Behavioral question
Tell me about a time you had to choose between shipping a novel mobile capability and fixing a core reliability or retention issue with incomplete data. How did you frame the tradeoff, what evidence did you use, and what was the outcome?
- Perplexity
- Behavioral
- Medium
Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.
Start a mock interview on this question · Mock interview from a job description
What this question tests
A behavioral question testing tradeoff reasoning between a novel growth feature and a core reliability or retention fix under incomplete data.
How to approach it
- Situation: a real choice between shipping a new mobile capability and fixing a known reliability or retention issue, with the data on the fix's impact incomplete.
- Task: state why the data was incomplete, for example the retention issue's root cause was only partially understood at decision time.
- Action: describe how the tradeoff was framed, for example estimating a conservative range of retention impact versus the new feature's expected upside, both clearly as estimates.
- Action: state the decision made and why, such as prioritizing the reliability fix because even the low end of its estimated impact outweighed the new feature's upside.
- Result: state what happened, whether the reliability fix's actual impact matched the estimate, and what the delayed feature's eventual reception was.
- Reflection: state what was learned about deciding with incomplete data.
What a strong answer includes
- Uses a conservative-range estimate for the uncertain fix rather than waiting for perfect data or ignoring the issue.
- States the actual decision rule, prioritizing when even the low-end estimate outweighs the alternative, showing real quantitative reasoning under uncertainty.
- Reports the outcome honestly, including whether the estimate was accurate, not just a story that confirms the decision was right.
Common mistakes
- Claiming certainty about impact that the incomplete data did not actually support.
- No real estimation method described, just an intuitive call dressed up as data-driven.
Likely follow-up questions
- How would you have decided differently if the estimate range had been wider?
- What would you do if the novel feature's opportunity cost turned out to be higher than expected?
More behavioral questions
- Tell me about a time you made a bet against a much larger competitor.Perplexity · Behavioral · Medium
- Tell me about a time you had to make a ship-or-no-ship decision on an AI-native workflow, platform, or collaboration feature with incomplete data. What uncertainty mattered most, how did you de-risk it, how did you align engineering, design, and research, and what was the outcome?Perplexity · Behavioral · Medium
- Tell me about a time you worked with users running real marketing workflows such as paid acquisition, lifecycle, CRO, or SEO and translated their operational pain points into a product, tooling, or automation improvement. How did you decide what to ship, what tradeoffs did you make, and how did you verify it created value for the user and the business?Perplexity · Behavioral · Medium
- How do you know when to cut corners to get a product out the door?Shopify · Behavioral · Medium
- If there are 3 different items on top priority for a release and the client is insisting on getting all 3 delivered in the same release. As a PM you know there is not enough engineering capacity. What will you do?6sense · Behavioral · Medium
- Tell me about a time you had to win over a group that didn't report to you with your vision.Meta · Behavioral · Medium
More questions from Perplexity
Learn the skill behind it
Chapters of the AI PM course that teach what this question tests.