Reducing the Speed Impact of Third-Party Tags
summary
Shows concrete fixes and measurable wins: use WebPageTest/SpeedCurve to spot connection and main‑thread contention, remove stray document.write flags in GTM (saved ~4KB and cut long tasks), proxy or server-side GTM at the CDN/edge, sequence nonessential tags with a Cloudflare Worker, try Partytown or web workers for heavy vendors, build a GitHub Action to watch GTM container versions, and measure with/without consent — one case removes a slow review widget and yields a 26% Android uplift worth ~1–2% of £250 million revenue.
AI-written summary · about
chapters (12)
- 0:00Welcome
- 3:32Third-party tags impact
- 8:03Mapping third-party tags
- 10:51Third parties in page load
- 14:45GDPR consent challenges
- 21:30Prioritizing third-party timing
- 25:22Lazy-loading third parties
- 28:12Off-main-thread edge execution
- 31:23Measuring performance impact
- 38:27Debugging main thread blocks
- 45:29Recommendations and case studies
- 50:07Q&A
related talks
46mChaperones and curfews: minimising 3rd party impact
45mWhy Fast Matters
53mThird Party Woes
48mThink Fast, First: Empowering a Performance Culture
41mSpeed at Scale: Web Performance Tips and Tricks from the Trenches
32mEverything you need to know about Web Performance
01h 04mNeed for speed 8, performance tuning of your web application
58mNeed for speed 8, performance tuning of your web application