You can't randomize identity — how do you prove the graph lifts ads? — Microsoft Ads Identity Data Scientist Interview

You can't randomize identity — how do you prove the graph lifts ads?

Experiment designHard

Problem. Design a study to measure whether the identity graph causally improves advertiser conversion outcomes. You cannot randomly assign people to "be resolvable."

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.