How do we do product at Spotify
summary
Spotify focuses on tiny, measurable bets: start with 7% monthly users 3% paying, spot a concrete failure (31k daily signups with 75% drop in a form field), combine quantitative logs with recorded mobile usability sessions and Jobs‑To‑Be‑Done interviews, brainstorm and prototype in Figma using the design system, ship small variations behind feature toggles, run experiments in A/B platform "Confidence", and push weekly releases (release every Friday). One experiment lifts registrations (≈+2%/day); most experiments fail but produce learnings that get documented and reused.
AI-written summary · about
chapters (21)
- 0:00Introduction
- 1:03Global user base and mission
- 3:09Why Spotify succeeded
- 5:20Anna’s daily workflow
- 7:11All Hands meeting
- 10:27User registration drop-offs
- 13:18Qualitative user insights
- 16:27Team brainstorming
- 19:16Prototyping design variations
- 21:54Fast engineering and weekly shipping
- 23:40A/B testing in production
- 26:55Experiment results and iteration
- 27:17Documenting and sharing learnings
- 29:49Continuous improvement
- 30:10Spotify’s org model
- 32:22Teams with data and insights
- 34:11Execution framework
- 37:05Design thinking process
- 39:14Secrets behind Spotify’s success
- 43:25Tips for product teams
- 46:34Q&A
related talks
29mFail Fast, Fail Smart, Succeed
48mAgile at scale at Spotify
29mThe Spotify model - Does copy paste implementation work
43mThe #Spotify model - Does copy/past implementation work?
01h 04mYou are not Spotify, and you don’t want to be
43mHow Autonomy Saved One of Spotify’s Most Loved Features
11mBreaking the (developer) silos
39mOwn your speed