Instacart partners with a local grocery store to introduce a “smart cart” in the physical store.
The smart cart UI lets shoppers:
-
search/browse items that exist in the current store and see in-store prices, and
-
also see items and prices from other nearby stores available on the Instacart app.
You are a Senior Data Scientist asked to evaluate whether this is a good product idea.
-
What are the key hypotheses (primary and secondary) for how this smart cart could impact the business? Consider potential positive and negative effects.
-
What metrics would you track? Specify at least:
-
a primary success metric,
-
2–4 supporting diagnostic metrics,
-
1–3 guardrail metrics.
Make sure you address tradeoffs like cross-store switching/cannibalization and user experience.
-
How would you design an experiment to measure causal impact? Include:
-
unit of randomization (user, trip, cart, store, day, etc.),
-
how you would handle interference/spillovers (e.g., shoppers seeing others use the cart),
-
required logging/instrumentation,
-
rough power/MDE considerations and how you’d estimate them.
-
If you cannot run a clean randomized experiment, propose a credible quasi-experimental alternative (e.g., diff-in-diff) and key assumptions to validate.