I desperately needed a company wiki.
Not a perfect knowledge-management strategy. Not a six-week comparison of tools. One place where we could start documenting processes in a form people could search.
Reonomy was adding people and operating habits quickly. We had put engineering work on a physical board because the company needed to see what was in progress. We were running grooming sessions and all-hands meetings to keep priorities from living inside a few conversations. The same problem existed with everything that was not a story card.
Email was easy to write and terrible to remember. Chat was fast and even easier to lose. A verbal explanation worked until the next person joined, asked the same question, and got a slightly different answer.
The company needed somewhere boring and obvious for the answer to live.
Searchable mattered more than elegant. A process nobody can find might as well not have been written. The first version did not need a beautiful hierarchy or a committee deciding the names of every section. It needed people to type what they knew and other people to find it later.
I also wanted the coding assessment setup handled. It was another small part of the same growth problem. Hiring more engineers meant turning a set of repeated actions into an actual process instead of reconstructing it for every candidate.
There was a danger in making the wiki itself into a product project. We could have debated permissions, integrations, structure, and every tool anybody had ever liked. That would give us another undocumented decision process and still no place to put the documentation.
My note was only three lines. I desperately need a company wiki. Let's pick one today. Also set up the coding assessment tool.
That was enough scope. Choose the shelf, start putting things on it, and let the organization emerge from the work we actually needed to remember.