Product Design for Self-Service Package Pickup Lockers at Amazon
Company: Amazon
Role: Product Manager
Category: Product Design & Strategy
Difficulty: medium
Interview Round: Other
Design a self-service Amazon locker: both the physical unit and the experience of everyone who uses it.
```hint Follow every handoff
List each party who touches a locker, such as the shopper, the driver who loads it and the site that hosts it, and ask what can go wrong at each handoff between them.
```
### Clarifying Questions
- Is this a new product designed from scratch, or a redesign of the lockers Amazon already operates?
- Should the locker handle only collecting orders, or also returns and sending packages?
- Which locations are in scope: dense city apartment buildings, stores and transit stations, or suburban and rural areas?
- Are we designing only the shopper's experience, or also the experience for delivery drivers and the sites that host lockers?
### What a Strong Answer Covers
- The goal for Amazon and for shoppers, including the failed, delayed or unsafe home deliveries a locker replaces.
- A chosen shopper segment and location type, with the pain points in their delivery journey.
- The end-to-end flow across checkout, driver drop-off and pickup, plus the key features of the physical unit.
- Failure handling: full lockers, oversized packages, lost codes, hardware or network faults, and uncollected packages.
- A primary metric with supporting metrics and guardrails for theft, wrong-package pickup and downtime.
### Follow-up Questions
- Lockers sit mostly empty in suburbs but are always full in city centers. What do you change?
- How would you protect high-value items, where a forwarded pickup code becomes a theft risk?
- Should Amazon open its lockers to packages from other carriers or retailers? What would you weigh?
Overview: A product design question from an Amazon product manager interview asking you to design a self-service pickup locker for Amazon orders. It tests choosing a target segment, designing the shopper, driver and host-site experiences, handling capacity and hardware failures, and defining success metrics with guardrails.
