Owners ask this in two forms: "is my site slow?" and "does it matter?" Both have measurable answers. Google publishes the thresholds it uses, the HTTP Archive publishes how the rest of the web scores against them, and Google and Deloitte measured what a tenth of a second is worth. This article puts those together and ends with a check you can run yourself in ten minutes, no account needed.

In brief
  • Three numbers: LCP under 2.5 s, INP under 200 ms, CLS under 0.1, measured for the slowest quarter of real visitors on mobile and desktop separately.[1]
  • Half the web fails on phones. 48% of mobile sites passed all three in 2025, up from 36% in 2023; desktop sat at 56%.[2]
  • Loading is the usual miss. Only 62% of mobile sites reach a good LCP; responsiveness and stability pass more often.[2]
  • Speed moves money. In a Deloitte study of 37 brands, a 0.1-second improvement went with 8.4% more retail conversions.[3]
  • Google says it matters for ranking. Core Web Vitals are the first question on its page-experience checklist.[4]

The three numbers Google actually uses

Google calls them Core Web Vitals. Each one measures something a visitor feels, and each has a published threshold for "good".[1]

MetricWhat the visitor feelsGood
Largest Contentful Paint (LCP)How long until the main photo or headline is on screen2.5 seconds or less
Interaction to Next Paint (INP)How long after a tap before something visibly happens200 milliseconds or less
Cumulative Layout Shift (CLS)How much the page jumps around while loading0.1 or less

The detail that trips people up is who is measured. Google judges a page by the 75th percentile of real visits, split by mobile and desktop.[1] In plain terms: three out of four visitors have to get a good experience, and the phone number counts separately from the laptop number. A site that feels instant on the office Wi-Fi can still fail, because the people who matter are on a two-year-old phone on a weak cellular signal.

How many sites pass

The HTTP Archive's 2025 Web Almanac measured millions of sites against those thresholds using real Chrome user data from July 2025.[2] On mobile, 48% of sites had good Core Web Vitals, up from 44% in 2024 and 36% in 2023. On desktop the figure was 56%, almost unchanged from the year before.[2]

48%
Share of websites with good Core Web Vitals on mobile in 2025. Desktop: 56%. Source: HTTP Archive Web Almanac 2025, Chrome UX Report data from July 2025.

The miss is almost always loading. Only 62% of mobile sites reach a good LCP, against 74% on desktop.[2] Responsiveness (INP) passes on about 77% of sites, and visual stability (CLS) on 81% of mobile pages.[2] So if your site fails, the first place to look is what has to download before the main image appears.

Two Almanac figures explain why. Only 15% of mobile pages pass Lighthouse's render-blocking resources audit, meaning 85% load scripts or stylesheets that hold up the first paint.[2] And about 16–17% of pages lazy-load their own LCP image, which tells the browser to wait on the one thing the visitor is waiting for.[2] Both are common defaults in page builders and theme plugins.

What a tenth of a second is worth

In 2020 Deloitte Digital, in a study commissioned by Google, studied 37 leading European and American brand sites across retail, travel, luxury and lead generation, covering more than 30 million user sessions. Where every page in the journey improved by 0.1 seconds across four speed metrics, retail conversions rose 8.4% and travel conversions 10.1%.[3] The study is observational, not a controlled experiment, so read it as "faster sites sold more", not as a guaranteed return. The direction, though, matches every other dataset on the subject.

Does it affect Google rankings?

Yes, as one factor among many. Google's page-experience documentation says its core ranking systems look to reward content that provides a good page experience, and the first item on its self-assessment list is "Do your pages have good Core Web Vitals?"[4] Speed will not lift a thin page above a good one. It does decide ties, and it decides whether the visitor who arrives stays long enough to call.

Why most small business sites are slow

Not the hosting. The page. A typical builder or theme site loads a page-builder framework, a slider, two or three font families from a CDN, a chat widget, an analytics tag, a cookie banner, a form plugin, and a hero image saved at 4,000 pixels wide. Each one is small on its own. Together they are several megabytes and a dozen render-blocking requests before the headline paints. The Almanac's median mobile page spends 1.9 seconds with the main thread blocked, which is why taps feel dead on phones.[2]

How we build for speed
Every site we ship is plain HTML and CSS: no builder, no plugin stack, no jQuery. Fonts are self-hosted so there is no third-party request and no render-blocking stylesheet. Images are sized to the slot they fill, and the hero image is never lazy-loaded. No external scripts, no cookies, no tracking pixels. We check every page at phone width and desktop width before it goes live. Rampfinity.com itself is built this way: no external scripts, no third-party fonts, no cookies. With nothing loaded from elsewhere, there is nothing to slow down later.

Check your own site in ten minutes

No login, no tools to install.

If the field data shows a red LCP, the fixes in order of payoff are: shrink and correctly size the hero image, remove render-blocking scripts and third-party fonts, and remove lazy-loading from the first image. If it shows a red INP, the cause is almost always JavaScript from plugins and widgets; removing is faster than optimising. If the site is a builder site with a dozen plugins, it is often cheaper to rebuild it as a plain page than to tune it, which is what our website service does, from $1,000 with the price agreed first.

Speed is one half of a site that wins work. The other is having a clear page for each thing you do, which we cover in why every service needs its own page.

Every site we build is plain HTML with self-hosted fonts and no third-party scripts, and it is checked at phone width before it ships. Tell us what you need and we will quote it in writing.

Get your free AI visibility score →

Frequently asked questions

What is a good page load time for a small business website?
Google's threshold is a Largest Contentful Paint of 2.5 seconds or less for 75% of real visits, measured on mobile and desktop separately. In practice, aim for the main headline or photo on screen within about two seconds on a phone over cellular.
How do I check if my website passes Core Web Vitals?
Paste your address into PageSpeed Insights and read the real-user section on the Mobile tab, or open the Core Web Vitals report in Google Search Console if your site is verified. Both use Chrome's real-visitor data.
Does website speed affect Google rankings?
Yes, as one factor among many. Google's page-experience guidance says its core ranking systems reward good page experience, and good Core Web Vitals is the first item on its checklist. Speed decides ties and keeps visitors on the page; it does not replace useful content.
Why is my website slow on mobile but fine on my computer?
Google measures phones and desktops separately, and phones are slower devices on weaker connections. Page-builder frameworks, plugins, third-party fonts and oversized images that a laptop absorbs easily can double or triple load time on a mid-range phone.
Will faster hosting fix a slow website?
Rarely on its own. Time to first byte is one part of loading, but most failures come from what the page loads after that: render-blocking scripts, third-party fonts, widgets and large images. Removing those usually does more than any hosting upgrade.
Why does every service need its own page? → How much time can AI automation save a small business? → How do you check whether AI recommends your business? →

Sources

  1. web.dev (Google). "Web Vitals." Updated Oct 31, 2024. https://web.dev/articles/vitals
  2. HTTP Archive. "Web Almanac 2025, Part II Chapter 7: Performance." Jan 15, 2026. https://almanac.httparchive.org/en/2025/performance
  3. web.dev (Google), summarising Deloitte Digital. "Milliseconds make millions." Jun 24, 2020. https://web.dev/case-studies/milliseconds-make-millions
  4. Google Search Central. "Understanding page experience in Google Search results." 2026. https://developers.google.com/search/docs/appearance/page-experience