People Search API Benchmark
When you ask a data API for “{role} in {country}”, how many of the returned people actually hold that role, at that seniority, in that country, and how deep and fast is the result? Each provider answers the same 10 frozen role + geo queries, and every person returned is judged blind.
Not measured yet. The first run of this sub-report will follow the method below.
Methodology
- 01Test set
A frozen, versioned set of 10 queries each fixes a role (with accepted title variants), a seniority bar, and a country. Each provider is asked for up to 25 matching people via its native people-search API, with the query expressed as faithfully as each API allows.
- 02Grading
Every returned person is judged blind, the judge seeing only current title, company and location and never the provider, on two checks: the current title matches the requested function AND seniority, and the location is in the requested country. Duplicate people across providers are judged once. Field fill-rates are deterministic.
What we measure
Questions about the People Search API Benchmark
What does the People Search API Benchmark measure?
It grades providers on 5 metrics: precision, precision@10, depth (avg/25), title fill and latency p50. It has not been run yet, so no provider is ranked on it.
How will the People Search API Benchmark be measured?
Every provider will answer the same frozen, verified test set (10 frozen queries · K=25 · set v2026-08-28.1) in the same window, graded by blind LLM judge. Each value will get a letter grade against the rest of the field, A for the top fifth down to D for the bottom fifth.