Sometimes you do not need a full build with a spec and a gate. You need to know who is out there, quickly, to see whether an idea is worth taking further.
Exploration. You have an idea for a segment and you want to know whether there are two hundred people in it or twenty thousand before you invest in a spec, a qualification loop and a copy angle. This answers that in one run.
It is not the skill to build a campaign from. The full build exists because a list that is going to receive real messages needs a written spec and a sample gate, and this has neither by design.
The two search routes have surprisingly little overlap. Running only the primary source and calling it the market consistently undercounts, particularly outside the largest English-speaking regions. The union is closer to the truth, and the overlap between them is a useful confidence signal on the rows where both agree.
This returns people, not addresses. Finding and verifying an address is a separate skill with its own cost profile, and joining the two would hide that cost inside what looks like a search.
Skills compound. These are the ones we usually install alongside it.