The week
The two days everybody picked
Twenty-three of thirty-four days on two weekdays, and nobody chose them. How a free choice produces a fixed pattern, and what it costs at both ends of the week.
Days this part is about, hour by hour
Monday, the third week
0 of 7Came in to make up a missed day
Tuesday, the fourth week
2 of 7Thursday, the first week
0 of 7Nobody from the team was in
three days · 21 hours logged · two of them were only possible there · two of the three days contained no such hour at all · three were made worse by being there
Two days from the quieter end of the week, and one from the full end.
A fair pilot starts with a shared baseline; teams can explore the platform before comparing what changes after a scheduling or process decision.
Teams can compare this approach with current guidance available from Owl Labs.
The whole quarter, by the day of the week it landed on
34 days in the building across five weekdays · 23 of them on just two of the days, Wednesday and Tuesday · the hours that needed the building fell hardest on Wednesday, seventeen of them · on Monday and Friday nothing at all required being there
Thirteen weeks under a three-day policy that named no days.
Nobody decided that this team would be in on Tuesdays and Wednesdays. There was no meeting about it, no message, and if you had asked in week one everybody would have said the days were up to them.
By week four the pattern was fixed, and by week eight it was binding — not by rule, but because the days had acquired consequences. The meetings had drifted onto them. The one-to-ones were there. Coming in on a Thursday had become slightly odd and noticeably pointless.
How it happens without anybody doing it
Four steps, none of which require coordination:
- A small initial asymmetry. Mondays and Fridays feel like edges. Slightly fewer people pick them in week one, for unrelated reasons.
- Observation. Everybody notices who was in when they were in, because that is the only information they get.
- Rational adjustment. Next week each person shifts a day towards where the people were. Independently, correctly, in their own interest.
- Lock-in. After three or four rounds the fullest days are full enough that the meetings move onto them, and the pattern is now structural rather than preferential.
Nothing in that is a mistake. Each step is the right move given what the person knew. The outcome is a schedule nobody chose and nobody can unilaterally change.
What it costs at the full end
Twelve Wednesdays and eleven Tuesdays in thirteen weeks. That is close to everybody, on the same two days, which is good for the first kind of hour and bad for everything else about the building.
- Rooms ran out by half past nine on both days. Nine rooms were booked across the quarter and six were obtained.
- Twenty-four hours of the quarter were calls taken without a room, and almost all of them were on the two full days.
- The desks on the home floor were gone by nine, which is why nine days were worked somewhere else in the building.
So the days with the most overlap are also the days with the worst conditions for everything that is not overlap. The two effects are produced by the same cause.
What it costs at the thin end
Monday twice, Friday once, in thirteen weeks. Zero hours between them that needed the building.
Anybody whose life makes Tuesday and Wednesday hard — a standing commitment, a school run, a shared car, a hospital appointment that recurs — does not get a slightly worse version of the office. They get a structurally empty one, at the same cost in travel, and they are fully compliant with the policy while getting none of what the policy is for.
That is the part that is hard to raise, because from the outside it looks identical to everybody else's arrangement. Three days a week, same as you.
The fix is embarrassingly small
Name the day. One sentence from whoever sets the week, and the coordination problem disappears, because the whole problem is that nine people are trying to guess each other's answer to a question nobody has been asked.
The three weeks it took
| Week | Mon | Tue | Wed | Thu | Fri |
|---|---|---|---|---|---|
| 1 | 1 | 2 | 3 | 3 | 1 |
| 2 | 1 | 3 | 4 | 2 | 0 |
| 3 | 0 | 4 | 6 | 2 | 0 |
| 4 | 0 | 6 | 7 | 1 | 0 |
| 8 | 0 | 7 | 8 | 1 | 0 |
People in the building from a team of nine, by week. The pattern is visible by week three and complete by week four. Nothing was announced, nothing was agreed, and nobody at any point made a decision about the shape of the week.
Why it cannot be undone from inside
Each person, in week eight, faces this: four of the five recurring meetings are now on Tuesday or Wednesday, most colleagues are there, and the thin days have been demonstrated to produce nothing.
Moving to a Thursday is therefore worse for that person, individually, even though a week in which half the team was on Thursdays would be better for everybody than the current one.
That is a coordination trap in the ordinary sense: every individual move away from the equilibrium is a loss for the individual making it, so nobody makes it, and the equilibrium holds however bad it is for the group.
The only thing that breaks it is somebody with the standing to say "from next month, this group is in on these days" — which is a sentence and costs nothing, and which no part of a day-count policy ever prompts anybody to say.