Improve biker delivery with metrics and levers

## Case: Optimize delivery performance for bike couriers You are a Data Scientist at a food-delivery marketplace (e.g., DoorDash/Uber Eats). Your team focuses on **bike couriers** in dense cities. Leadership asks: **“What should we optimize for, and how would you do it?”** ### Context Bike delivery performance is affected by many external factors: - **Geography & built environment:** hills, bridges, plazas vs high-rise malls, downtown vs suburbs, parking/locking availability. - **Weather & traffic conditions.** - **Merchant attributes:** cuisine type (hot/cold, wet/dry), packaging, order prep speed/variance, how orders are handed off to couriers. - **Customer demand:** urgency, distance to merchant, order size/quality sensitivity. - **Marketplace interactions:** merchant–courier and courier–customer **supply/demand matching**. ### Task 1. Propose a **clear objective** for “better” biker delivery (what are you trying to improve?). 2. Define a **metrics framework** including: - Primary (north star) metric(s) - Diagnostic metrics (to explain movement) - Guardrail metrics (to prevent harm) Consider perspectives of **couriers, customers, and merchants**. 3. List **key data/features** you would need (and any data quality risks). 4. Suggest **actionable levers/ideas** (product, ops, routing/matching, merchant experience) you would test or roll out. 5. Outline an **experimentation/causal plan** to validate impact given strong confounding from weather, geography, time of day, and demand fluctuations. ### Output Provide the metric definitions (with formulas where helpful), and a concrete plan of analysis/experiments you would run to decide what to ship.