Diagnose Delivery Delays: Key Metrics and Experiment Design
Overview
This question evaluates a data scientist's diagnostic analytics and experiment-design competencies, including root-cause analysis, metric and segment specification, and causal inference using event-level timestamps, geolocation, ETA predictions, and operational supply-demand data.
DoorDash
Aug 4, 2025, 10:55 AM
Data Scientist
Technical Screen
Analytics & Experimentation
0
0
Diagnosing Delivery Delays and Validating a Fix
Context
You are the PM for a consumer delivery marketplace that has seen an uptick in delivery delays. You have access to event-level timestamps (order placed, merchant accepts, courier assigned, courier arrives at merchant, food ready, pickup, drop-off), geolocation, ETA predictions, and operational data on courier supply and demand.
Task
- Diagnose the root causes of delivery delays.
- Specify the key metrics and segments you would analyze.
- Design an experiment to validate one proposed solution to reduce delays.
Guidance
- Consider a stage-level funnel from order to delivery.
- Examine courier supply vs. demand imbalance, restaurant prep times, geography/time-of-day, and batching/dispatch logic.
- Propose an A/B test using control vs. treatment regions.
Solution
Login required
Show