Statistics · Statistics
What Is the Average Website Load Time in 2026?
Short answer
There is no single average website load time that fits every site, because Google now judges speed with three separate measures. In July 2025 data, 48% of mobile sites and 56% of desktop sites passed all three Core Web Vitals. A good Largest Contentful Paint is 2.5 seconds or less, measured at the 75th percentile of real visits.
Key takeaways
- In 2025, 48% of mobile sites and 56% of desktop sites passed all three Core Web Vitals.
- A good Largest Contentful Paint is 2.5 seconds or less at the 75th percentile of real visits.
- The median home page weighed 2.6 MB on mobile and 2.9 MB on desktop in July 2025.
- Mobile pages under 1 MB passed Core Web Vitals at 57%, against 30% for pages of 5 MB or more.
- Images are the largest content element on 76% of mobile pages and 85.3% of desktop pages.
- The median mobile page made 72 requests in 2025, and the median desktop page made 77.
What is the average website load time?
There is no single average load time for the whole web, because speed now depends on the metric, the device and the page.
The closest public answer comes from the Web Almanac 2025 Performance chapter, which used Chrome UX Report field data from July 2025. It found that 48% of mobile sites and 56% of desktop sites passed all three Core Web Vitals in 2025.
That pass rate is the headline number to use. It covers loading, responsiveness and visual stability together, not one lab run on one connection.
A site that passes is not average in the usual sense. It is in the faster half of the web, and the gap is wider on mobile than on desktop.
How is load time measured in 2025 and 2026?
Load time is measured from real Chrome users, not from a single test in a lab.
Google's Web Vitals guide explains that Core Web Vitals are field metrics, measured from real visits. The Chrome UX Report collects that anonymised real-user data and powers PageSpeed Insights, Chrome DevTools and the Search Console Core Web Vitals report.
Three metrics make up the set. Largest Contentful Paint covers loading, Interaction to Next Paint covers responsiveness, and Cumulative Layout Shift covers visual stability.
Each metric is judged at the 75th percentile of page loads, split by mobile and desktop. That means a page passes only when at least 3 in 4 real visits meet the target on that device type.
How fast is a good load time?
A good load time means a Largest Contentful Paint of 2.5 seconds or less at the 75th percentile.
The Web Almanac 2025 Performance chapter puts the bands at good under 2.5 seconds, needs improvement from 2.5 to 4.0 seconds, and poor over 4.0 seconds. The same bands appear in Google's Web Vitals guide.
First Contentful Paint uses a different scale. A good score is under 1.8 seconds, needs improvement is 1.8 to 3.0 seconds, and poor is over 3.0 seconds.
The other two thresholds are an Interaction to Next Paint of 200 milliseconds or less, and a Cumulative Layout Shift score of 0.1 or less.
If you want a single target for WordPress, start with LCP, then check the other two. Our guide to how fast a WordPress site should load puts those targets in a WordPress setting.
| Metric | Good | Needs improvement | Poor |
|---|---|---|---|
| Largest Contentful Paint | 2.5 seconds or less | 2.5 to 4.0 seconds | Over 4.0 seconds |
| First Contentful Paint | Under 1.8 seconds | 1.8 to 3.0 seconds | Over 3.0 seconds |
| Interaction to Next Paint | 200 ms or less | 200 to 500 ms | Over 500 ms |
| Cumulative Layout Shift | 0.1 or less | Over 0.1 to 0.25 | Over 0.25 |
How many websites pass Core Web Vitals in 2025?
In 2025, 48% of mobile sites and 56% of desktop sites passed all three Core Web Vitals.
The Web Almanac 2025 Performance chapter reports those rates from July 2025 Chrome UX Report data. A site passes only when it is good on LCP, INP and CLS at the same time.
Loading is the weakest link on mobile. Good LCP was 62% on mobile and 74% on desktop, while poor LCP was 13% on mobile and 7% on desktop.
Responsiveness and stability scored higher. Good INP was 77% on mobile and 97% on desktop, and good CLS was 81% on mobile and 72% on desktop.
| Measure | Mobile good in 2025 | Desktop good in 2025 | Source and year |
|---|---|---|---|
| All three Core Web Vitals | 48% | 56% | Web Almanac 2025 |
| Largest Contentful Paint | 62% | 74% | Web Almanac 2025 |
| Interaction to Next Paint | 77% | 97% | Web Almanac 2025 |
| Cumulative Layout Shift | 81% | 72% | Web Almanac 2025 |
If your site fails, start with LCP on mobile. That is where most sites lose the pass, and where a slow hero image or server shows up first. See how to pass Core Web Vitals in WordPress for the fix order that follows from these numbers.
How heavy is the average web page in 2025?
The median home page weighed 2.6 MB on mobile and 2.9 MB on desktop in July 2025.
The Web Almanac 2025 Page Weight chapter reports a median mobile home page of 2,362 KB, up from 845 KB in July 2015. That is a rise of 202.8% in 10 years, while the median desktop home page rose 110.2% over the same period.
Growth did not stop in 2025. The median home page grew 8.4% year over year on mobile, from 2.4 MB in 2024, and 7.3% on desktop, from 2.7 MB in 2024.
Images and JavaScript take most of the bytes. On a median mobile home page in 2025, images were 911 KB and JavaScript was 632 KB, with fonts at 122 KB, CSS at 77 KB and HTML at 22 KB.
Desktop pages were heavier in the same parts. The median desktop home page used 1,058 KB of images and 697 KB of JavaScript, with fonts at 139 KB, CSS at 82 KB and HTML at 22 KB.
The median page also made 72 requests on mobile and 77 on desktop in 2025. Requests grew 9% year over year on mobile and 8% on desktop, so pages are adding files as well as bytes.
How does page weight change the chance of passing?
Lighter pages pass far more often than heavy pages, on both mobile and desktop.
The Web Almanac 2025 Page Weight chapter found that 57% of mobile home pages under 1 MB passed all Core Web Vitals, against only 30% at 5 MB or more. On desktop, the same split was 70% under 1 MB against 38% at 5 MB or more.
The pattern holds for LCP alone. The lightest pages had good LCP on 82% of desktop pages and 72% of mobile pages, while pages of 5 MB or more fell to 57% on desktop and 44% on mobile.
Pages in the middle tell the same story. Pages of 2 to 3 MB, close to the median weight, had good LCP on 71% of desktop pages and 59% of mobile pages.
Weight is linked to the pass rate, not a part of the score itself. Google does not grade page weight directly, but heavy pages tend to carry the images, scripts and video that slow LCP and block the main thread.
Why is mobile slower than desktop?
Mobile is slower because the same page meets weaker devices, weaker networks and larger image loads.
The Web Almanac 2025 Performance chapter shows the gap in every loading measure. Good LCP was 62% on mobile against 74% on desktop, and poor LCP was nearly double on mobile, at 13% against 7%.
The main thread gap is wider still. Median Total Blocking Time was 1,916 milliseconds on mobile in 2025, up 58% from 1,209 milliseconds in 2024. Desktop median blocking time was 92 milliseconds, up from 67 milliseconds.
At the 90th percentile, mobile users faced over 7.5 seconds of blocking time before the page became fully interactive. That is a lab measure, not a field pass rate, but it shows why taps can feel slow even when the page looks loaded.
Images explain much of the LCP gap. An image is the largest content element on 76% of mobile pages and 85.3% of desktop pages, so the hero image often decides the LCP time. Our guide to how to improve LCP in WordPress starts with that element for the same reason.
How has load time changed since 2021?
The web has improved on Core Web Vitals since 2021, but the gains are uneven and pages keep getting heavier.
The Web Almanac 2025 Performance chapter reports mobile pass rates of 36% in 2023, 44% in 2024 and 48% in 2025. Desktop rose from 48% in 2023 to 55% in 2024, then only to 56% in 2025.
Stability improved most over the longer run. Good CLS on desktop rose from 62% in 2021 to 72% in 2025, while mobile reached 81% in 2025.
Responsiveness also improved in 2025. Good INP on mobile rose from 74% in 2024 to 77% in 2025, and the top 1,000 mobile sites jumped from 53% to 63% good INP in the same year.
The counter-trend is weight. The Web Almanac 2025 Page Weight chapter reports median video bytes of 315 KB per page in 2025, up 28% from 246 KB in 2024. Median unused JavaScript was 251 KB on mobile and 280 KB on desktop, measured uncompressed.
That mix explains the headline: field scores are better, yet the median page is heavier and the median mobile blocking time rose sharply. Faster devices and browsers are masking heavier pages, not removing the weight. For the search side of the same trend, see does page speed affect SEO.
How do you check your own load time?
Check field data first, then use a lab test only to find the cause.
Google's Web Vitals guide lists PageSpeed Insights and the Search Console Core Web Vitals report as field tools, with Chrome DevTools and Lighthouse as lab tools. Field data covers the last 28 days of real visits, while a lab test is one simulated load.
Open PageSpeed Insights, enter your URL, and read the mobile result first. The report shows the 75th percentile for LCP, INP and CLS, plus a pass or fail for all three together.
Then open the Core Web Vitals report in Search Console. Google's Search Central docs on Core Web Vitals describe the same field assessment, grouped by URL pattern, so you can see which templates fail instead of single pages. The report is at Google Search Central.
If a page has too little traffic for field data, use the lab result as a guide only. It can show a slow hero image or a blocked main thread, but it cannot prove a pass. For the steps after the check, see how to speed up WordPress.
What mistakes should you avoid?
Do not quote a single average load time as a target for your site.
Averages hide the device split, the page type and the metric. Home pages pass less often than inner pages, by 11 points on mobile and 14 points on desktop in the 2025 Almanac data, so a home page average will mislead you on a product or post page.
Do not lazy-load the hero image. About 16% to 17% of pages lazy-load their LCP image, according to the Web Almanac 2025 Performance chapter, and that delays the exact content the visitor is waiting to see.
Do not rely on a lab score alone. A lab test uses one device and one connection, while the pass uses the 75th percentile of real visits over 28 days.
Do not chase a single number while ignoring weight. JPG was still 57% of LCP images in 2025, WebP rose only 4 points since 2024, and only 17% of mobile pages with LCP images used fetchpriority=“high”, up from 15% in 2024.
Frequently asked questions
What is a good average website load time?
A good target is a Largest Contentful Paint of 2.5 seconds or less at the 75th percentile of real visits. That is Google's good threshold for LCP. In 2025, 62% of mobile sites and 74% of desktop sites met it, so it is a realistic target, not an elite one.
Is 3 seconds a good load time?
It depends on the metric. For LCP, 3 seconds falls in the needs improvement band, which runs from 2.5 to 4.0 seconds. For First Contentful Paint, 3 seconds is the edge of poor, since poor starts over 3.0 seconds. Aim for LCP under 2.5 seconds instead.
Why does my site feel slow even with a good PageSpeed score?
A lab score is one simulated load on one device and connection. Core Web Vitals use field data from real visitors over 28 days, at the 75th percentile. Your visitors may use slower phones and networks than the lab, and blocking JavaScript can delay taps after the page looks loaded.
Does page weight really affect load time?
Yes, in the published data. In 2025, 57% of mobile home pages under 1 MB passed all Core Web Vitals, against 30% of pages of 5 MB or more. The link is a correlation in the Almanac data, not a direct score, but lighter pages pass at nearly double the rate.
How often should I check my site load time?
Check field data monthly, and after any theme, plugin or image change that affects the home page or main templates. Field data covers 28 days, so a fix needs about four weeks to show fully in the report. Use a lab test right after a change to spot a new problem early.
Sources
- Performance | 2025 | The Web Almanac by HTTP Archive HTTP Archive
- Page Weight | 2025 | The Web Almanac by HTTP Archive HTTP Archive
- Web Vitals web.dev
- Chrome User Experience Report Chrome for Developers
- Understand Core Web Vitals Google Search Central
Is your product missing, or something out of date?
Tell us what to add or correct. We read every note and reply when we need a source. A message does not buy a ranking.
Thanks — we review requests within 3 business days.
