Evaluate Dating App Product Changes

You are a Data Scientist at Grindr, a location-based dating and social discovery app. The product is considering several ranking, recommendation, and monetization changes. Assume the app has free users, paid subscribers, profile browsing through a Grid page, messaging, and a notion of a successful connection, such as two users mutually engaging and having a meaningful conversation. Answer the following product analytics and experimentation questions. For each one, define the goal, primary metrics, secondary metrics, guardrail metrics, experimental design, key segments, and major risks or biases. 1. **Reduce the profile browsing limit**: The CEO wants to reduce the maximum number of profiles a user can browse or view in a session from 100 to 80. How would you determine whether this change is good or bad? 2. **Choose how many recommendations to show**: The Machine Learning team has built a model that recommends users whom each person is most likely to connect with successfully. How would you decide how many recommended users to show to each person? What product surfaces could use this model? 3. **Test a paid high-connect-probability filter**: The Machine Learning team has built a model that predicts the probability that two users will connect. A Product Manager wants to add a paid Grid-page filter that allows subscribers to directly filter for profiles with the highest predicted connection probability. How would you evaluate this with an A/B test? 4. **Evaluate a paid “heart” feature**: Grindr launches a feature that lets users send a special “heart” to another user, similar to a premium like or rose. The first heart is free, and additional hearts are paid. How would you measure whether this feature is successful? If adoption is low immediately after launch, how would you investigate?