I’d combine quantitative analysis with qualitative feedback. First, I’d map out key user actions and analyze retention curves based on when those actions occur. Using funnel analysis and cohort retention , I’d look for inflection points that strongly correlate with longer engagement. On the qualitative side, I’d conduct user interviews or run in-app surveys right after key moments to ask, “What made this feel valuable?” or “What would make you come back?” Combining both views helps identify not just where users stay, but why. I’d then use that insight to accelerate users to that moment faster.

I haven’t run a growth or retention experiment yet, but I’d design one around personalized re-engagement. For a music app like Spotify, I’d A/B test push notifications: one group gets none, another gets a generic playlist reminder, and a third gets a personalized message. I’d use tools like Amplitude or Mixpanel for analytics, Segment for event tracking, and OneSignal or Firebase for notifications. The goal would be to see if behavioral personalization drives higher retention and engagement.

Spotify is my favorite Gen Z-focused app. Whether I'm deep into coding, training ML models, or decompressing, Spotify’s recommendations adapt surprisingly well to those contexts. I also love how it's social without being overwhelming. Sharing playlists with friends or discovering what someone else is listening to feels low-pressure but still connected.