Why we picked a 15-minute focus timer, not 25.

The decision, in one paragraph.

Koda ships with the focus timer set to 15 minutes per chunk, followed by a real break. Not 25 (Pomodoro), not 45 (a school period), not 5 (a token), not "until the worksheet is done." 15 is the number the common developmental-attention rule of thumb supports for a typical K-5 kid on a non-preferred task, the number the ADHD picture supports from a different angle, and the number we kept landing on while watching kids work. (Caveat up front: the rule of thumb below is widely cited in teacher-training and pediatric occupational-therapy materials but isn't a peer-reviewed law of nature — treat the supporting numbers as design-doc evidence, not clinical claims.) Parents can change the chunk length — the slider runs from 5 to 25 minutes per child — but the default is 15, and we want to explain why.

Where Pomodoro's 25 comes from.

The Pomodoro Technique was developed in the late 1980s by Francesco Cirillo, a university student trying to study without losing the next hour to his own brain. He used a tomato-shaped kitchen timer — pomodoro in Italian — and settled on 25 minutes as the work interval. That's all the source material confirms; there's no developmental-psychology basis for the number. It happens to be a useful chunk for an adult doing focused study or knowledge work, and the technique survives because that pairing is real for grown-ups. None of that machinery transfers cleanly to a 4th-grader. Picking 25 because Pomodoro picked 25 would be cargo-culting an adult productivity convention onto a kid's brain.

The rule of thumb that pointed us at 15.

Sustained attention — the cognitive subsystem that lets you keep doing one effortful thing in the face of distraction — develops slowly through childhood and into the late teens. The heuristic that recurs in teacher-training materials and pediatric occupational-therapy guidance, usually framed as "typical sustained focus on a task," is roughly 2-5 minutes per year of age, with wide individual variation. The high end of that range assumes the task is engaging — a preferred activity, a game, a hobby. For non-preferred tasks like a math worksheet, the same materials typically land closer to the bottom of the range: somewhere around 2-3 minutes per year of age. For a 6-year-old that's roughly 12-18 minutes; for a 10-year-old, 20-30. Homework after school, after six hours of school-day attention regulation, sits at the low end of that low end.

15 minutes lands inside that low-end window for most of the K-5 band. A 1st-grader can usually manage 15 if the task is sized right; a 5th-grader could go longer but usually doesn't need to. We picked the chunk that holds across the band rather than tuning per grade — the per-kid slider does the per-grade work, and a uniform default makes the parent portal less of a puzzle.

The ADHD angle, which is the angle that broke the tie.

For many ADHD kids, 25 isn't a slightly-too-long version of the right answer. It's a qualitatively wrong shape. Attentional lapses tend to compound as time on task grows, and the gap between "still engaged" and "no longer engaged" is shorter than for a neurotypical peer. In our testing, we frequently observe a child who started a 25-minute chunk well spend the back half of it re-entering rather than doing math — and the experience that lands isn't "I worked," it's "I failed to focus." That experience compounds.

We wrote at more length about math homework with ADHD; the short version: 15 is short enough that the kid finishes more chunks, ends each one before re-entry gets hard, and gets a real break before the next one. Three 15s with breaks tends to produce more actual math than one 45 — because the alternative to a focused 15-minute chunk usually isn't a focused 45-minute one; it's a 45-minute chunk where the kid disengages partway through.

Why not 5 or 10 minutes.

Going the other direction has a different failure mode. A 5-minute chunk is too short to get into anything — the kid spends most of it settling, and the timer goes off the moment they're focused. 10 minutes is workable for very young kids or kids in a rough patch, but the kid leaves most chunks without finishing a problem, which produces the sense that math is a thing you start and stop. 15 is the smallest number we tested where most kids could finish a small unit of work — two or three problems, or one harder problem with the visualization step — inside a single chunk. Finishing matters. The feeling at the end should be "I did a thing," not "I was interrupted."

The break is half the design.

A 15-minute chunk only works if the break after it is real. Koda's default break is 3 minutes, with a soft cue that the next chunk is ready when the kid is. The break screen has no math on it, no XP, no avatar unlocks. It just sits there. We've watched too many "break" screens that are actually a different kind of engagement — a mini-game, a leaderboard, a re-engagement loop — and what they do is move the cognitive load sideways instead of unloading it. For many kids, especially when re-entry is already hard, a mini-game in the break window is more likely to extend the transition than to reset attention. The break is for the kid to get up, drink water, fidget. The point of the timer isn't to extract maximum minutes per session; it's to keep the kid's relationship with math from getting ground down. This isn't a speed timer either — Koda doesn't grade how many problems fit into 15 minutes, and the chunk ending isn't a failure state, a point we go into more in the math-anxiety note.

The dials we leave open.

Defaults are the strongest design choice a piece of software makes, but they aren't the only one. We let parents change three things from the portal:

• Chunk length, per child, from 5 to 25 minutes. A kid who can comfortably do 20 should do 20; a kid having a rough week can drop to 10 without it feeling like a regression. Per-child, not per-household — a 1st-grader and a 5th-grader will have different numbers.
• Break length, from 2 to 10 minutes. Some kids need a longer reset; some lose re-entry if the break stretches.
• The transition warning. Koda surfaces a 2-minute and 30-second warning before any chunk ends. You can turn either off (some kids find the warning itself stressful) or extend them — we leave both on by default, and we'd recommend keeping them on for ADHD kids specifically. (Spoken cues — "two more problems and then a break" — are on the roadmap; visual warnings ship today.)

What we don't let you change: there's no "no breaks" mode. Even the longest chunk surfaces a break at the end. We know the temptation — the homework pile is high, the evening's short, the kid is finally focused and you don't want to break the spell. We've watched what happens on the other side of skipping breaks enough times to be confident this is one place to take the choice off the parent's plate. Shortening the chunk is the right move when the night is short; skipping the break is a cost that lands on the kid in the next session, not this one. So the slider goes down to 5 minutes, but the break always comes.

What this looks like at the kitchen table.

A typical evening: your child taps their avatar to start. The timer reads 15:00. They work on the worksheet; Koda watches the paper, says small things when the work warrants it, says nothing the rest of the time (we wrote about when Koda interrupts separately). With 2 minutes left, a soft visual warning appears. When the timer hits zero, the break screen appears. The kid stands up, comes back in 3 minutes. The next chunk starts when they tap to begin. On a good evening, that's two chunks. On a hard one, one. The metric isn't the number of chunks — it's whether the kid finished each one feeling like they did a thing.

What would change our minds.

A few things would push us off 15:

• Field data from Koda families showing that a different default produces more on-task time across the K-5 band without increasing the rate of unfinished chunks. We'll know this within the first six months of shipping.
• A well-powered modern study of sustained attention during structured non-preferred tasks at K-5 that lands the typical-attention window somewhere other than where we put it. The literature here is older than we'd like.
• Strong per-grade variation in what works. If 1st-graders cluster at 10 and 5th-graders at 20, we'd switch to a per-grade default with the per-child override on top.

What 15 minutes is for.

Picking 15 wasn't an obvious call. The Pomodoro default is in everyone's head; school periods are 45 minutes; many homework apps don't have a timer at all. We picked the number the developmental-attention rule of thumb, the ADHD picture, and our own observations kept pointing at, then we let parents move it. The default is the part you didn't have to fight for; the override is for the kid sitting at your table.

If you'd like to know when Koda ships, the waitlist is here. The related notes on why we don't use streaks and how Koda decides when to interrupt cover the rest of the quiet-by-default design.