analyze_page_speed runs Google PageSpeed Insights on a URL and returns two views of it: a Lighthouse lab test, and field data from real Chrome users (the Chrome UX Report) where Google has enough of it.
No Google account is involved: PageSpeed Insights fetches the page itself, so there is nothing to connect.
Set up
- AdLoop Cloud
- Self-hosted
Nothing to set up. Include the
web toolset in your API key, or leave the key unrestricted (toolsets).What you get
- Performance score from 0 to 100.
- Lab metrics: Largest Contentful Paint, Cumulative Layout Shift, Total Blocking Time, First Contentful Paint.
- Field metrics over the last 28 days: the 75th percentile for LCP, Interaction to Next Paint and CLS, each rated fast, average or slow, plus an overall rating.
- Top opportunities: the largest Lighthouse improvements with their estimated savings.
- Insights that flag a score below 50, a slow real-user rating, or an LCP above the 2.5 second threshold.
Behaviour worth knowing
- Mobile by default, because that’s where most paid traffic lands; pass
strategy="desktop"for the other view. - A run takes 10 to 30 seconds. That’s Lighthouse loading the page, not a hang.
- Field data beats lab data. Lab numbers come from a simulated load. When the two disagree, trust the real-user numbers. No field data means the page has too little Chrome traffic, not that it’s fast.
- Public pages only. Google fetches the URL from its own servers, so localhost, staging behind a login or pages blocked by a firewall can’t be measured.
- It pairs with
landing_page_analysis: a page with paid traffic, no conversions and a poor score needs speed work before new ad copy.
Tools
analyze_page_speed: PageSpeed Insights for one URL
web.