Measuring Time to Visually Complete at Dropbox
summary
Dropbox ships an open source library called dropbox-ttvc that measures Time to Visually Complete in real users by combining DOM and visibility signals rather than lab video. It tracks visual changes as DOM mutations and image loads, filters only viewport changes, uses MutationObserver + IntersectionObserver and a wrapped requestIdleCallback to wait two consecutive idle seconds, then reports the last visible-change timestamp. It supports SPA navigations, is on npm/GitHub, is beta but matches lab tests.
AI-written summary · about
chapters (11)
- 0:00Introduction to TTVC
- 1:03Why web performance matters
- 3:51Page load lifecycle explained
- 5:35What is Time to Visually Complete
- 7:39Problems with Largest Contentful Paint
- 9:28Limitations of lab testing
- 10:30How TTVC works with browser APIs
- 11:33Library implementation details
- 15:25Using the TTVC library
- 17:30Open source release and next steps
- 18:55Q&A session
related talks
01h 02mWeb Performance Tuning with browser APIs
05mMeasuring what matters
35mWeb Application Performance Testing from Users Perspective
19mPerceived Performance, And Why It Matters
17mVisualising Front-End Performance Bottlenecks
11mFixing common Web Vitals issues
50mWeb Performance APIs You (probably) Didn't Know Existed
54mThe Latest in Metrics & Measurement