Metrics question

Perplexity cares about product-led growth and data-driven flywheels. For a new AI productivity feature inside Search or Computer, what user actions would you optimize first to drive activation, retention, and learning effects, and how would you know the feature is creating a real flywheel rather than a one-time novelty spike?

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

Growth thinking for AI products: can you identify the right leading actions to optimize and design a test that distinguishes a real flywheel from a one-time novelty spike.

How to approach it

  1. Pick the action that is closest to habitual value creation, for example saving or reusing a prior search or workflow, rather than just first use.
  2. Prioritize actions with compounding value: each use should make the next use better or faster, for example connectors or personalization that improve with more data.
  3. Optimize activation around getting a user to that compounding action quickly, and retention around whether they return to use it again within a defined window.
  4. Define a flywheel signal specifically: usage frequency or session depth increasing over successive weeks for the same cohort, not just aggregate growth.
  5. Distinguish novelty from flywheel by tracking a cohort's week-over-week usage curve; a spike that decays to baseline within a few weeks is novelty, sustained or increasing usage is the flywheel.

What a strong answer includes

Common mistakes

Likely follow-up questions

More metrics questions

More questions from Perplexity

Learn the skill behind it

Chapters of the AI PM course that teach what this question tests.

Preparing for a specific role?

Book summaries for this kind of question

Browse all 4,000+ questions in the bank