Spotify execution case: triaging and root-causing a listener's connection problem

You are a PM at Spotify, and a user has reported a connection issue. How would you go about diagnosing it? ```hint Size it first Before chasing causes, decide whether this is one person's setup or the first visible sign of a wider pattern. Everything you do next depends on that answer. ``` ### Clarifying Questions - What does "connection" mean in the report: the app cannot reach the service (offline banner, playback will not start), the user cannot sign in, or Spotify Connect cannot find or control a speaker, TV or car? - Is this a single report, or are similar reports arriving through support, app reviews or social media? - Which device, platform, app version and network was the user on, and is the problem constant or intermittent? - Can we look at this user's client logs and error codes, and what consent does that require? ### What a Strong Answer Covers - Pinning down the symptom and the user's context before forming hypotheses. - Sizing the problem: one user versus a cohort, using telemetry and support signals cut by platform, app version, region, network and device. - A structured split between internal and external causes, checked in an order driven by what changed recently and what is cheapest to rule out. - Mitigation and communication: who is pulled in, how affected users are protected while the cause is found, and how the loop is closed with the reporter. - Prevention: the monitoring or alerting change that would catch the next occurrence before users report it. ### Follow-up Questions - Your connection error dashboards are flat, yet complaints about dropped connections are climbing on social media. What do you do? - The cause turns out to be a firmware update from a third-party speaker maker. What can Spotify do, and how do you weigh it against roadmap work? - How do you choose between rolling back the latest app release and shipping a hotfix?