A time-varying marketplace change — why reach for a switchback?

Experiment designHard

Problem. You want to test a change to how Airbnb allocates search demand in a market (a matching/ranking policy) where treating some users while others get the old policy would badly interfere. Why might a switchback design help, and what are its pitfalls?

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.

← Randomize the guest or the listing? The choice changes your answerA promo lifts bookings in treated markets — created or just shifted? →