top of page

Half of Websites Still Fail Google's Core Web Vitals on Mobile. Speed Won't Rescue Weak Content, but It Settles Close Calls

Nov 14
3 min read

Updated: 5 days ago

Just under half of websites pass all three of Google's Core Web Vitals on mobile. The 2025 Web Almanac, published by HTTP Archive in January 2026 using Chrome UX Report data from July 2025, puts the mobile pass rate at 48%, up from 44% a year earlier. Desktop does slightly better at 56%. Put the other way round, more than half of sites give mobile visitors a slower, jumpier experience than Google considers good.


Those three checks are not abstract. Largest Contentful Paint measures how long the main content takes to appear, and the threshold for good is 2.5 seconds. Interaction to Next Paint measures how quickly a page responds to a tap or click, with a target under 200 milliseconds. Cumulative Layout Shift tracks how much the page jumps about while it loads, and should stay below 0.1. Google's own Core Web Vitals guidance, last updated in December 2025, sets those numbers out plainly.


Loading speed is where most sites fall down. Only 62% of mobile origins met the LCP threshold in the Almanac data, against 77% for INP and 81% for layout shift. On desktop the order changes, with LCP rising to 74% and layout shift the weakest at 72%. On an engineering or manufacturing site the usual causes are familiar: oversized hero photos of the factory floor, video that loads on every page whether anyone plays it or not, and sliders and PDF viewers bolted onto a template that was never built for them.


Before anyone commissions a rebuild, there is a caveat worth taking seriously. Google does not say speed wins rankings. Its page experience documentation, updated in September 2026, states that "Google Search always seeks to show the most relevant content, even if the page experience is sub-par", and that there is no single page experience signal. A fast page with nothing useful on it will not outrank a slower one that answers the buyer's question. Anyone selling Core Web Vitals work as a quick ranking fix is overselling it.


Where it does count is in close calls. The same guidance says page experience becomes more of a factor when several pages are similarly helpful. For a specialist supplier competing with three or four firms that have comparable capability pages, that is often exactly the position. It matters away from Google too. A procurement engineer opening your site on a phone between meetings, or someone checking a datasheet from a customer's site, is unlikely to wait for a heavy banner image before heading back to the search results.


The Almanac figures cover millions of websites worldwide, not UK B2B firms in particular, so they describe the web in general rather than your sector. The only way to know where your own site stands is to check it. Google's free PageSpeed Insights tool and the Core Web Vitals report in Search Console both show field data from real Chrome users where a site has enough traffic, and PageSpeed Insights adds a lab test either way. Many small industrial sites will not have enough visits for field data, which is worth knowing before reading too much into one lab score.


The fixes are usually unglamorous. Compress and correctly size images, stop video loading until someone presses play, remove third-party scripts nobody remembers adding, and test on a mid-range phone over mobile data rather than the office connection. Most of this can be done inside a website builder without starting again.


Speed is the smaller half of a bigger question, which is whether your site appears in search at all. We covered that in Why Most UK Engineering and Manufacturing Websites Never Show Up in Search. At Nebula, website work for technical SMEs starts with the content buyers actually come for, specifications and proof of past work, and then makes sure the page carrying it loads quickly on whatever device the buyer is holding. Speed on its own rarely wins the work. A slow page can still lose it.

Comments


Still have questions?

Get in touch and we'll answer directly.

bottom of page