Experiment hypotheses to double Facebook Shop's weekly actives despite low retention
Company: Meta
Role: Product Manager
Category: Product Design & Strategy
Difficulty: medium
Interview Round: Other
Facebook wants twice as many people using Facebook Shop each week. Retention is the backdrop: you are told that 6% of Facebook users are still active on Shop in their first week, and only 0.4% are still active three weeks in. As the product manager, how would you get weekly active users to double? Spend most of your answer on the hypotheses you would test and the experiments you would build around them.
```hint Split the weekly actives
Before proposing features, sort this week's active users by when they first came to Shop, and use the two retention figures to judge how much each group could realistically add.
```
### Clarifying Questions
- Does "Facebook Shop" mean the storefronts businesses run on Facebook, or Marketplace, where people buy and sell locally?
- What counts as active: opening Shop, viewing a product, or buying something?
- Are the 6% and 0.4% measured on a cohort of first-time Shop visitors or on all Facebook users, and when does "week one" start?
- By when should weekly actives double, and which parts of Facebook, such as Feed engagement or ad revenue, must not suffer for it?
### What a Strong Answer Covers
- A reading of the two retention figures and of what counts as active, and what they imply about how a doubling could be reached
- A chosen user segment and where in its shopping journey people drop off, with the reasoning
- Specific hypotheses, each tying a product change to a user behavior and a measurable outcome
- An experiment plan for the priority hypotheses: unit of randomization, primary metric, guardrails, and a duration and sample size suited to a low baseline
- Prioritization across the hypotheses and the risks of raising the number without real shopping value
### Follow-up Questions
- Your top experiment lifts week-one return, but week-three return does not move. What do you conclude, and what do you test next?
- How would you test a change where buyers and sellers influence each other, so a split by user could bias the result?
- If doubling by the deadline looks out of reach, what would you tell leadership, and what target would you propose instead?
Overview: A product growth question about doubling weekly active users of Facebook Shop when only 6% of users stay active in their first week and 0.4% after three weeks. It tests how a candidate reads retention data, forms testable hypotheses, and designs experiments with sound metrics, guardrails and sample sizes.