Metrics Framework for Instagram's Account Verification Badge

You are the product manager who owns verification on Instagram: the badge shown next to an account's name to confirm that the account really belongs to the person, creator, brand or organization it claims to represent. How would you measure whether this feature is successful? Explain the goal the feature serves, the metric you would treat as the main measure of success, and the supporting and guardrail metrics you would monitor alongside it. ```hint Who is the badge for? A badge helps the account that holds it and also the people who see it next to a name in search, comments or messages. Decide whose problem verification mainly solves before choosing what to count. ``` ### Clarifying Questions - Is verification granted free to notable accounts after a review, sold as part of a paid subscription, or both? That decides whether revenue belongs among the metrics. - Which accounts can apply (public figures, creators, businesses, or any person who passes an identity check), and in which countries? - Is the main purpose to protect people from impersonators, to help people find the authentic account, or something else such as a status signal for creators? - Are we measuring the health of the feature as it stands, or evaluating a change such as new eligibility rules? ### What a Strong Answer Covers - A goal for verification tied to trust on Instagram, set before any metric is named. - The value to verified account holders and to the viewers who rely on the badge, with metrics for each and a funnel from applying to being verified. - One primary metric chosen over alternatives such as the number of badges issued, with a stated reason. - Guardrails on integrity: wrongful approvals and rejections, impersonators that still get through, review cost, and dilution of what the badge means. - A way to separate the badge's effect from the fact that verified accounts are already different from other accounts. ### Follow-up Questions - Accounts receive more follows after they are verified. How would you tell whether the badge caused that increase? - If verification became part of a paid subscription, which metrics would you change and which guardrails would matter more? - Impersonation reports fall, but so do applications for verification. How do you read that?