Biker Dasher - Case Study: Analyze Bike Courier Efficiency Metrics

Question

## Round 1 - Data Science Case Study ## Problem DoorDash is piloting a bike courier program ("Biker Dashers") in a dense urban market. You have one week of delivery data. Evaluate whether the program is performing well and recommend whether to expand, hold, or cut the program. **Data Available:** - `deliveries(delivery_id, dasher_id, dasher_type, pickup_time, delivery_time, distance_km, order_value, tip)` - `dashers(dasher_id, dasher_type, city, start_date)` **Your analysis should cover:** 1. **Efficiency metrics:** Average delivery time by dasher type (bike vs. car). Is bike faster for short distances? 2. **Coverage gap:** What fraction of deliveries are beyond typical bike range (e.g. > 3 km)? 3. **Economics:** Compare earnings-per-hour for bike vs. car dashers including tips. Are bikers fairly compensated? 4. **Customer impact:** Is there a significant difference in delivery time variance? Do customers of bikers rate orders differently? ``` Sample insight: Bike Dashers: avg 18 min, car Dashers: avg 24 min for distances < 2km Bike coverage: 68% of all deliveries in pilot zone are < 2km -> Recommendation: expand in high-density zones ``` ## Follow-ups 1. What confounding variables might make bike dashers look artificially faster? 2. How would you run an A/B test to cleanly isolate the impact of bike vs. car assignment? 3. How would weather data change your analysis? 4. What guardrail metrics would you monitor to catch problems early in an expansion?

Biker Dasher - Case Study: Analyze Bike Courier Efficiency Metrics

Question Details

Full Details

Topics

More from DoorDash