PageSpeed scores up since November 2, 2026: Lighthouse 13.6 changed, your site may not have
Partner · Strategy, advertising and measurement
Since early November 2026, PageSpeed Insights (Google's free tool that scores a page's speed out of 100) has been running on Lighthouse 13.6, released on October 27. This version recalculates simulated speed, in particular the first paint and the paint of the largest element, in a way that is closer to what Chrome really does. The result: many pages get a better score without a single line having changed, especially sites that load a lot of JavaScript. The score goes up, the real speed your visitors experience does not move. Before congratulating anyone (or paying for a gain), check the version number at the bottom of the report and compare the field data, which has not changed.
What changed
Lighthouse is the engine that calculates the performance score in PageSpeed Insights. It does not measure your site on your visitors' devices: it loads the page once, in a controlled environment, then simulates a slow mobile connection to estimate paint times. Version 13.6, released on October 27, 2026, changes that simulation. The release notes describe request modelling "updated to improve simulation accuracy," the use of the render-blocking status provided by Chrome itself to decide which resources delay the page, and a calculation of the largest element (the LCP) aligned with the first paint when that element is text.
Barry Pollard, who works on web performance at Chrome, summed it up on November 2: version 13.6 "improves the simulated FCP and LCP resulting in more accurate, usually better, scores." The change is "more likely to impact JS-heavy sites". The same day, he confirmed that the version was live in PageSpeed Insights. It will reach Chrome's developer tools with version 159, scheduled for December 8, 2026.
The fix goes back to June 2025. Ivan Akulov, who works at Framer (a website building tool), had shown that Lighthouse penalized file preloading: a page whose real first paint took about one second received a much higher simulated LCP, because the simulation assumed that every request started before the paint was delaying it. Removing those preloads raised the score while the real experience got worse. That is the flaw that 13.6 corrects.
What it means for you
The PageSpeed score is a lab estimate. Google says so on the page that explains the tool: good lab data "does not necessarily mean real-user experiences will also be good". The field data comes from the Chrome User Experience Report (CrUX) and appears at the top of the same report, under "Discover what your real users are experiencing." It covers the last 28 days and the 75th percentile of visits. It did not change on November 2. That field data, the Core Web Vitals, is what counts in Google's ranking, not the score out of 100: the Core Web Vitals entry in our glossary explains the difference.
The score itself combines five measures: first paint (10%), speed index (10%), largest element (25%), blocking time (30%) and visual stability (25%). Version 13.6 mostly affects the first two and the LCP. A site that already scored 95 will not gain much; a script-heavy site, built with a page builder or as a single-page application, can gain several points at once.
What we recommend
| Your situation | What we recommend |
|---|---|
| Your score went up last week without anyone touching the site | Look at the footer of the PageSpeed report: it shows the Lighthouse version. If the old report said 13.5 and the new one 13.6, the gain comes from the tool. Compare the field data for the last 28 days instead. |
| A vendor presents this gain as the result of their work | Ask for two reports of the same page, before and after, with the version number visible, then the real Core Web Vitals. A real fix shows in both. |
| Your score is still low even after 13.6 | The problem is not the simulation. Fixes are prioritized as we explain in eCommerce site speed: prioritizing fixes for a large catalogue and, for WordPress, on our WordPress speed optimization page. |
| You compare your site to a competitor | Run both tests the same day, with the same version. An October report and a November report no longer compare. |
A PageSpeed score is read together with its version. Lighthouse 13.6 corrects a flaw known for a year and a half, which punished good loading practices, especially on JavaScript-heavy sites. Pages go up, the speed your customers experience does not change. Keep the field data as your reference and note November 2, 2026 in your tracking: it is the date from which lab scores no longer compare with earlier ones.
Monitoring your site's speed is part of our website maintenance service.
Frequently asked questions
My PageSpeed score went up ten points in November 2026. Is my site faster?
Not necessarily. PageSpeed Insights changed its Lighthouse version in early November 2026, and the new simulation generally gives better scores to the same pages. Look at the field data for the last 28 days, at the top of the report: if it has not moved, your site's speed has not changed.
Will Google rank my site higher because of this new score?
No. Ranking takes into account the Core Web Vitals measured on real visitors, not the lab score out of 100. Those real measurements are not affected by the Lighthouse update.
How do I know which Lighthouse version produced my report?
The version number appears at the bottom of every PageSpeed Insights report, in the technical notes. Since early November 2026, it should read 13.6.0. Reports produced before that carried 13.5 or an earlier version.
- GitHub, Lighthouse v13.6.0, release notes, October 27, 2026. Source of the date, the simulation changes, the use of Chrome's render-blocking status, the text LCP calculation and the arrival in Chrome 159 and in PageSpeed Insights.
- Barry Pollard (Chrome), post on Bluesky, November 2, 2026. Source of the confirmation that 13.6 is live in PageSpeed Insights, of the "usually better" scores and of the effect on JavaScript-heavy sites.
- Chromium Dash, Chrome release schedule, milestone 159, accessed November 3, 2026. Source of the December 8, 2026 stable release date.
- GitHub, Lighthouse Punishes Good Asset Preloading (issue 16539), Ivan Akulov, opened June 12, 2025, closed October 8, 2026. Source of the original flaw and the preloading example.
- Google, About PageSpeed Insights, accessed November 3, 2026. Source of the distinction between lab data and field data, the 28-day window and the 75th percentile.
- Chrome for Developers, Lighthouse performance scoring, accessed November 3, 2026. Source of the weight of the five measures in the score.
- Search Engine Journal, PageSpeed Insights Scores May Rise After Lighthouse Update, Matt G. Southern, November 2, 2026. The article that flagged the change; the facts were verified against sources 1 to 6.
Gabriel almost always takes your first call and carries out your audit. He builds the strategy starting from your growth goal: where to put your budget, which market to test and how to connect each lead to a real sale in your CRM. He mainly leads engagements for three goals: Optimize the profitability of your digital campaigns, Develop a new market, and Generate demand and growth. With Geneviève, he also works on organic search (SEO), AI visibility (GEO) and conversion rate optimization (CRO). The sales a Google Ads or Meta Ads campaign brings in depend on the page that receives the click. He writes mainly about marketing strategy, paid advertising and measurement.
About Falia →
Strategy
SEO
Paid advertising
Conversion
AI visibility
Web design