PageSpeed Insights is now using Lighthouse 13.6.0, which was released Oct. 27. This type of Lighthouse changes how it simulates how lengthy a leaf takes to load.
Chrome’s Web Performance Developer Advocate, Barry Pollard, has stated on Bluesky that the updated type of PageSpeed Insights alongside Lighthouse 13.6.0 is now available. He stated this improves the simulated FCP (First Contentful Paint) and the simulated LCP (Largest Contentful Paint), the two of which Pollard says volition be additional exact than before and mostly outcome in higher scores.
A leaf may obtain a distinct PageSpeed Insights mark without item on it changing, because of the alter in the test. Pollard additionally stated that websites heavily reliant on JavaScript would be additional apt to see an effect from the updated test.
Pollard Points To JavaScript-Heavy Sites
Pollard quoted a developer who had posted concerning how single-page apps (SPAs) had seen a big betterment in their results since the Lighthouse update, citing react.dev. He opened his article stating, “True!”, afterward added:
“Note this is not SPA-specific (and unrelated to another SPA things we’ve been operating on), but additional apt to effect JS-heavy sites.”
The experimental gentle navigation metrics in Chrome that I wrote concerning rear in October are a distinct way of measuring person experience. The data comes from genuine Chrome users through the Chrome User Experience Report (CrUX) and Lighthouse uses a simulated test.
What Changed In The Simulation
By default, Lighthouse volition burden a leaf at complete speed and afterward evaluation how it would burden on a slower connection. The throttling docs say that this simulated throttling matches PageSpeed Insights’ setup.
The publish notes for type 13.6.0 of Lighthouse catalog multiple changes to Lantern, which is the simulation rearward those estimates, aimed at improving its accuracy.
Lighthouse 13.5 used all request’s fetch precedence to decide whether or not it blocked rendering. In 13.6, Lantern uses Chrome’s own render-blocking tag for all petition (and falls rear to using precedence lone whenever that tag is missing).
A second alter applies whenever the LCP component is content that appears in the identical example as the page’s archetypal content. Lantern now estimates that LCP from the identical requests and leaf tasks it uses for FCP alternatively of from a set that can contain all petition made before LCP.
Ivan Akulov showed in a GitHub issue how the old simulation handled a test leaf created alongside Framer whose lone satisfied complete the fold was text, so one component counted as the two FCP and LCP. Lighthouse reported much afterward LCP than FCP for that text. Even although preloading the page’s JavaScript modules improved interactive period for the genuine page, they caused simulated LCP to be worse.
Pollard thanked Akulov in a reply, saying he “kicked off several of these improvements.” The Lighthouse squad closed the matter as completed in the Lantern upgrade listed in the 13.6 publish notes.
The ARD Audit Now Checks For ard.json
Lighthouse 13.6 has additionally updated the Agentic Resource Discovery (ARD) audit from Lighthouse 13.5, which I covered in September. When Lighthouse 13.5 was released, it looked for an ARD manifest under the old ai-catalog names equal although the spec had changed to ard.json. So if your website lone used the new name but you didn’t contain a row in your robots.txt document called Agentmap, it would display as Not Applicable.
With this update, Lighthouse 13.6 volition appearance for an Agentmap line, a rel="ard" nexus tag or Link header and /.well-known/ard.json previous to looking for the ai-catalog names. And if Lighthouse 13.6 finds the manifest lone at an ai-catalog location, it volition note that ARD consumers are not required to inspect there.
Lab Scores And Field Data
The achievement mark listed for your leaf in PageSpeed Insights is according to data generated by Lighthouse. According to Google’s PageSpeed Insights documentation, lab data are results of a simulated burden using a sole equipment operating on a fixed set of network conditions. They note that “having fine lab data does not necessarily average real-user experiences volition additionally be good.”
Field data (also shown in the identical report) represents genuine visits to your website inside the final 28 days, as reported by CrUX. Each PageSpeed Insights study includes the type of Lighthouse used to create results, next to the emulated device.
Why This Matters
PageSpeed Insights scores from before and following the Lighthouse 13.6 update weren’t measured alongside the identical simulation. Part of a higher lab mark on a leaf you fair optimized may arrive from the new version, and charts that track lab scores complete period may display a jump at the toggle to 13.6.
Before reporting an improvement, inspect that the two reports name the identical Lighthouse version, and difference the site data.
Looking Ahead
Lighthouse’s publish notes say its squad expects 13.6 to container in the DevTools of Chrome 159, which is now in Chrome’s desktop Dev channel and in Canary. As of Nov. 2, Chrome’s release schedule lists Chrome 159 for stable publish on Dec. 8.
An open issue on Lighthouse’s GitHub is additionally looking into cases anywhere Lantern’s simulated LCP runs far complete measured LCP, using ample JavaScript and SPA sites as examples.
Featured Image: Remo_Designer/Shutterstock
.png)
