How would you prioritize and test features?

In an eBay Product Manager interview, explain how you would prioritize and manage a backlog, execute a product initiative from idea to launch, measure impact, stay customer-focused, and use experiments or A/B tests to make decisions.

Use an eBay marketplace example involving buyers, sellers, design, engineering, and analytics.

Constraints & Assumptions

• Consider both sides of the marketplace: seller activation and buyer trust.
• Do not optimize a single metric if it harms marketplace quality.
• Use a structured prioritization framework, but explain judgment beyond the score.
• Include experimentation, launch readiness, and guardrail metrics.

Clarifying Questions to Ask

• What business goal is the team optimizing this quarter?
• Which user segment is in scope: first-time sellers, power sellers, buyers, or marketplace operations?
• What constraints exist around engineering capacity, trust and safety, legal, or analytics instrumentation?
• Is experimentation feasible, or do marketplace/network effects require a phased rollout?

Part 1 - Frame The Product Goal

How would you define the objective and target users?

What This Part Should Cover

• A clear objective, such as improving first-time seller listing completion.
• Primary and secondary stakeholders.
• Success criteria and guardrails from the start.

Part 2 - Prioritize The Backlog

How would you prioritize possible features or improvements?

What This Part Should Cover

• Inputs from funnel data, research, support tickets, marketplace quality signals, and stakeholder requests.
• A prioritization method such as RICE or weighted scoring.
• Explicit tradeoffs when a strategically important item does not have the highest score.

Part 3 - Execute From Idea To Launch

How would you manage design, engineering, analytics, and stakeholder execution?

What This Part Should Cover

• MVP versus follow-on scope, owners, milestones, dependencies, and risk log.
• Instrumentation and QA readiness.
• Blocker handling and communication cadence.

Part 4 - Measure And Experiment

How would you measure impact and use experiments or A/B tests?

What This Part Should Cover

• Primary metrics, secondary metrics, and guardrails.
• A/B test or phased rollout design.
• Segmentation by device, seller type, category, geography, or experience level.
• Decision rules for ship, iterate, or roll back.