Choose and justify a north-star success metric for Meta's Facebook News

Quick Overview

A product metrics question asking which success metric a product manager for Facebook News at Meta should track and why. It tests tying a metric to the product's goal for readers and publishers, defining it precisely, comparing it with alternatives such as clicks or time spent, and adding integrity guardrails.

Choose and justify a north-star success metric for Meta's Facebook News

Company: Meta

Role: Product Manager

Category: Analytics & Experimentation

Difficulty: medium

Interview Round: Other

You are the product manager responsible for Facebook News at Meta. Which metric would you use to judge whether the product is succeeding? Define it clearly and explain why you chose it. ```hint Ask what a good visit leaves behind Opens, clicks and minutes are easy to grow with sensational headlines. Think about what a reader and a publisher would each count as a good outcome of a visit, and what could go wrong if the team optimized your metric hard. ``` ### Clarifying Questions - Does "Facebook News" mean only the dedicated news tab, or all news stories people see in their Feed as well? - Which outcome should the metric favor first: readers staying informed, publishers gaining an audience, or Meta's overall engagement? - Should the metric work across every country where the product runs, or is there one market to focus on? ### What a Strong Answer Covers - A goal for Facebook News tied to Meta's mission and to what both readers and publishers want from it. - One precisely defined success metric (who counts, which action counts, over what window) and why it beats alternatives such as clicks or time spent. - Supporting metrics that break the success metric down and cover the publisher side. - Guardrails for misinformation, sensationalism, source diversity and effects on the rest of Facebook. - The known limits of the metric and how gaming would be detected. ### Follow-up Questions - Your metric rises after a ranking change, but surveys say people feel less well informed. What do you do? - How would you measure the value Facebook News creates for publishers? - How would you keep a big news event, such as an election, from distorting your read of the metric?

Overview: A product metrics question asking which success metric a product manager for Facebook News at Meta should track and why. It tests tying a metric to the product's goal for readers and publishers, defining it precisely, comparing it with alternatives such as clicks or time spent, and adding integrity guardrails.

Meta logo
Meta
Aug 8, 2025
mediumProduct ManagerOtherAnalytics & Experimentation
0
0

You are the product manager responsible for Facebook News at Meta. Which metric would you use to judge whether the product is succeeding? Define it clearly and explain why you chose it.

Clarifying Questions Guidance

  • Does "Facebook News" mean only the dedicated news tab, or all news stories people see in their Feed as well?
  • Which outcome should the metric favor first: readers staying informed, publishers gaining an audience, or Meta's overall engagement?
  • Should the metric work across every country where the product runs, or is there one market to focus on?

What a Strong Answer Covers Guidance

  • A goal for Facebook News tied to Meta's mission and to what both readers and publishers want from it.
  • One precisely defined success metric (who counts, which action counts, over what window) and why it beats alternatives such as clicks or time spent.
  • Supporting metrics that break the success metric down and cover the publisher side.
  • Guardrails for misinformation, sensationalism, source diversity and effects on the rest of Facebook.
  • The known limits of the metric and how gaming would be detected.

Follow-up Questions Guidance

  • Your metric rises after a ranking change, but surveys say people feel less well informed. What do you do?
  • How would you measure the value Facebook News creates for publishers?
  • How would you keep a big news event, such as an election, from distorting your read of the metric?