Metrics question

You are a PM at Meta. How would you set goals for and measure the success of verified accounts?

Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.

What this question tests

Tests goal-setting for a trust and safety-adjacent feature, balancing platform integrity goals with user and business incentives.

Clarify purpose: verification signals authenticity for public figures and businesses, reducing impersonation and helping users trust who they're engaging with.
Set the north star as reduced impersonation-related reports for verifiable entities, since that's the feature's core integrity purpose.
Add supporting metrics: application-to-approval rate and time, and engagement lift on verified accounts versus unverified lookalikes.
If monetized as a paid subscription, add paid conversion and retention as a separate business metric.
Add a guardrail: false-positive verification rate and verification process fraud rate, since a compromised system undermines the entire trust signal.

Separates the platform-integrity goal, reducing impersonation, from the business or monetization goal, correctly identifying they can conflict.
Names guardrails specifically against verification fraud, the biggest risk to the feature's credibility.
Treats trust and safety and monetization as distinct metric tracks rather than one blended number.

Setting only a monetization or subscription metric without acknowledging the original trust and safety purpose of verification.
Ignoring the guardrail risk of fraudulent verification undermining the whole system's credibility.

How would you balance the trust and safety goal against a monetization goal if they conflict?
How would you detect and handle verification fraud at scale?