What you share
The handbook nobody has read
A written team charter is a row on the pyramid with a width of zero until somebody has used it. What makes a shared document real, and what a dead one costs.
The people this page is about, and which of them have ever worked together
- 01The team's managerEvery working pair in this grid includes this person
- 02The designer
- 03The data analyst
- 04The product managerWrites the brief everybody reads
four people · 6 possible pairs · 3 of them have worked on something together · 1 has only been in the same meeting · 2 have never done either
Four of the eight, and only the pairs that run between those four.
For a distributed group, teams can review this solution to review where work travels, where it stalls, and whether responsibility is concentrated in one person.
Teams can compare this approach with the practical guidance available from Strategy+business.
Somebody once spent a fortnight writing the team's working agreement. How decisions get made. What the review standards are. What the team values. Where things live. It exists, it is linked from the channel description, and it has a table of contents.
Asked to put it on the pyramid of what the team shares, you would write the row and then have to decide its width, and the honest answer is zero. Nobody has opened it since the month it was written. It is not a thing held in common; it is a thing that exists.
Why a zero row is worth refusing
The sign used on this site declines to draw a row with a width of zero, and the reason is specific: a bar of no width looks like a bar that went missing rather than like a nought. On a pyramid sorted by reach it would sit at the bottom as a hairline and read as a rendering fault.
The same thing happens in conversation. A document that nobody uses does not register as an absence. It registers as a thing that has been dealt with, which is worse than nothing, because the slot marked "shared understanding of how we work" is now occupied and nothing else will be put in it.
What makes a document a real row
Three tests, and a document needs all three.
- Somebody has consulted it to settle something. Not read it on arrival — consulted it, during a disagreement or a decision, and been answered. One such occasion is worth more than fifty first-day readings.
- It has been changed by somebody other than its author. A document with one contributor is one person's opinion that has been formatted. A document three people have edited has had three people's disagreement resolved into it, and that resolution is the shared thing.
- It gets cited by people who were not there when it was written. This is the strongest signal and the slowest to arrive. When a member who joined after it was written says "the handbook says we do it this way", the document has become a thing the team holds rather than a thing it has.
Why handbooks die
Almost always for one of three reasons, and none of them is that the writing was bad.
It was written before the questions. A charter composed at the start of a team describes what people imagined would be contentious, which is reliably not what turns out to be contentious. Four months later the real arguments are about something it does not mention, and the document is irrelevant rather than wrong.
It was written by one person. Usually the manager, usually in a week when they had time, usually well. It records one person's view of how the team works, and the other seven did not have to agree with any of it, so none of them are bound by it in their own minds.
It has no occasion. Nothing in the normal week causes anybody to open it. A document without a moment that requires it decays at the speed of memory.
Making one that has width
The approach that survives is backwards from the usual one. Do not write the document and then look for uses. Wait for the second time an argument recurs, and write down only that.
A real working agreement grows a paragraph at a time, each paragraph added on the day somebody needed it, each one written by whoever had the argument. After six months it is four pages of settled questions, every one of which somebody cared about, and it is cited because every line in it was produced by a moment that will recur.
This is slower and it produces a worse-looking document. It also produces the one thing a handbook is for, which is that two people who have not worked together can find out how the other will expect things to be done without having to ask.
What to do with the dead one
Do not delete it, and do not rewrite it either.
Deleting makes the problem invisible again. Rewriting produces a second document with the same fate, usually written by the same person, and the team now has two.
The cheapest useful move is to cut it down to the three or four paragraphs that somebody has actually used, delete the rest, and put the date at the top. A one-page document with a recent date gets read; a twelve-page one with no date does not, and the eight pages that were removed were the ones describing how the team imagined it would work before it had tried.
The row it eventually becomes
A living document is one of the few rows on the pyramid that can be true of all eight and still produce pairs, which makes it unusual. The manager row and the all-hands row are wide and inert. A document that people edit creates occasions: somebody changes a paragraph, somebody else disagrees, and the two of them settle it.
That is a small pair, formed around a sentence. It counts, by the bar used here, if the output changed — and on a document that is being argued over, it usually did.