Why is a user-level A/B biased in a marketplace, and how do you fix it?
Experiment designHardProblem. You test a feature that boosts demand (say, a discount shown to some guests). You randomize guests 50/50 and measure bookings. Why is the estimate biased, and how do you design it correctly?
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.