Prove the identity graph actually improves ad measurement

Measure impactHard

Problem. Leadership asks: does the identity graph actually improve advertiser outcomes, or is it expensive infrastructure we assume works? You can't randomly assign people to "be resolvable." How do you measure its impact?

Before you reveal: say your answer out loud, as if you were in the real interview — get your reasoning across clearly first. There is no single correct answer: reading what the interviewer is really after and defending your own thinking is what makes an answer strong.

← Ship a higher-recall matching model that adds false merges?Extend identity resolution to Copilot — define and size success →