PageSpeed Insights has quietly switched its lab measurement engine to Lighthouse 13.6.0, the version released October 27. The change matters because the same page can now receive a different performance score without a single code update.
Barry Pollard, Chrome’s Web Performance Developer Advocate, confirmed the update on Bluesky. The refreshed simulation, he said, makes First Contentful Paint and Largest Contentful Paint more accurate and generally pushes scores higher.
Why JavaScript-heavy pages feel it first
The biggest swing is likely on JavaScript-heavy sites. Pollard pointed to reports that single-page apps saw large improvements, then clarified the effect is broader than SPAs. He described it as "more likely to impact JS-heavy sites." That includes any page where client-side rendering, hydration or heavy scripts shape what loads first.
For growth teams and digital marketers, this is not a ranking factor change by itself. But if you use lab scores to judge page speed or CRO, the new simulation changes the baseline.
What actually changed under the hood
Lighthouse uses Lantern to simulate slower network conditions. In version 13.5, Lantern used each request’s fetch priority to decide whether it blocked rendering. Version 13.6 now reads Chrome’s own render-blocking label, with a fallback to priority when the label is missing.
There is also a fix for pages where the same text element is both FCP and LCP. Lantern now estimates that LCP from the same requests and tasks it uses for FCP, instead of treating every request before LCP as a blocker. This removes a known distortion, including on Framer-built test pages reported by Ivan Akulov.
Before you celebrate a speed win
If your PageSpeed score jumps after an optimization, part of that leap may come from the new simulation rather than your change. Here is the quick check:
- Confirm both reports show the same Lighthouse version.
- Compare CrUX field data from real users over the last 28 days.
- Treat any score jump at the version switch as a measurement change.
- Watch JavaScript-heavy landing pages and apps first.
Lab data is a simulated single-device test. Google’s own documentation warns that good lab data does not guarantee good real-user experience. Field data from CrUX is the reality check.
One more audit quietly changed
Lighthouse 13.6 also updates the Agentic Resource Discovery audit. It now looks for an Agentmap line, rel=”ard” link tags or /.well-known/ard.json before the older ai-catalog names. If your site only uses the new ard.json spec, the audit should now see it.
What comes next
Lighthouse 13.6 is expected to reach DevTools in Chrome 159, with a stable release around December 8. If you report on site speed, version your reports now. A higher score is great, but only if you know whether the page improved or the ruler changed.
Source: Search Engine Journal
