DoorDash — Consumer & Growth — A/B Test & Causal Inference Questions

Role context: Data Scientist, Consumer & Growth Analytics · Est. study time: 45 min · 4 questions

How experimentation works here

Consumer growth at DoorDash runs on DoorDash's Curie / Dash-AB experimentation platform, but the hard parts are causal:

  • Many features A/B cleanly, but a change to the OEC must be orders and retention, not one funnel metric, with unit-economics and marketplace guardrails.
  • Marketing channels and promos can't always be user-randomized, so incrementality comes from geo holdouts / diff-in-diff and holdouts.
  • Effects are heterogeneous (promos help lapsed users, waste on regulars), so target with conditional effects.
  • Order metrics are noisy, so variance reduction (CUPED) matters for power.

For the fundamentals — p-values, power, error types, distributions — see the Probability & Statistics section.

Each answer is a coaching walkthrough: a Sample answer (clarify → approach → a simulated back-and-forth → a clear call), then a Deep dive with illustrative example, then a Grading rubric.

Questions (4)

Design an A/B test for a checkout change — what's the OEC?Experiment designMediumYou can't A/B a marketing channel — measure its incrementalityCausal inferenceHardWho should get the promo? Estimating heterogeneous effectsHeterogeneous effectsHardYour order-rate experiment is underpowered — how do you fix it?Variance & ratio metricsMedium