Driving App Reliability with Real Time Monitoring and SLIs
summary
Claims a deployable monitoring stack: a KMM library with Firebase + Google Cloud (BigQuery) that tags errors with a session ID per session so feature availability and user frustration are measurable. Defines SLI/SLO math (example targets: 95% SLO and 99% core-feature target), shows dashboards where availability dips below 80% and one feature has 19% crash‑free but 60% availability, and adds a simple frustration score 0–5. Implementation finds root causes and enables fixes within days.
AI-written summary · about
chapters (14)
- 0:00App reliability challenges
- 1:52Mobile metrics and blind spots
- 3:39Crashes to user experience
- 4:43SLIs and objectives explained
- 6:11Measuring feature availability
- 8:14Cross-platform monitoring framework
- 10:39User frustration metrics
- 13:28Analyzing monitoring data
- 16:59Data-driven decisions
- 24:02Team effort for reliability
- 25:48Metrics for markets and architectures
- 27:14Error budgeting
- 29:47Balancing reliability and growth
- 30:37Q&A
related talks
44mEfficient Android app monitoring
17m“How’s the App Doing?” Bringing Mobile Into Your Reliability Picture
32mChasing performance issues methodically
14mBeyond the Basics: Performance Monitoring and User Experience for Mobile App Growth
37mScaling Performance and Availability Monitoring across distributed squads
38mImplementing SRE practices: SLI/SLO deep dive
44mDriving App Stability & Performance Zero to Hero
37mRelia...bility?