Identify Key Drivers of Delivery Decline in Los Angeles
Company: DoorDash
Role: Data Scientist
Category: Analytics & Experimentation
Difficulty: hard
Interview Round: Technical Screen
##### Scenario
DoorDash sees a 10% drop in the number of completed deliveries in Los Angeles week-over-week.
##### Question
How would you identify, attribute, and quantify the key drivers behind the 10% delivery decline in LA?
1. List the potential root causes and propose testable hypotheses across the supply, demand, merchant, product, external, and measurement dimensions.
2. Outline an analytical plan to test each hypothesis — the data you would examine, the segmentation/cuts you would apply, and the method you would use.
3. Quantify and attribute the −10% to its contributing drivers (e.g., via funnel/marketplace decomposition).
4. If controlled experiments (A/B tests) are infeasible, describe the alternative approaches you would use to establish causal inference and validate your findings.
##### Hints
Think funnel analysis, supply vs. demand decomposition, segmentation by time/location/cohort, seasonality, competitor impact, and synthetic-control or time-series causal methods (DiD, BSTS/CausalImpact, event study, IV).
Quick Answer: A DoorDash data scientist technical-screen case: diagnose a 10% week-over-week drop in completed Los Angeles deliveries. It tests funnel and supply/demand decomposition, hypothesis generation across demand, supply, merchant, product, external, and measurement causes, and causal attribution via quasi-experimental methods (DiD, synthetic control, BSTS, event study, IV) when A/B testing is infeasible.