Website Speed Statistics 2026: Core Web Vitals, Page Weight and Sales

These website speed statistics for 2026 show where real sites stand on Google's Core Web Vitals, how heavy pages have become, which platforms are fastest, and what speed does for sales. Every number links to its source and gives the period it covers. Most come from HTTP Archive and Google's Chrome UX Report, which measure millions of real sites using data from real Chrome visitors.

We wrote this for US small business owners and the people who build their sites. After each group of numbers we say what it means in practice. Figures were checked on October 23, 2026.

The key numbers at a glance

About 63% of US websites pass Google's Core Web Vitals on mobile, which means more than a third do not. That figure comes from Chrome's real-user data for September 2026. Here are the headline figures, each with its source and the period it covers.

Two notes before the detail. First, "pass" means a site is good on all three metrics at the 75th percentile of real visits, which is Google's own test. Second, the Tech Report and the Web Almanac measure in different months and slightly different ways, so their numbers differ. We give both and say which is which.

How many websites pass Core Web Vitals?

In September 2026, 63.3% of US websites passed Core Web Vitals on mobile and 65.4% passed on desktop. We calculated those shares from the official counts in the HTTP Archive Tech Report: 815,622 of 1,288,504 US origins on mobile, and 840,976 of 1,284,986 on desktop. An origin is a site's domain, such as https://example.com.

Worldwide the numbers are lower. The same data source gives 52.8% for mobile and 60.8% for desktop across all countries. Google's own Chrome UX Report release notes for the September 2026 dataset, published October 13, 2026, report that 55.9% of 18,376,044 origins had good Core Web Vitals across all devices combined.

The trend is slow improvement. The Web Almanac 2025 performance chapter, which uses July 2025 data, says good Core Web Vitals "reached 44% on mobile and 55% on desktop" in 2024 and "further improved in 2025 to 48% for mobile websites and 56% for desktop websites." The Chrome release notes show the all-device figure moving from 55.6% in August 2026 to 55.9% in September 2026.

What this means for you

If your site fails, you are in a large group, but not the majority. Roughly two in three US sites now pass on mobile. That makes failing a small but real disadvantage against local competitors who pass. It also means passing is achievable on ordinary budgets: most passing sites are not run by big tech teams.

Which metric fails most often?

Loading speed, measured by Largest Contentful Paint (LCP), is the metric most sites fail. In the US in September 2026, 74.8% of origins had good LCP on mobile, against 87.3% with good responsiveness (INP) and 85.1% with good layout stability (CLS). Those are our calculations from the Tech Report counts.

The Web Almanac tells the same story with July 2025 world data. It says "74% of desktop pages achieve a 'good' LCP score compared to 62% on mobile, with mobile also showing nearly double the rate of 'poor' experiences (13% versus 7%)." Mobile INP improved, "with 77% of websites achieving good scores, up from 74% in 2024", while desktop INP sat at 97%. For CLS, "72% of desktop pages and 81% of mobile pages" were good. That CLS result is one of the few places where phones do better than desktops.

Why LCP is the hard one

The largest thing on screen is usually a picture. The Web Almanac found that "85.3% of desktop pages and 76% of mobile pages" have an image as their LCP element. It also found that about 16% to 17% of pages lazy-load their LCP image, a figure "that has held steady since 2024." Lazy-loading the main hero image tells the browser to wait before fetching the one picture that matters most. Removing the lazy-load attribute from that one image is usually a one-line fix.

Server speed is the other half. Only 54.8% of US origins had a good Time to First Byte on mobile in September 2026. The Web Almanac's July 2025 figure was 44% of mobile sites worldwide. A slow server delays everything after it, including LCP.

What this means for you

Fix loading before anything else. In practice that means a properly sized hero image that is not lazy-loaded, decent hosting, and caching. Our Core Web Vitals guide explains each metric, and how to improve website speed walks through the fixes in order.

Which website platforms are fastest?

Hosted builders pass Core Web Vitals far more often than WordPress. In September 2026, 88.7% of US Wix sites, 85.5% of Shopify sites and 82.9% of Squarespace sites passed on mobile, against 62.0% of US WordPress sites. These are our calculations from the Tech Report's US counts for each platform.

The Web Almanac's CMS chapter, which uses worldwide mobile data, puts WordPress lower still: it "remains among the lowest-ranked with a pass rate of 45%." It also reports that Wix "saw a significant jump from 55% to 74%" and that Duda led with 85%. The US numbers are higher across the board because US sites, on average, pass more often than sites worldwide.

Why WordPress lags

WordPress is not slow by design. It is slow by default on cheap hosting, with heavy themes and many plugins. The clearest sign is server response time. Only 32.2% of US WordPress sites had a good Time to First Byte on mobile in September 2026, against 54.8% of all US sites. Responsiveness is not the problem: 92.6% of US WordPress sites had good INP on mobile.

WordPress also runs a large share of the web, so it includes millions of old, unmaintained sites. The Web Almanac says WordPress powers "roughly 64% of CMS-driven sites" in 2025. A big, mixed population drags down the average. A well-built WordPress site on good hosting can pass easily.

The right column is our explanation, not data from the source.

What this means for you

Platform choice matters less than how the site is built and hosted, but the defaults are real. If you are on WordPress, check your hosting and your plugin count before anything else. If you are choosing a platform now, our guide on how to choose web hosting covers the trade-offs.

How heavy are web pages in 2025 and 2026?

The median mobile home page weighed 2.6 MB in July 2025, up 8.4% from 2.4 MB a year earlier. The Web Almanac 2025 page weight chapter gives the median desktop page as 2.9 MB, up 7.3%. Pages keep getting heavier every year.

One caution: the same chapter also says the median mobile home page was "2,362 KB" in July 2025 when it compares against 2015. That conflicts with its 2.6 MB (2,559 KB) figure elsewhere. We use 2.6 MB because it appears in the chapter's main breakdown, but the source is not consistent.

Source: Web Almanac 2025, page weight chapter. The chapter gives a separate median of 646 KB of JavaScript for mobile pages in another breakdown.

JavaScript is the cost you feel

Images make up the most bytes, but JavaScript does the most damage on phones, because the phone has to run it. The Web Almanac reports that "the median TBT on mobile increased to 1,916 milliseconds in 2025, up 58% from 1,209 milliseconds in 2024." Total Blocking Time is the time the page is busy and cannot respond to a tap. Desktop rose from 67 to 92 milliseconds, which shows how much weaker phone processors are.

What this means for you

Count your scripts. Chat widgets, tag managers, review widgets, heatmaps and ad pixels each add code a phone must run. Remove the ones nobody uses. Then compress and resize images, which is usually the biggest and cheapest win. Our website speed optimization guide covers both.

How much traffic comes from phones?

In the US, phones and desktops now split traffic almost evenly. StatCounter reports that in September 2026, 49.1% of US web traffic came from mobile, 48.47% from desktop and 2.42% from tablets. Worldwide, mobile was 58.99%.

Browser share matters for how speed is measured. In the US in September 2026, StatCounter put Chrome at 53.82% and Safari at 30.42% across all devices. On phones alone, Chrome had 47.44% and Safari 46.38%. Google's Core Web Vitals field data comes from Chrome users, so the experience of iPhone users on Safari is not in those numbers. A site that fails in Chrome is very likely slow in Safari too, but you should test on a real iPhone as well.

What this means for you

Half your visitors are on a phone, and Google judges your site mainly on its mobile version. Design and test for the phone first. Check your pages on a mid-range Android phone, not just a new iPhone.

Does site speed affect Google rankings?

Yes, but only a little, and only as one signal among many. Google's page experience documentation, updated September 22, 2026, says "Core Web Vitals are used by our ranking systems." It also says "there is no single signal" and that good scores do not guarantee "that your pages will rank at the top of Google Search results." Google adds that it "always seeks to show the most relevant content, even if the page experience is sub-par."

The thresholds come from web.dev: LCP within 2.5 seconds, INP of 200 milliseconds or less, and CLS of 0.1 or less, measured at the 75th percentile of page loads. INP replaced First Input Delay as a Core Web Vital on March 12, 2024, according to the Google Search Central blog.

What this means for you

Do not expect a speed fix alone to move you from page three to page one. Content and relevance still decide most rankings. Speed matters more for what happens after the click: whether the visitor stays and contacts you. The SEO case is a bonus on top of the sales case.

Does a faster site make more money?

The best-known study says yes. In "Milliseconds Make Millions", a 2020 report commissioned by Google and prepared by Deloitte, a 0.1 second improvement in mobile site speed came with "an 8.4% increase in conversions with retail consumers" and "a 10.1% increase in conversions with travel consumers." For lead generation sites, "bounce rate improved by 8.3%."

Read that study with care. It used data from 37 brands, mostly in Europe and the US, collected from October 28 to November 24, 2019. The report states that "Deloitte LLP did not audit nor validate this data." It was paid for by Google, which sells speed tools and benefits from a faster web. The direction of the finding is consistent with other evidence, but the exact percentages are from one study six years ago.

The famous 53% statistic

You will see "53% of mobile visitors leave a page that takes more than three seconds to load" on many sites. The original Think with Google article is no longer online. The figure does appear in Google's Mobile Site Speed Playbook, which says "fifty-three percent of people will leave a mobile site if it takes more than three seconds to load." Its footnote dates the data to March 2016, from aggregated Google Analytics data. It is a decade old and was measured on a very different web. We would not build a business case on it.

Company case studies

These are large companies with real traffic, published by Google. Results on a small business site will be smaller in absolute terms and harder to measure.

What this means for you

Treat speed as a sales issue, not a technical score. A local service business does not need a 50-page performance plan. It needs a site that loads its main message fast on a phone and has a contact form that works. That is usually a few days of focused work, or part of a redesign if the site is old.

How many sites still miss the basics?

Most sites now get the basics right, but not all. The Web Almanac 2025 SEO chapter found HTTPS on "91.7% of desktop pages and 91.5% of mobile pages." So roughly one page in twelve still loads over plain HTTP. That matters more now. Chrome asks visitors for permission before it opens a public HTTP page, a change Google confirmed was on for all users on October 21, 2026, as we explain in our guide to Chrome HTTPS by default.

Other basics from the same chapter: the viewport meta tag, needed for a site to work on phones, appears on 95.2% of mobile pages. Title tags are on 98.54% of mobile pages. Meta descriptions are on only 67.2%. The performance chapter adds that "62% of pages fail to set dimensions on at least one image" on mobile, a common cause of layout shift.

What this means for you

Check the cheap things first. HTTPS, a viewport tag, image sizes and page titles cost almost nothing to fix and are still missing on a meaningful share of sites. Our 2026 website audit checklist lists them in order of value.

How we compiled this

We checked every figure against its source on October 23, 2026. Our rules:

• Primary sources only. HTTP Archive, Google's Chrome UX Report, Google's own documentation and case studies, and StatCounter. We dropped any statistic we could only find on a blog that did not name its origin.
• Tech Report shares are our calculation. The HTTP Archive Tech Report publishes counts of origins tested and origins with good scores. We divided one by the other and rounded to one decimal place. The data is from the September 2026 Chrome UX Report dataset, filtered to the United States unless we say "worldwide".
• We kept conflicts visible. Where a source gave two different numbers for the same thing, we said so.
• We excluded vendor surveys with no stated method, and speed statistics older than 2019 except where we name and date them as old.
• Field data, not lab scores. All Core Web Vitals figures are from real Chrome users, not Lighthouse lab tests. A Lighthouse score and a pass or fail in the field are different things.

We will update this page when the next Chrome UX Report dataset and the next Web Almanac are published.

What to do with these numbers

Find out which group your site is in, then fix the cheapest problems first. Here is a short checklist for a small business site:

• Open Google Search Console and check the Core Web Vitals report for mobile.
• Run your home page and your main service page through PageSpeed Insights and look at the field data section, not just the lab score.
• Make sure the hero image is compressed, has width and height set, and is not lazy-loaded.
• Remove scripts and plugins you no longer use.
• Check your server response time. If it is slow, talk to your host or move.
• Confirm the whole site loads on HTTPS with no warnings.
• Test the contact form on a real phone.

If the list turns into a rebuild, use our free website cost calculator for a planning range, or read our guide to small business website cost. And if you want a second opinion on what is worth fixing, contact us for a free, fixed-scope quote.

Speed is one half of how real people experience a site. Accessibility is the other. Our web accessibility statistics for 2026 cover that side.