Experimentation and causality

Interview question

seniorlevel
15 minsuggested time
2Data scientist, Data analyst

An airline launches a lounge-subscription marketing campaign in one city (Izmir) and cannot randomise users. There are 20 comparable cities that did not get it. Weekly bookings in Izmir rose 8% after launch, but all cities have seasonal swings. The marketing lead asks "did the campaign work, and by how much?" What design do you propose and how will you express confidence?

Tests: Designing a single-treated-unit causal analysis: synthetic control or comparable methods, pre-period fit, placebo inference, and honest limits on what one market can show.

day 5