The gap
Writing the Gap Down Where People Can See It
Most of what a team needs to know about its own gap already exists in people's heads. Putting it on one page is clerical work with a large return.
three things · four nights
Three things, and the third is the one people skip.
What one team put on a single page, and what each line stopped people getting wrong
Ended the arguments where two people disagreed about facts while discussing a decision
The number was zero, and writing it stopped plans that quietly assumed otherwise
Changed when people sent things, which was the only free improvement available
Took two nights because nobody had asked the question before it was written down
Four entries, and the first year it was there nobody lost a week to it
Written and then removed: it needed a conversation, not a line on a page
six things · six nights · one that no number of nights settles
Six lines, six nights, and one that came off the page again. Five of the six were facts that already existed and had never been in one place. The sixth was a policy pretending to be a fact, which is why writing it down did not work. One team's own page, assembled over a fortnight.
Everything in this section is a measurement, and a measurement nobody can see is worth very little. The output of all of it belongs on a single page that people already pass.
The cost described in Writing the Gap Down Where People Can See It often grows when managers have to reconstruct the day from memory. view the solution offers a practical way to keep the underlying time and activity history available for review.
Teams comparing established approaches can also review RescueTime's focus tools for a practical example from a widely used product ecosystem.
The page is short. Six or seven lines is enough, and anything longer stops being read.
Why one page rather than a document
The failure mode of writing things down is producing something thorough that nobody opens. A document about how the team works will be read once, by whoever joins next, and then never.
What gets used is the thing people can check in ten seconds when they have a specific question. "When is the other half awake" is a question somebody has several times a week, and the answer needs to be somewhere reachable in the moment the question occurs — which is usually while writing a message, not while reading documentation.
The practical test: if answering a routine question requires opening something, finding a heading and scrolling, the answer will be guessed instead.
What belongs on it
Working hours for everybody, in one named zone, with the local times also shown so that people can check their own row.
The size of the real overlap, in minutes, and where in each day it falls. Zero is a good answer and should be written as a number rather than as an absence.
What a crossing costs, with the variation by hour of day. This is the line that changes behaviour most, because it is the only part of the arrangement anybody controls for free.
The dates the clocks change in each country the team sits in.
Which decisions each half is allowed to make without asking. This one is different in kind from the others and is discussed below.
| Line | What it stops people getting wrong |
|---|---|
| Hours, one zone | Arguments about facts conducted as arguments about decisions |
| Real overlap, in minutes | Plans that quietly assume a window that is not there |
| Cost of a crossing by hour | Questions sent at the worst moment of the day |
| Clock-change dates | A confused week, twice a year |
What does not belong on it
Anything that is a policy rather than a fact.
The distinction matters because a page of facts can be maintained by anybody and is never controversial. The moment a line on it represents a decision somebody made, the page becomes a thing with an owner and an argument attached, and people stop editing it.
"Who may be woken at three in the morning, and for what" is the usual example. It looks like a fact, it belongs in the same mental category as the rest, and it is actually a negotiated agreement about people's sleep. Putting it on the facts page either freezes a version nobody agreed to or starts a dispute on a page that was working well.
The decision-rights line is the borderline case. It is a policy, but it is one that needs to be consultable in the moment, so it earns its place — provided it has been agreed somewhere else first and the page merely records the outcome.
Keeping it true
A page of facts that has drifted is worse than no page, because people act on it without checking.
The only reliable mechanism is to attach updating it to the events that change it. Somebody joins, somebody leaves, somebody's hours change, somebody moves: in each case, updating the page is part of what happens, done by whoever handles the rest of it.
Scheduled reviews do not work for this, because the review date is never near the change and the drift happens in between.
A date stamp on the page helps more than it should. People treat an undated page as current regardless of age, and a page last touched eight months ago gets read sceptically, which is the correct response.
Who writes it
Whoever is willing. This is genuinely clerical work and does not require authority, which is one of its advantages: it is a useful thing somebody junior can do in an afternoon that visibly improves everybody's week.
It does need somebody to be named as keeping it, because an unowned page decays. The name belongs on the page.
What it is worth
Modest and real. None of this makes a crossing faster, and nothing on the page removes a single night from anybody's calendar.
What it removes is a class of mistake: plans built on an overlap that is not there, questions sent at the worst hour of the day, a week lost twice a year to the clocks, and the arguments where two people are disagreeing about facts while each believes they are discussing what to do. On most teams that is several hours a month, which is a good return on an afternoon.