In the world of backend distributed systems, the process through which production data to used to answer questions about how your app is performing is called Observability (or o11y for short). The combination of telemetry and metadata goes way beyond monitoring p50s, and allows you to answer questions you hadn't anticipated at instrumentation time. On Android, most devs rely on benchmarking and profiling tools in internal environments to understand performance problems. In production, few have access to anything more than what the Android Vitals dashboard offers. This is partially because the infrastructure to support o11y workflows had been costly to setup and maintain. But with the rise of the open-source o11y platform OpenTelemetry as well as o11y vendors with mobile expertise, this is quickly changing. This talk will be a practical intro (complete with examples) to Observability for a group of devs who, for the most part, have not truly experienced it. It will cover: - How observability is different from crash monitoring and traditional RUM (Real User Monitoring) tools - Using practical use-case, show what you can achieve with o11y when you focused on users - How Android devs can get started in the world of user-focused observability, from startups to big-tech alike - Introduction to the OpenTelemetry project from CNCF and the newly proposed Kotlin API and SDK