Measuring and Improving Performance- Insights from One of Cash App's Main Screens
summary
Built a central applet provider model that timestamps when each applet enters loading and finishes in adopted, unadopted, unavailable or error states, enabling per‑applet load times and an overall screen load time measured from the tab tap to interactable state. Data flows into either in‑house analytics or considered DataDog for dashboards and automated alerting. Found Bitcoin applet needs more CPU for 24h graphs, staleness windows are inevitable, beta fleet gives early signals, and so far monitoring infrastructure is in place but concrete performance wins are still forthcoming.
AI-written summary · about
chapters (17)
- 0:00Introduction and background
- 1:14What we’ll cover today
- 3:19Why we needed better metrics
- 4:41Key performance metrics
- 6:05Applet load time tracking
- 9:14Customer experience breakdown
- 10:37Choosing monitoring tools
- 12:22In-house analytics vs DataDog
- 14:07Setting up automated alerts
- 15:31Real customer data in beta
- 16:32Building clear dashboards
- 17:39Architecture challenges
- 18:44Customer-based measurement
- 20:08Platform consistency challenges
- 21:11Discoveries on feature performance
- 22:57Lessons learned and next steps
- 24:20Future monitoring plans
related talks
19mBeyond the Basics: Performance Monitoring and User Experience for Mobile App Growth
18mBeyond the Basics: Performance Monitoring and User Experience for Mobile App Growth
39mAndroid App Performance in a Nutshell
17mBeyond the Basics: Performance Monitoring and User Experience for Mobile App Growth
09mAndroid App Excellence Summit in 9 minutes
14mBeyond the Basics: Performance Monitoring and User Experience for Mobile App Growth
41mModern App Performance
39mApp performance