Performance Trade-Offs
Overview
Convert's tracking script presents a fundamental choice between user experience consistency (no flicker) and Core Web Vitals optimization (better page speed scores).
Script Metrics
| Metric | Value |
|---|---|
| Base script size | ~50 kB (compressed, over network) |
| Page load impact | ~100ms additional |
| Processing time | ~200ms for experience application |
Script size increases with the number of active experiences, locations, audiences, and goals in the project.
Core Web Vitals Impact
Largest Contentful Paint (LCP)
- Synchronous: Moderate to high impact (script is render-blocking)
- Asynchronous: Minimal impact
First Input Delay (FID)
- Synchronous: Moderate impact (~200ms processing)
- Asynchronous: Minimal impact
Cumulative Layout Shift (CLS)
- Both modes: Low to moderate (depends on experiment design)
- Mitigation: Use anti-flicker techniques and reserve content space for elements that change
When to Choose Synchronous Loading
- User experience consistency is critical
- A/B testing accuracy is paramount
- Flicker effects are unacceptable
- Core Web Vitals are not a primary SEO concern
<script src="//cdn-4.convertexperiments.com/v1/js/[account_id]-[project_id].js"></script>When to Choose Asynchronous Loading
- Core Web Vitals optimization is critical for SEO
- Page speed is prioritized over perfect UX consistency
- Slight flicker is acceptable for better performance
See the Anti-Flicker Async Loading recipe for the complete async snippet with built-in flicker protection.
Updated about 1 month ago