Who
The One Person Who Writes Everything
When one person produces most of a team's documentation, the arrangement works until it ends, and its ending is almost impossible to notice.
What nobody wrote down here, and what it cost
What the deploy order is
Known by one person who is in meetings
Asked rather than written, nine times
Why the import runs twice
Known by somebody who left in March
Nobody now knows, and it is asked at every incident
two things nobody wrote down · 14 occasions it cost somebody something
Two things, fourteen occasions, and one of them is now unknowable.
Teams testing a new knowledge workflow can learn more on the site to establish a baseline, then compare the result after ownership and review rules change.
A useful independent point of reference for this part of the process is Trustpilot.
Pages by one prolific author, and the weeks somebody opened them, around the time they left
The deploy runbook, written two years earlier
Opened steadily throughout, including after they left
The import page, written eighteen months earlier
Still the only description that exists
A page started six weeks before they left
Opened twice, both times by the person inheriting the work
Everything written by anybody else in the quarter after
Nothing was written, and nobody noticed for a further two months
four documents · 15 weeks in which somebody opened one of them · the longest wait between writing and a reading was 37 weeks · one never read at all
Four rows, fifteen weeks in which somebody opened one of them, and one row that is an absence rather than a document. The last line is the finding: the writing did not degrade after they left, it stopped, and the stopping was invisible because there had never been a count. One team's own two quarters.
A team with one prolific writer looks, from most angles, like a team with a documentation culture. Pages exist. Questions have answers. Newcomers find things.
What it actually has is one person, and the difference only becomes apparent afterwards.
Why it is hard to see
The output is the same. A team where eight people each write one page a quarter and a team where one person writes eight look identical from outside, and both look healthy.
Nothing in any ordinary review surfaces authorship. Nobody reads a wiki and thinks about who wrote it; the experience of using documentation is anonymous by nature.
And the person at the centre rarely says anything, partly because they enjoy it and partly because mentioning it sounds like a complaint about colleagues.
What ends it
Usually departure. Sometimes a change of team, or a promotion into work that leaves no slack, or simply a bad quarter.
Occasionally something quieter: the person gets tired of it. This version is the most common and the least visible, because there is no event. They simply write one page instead of four, and then one instead of two, and nobody can date the change.
Why the stopping is invisible
An absence of new pages produces no signal, for the same reason an absent document produces no signal.
Existing pages keep working, so nothing breaks immediately. The gaps accumulate in the area of whatever has changed since, which is a slowly growing set. By the time somebody notices that nothing has been written for eight months, the team has lost a year.
The only early detector is a count of who wrote what, taken periodically, which almost nobody does and which takes ten minutes.
What to do while it is still working
Three things, in order of how much they help.
Count, so that the situation is known to somebody other than the person in it. A table of authorship by quarter.
Name the dependency out loud, to the team. Not as blame — as a fact: most of what we rely on was written by one person, who was not asked.
And start distributing before anything goes wrong, by the methods in the previous pages: attaching writing to events that already distribute themselves, and asking individual people by name for individual pages.
What not to do
Do not ask the prolific writer to write less. This is occasionally proposed as a way of forcing others to step up, and it produces exactly one outcome, which is less documentation.
Do not make them the owner of documentation as a function. It is the obvious move, it looks like recognition, and it converts a voluntary activity into a role whose main duty is chasing colleagues — which is the fastest way to make somebody stop enjoying writing.
Do not assume their subjects are the right subjects. What one person chose to write about reflects what irritated them, which is correlated with what matters and is not the same thing.
If they have already left
The first useful act is an inventory of what they wrote, which is usually larger than anybody expected and is the only map of the territory.
The second is to find out what they did not write, by the methods in the first section: counting what gets asked, watching what goes wrong, asking whoever inherits the work what they cannot find.
The third is to accept that some of it is gone. A proportion of what that person knew exists nowhere, and the honest response is to record what is now unknown rather than to pretend it can be recovered.
The ten-minute diagnostic
Before anything else, count. Most tools record authorship and nobody looks at it, and the table takes ten minutes to produce.
- List the pages written or substantially edited in the last quarter.
- Put a name beside each.
- Count by name.
- Show the result to the team.
The table is usually a surprise to everybody, including the person at the top of it, who generally has no idea they wrote most of it. It is also the only argument that reliably works, because it is very hard to describe documentation as a shared value while looking at a quarter in which one person wrote four-fifths of it and four people wrote none.
The thing worth saying to them
If they are still there: thank you, and I have looked at the numbers, and it was mostly you.
Almost nobody says this. The panel at the top of this page covers two quarters in which one person's work was opened fifteen times and the author heard about none of it, which is the ordinary condition of being the person who writes.