Asking the Same Thing Twice

Why it happens

Not carelessness. An answer arrives in the middle of something else, you understand it, you use it, and three weeks later you need it again and it is gone.

Spoken answers are worse than written ones, and answers in a call are worst of all. An answer given on a Tuesday call is almost always lost by the following month unless somebody wrote it down, and the person who gave it has no idea it is gone.

What to do with an answer

Three things, in ascending order of value.

Write it in your own notes. Thirty seconds, immediately, in your own words. Your own words matter: an answer you have restated is one you have understood, and the restating catches the cases where you have not.

Write it where the next person will look. The handbook, the repository, the channel. This takes two minutes and is the single most valuable thing a new person does in their first quarter, for reasons covered below.

Tell the person who answered that you have written it down. One line, and it converts a favour into a contribution.

Thanks — I have put that in the runbook so the next person does not have to ask.

Why writing it down is your best available contribution

Because you are the only person who can see that it was missing.

Everybody else in the organisation knows the answer. They cannot tell that it is undocumented, because for them the knowledge is simply present. You can tell, because you just spent an afternoon failing to find it.

That visibility lasts about eight weeks. Afterwards you also will not be able to see it.

What it looks like in practice

Four lines, in the obvious place, with no ceremony.

Staging database access. The password rotates on Mondays and is in the shared vault under "staging-db". You also need to be in the staging group — ask whoever is on call to add you.

Written up after asking in October.

The last line is worth including. It dates the answer, which is the thing that makes a document trustworthy a year later, and it is one of the few pieces of record-keeping nobody objects to.

If you ask twice anyway

It happens and it is survivable. Say so.

I think I asked you this a few weeks ago and I have lost the answer — sorry. Where does the on-call rota live? I will write it down properly this time.

That is far better than pretending not to have asked, which the other person will usually remember. And "I will write it down this time" converts the second ask into something that will not become a third.

The asymmetry worth remembering

Of the six answers in the panel above, four were written down and two were not. Both of the lost ones had to be obtained again, and one cost more the second time, because the person who had answered first had moved on and the knowledge had to be reconstructed by two people instead of recalled by one.

Thirty seconds of writing, against an hour of reconstruction. That is the whole argument, and it holds for every answer you will be given this quarter.

Where to write it down

Three places, and the third is the one that counts.

Your own notes. Always, immediately, in your own words. Takes fifteen seconds and catches the cases where you did not quite understand.

The thread or channel where you asked. A one-line summary of the answer, posted under your own question. Costs nothing and makes the answer findable by search, which is how the next person will look.

The place the next person would look first. The handbook, the repository, the runbook. This takes two minutes and is the only one of the three that reaches somebody who does not know to search.

Most people do the first and nothing else. The second is nearly free and worth making a habit; the third is worth doing for anything that cost you more than an hour to find out.

The answers worth writing up properly

Not all of them, and the distinction is simple.

Anything that cost you more than an hour to get is worth four lines in a findable place. Anything that took four minutes goes in your own notes and nowhere else.

That threshold keeps the habit sustainable. A new person who writes up every answer they receive spends their first month documenting rather than working, and produces a quantity of material nobody will read.

Six or eight properly written entries over a quarter, each covering something that genuinely cost somebody an afternoon, is a real contribution. Forty short entries about things anybody could have worked out is a mess, and it makes the six useful ones harder to find.