Launch Decision for Facebook Reactions: Metrics, Experiment Design and Trade-offs
Company: Meta
Role: Product Manager
Category: Product / Decision Making
Difficulty: medium
Interview Round: Other
You are a product manager at Meta. Assume "Reactions" means extending the Like button on Facebook posts: instead of only liking a post, people can choose from a small set of emoji-style responses that express other feelings. Walk through how you would decide whether to launch Reactions to everyone.
```hint Decide what a win looks like before the data arrives
A new way to respond can pull activity away from the responses people already use instead of adding to them. Before you look at any results, work out which of those shifts you would accept and which would stop the launch.
```
### Clarifying Questions
- Which surfaces are in scope for the launch: posts in the feed only, or also comments, Groups, Pages and Messenger?
- What problem is Reactions meant to solve: giving people a way to respond to posts where a Like feels wrong, such as sad news; increasing feedback on posts; or giving feed ranking a richer signal?
- Has an experiment already run, or am I designing it? If it ran, which results are available?
- Does Like stay the one-tap default, with the other reactions one extra gesture away?
### What a Strong Answer Covers
- A goal for Reactions grounded in the gap a single Like button leaves, considering both the person responding and the person who posted.
- A primary success metric plus guardrails for what Reactions could cannibalize or harm, such as comments, shares, posting and feed quality.
- An experiment design that handles network effects (friends in the other test group see each other's reactions) and novelty effects.
- A decision rule agreed before results arrive, including how to resolve mixed outcomes, followed by a rollout and monitoring plan.
### Follow-up Questions
- Reactions per user rise, but comments per post fall. Do you launch, and what evidence would change your mind?
- Should feed ranking treat an Angry reaction the same as a Love reaction? What risks follow from each choice?
- Six months after launch, how would you tell whether Reactions made Facebook better for people rather than simply adding taps?
Overview: A product execution exercise asking how a Meta product manager would decide whether to launch Reactions, the emoji-style alternatives to the Like button on Facebook posts. It tests goal setting, success and guardrail metrics, experiment design under network effects, and making a launch call when results are mixed.