Nothing at home in an hour,
everything in the library in thirty minutes

Same task, same you, same day. At your own desk you sit down and an hour later you are still on the first line. You go to the library, speak to nobody, are supervised by nobody — and the work moves.

This has a name: body doubling, working in the presence of another person. It has been a practice in the ADHD community for years, and most people discover it before they ever hear the term. This page covers why it works, how much of that is actually known, and what separates a room that helps from one that does not.

The person beside you is not checking on you. They cannot see your screen and they will not ask whether you finished. Something changes anyway, and the most plausible account has three parts:

A start that comes from outside you. Alone, deciding when to begin is also your job, and that decision can be postponed indefinitely. When a shared timer starts for everyone at once, the decision stops existing.

Mild accountability. Nobody is asking, but somebody is there. That is too little to feel like pressure and enough to hold your attention.

A bounded stretch of time. The room has an end. "I'll work today" is vague; "these 25 minutes" is not.

To be straight about it: the evidence that body doubling works is largely lived experience, and the controlled literature on it is thin. Being widely used does not make it proven. Whether it helps you is something a week of trying will answer better than we can — you do not have to take our word for it.

You can do this with a two-person video call and no app at all. Without these four, it usually falls apart:

1. One timer, for everyone. The room should start on a shared clock, and it should be obvious when the break comes.

2. Chat between sessions, not during. No messages while the timer runs. What you want to say can wait 25 minutes; if it cannot, it was not the work.

3. Keep it small. A room of thirty feels busy but nobody knows anybody. Three to eight is usually the sweet spot.

4. Know what you are working on before you arrive. Entering a room to "do some stuff" lands you exactly where sitting alone did. Break the goal into steps first, then join.

The fourth is the one people skip. Working alongside someone solves the starting problem; it does not solve the not knowing what to do problem. Those are two different problems, and the second one is settled before you enter the room.

You enter a room with a six-digit code. A friend sends you the code and you are in; there is no public directory, no discover tab, no matching with strangers. The room belongs to whoever has the code.

There is no camera in the room. That is not a setting — the feature has no video at all. You see the others as one of three states: Focusing, Warming up, Idle. A status rather than a face, which is the part of body doubling that does the work anyway.

The timer is shared: whoever opened the room starts it, everyone starts at the same moment, 25 minutes by default. Alongside it there is short chat and a high five — when someone completes their goal, the room sees it. A room holds up to 32 people, though remember the third rule above.

Joining is free. Opening a room requires PRO — we are not hiding that, because joining a friend's room on the free plan and using it for months is a perfectly normal way to use this.

Unembellished, in the order it actually happens:

If your connection drops, the app reconnects on its own and catches up on what it missed; you are not thrown out of the room. Going through a tunnel does not end your session.

Nothing. Presence and chat inside a room are processed in memory only, for as long as the room lives. We keep no permanent chat history; when the room ends, the room's data goes with it.

The room itself is not permanent either: a few minutes after the last person leaves, it closes and its code stops working. You cannot come back tomorrow to the same room with the same code — you open a new one.

This paragraph says the same thing as our privacy policy. A marketing page that sounds more optimistic than the policy is the worst thing to trust.