Define the North Star metric for Marketplace serving delivery partners
- Metrics
- Amazon
- Medium
- 10 min
Problem Statement Description
Product context: Amazon is a commerce, logistics, media, devices, and cloud company; its products include Marketplace, Prime, Prime Video, Alexa devices, ads, fulfillment, and AWS.
Amazon Marketplace depends on delivery partners to reliably move customer orders from fulfillment or seller handoff points to end customers. These partners may include individual drivers, delivery service providers, and logistics operators who accept routes, pick up packages, complete deliveries, handle exceptions, and manage earnings or operational capacity through Amazon systems.
You are asked to define a North Star metric for this marketplace experience from the delivery-partner side. The metric should capture whether the product is creating sustained value for delivery partners while supporting Amazon’s broader marketplace promise of fast, reliable, convenient delivery for customers and sellers.
This is a metrics interview question, so focus on how you would define the metric precisely, what behavior it should reflect, how it would be measured, and how it would guide product and operational decisions. Avoid jumping directly to a dashboard; first clarify the user, the core value exchange, and the business outcome the metric is meant to align.
The experience should consider:
- The delivery partner workflow, including onboarding, offer discovery, route acceptance, pickup, delivery completion, exception handling, and payout visibility.
- The exact metric definition, including numerator, denominator, time window, eligibility rules, and whether the unit is a partner, route, package, delivery block, or marketplace transaction.
- How to distinguish marketplace health from pure logistics output, especially where volume growth could mask partner dissatisfaction or poor economics.
- Instrumentation needed across partner apps, route assignment systems, delivery scans, support contacts, cancellations, earnings, and customer delivery outcomes.
- Cohorts such as new vs. tenured partners, geography, delivery model, route type, package density, peak vs. non-peak periods, and partner capacity level.
- Guardrail metrics for customer experience, delivery reliability, safety, partner earnings quality, churn, cancellations, support burden, and operational cost.
- How the metric would be used by product, operations, and marketplace teams to make prioritization, experiment, and scaling decisions.
- Risks of gaming, seasonality, supply-demand imbalance, and cases where optimizing the metric could harm customers, partners, or Amazon economics.
Your goal is to present a clear, decision-useful North Star metric framework for Amazon’s delivery-partner marketplace: one that reflects real partner value, connects to marketplace liquidity and customer reliability, can be instrumented credibly, and is paired with the right supporting and guardrail metrics.
What this question tests
- Metric Definition
- Instrumentation
- Counter-metrics
- Decision Quality
Practise this question under interview conditions. Answer it out loud against a timer with an AI interviewer that asks follow-ups, then review the scored report.
Related Metrics questions
- What metrics would you use to evaluate a new Alexa feature for familiesAmazon · Metrics · Hard
- How would you detect unhealthy growth in Prime VideoAmazon · Metrics · Hard
- What metrics would you use to evaluate a new Alexa feature for familiesAmazon · Metrics · Medium
- How would you detect unhealthy growth in Prime VideoAmazon · Metrics · Medium
- Design an experiment to measure whether Logistics improved outcomes for small merchantsAmazon · Metrics · Medium
- Create a metric dashboard for the executive team reviewing AWSAmazon · Metrics · Medium
All Metrics questions · Product manager interview questions by skill area