“How do we know when a process is worth documenting?”
I have a simple guideline. It's inspired by a quote from Ian Fleming's novel Goldfinger of James Bond lore. Fleming writes:
Once is happenstance. Twice is coincidence. The third time, it's enemy action.
If one person needs to do something, wait before documenting it. Twice even… sit tight. But if three or more people need to follow the same process right now or very soon (within the current quarter of work), that’s a good candidate to document. Documentation exists to solve common problems; less than three instances isn’t common enough to be worthwhile work to warrant the time it takes to write and maintain. Documentation is a tool for scale; spending time codifying one-offs (and two-offs) isn't working at scale, but three or more starts to get you in that territory.
If it’s good enough for 007, it’s good enough for me.
read this next
A full line doesn’t always mean the right line.
