A countdown big enough for the whole room to see

A visible timer changes a room. Announcing that an exercise has ten minutes achieves very little, because everyone's sense of ten minutes is different and the confident talkers will use most of it. Putting the ten minutes on the wall moves the pressure off the facilitator and onto something nobody can argue with, and groups self correct against it without being asked.

That is why timeboxing works in a workshop and why it usually fails without a clock. The point of a box is not efficiency, it is that a group with unlimited time will spend it on whatever is most comfortable to discuss rather than on what the session is for. A timer is the cheapest structural intervention available, and it does not require anybody to be the person interrupting.

Drift is worth caring about for anything that runs in sequence. A naive timer loses a fraction of a second per tick, which nobody notices in ten minutes and everybody notices across a day of stacked sessions when the afternoon has quietly moved. Timing against the clock rather than counting ticks avoids it, and this one does.

The pomodoro side is the same idea pointed at one person. Twenty five minutes of work and five of rest is arbitrary, and the arbitrariness is the useful part: a fixed boundary you did not negotiate is easier to respect than one you set by feel. The breaks matter as much as the work, and skipping them is the most common way people conclude the technique does not suit them.

The session log is the part that turns this from a stopwatch into something more useful. Exporting how long things actually took, rather than how long they were planned to take, is how a workshop agenda gets better the second time it is run. Almost every session has one exercise that always overruns, and it is rarely the one anyone expects.

It runs entirely in your browser, works with no connection once loaded, and stores nothing, which is what you want from something being projected in front of a client.