Amazon Questions
Tell me about a time you had to make a decision with incomplete information.
MediumBehavioralLeadership Principles2-3 minutes
Model Answer
Situation: At my previous company, we needed to decide whether to migrate our monolithic application to microservices while facing a critical scaling issue during peak traffic season.
Task: As the tech lead, I had to make the architecture decision within two weeks, but we didn't have complete performance data for all services and the team had limited microservices experience.
Action: I took a data-driven approach despite the gaps. First, I identified the three most critical bottleneck services using the data we did have. I ran targeted load tests on those services to project scaling behavior. I consulted with two engineers who had microservices experience at their previous companies. Instead of a full migration, I proposed a hybrid approach: extract only the three bottleneck services into microservices while keeping the rest monolithic.
Result: This decision reduced our p99 latency by 62% during peak traffic and avoided a 6-month full migration. The incremental approach also let the team build microservices skills gradually. Six months later, we had migrated two more services based on actual need rather than speculation.
Common Mistakes
- 1.Giving a hypothetical answer instead of a real example
- 2.Failing to explain what information was missing and why it mattered
- 3.Not quantifying the outcome or impact of the decision
- 4.Describing a team decision without highlighting your individual role
- 5.Forgetting to mention what you learned or would do differently
Follow-up Questions
- What was the biggest risk of your decision, and how did you mitigate it?
- If you had to make the same decision again with the same information, would you change anything?
- How did you get buy-in from stakeholders who wanted to wait for more data?