fatigue (d = 0.35), both with p < .001. Performance landed at d = 0.16, p = .17, and the confidence interval crossed zero. Subgroup checks found performance effects only when the task was less cognitive. A meta-regression found that a longer break went with a greater performance boost.[6]
A 5-minute Pomodoro break sits inside their definition of a micro-break. The studies were not a test of Cirillo’s rules, the indivisible interval, or the daily sheet. It is fair to say that a short pause can leave you less tired. It is not fair to say the literature has proved that 25 minutes is the best length for programming, writing, or study. For hard cognitive work, this review did not find a clear performance benefit.
What about 52/17, 90 minutes, and Flowtime?
52/17, a 90-minute block, and Flowtime are other schedules people use when 25 minutes feels wrong. DeskTime’s 52/17 figure is an observation of its own users, and later years did not repeat the same ratio. A 90-minute ultradian claim is weaker than popular articles imply. Flowtime, described in 2016, drops the work timer and lets you stop when you need to.
| Schedule | Work | Break | What it actually is |
|---|---|---|---|
| Pomodoro | 25 min | 3–5 min, then 15–30 after four | Cirillo’s traditional unit of work[5] |
| 52/17 | 52 min | 17 min | DeskTime’s 2014 snapshot of its most productive users[7] |
| 90-minute block | 90 min | 20 min in ntimer | A preset you can try; the biology is unsettled[8] |
| Flowtime | Until you stop | You choose | A 2016 method with suggested, not required, breaks[9] |
DeskTime, which sells time-tracking software, wrote that in 2014 the top 10% of users it treated as most productive worked about 52 minutes and then broke for 17 during one workday. The same write-up says a 2021 repeat came out at 112 minutes of work and 26 of break, and that February 2020, just before pandemic lockdowns, was 80 and 17. The ratio moved. It is not a controlled experiment and it is not a law of attention.[7] Use 52/17 when your tasks are long enough that 25 minutes is mostly restart cost, and check whether you are actually finishing.
The 90-minute idea is often pinned on Nathaniel Kleitman’s basic rest-activity cycle. A review of biological clocks notes that Kleitman proposed that cycle in the 1960s, then records later negative findings, including a 1995 study that found no 1.5-hour rhythm in cognitive performance. The same review says some authors think the evidence for the cycle may be an artifact of how studies were run and calculated.[8] ntimer’s Ultradian preset is 90 minutes of focus and 20 of break. That is a schedule you can run. It is not a claim that your brain must work in 90-minute chapters.
Flowtime, posted by Urgent Pigeon on 17 August 2016, has one hard rule: unitasking. You do not set a timer on the work. You stop when you need a break. Suggested rests, which the post says are not rules, are about 5 minutes after less than 25 minutes of work, 8 after roughly 25 to 50, 10 after roughly 50 to 90, and 15 after more than that.[9] ntimer’s Flowtime preset is simpler and different: the break is one fifth of the focus time you just logged. If you want the 2016 table, set the break yourself. If you want a proportional break without a chart, use the preset.
How do developers use a Pomodoro?
Developers use a Pomodoro to fence one bug, one review, or one slice of a feature so chat, mail, and the next idea do not eat the interval. Cirillo was already using the method with software teams in the 1990s. The useful part for code is the interruption rule: a broken interval does not count, so you notice how often the day is actually split.
The mentor years are Cirillo’s own account, not a lab study of programming speed. He writes that clients were struggling to stay focused, that one team tried the method, and that other teams followed.[2] The paper’s rule still fits a coding session. If a production alert voids the Pomodoro, you start another one after the alert, and the void is information: this morning was interruptions, not feature work.[5] Nobody has a clean public trial that says “Pomodoro raises pull-request throughput by N percent,” and you should not invent one.
A practical split: use 25 minutes for a review, a repro, or a test you can finish inside one ring. Use a longer preset when you are inside a design you should not context-switch out of. Pair the timer with a short map of the feature when the work has several branches. A mind map is one way to see those branches before you start the clock.
How do you pair a Pomodoro with an AI agent?
You pick the work, the agent starts the timer, and a heartbeat every few minutes tells ntimer the session is still the thing you are doing. The timer stays open while heartbeats arrive. After a stretch of silence it closes at the last heartbeat, so an abandoned run does not bill the rest of the afternoon. You still decide what counts as the task.
Connect Claude Code once, then ask it to start a Pomodoro on a named work. The remote server is https://api.ntimer.app/mcp. The command name you add is ntime.
claude mcp add --transport http -s user ntime https://api.ntimer.app/mcp
claude mcp login ntimeTools the agent can call include ntime_start, ntime_pause, ntime_resume, ntime_stop, ntime_status, and ntime_heartbeat. Heartbeats are meant to arrive every few minutes. If they stop, the open timer closes at the last heartbeat once idleMinutes of silence have passed. The default is 30 minutes. ntime_focus is there when you want the agent to look at the current focus rather than guess. The setup steps and the rest of the tool list are on the MCP page.
This does not make the agent the author of your plan. You still void an interval if you were not actually working, the same way Cirillo voids an interrupted Pomodoro. The agent is a pair of hands on the timer and the log.
How do you track Pomodoros as billable time?
Count finished intervals as a record of attention, then translate the minutes into hours on a rate. One Pomodoro is 25 minutes of focus, not one billable hour. Four traditional work intervals are about 100 minutes before the long break. Log the time you actually spent, attach a rate, and let the month become an invoice only after you have checked the entries.
The X’s in Cirillo’s sheet are a tally of protected intervals.[5] Clients usually buy hours, so you convert. If you rounded every Pomodoro up to an hour you would overbill. If you dropped the breaks and also dropped the voids, you would hide the interruptions that ate the day. Keep the work minutes and the reason.
Sessions land on the device first and sync later, so a timer still records when you are offline. The worklog stores the time against a work. Rate profiles turn those hours into money: the Free plan includes 3 profiles. Pro turns a month into an invoice in EUR or USD and can send it by mail. Web, iPhone, and macOS all read the same log once it has synced.
If you want the agent to file the session, ntime_log writes a finished block and ntime_report reads the period back. Editing a mistaken entry is ntime_edit_session. None of that changes the rate you agreed with the client. When you are ready to keep the log, create an account and run one real afternoon before you trust a month of totals.
Questions
Do you need a tomato-shaped timer?
No. Cirillo used a tomato-shaped kitchen timer, and the Italian word for tomato is pomodoro. Any timer that rings at 25 minutes works. The method is the plan, the single task, the break, and the mark on the sheet.
What should you do when a Pomodoro is interrupted?
Void it and start a new 25-minute interval. Cirillo’s rule is that a Pomodoro cannot be split or partly saved. A quick question that truly pulls you off the task makes that round not count. The void is useful: it shows how broken the day was.
Is 25 minutes better than 52?
Neither length has a trial that crowns a winner for every job. 25 minutes is the traditional Pomodoro. 52/17 is one year’s pattern among DeskTime users, and later DeskTime snapshots were different. Pick a length, keep it stable for a week, and judge it by finished work.
Can you run two focus timers at once?
You can, if they belong to different works. ntimer allows one timer per work, with 2 running timers on Free and up to 10 on Pro. Two timers on the same task only split your attention. The Pomodoro rule is still one activity until the ring.
Does one Pomodoro equal a billable hour?
No. A traditional Pomodoro is 25 minutes of work. Bill the minutes you will stand behind, usually the focus time and not the break, at the rate on that work. A tally of rounds is a check on the timesheet, not a substitute for it.
Sources
- Pomodoro Technique - Time Management Method, Francesco Cirillo. accessed Nov 21, 2026. Back
- Francesco Cirillo - Pomodoro Technique, Francesco Cirillo. accessed Nov 21, 2026. Back
- Francesco Cirillo, Francesco Cirillo. accessed Nov 21, 2026. Back
- Pomodoro Book - Pomodoro Technique, Francesco Cirillo. accessed Nov 21, 2026. Back
- The Pomodoro Technique, Francesco Cirillo, 2006. accessed Nov 21, 2026. Back
- Give me a break! A systematic review and meta-analysis on the efficacy of micro-breaks, PLOS ONE, 2022. accessed Nov 21, 2026. Back
- Does the 52-17 rule really hold up?, DeskTime, 2024. accessed Nov 21, 2026. Back
- Biological clocks and the practice of psychiatry, Dialogues in Clinical Neuroscience, 2007. accessed Nov 21, 2026. Back
- The Flowtime Technique, Medium, 2016. accessed Nov 21, 2026. Back