Style Guide
We spent a quarter trying to get AI industry right, and the useful lessons were not the ones we expected.
The advice worth ignoring
What looks like a process problem is frequently an ownership problem. Where a design is obvious the prose is short, so the length of an explanation is a reasonable proxy for where to look next.
The first thing to establish is what you are actually optimising for. When responsibility is spread across a group, the work that falls between the named parts is the work that does not happen.
Documentation is a symptom: you write it where the design is unclear. If you learn on Friday what you assumed on Monday, the assumption never has time to become an architecture.
Begin with the obvious one
Nobody gets credit for the work that did not need doing. A small improvement applied consistently beats a dramatic one applied once, which is unsatisfying advice precisely because it is correct.
Consider the failure mode rather than the success case. Most disagreements that present as strategic turn out, on inspection, to be two people using one word for two things. We ran both approaches in parallel for six weeks. The difference was smaller than the cost of the debate about it.
Simplicity is not the absence of work. It is the result of it.
— Overheard in a retrospective
The habit that compounds
The second-order effects arrive about a quarter after the first-order ones. The stated constraint is usually a proxy for a real one nobody wants to say aloud, and optimising the proxy is wasted effort. The version of this that works fits on an index card. The version that fails needs an onboarding session.
The tooling question is downstream of the constraint question. A team that changes approach every quarter pays a coordination tax that routinely exceeds whatever the change was meant to fix. One team we spoke to cut their review stage entirely and found throughput unchanged, which told them something the metrics had not.
The one that only matters at scale
Most of the difficulty lives at the boundaries, not in the middle. The things that are easy to count are rarely the things that matter, and once a number reaches a dashboard it starts shaping behaviour whether or not it deserves to.
Feedback loops shorter than the planning cycle change everything. Handoffs between people who each hold a coherent local picture and no shared one produce most of the pain later attributed to tooling. In practice the answer showed up in the calendar before it showed up in the dashboard.
A shared definition of "done" removes more friction than any tool. They are decisions made quickly, defended slowly, and built upon for six months before anyone recalculates. The clearest signal was that people stopped asking where things were.
What to do first
Measurement is usually where this falls apart. The decision is usually cheap and reversible; the execution is where the cost lives, and that is where the argument should have happened. Set a date at which you will stop, and write down in advance what would make you stop earlier.
The checklist we ended up with:
- Prefer the reversible option when the evidence is thin
- Name one person accountable — not a group
- Agree on what "done" means, in writing, before starting
- Decide in advance what would make you stop
- Review the numbers monthly; change the targets rarely
The short version: decide what you are optimising for, write it down, and revisit it when the answer stops feeling obvious.