The hours
Counting your own fortnight
Ten days, four letters, about three minutes of effort in total — and the four questions the result will answer that nothing else will.
Days this part is about, hour by hour
Wednesday, the second week
3 of 7Planning day, the densest of the quarter
Thursday, the third week
0 of 7Seven hours of solo work at a hot desk
Wednesday, the fourth week
3 of 7The new starter's first day in the building
three days · 21 hours logged · six of them were only possible there · one of the three days contained no such hour at all · two were made worse by being there
Three of the twelve days drawn in full on this site.
For teams moving between home and office, project time tracking for flexible work offers a practical way to compare time, attendance, and work patterns without treating presence as the result itself.
The broader workplace question can also be compared with resources from AI Now Institute.
Everything on this site comes from one person's record. The obvious objection is that one record is one record, and the obvious answer is that you can make another one in a fortnight.
What to write down
For each working hour, one letter, and four words so you can recognise it later. Nothing else — no ratings, no categories, no reflection.
| Letter | Means | Typically |
|---|---|---|
n |
Could only have happened here | A wall, a hardware thing, somebody who sat down beside you |
b |
Possible anywhere, better here | The conversation that went faster in a room |
s |
Identical anywhere | Solo work, which is most of it |
w |
Made worse by being here | A call from a corridor, focus work in an open room |
Include the days at home as well. Half the value of the exercise is the
comparison, and a record of office days alone cannot tell you whether
the w hours would have been s hours elsewhere.
Write down the journey too: the time you left and the time you got back, both ways. Not the scheduled journey — the actual one.
The four questions it answers
How many hours in the quarter actually needed the building? Most people guess high by a factor of three or four, and the guess is sincere, because the memorable hours are overwhelmingly the ones that did.
Which day do they fall on? They will cluster. If they cluster on a day you are not routinely in, you have found something immediately actionable and you have the sentence to go with it.
What is the ratio of travel to the hours that needed the room? On the record here it was two hours and twenty-six minutes of travel for each hour that required the building. Whatever yours is, it will be a number, and a number is a different kind of thing from an opinion in a conversation about this.
How large is the fourth column? This is the one that surprises people. If a fifth of your office hours are worse than the same hours elsewhere, that is worth knowing before you next argue about days.
What it is not evidence of
It is not evidence about anybody else. A manager with a different job
will have a different distribution, and a person four months into a
role will have far more n hours than somebody four years in, for good
reasons.
It is also not evidence that the policy should change. It is evidence about what your own quarter contained, which is the thing you are currently guessing at, and which you need before you can say anything useful about what should happen instead.
When to stop
Two weeks. Not longer.
A fortnight is enough to see the shape, and the shape is the whole output. Running it for a quarter produces a slightly better number and a much worse chance of finishing, and an unfinished record answers none of the four questions.
What to do in the second week
Nothing different. The temptation after five days is to start adjusting — choosing better days, arranging things, making the record look better — and it should be resisted for one more week.
The reason is that you are trying to measure the arrangement as it normally runs. A fortnight of deliberately well-chosen days measures what is possible, which is a different and less useful question than what is happening.
Adjust afterwards. The record is the baseline, and you will want it later when somebody asks whether the change made a difference.
The four numbers, written out
At the end, five minutes of arithmetic:
| Hours logged in the building | |
Of which n |
|
| Travel, door to door, median | |
| Travel, total for the fortnight | |
n hours by weekday |
The last row is the one to fill in carefully, because it is the one that leads to a request somebody can say yes to.
What people find
Three results recur often enough to be worth warning about in advance, so that none of them is a shock.
The total is lower than expected. Usually by a factor of three or four. This is not a sign that your job is unusual; it is the normal gap between memory and record on this subject.
The travel is longer than you would have said. By fifteen to thirty per cent, consistently, because the walks and the buffer and the ordinary delay are not in anybody's stated number.
One weekday carries most of the value. Almost always. Which one varies by team and is rarely the one you would have guessed.