Microsoft — Ads Identity — Product Case Questions

Role context: Senior Applied Scientist, Ads Data Management Platform · Est. study time: 45 min · 5 questions

How these map to the role

These are identity-infrastructure cases: an ambiguous "is the graph healthy / did it help" question you turn into precision/coverage metrics and a clean measurement, then a recommendation. The recurring instinct: anchor on precision at the needed coverage (not match rate), reason about the asymmetric cost of false merges vs false splits, prove impact through downstream advertiser outcomes when you can't randomize identity, and treat privacy/consent as a hard constraint.

Each answer is a coaching walkthrough: a Sample answer (clarify → approach → a simulated back-and-forth → a clear call), then a Deep dive with illustrative example, then a Grading rubric.

Questions (5)

Cross-device coverage dropped 10% — what happened?Diagnose a problemMediumAdvertisers say frequency caps are being blown — why?Diagnose a problemHardShip a higher-recall matching model that adds false merges?Launch or notHardProve the identity graph actually improves ad measurementMeasure impactHardExtend identity resolution to Copilot — define and size successMeasure successHard