##### Scenario A food-delivery platform plans to let couriers (dashers) sign up to deliver by bicycle instead of cars. ##### Question Why do we want dashers to sign up for bikes? State the primary business goal and supporting hypotheses. Which metrics would you track to evaluate the bike program? Include success, guardrail, and driver metrics. Before full rollout, how would you design the experiment? Detail randomization unit choice, pre-launch checks, and analysis plan. ##### Hints Frame hypotheses, define KPIs (e.g., delivery time), choose randomization (dasher vs market), plan power, logging, and guardrails.

Evaluates marketplace analytics for launching bicycle delivery in dense urban areas. Strong answers define business goals, success and guardrail metrics, mechanism metrics, and an experiment design that accounts for courier behavior, marketplace interference, safety, weather, and unit economics.

How do I approach Analytics & Experimentation interview questions?

Analytics & Experimentation questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master analytics & experimentation interviews.

What difficulty level is this interview question?

This is a medium difficulty Analytics & Experimentation question, commonly asked during Technical Screen rounds at DoorDash.

What role is this question designed for?

This question is commonly asked for Data Scientist candidates at DoorDash during technical interviews.

Evaluate Impact of Bicycle Deliveries on Efficiency and Costs

A food-delivery marketplace plans to let couriers sign up to deliver by bicycle in addition to cars. The company wants to evaluate whether bike delivery improves marketplace efficiency and unit economics in dense urban areas.

Constraints & Assumptions

Assume bicycle delivery is most relevant for short-distance, dense, urban trips with parking or traffic constraints.
The program may affect couriers, customers, merchants, marketplace matching, safety, and cost.
A clean experiment may require geo or time clustering because supply and demand interact.
Include success metrics, guardrails, mechanism metrics, and an experiment design.

Clarifying Questions to Ask

Are bikes, e-bikes, and scooters all included, or only pedal bikes?
Which markets and order types are eligible?
Can couriers switch between car and bike modes?
Are there safety, weather, legal, or equipment constraints?

Part 1 - State Goals and Hypotheses

Why do we want couriers to sign up for bikes? State the primary business goal and supporting hypotheses.

What This Part Should Cover

Primary goal such as faster and cheaper delivery for bike-suitable orders in dense areas.
Hypotheses around travel time, parking time, delivery reliability, courier utilization, fulfillment rate, contribution margin, and sustainability.
Recognition that bike delivery may only help specific segments.

Part 2 - Define Metrics

Which metrics would you use to evaluate the program?

What This Part Should Cover

Primary metrics such as delivery time, on-time rate, cost per delivery, contribution margin, fulfillment rate, and courier active-hour productivity.
Secondary metrics for customer experience, merchant readiness, courier earnings, order assignment, and peak-hour reliability.
Guardrails for safety incidents, cancellation, lateness tails, courier churn, customer complaints, food quality, and regulatory issues.
Mechanism metrics such as parking time, distance, route density, batching, idle time, and mode eligibility.

Part 3 - Design the Experiment

Before full rollout, design an experiment to evaluate impact.

What This Part Should Cover

Randomization unit choice, such as market, geo/time cluster, courier, or order, and the trade-offs for interference.
Pre-launch checks for logging, eligibility, safety configuration, power, treatment exposure, and marketplace balance.
Analysis approach, non-compliance handling, variance reduction, heterogeneity by distance/weather/market, and stopping rules.

What a Strong Answer Covers

A strong answer treats bike delivery as a marketplace intervention, defines segment-specific metrics, designs around interference, and balances speed and cost gains against safety and customer experience guardrails.

Follow-up Questions

How would you handle couriers who switch modes during the experiment?
What if bike deliveries are faster but couriers earn less per hour?
How would weather affect your experiment design?