A marketplace feature can't be user-randomized — switchback design

Experiment designHard

Problem. You want to test a marketplace change (e.g., a pricing or matching tweak). Why is a rider-level A/B biased, and how would you design it instead?

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.

Build an MMM you can trust — calibrate it with experiments →