Method
One property per trade. Each one answers a working question, in the order the work happens.
The information problem in most trades is not scarcity. What you need to know is public. It's split across a hundred sources, written for a hundred different readers, and none of it says what comes first.
So we go vertical. One trade at a time. We read everything, throw out most of it, and publish what's left in the order the work happens.
Four filters. A real trade with real budgets. Tooling fragmented enough that no default exists. Search demand that compounds instead of spiking. And the scatter has to be expensive — assembling it yourself costs months, a bad hire, or a wrong contract.
Pass all four and the vertical gets a domain and a deadline. Pass three and it goes on the watchlist. Cheap scatter is not a wedge. If a beginner can reach competence in a weekend of reading, we don't build it.
A list of everything decides nothing. A summary answers your question once and disappears. Neither one is a place you come back to on Tuesday.
Every list here ranks. Every page names a default. And every property is built as records you can filter, compare and track, not prose you read once.
The method
Rung 1
Every trade has a real job under the title. The knowledge to do it is public and it's everywhere — a forum thread from 2019, a vendor blog with a sales pitch attached, one good podcast episode. We read all of it and publish what survives.
Rung 2
Directories optimize for coverage. Coverage doesn't run a firm. We cut the list, name a default, and say what to buy first and what to skip until month six.
Rung 3
The last rung is software. Not a description of the work — the thing that produces it. The model. The memo. The resume.