Using Analytics to Improve Quality
summary
Using production telemetry to change testing choices and code fixes reduces crashes: the Wikipedia (Qix) app cuts a ~9% crash rate down to 0.4% crash rate after six releases and a first-release 38% drop from 25 lines of defensive code. Work uses Android Vitals, Firebase/Crashlytics, heat maps and session analytics, reproduces device-specific faults (codec, battery, nulls), and suggests test targeting (about 8 devices for 80% coverage). Also covers CI checks for analytics firing and privacy tradeoffs.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:20Analytics
- 2:02Are analytics useful
- 2:38Learning from history
- 13:03Mobile analytics overview
- 17:29Mobile analytics examples
- 25:06What are we trying to achieve
- 28:14Using analytics to improve testing
- 30:54How many devices are enough
- 32:58Getting access to analytics
- 34:26QA Session
- 39:03Automated Testing
- 41:00Crash Rates
- 42:14Apples Data Privacy Policy
- 43:31Tracking Fields in Analytics
related talks
55mBuilding Quality with Firebase
23mMobile App Crashes in Production: Lessons Learned from a Tester’s Viewpoint
14mRoundtable: Ensuring Game Stability: How To Find, Monitor, & Fix Issues With Firebase Crashlytics
46mBoost the Quality of Your App with Firebase
03mSummit Report: Justin and Ali on Crash Reporting
30mImprove app performance and stability with Firebase
01h 00mBoost the quality of your app with Firebase
38mImprove your app and game quality with Android vitals