You have tried to organise your calendar. You have read both terms — time blocking and timeboxing — in articles that use one for the other, and started doing the same, because nobody explained the difference.
And even with the calendar organised, something does not add up. Either the slot arrives and you do not start. Or you start, sink in, and at 2pm you are still on the thing that was meant to take forty minutes.
Two different problems — and each term solves one.
One answers "when", the other answers "until when"
Time blocking gives the task a place on the calendar: 9 to 11, the report. It is allocation: it decides what occupies each stretch of the day. You know this one.
Timeboxing is a different operation. You do not reserve a slot for the task; you set how much it may consume, at most. When the time is up, it is up — even if the task is not.
On the calendar the difference is invisible. "9–11, report" is a block if, at 11, you carry on until it is done. It is a box if, at 11, the report is what ends, in whatever state it is in. A block is a place. A box is a ceiling.
The idea came from engineering and barely left
Timeboxing was born in 1990s software development, inside DSDM. In the method's handbook, now maintained by the Agile Business Consortium, the rule is explicit: the duration of a timebox is fixed at kick-off, before detailed planning exists. First you decide how much time; then what fits.
And when it does not fit? Requirements are sorted into what the delivery must have, should have and could have — and as the deadline tightens, the "could haves" go first. Time does not stretch; scope does. Scrum and XP inherited the rule as fixed-length sprints.
Now look at what you do with your own tasks: you fix the scope — "I am going to finish this" — and let the time vary. The inversion engineering made thirty years ago is the one nobody makes in their own week.
Are you the type who never starts, or the type who never stops?
If you are the type who looks at the important task, knows what to do and simply does not begin — pushing it to the afternoon, and from the afternoon to tomorrow — your problem is entry. You need the block: without a set time, the task is never now.
If you are the other type — you start, sink in, and look up having spent the afternoon on a half-hour task that came out flawless and delayed five others — the block does nothing for you. It opens a door, and getting in was never the problem. Getting out is. You need the box.
Almost every article about calendars talks to the first type. That is why the advice never fitted you.
The ceiling works because estimation never will
Estimating time is a skill nobody has. In the classic study on this, students predicted they would take 33.9 days to finish their thesis and took 55.5. Asked for the most pessimistic scenario — "if everything went as poorly as it possibly could" — the average was 48.6 days. Still below reality.
Even your catastrophic estimate is optimistic, and no technique fixes that. A box does not improve the forecast — it caps the damage when the forecast is wrong.
There is a second effect. Sophie Leroy showed that when we switch tasks we carry attention residue: part of your head stays on task A while your body is on task B. The useful finding came next — under time pressure to finish A, disengagement gets easier and people perform better on what comes after.
How to put a ceiling on a task
The ceiling you never close
Worth saying what does not exist: there is no controlled study showing that timeboxing, as a personal technique, raises productivity. The most cited piece is Marc Zao-Sanders' article in the Harvard Business Review, which puts timeboxing top of a list of 100 productivity habits and reports that his own productivity "at least doubled". A personal account and an opinion ranking, not an experiment.
And the practical limit: the box depends on somebody closing it, and that somebody is you — at the minute you are deepest in and least willing to stop.
Why this matters for what I built
I spent years thinking my problem was not starting, when it was not stopping. No block fixed that: a block opens a door, and what I lacked was someone knocking on it to say the time was up.
That is the piece I wanted to put into ASB: an assistant that holds the ceiling you agreed to, shows up when it arrives and asks if it can close.