Duda sites passed all three Core Web Vitals on mobile 85.8% of the time in September 2026, Adobe Experience Manager sites 33.9% and Angular sites 16.5%, against 52.8% for the web as a whole. Those figures come from the HTTP Archive's technology report, which reads Chrome's real-user data for 8.57 million origins. The platform a site runs on predicts its pass rate better than almost anything else in the public data.
This page sets out pass rates by CMS, by JavaScript framework, by industry proxy and by country, with the sample size and the year-on-year change next to each. Every platform figure was read from the HTTP Archive report API on 2026-10-28; our Website Speed Statistics 2026 post covers the overall trend since 2020 and the business case for speed, so it is not repeated here.
How these pass rates are measured
Core Web Vitals are three measures taken from real Chrome users: Largest Contentful Paint (LCP) for loading, Interaction to Next Paint (INP) for responsiveness and Cumulative Layout Shift (CLS) for visual stability. web.dev sets the good thresholds at an LCP of 2.5 seconds or less, an INP of 200 milliseconds or less and a CLS of 0.1 or less, judged at the 75th percentile of page loads and segmented by mobile and desktop.
The HTTP Archive's technology report counts an origin as passing when it is good on all three. Its own definition adds one leniency: if an origin has no INP data, it is assessed on the other two. Origins only appear if they are public, indexable and popular enough for the Chrome UX Report, which does not disclose its traffic cut-off. Technologies are detected with Wappalyzer on one page per site, usually the home page, during a crawl on the second Tuesday of each month.
Two consequences follow. A platform's pass rate describes the sites that use it, not the platform in a lab. And a small site with little traffic may not be in the data at all.
Pass rates by CMS and site builder
| Platform | Mobile origins | Mobile pass | Desktop pass | Mobile, September 2025 |
|---|---|---|---|---|
| Duda | 42,026 | 85.8% | 87.8% | 85.1% |
| Wix | 177,709 | 80.9% | 87.1% | 73.9% |
| TYPO3 CMS | 40,300 | 80.3% | 72.3% | 79.0% |
| Shopify | 445,297 | 76.2% | 79.5% | 76.2% |
| Webflow | 59,483 | 71.3% | 82.6% | 64.0% |
| Squarespace | 89,052 | 71.0% | 76.1% | 65.0% |
| Framer Sites | 11,319 | 70.8% | 87.9% | 64.0% |
| Craft CMS | 10,357 | 67.5% | 68.9% | 65.4% |
| Drupal | 93,193 | 64.7% | 66.1% | 61.7% |
| HubSpot CMS Hub | 13,904 | 60.0% | 75.1% | 51.9% |
| Joomla | 83,029 | 57.4% | 59.7% | 56.0% |
| Sanity | 10,586 | 52.4% | 68.6% | 47.6% |
| WordPress | 2,658,175 | 48.3% | 53.7% | 45.0% |
| Tilda | 49,085 | 47.9% | 51.2% | 48.2% |
| Weebly | 17,820 | 41.1% | 55.3% | 47.1% |
| Sitecore | 5,890 | 39.3% | 48.7% | 37.2% |
| Contentful | 9,212 | 38.1% | 55.5% | 36.3% |
| Adobe Experience Manager | 19,174 | 33.9% | 44.3% | 35.4% |
| All origins | 8,569,446 | 52.8% | 60.8% | 50.5% |
Source: HTTP Archive Core Web Vitals Technology Report, all ranks, all geographies, September 2025 and September 2026, read through the report API on 2026-10-28
The year brought gains for most platforms. HubSpot CMS Hub rose 8.1 points on mobile, Webflow 7.3, Wix 7.0, Framer 6.8 and Squarespace 6.0, while the web as a whole gained 2.3. Weebly fell 6.0 points and Adobe Experience Manager 1.5. Shopify did not move.
The 2025 Web Almanac's CMS chapter, built on 2025 data, shows the same order: Duda at 85% and TYPO3 at 79% on mobile, WordPress at 45%, and Weebly as the only major platform that declined. It also found that WordPress runs about 64% of all CMS sites on mobile, so WordPress's figure is close to the web's own average almost by arithmetic.
Sanity and Contentful are headless: they hold the content, and a separate front end, often Next.js, renders the page. Their pass rates mostly describe those front ends.
Which metric each platform fails on
| Platform | LCP | INP | CLS | Time to first byte |
|---|---|---|---|---|
| Shopify | 87.7% | 86.7% | 92.4% | 83.8% |
| Wix | 84.8% | 92.6% | 97.9% | 74.3% |
| Framer Sites | 83.4% | 80.3% | 94.2% | 69.3% |
| Webflow | 78.6% | 88.0% | 93.6% | 73.1% |
| HubSpot CMS Hub | 78.3% | 86.2% | 81.2% | 67.3% |
| Drupal | 76.1% | 89.1% | 88.6% | 45.6% |
| WordPress | 55.4% | 90.3% | 87.3% | 22.9% |
| Next.js | 53.5% | 59.7% | 73.5% | 45.6% |
| Adobe Experience Manager | 52.3% | 66.8% | 72.3% | 44.1% |
| All origins | 65.6% | 79.4% | 83.4% | 45.9% |
Source: HTTP Archive Core Web Vitals Technology Report, September 2026; time to first byte is not a Core Web Vital, good is 0.8 seconds or less
The failures have different causes. WordPress sites are responsive once loaded, 90.3% good INP, but only 22.9% get the first byte back within 0.8 seconds, and LCP follows. That points at hosting and caching, not the editor. Hosted builders serve from their own CDNs and score 69% to 84% on time to first byte. Next.js and Adobe Experience Manager lose on responsiveness, with INP good on fewer than seven in ten mobile origins, which is the cost of large JavaScript bundles running on a phone.
Busy sites narrow the gap between platforms
| Platform | All origins | Top 1 million | Top 1 million origins counted |
|---|---|---|---|
| Wix | 80.9% | 52.5% | 823 |
| Framer Sites | 70.8% | 42.3% | 359 |
| Webflow | 71.3% | 49.8% | 2,635 |
| Squarespace | 71.0% | 56.3% | 465 |
| Astro | 71.1% | 54.5% | 3,087 |
| HubSpot CMS Hub | 60.0% | 45.2% | 704 |
| Shopify | 76.2% | 63.7% | 47,954 |
| Drupal | 64.7% | 58.6% | 11,658 |
| WordPress | 48.3% | 50.1% | 110,537 |
| Next.js | 35.7% | 25.5% | 39,764 |
| Contentful | 38.1% | 28.0% | 2,259 |
| All origins | 52.8% | 46.6% | 747,521 |
Source: HTTP Archive Core Web Vitals Technology Report, Top 1M rank filter, September 2026
Among the top 1 million origins, Wix drops 28.4 points and Framer 28.5. Webflow drops 21.5. WordPress is the only platform in the table that does better among busy sites. Desktop holds up better: Webflow passes on 72.3% of top 1 million desktop origins and Framer on 74.5%.
The pattern is consistent with the explanation in our Website Speed Statistics 2026 post: busy sites carry more tags, chat, personalisation and consent tooling, and those scripts hurt INP most. Busy Framer sites have good INP on only 54.2% of mobile origins, and busy Webflow sites on 68.5%. For a B2B company with real traffic and a marketing stack, the top 1 million column is the fairer benchmark.
Pass rates by JavaScript framework
| Framework | Mobile origins | Mobile pass | Desktop pass | Mobile, September 2025 |
|---|---|---|---|---|
| Hugo | 6,043 | 78.2% | 83.5% | 70.2% |
| Astro | 47,277 | 71.1% | 84.6% | 61.6% |
| Svelte | 92,950 | 53.6% | 65.7% | 51.3% |
| React | 1,283,019 | 43.9% | 54.7% | 42.7% |
| Gatsby | 9,278 | 43.7% | 59.5% | 45.2% |
| Vue.js | 609,131 | 36.2% | 50.9% | 35.8% |
| Next.js | 395,345 | 35.7% | 56.9% | 28.1% |
| Remix | 4,547 | 32.3% | 53.4% | 28.9% |
| Nuxt.js | 93,276 | 28.8% | 42.2% | 24.8% |
| Angular | 116,296 | 16.5% | 27.2% | 14.7% |
Source: HTTP Archive Core Web Vitals Technology Report, September 2025 and September 2026
The frameworks that ship static HTML by default, Hugo and Astro, pass most often. The ones that ship an application to the browser pass least, and the gap is widest on mobile: Next.js passes on 56.9% of desktop origins and 35.7% of mobile ones. Next.js gained 7.6 points in a year, more than any other framework except Astro and Hugo, but it still trails the web average by 17.1 points on mobile.
A bias to state: we build marketing sites in Webflow and applications in Next.js, so both sit in our own stack. The Next.js figure includes logged-in apps, dashboards and stores alongside marketing pages, which pulls it down. It is still the honest reason we do not default to Next.js for a marketing site.
Ecommerce platforms. The same split appears in commerce. Shopware passed on 84.0% of mobile origins, Wix eCommerce on 79.5% and BigCommerce on 62.5%, against Magento at 44.6%, WooCommerce at 39.3% and Salesforce Commerce Cloud at 38.4%. WooCommerce has the weakest server response of any large platform here: 12.1% of mobile origins have a good time to first byte. The Web Almanac's 2025 ecommerce chapter found the same ranking, with Shopify at 76% and WooCommerce at 35% on mobile.
Page builders on WordPress. Elementor, the largest, passes on 36.3% of its 877,513 mobile origins. Divi passes on 41.4%, WPBakery on 38.7% and the native block editor on 52.0%.
Pass rates by industry, using platforms as proxies
The Chrome UX Report has no industry field. Its documented dimensions are form factor, country and network round-trip time, so no public dataset gives a true Core Web Vitals pass rate for law firms, SaaS companies or hotels. The nearest proxy is a technology that only one industry uses. The table below uses that proxy, so read each row as "sites in this industry that use this tool", not as the industry itself.
| Industry | Platform used as proxy | Mobile origins | Mobile pass | Desktop pass |
|---|---|---|---|---|
| Online retail | Shopify | 445,297 | 76.2% | 79.5% |
| Online retail | WooCommerce | 542,541 | 39.3% | 39.3% |
| Restaurants | Popmenu | 3,375 | 71.6% | 63.3% |
| Restaurants | Bentobox | 3,804 | 19.4% | 66.1% |
| Healthcare booking | Zocdoc | 516 | 56.0% | 67.3% |
| Hotels | Cloudbeds | 2,262 | 56.1% | 62.0% |
| Hotels | The Hotels Network | 4,495 | 36.4% | 41.2% |
| Employers hiring | Lever | 663 | 53.5% | 67.8% |
| Wellness and fitness | MindBody | 634 | 53.3% | 46.8% |
| Nonprofits | Donorbox | 1,901 | 52.7% | 58.2% |
| Online courses | Teachable | 1,795 | 36.6% | 49.6% |
| Online courses | Thinkific | 1,101 | 22.9% | 35.4% |
| Real estate | iHomefinder IDX | 287 | 27.9% | 27.0% |
| Car dealers | PureCars | 1,876 | 2.8% | 17.0% |
Source: HTTP Archive Core Web Vitals Technology Report, September 2026; platform descriptions from the report's technology list. Proxy data: one tool per industry, not a sample of the industry.
Three readings hold up despite the proxy. The platform still matters more than the industry: two restaurant platforms sit 52 points apart on mobile. Bentobox's mobile problem is layout shift, with only 22.8% of origins good on CLS while 90.1% are good on LCP. Car-dealer sites on PureCars fail almost everywhere on mobile, with 7.0% good INP and 15.0% good CLS, the worst result of any platform we read. Property search is heavy too: iHomefinder sites have good LCP on 36.0% of mobile origins.
For B2B companies there is no clean proxy. HubSpot CMS Hub, at 60.0% on mobile, is the closest thing to a B2B marketing-site platform in the data, and we would treat it as a rough benchmark, not a sector figure.
Pass rates by country
| Country | Mobile origins | Mobile pass | Desktop pass |
|---|---|---|---|
| Sweden | 77,176 | 72.1% | 72.4% |
| Japan | 606,580 | 71.4% | 76.9% |
| Germany | 586,975 | 69.2% | 73.0% |
| United Kingdom | 412,823 | 67.6% | 69.3% |
| Canada | 253,874 | 67.4% | 66.8% |
| Netherlands | 221,157 | 66.6% | 72.1% |
| Poland | 273,724 | 63.5% | 69.8% |
| United States | 1,288,504 | 63.3% | 65.4% |
| France | 417,781 | 62.6% | 69.5% |
| Serbia | 56,127 | 61.9% | 67.5% |
| Italy | 376,767 | 53.6% | 65.6% |
| Spain | 335,434 | 53.1% | 64.2% |
| Australia | 166,124 | 50.7% | 56.1% |
| Mexico | 224,290 | 40.5% | 55.4% |
| Brazil | 598,869 | 33.7% | 55.4% |
| India | 745,962 | 31.6% | 50.9% |
| Indonesia | 383,720 | 29.4% | 53.6% |
Source: HTTP Archive Core Web Vitals Technology Report, geography filter, September 2026
Country here is where the visitors are, not where the site is hosted. The same site is measured separately for each country its visitors come from, and visitors on slower phones and networks pull a site down. A site with mostly US and UK visitors should expect a mobile pass rate in the mid 60s as the local norm; one selling into India or Brazil should plan for real devices that are far slower, with good mobile LCP on only 41.2% of Indian origins. Australia is the outlier among rich markets, with 36.8% good time to first byte, which suggests distance from servers rather than devices.
What the ranking effect is worth
Google's page experience documentation, updated in September 2026, says Core Web Vitals are used by its ranking systems, and that Search always tries to show the most relevant content even when page experience is poor. Treat passing as a tie-breaker, not a lever. The commercial case for speed rests on conversion, and our Website Speed Statistics 2026 post sets out the controlled tests behind it.
What this means for choosing a platform
Pick the platform with your real traffic in mind. On all origins, hosted builders such as Webflow, Wix and Shopify pass far more often than WordPress or enterprise CMSs. Among the top 1 million sites the lead of Webflow, Wix and Framer narrows to a few points or reverses, because by then the scripts on the page decide the result, not the platform under it. Set a field target in the brief, an LCP of 2.5 seconds or less and an INP of 200 milliseconds or less at the 75th percentile on mobile, and budget the third-party scripts against it. We build marketing sites in Webflow, so read the platform rows with that bias in mind; our marketing website service page explains how we scope a build with a speed budget in it.
