Design Experiments to Measure Promotion Scheduling Impact
Company: DoorDash
Role: Data Scientist
Category: Analytics & Experimentation
Difficulty: medium
Interview Round: Onsite
##### Scenario
Platform is releasing flexible promotion scheduling (time-of-day deals, merchant-funded discounts, broader eligibility).
##### Question
What business goals and success metrics should be set for this feature? How would you design and monitor an experiment to assess its impact? How would you use pre-period data when interpreting results, and would you ramp 80/20 or 50/50? Why?
##### Hints
Define primary KPI (incremental GMV, margin, merchant adoption); apply CUPED or diff-in-diff; weigh risk vs. speed when choosing ramp-up.
Quick Answer: This question evaluates a data scientist's competency in experimental design, causal inference, metric definition, interpretation of pre-period data, and selection of ramping strategies for a promotion scheduling feature.