Investigate Declining Successful Orders
Company: DoorDash
Role: Data Scientist
Category: Analytics & Experimentation
Difficulty: hard
Interview Round: Technical Screen
DoorDash observes that the number of **successful orders per day** in **Los Angeles** has declined over the last few weeks. Assume a successful order is one that is placed, paid for, accepted by the restaurant, assigned to a dasher, and delivered without cancellation.
As a Data Scientist, explain how you would investigate this decline.
Your answer should cover:
1. How you would clarify the metric definition, baseline, and key cuts of the data.
2. At least one plausible hypothesis from each side of the marketplace: **customer**, **dasher**, and **restaurant**.
3. If all three sides appear to contribute to the decline, how you would **quantify the impact** of each hypothesis on the overall metric drop.
4. How you would use a **funnel analysis** to diagnose the issue, including important stages, likely failure points, and segmentations.
5. One product or operational improvement you would propose, and how you would design an **A/B test** to evaluate it. Be explicit about the experiment unit, primary metric, guardrail metrics, and any marketplace interference concerns.
Quick Answer: This question evaluates a data scientist's ability to define and validate metrics, generate and prioritize marketplace hypotheses across customer, delivery partner, and restaurant sides, perform causal attribution and funnel analysis, and design controlled experiments for product or operational fixes.