Web Design Statistics 2026: Accessibility, Page Weight, Mobile and Pay
Most websites are still hard to use for disabled people, heavy and slow on phones. WebAIM found detectable WCAG 2 failures on 95.9% of the top 1 million home pages in February 2026, up from 94.8% in 2025. The median mobile home page weighed 2,559 KB in July 2025, according to HTTP Archive, and images were about 35.6% of that, calculated by StatisticsGlobal. Only 48% of mobile websites had good Core Web Vitals in 2025. In the US, web developers and digital designers earned a median $99,520 in May 2025, according to the US Bureau of Labor Statistics.
Key statistics
- Detectable WCAG 2 failures were found on 95.9% of the top 1 million home pages in 2026, according to WebAIM.
- The median mobile home page weighed 2,559 KB in July 2025, according to HTTP Archive.
- Images made up about 35.6% of the median mobile home page weight in 2025, calculated by StatisticsGlobal from HTTP Archive data.
- Good Core Web Vitals were reached by 48% of mobile websites in 2025, up from 44% in 2024, according to HTTP Archive.
- Web developers and digital designers in the US earned a median $99,520 a year in May 2025, according to the US Bureau of Labor Statistics.
Web accessibility statistics
WebAIM scans the home pages of the top 1 million websites every February. HTTP Archive runs Lighthouse accessibility audits on millions of sites each July. Both use automated tests, which catch only part of the problems a disabled visitor meets.
Web accessibility went backwards in 2026. WebAIM found detectable WCAG 2 failures on 95.9% of top home pages, up from 94.8%. Errors per page rose from 51 to 56.1. That ends six years of small gains. Low-contrast text did most of the damage. It jumped 4.8 points in a year, back to where it was in 2022.
Pages are getting bigger and more complex. The average home page had 1,437 elements and 66.6 images in 2026. ARIA code, which is meant to help screen readers, was on 82.7% of pages. Yet pages with ARIA had 40.7% more errors than pages without it. WebAIM links the rise to heavier frameworks and AI-assisted coding. Tools matter too. WordPress home pages had 60% more errors than Squarespace ones, though WebAIM warns the tool is not always the cause. See our WordPress statistics for how widely it is used.
The two big sources disagree on how bad contrast is. HTTP Archive found only 30% of sites passed its contrast audit in 2025, which implies about 70% fail. WebAIM found low-contrast text on 83.9% of home pages. The 13.9-point gap comes from different tools, samples and years. Both agree that most sites fail. HTTP Archive also found 67% of sites remove default focus outlines, which keyboard users rely on.
- Low-contrast text
- Missing alt text
Source: WebAIM, The WebAIM Million 2026.
-
Detectable WCAG 2 failures were found on 95.9% of the top 1 million home pages in 2026, according to WebAIM.
-
Detectable WCAG 2 failures were found on 94.8% of the top 1 million home pages in 2025, according to WebAIM.
-
The top 1 million home pages averaged 56.1 detectable accessibility errors each in 2026, up from 51 in 2025, according to WebAIM.
-
Low-contrast text was found on 83.9% of the top 1 million home pages in 2026, according to WebAIM.
-
Images with missing alternative text were found on 53.1% of the top 1 million home pages in 2026, according to WebAIM.
-
Missing form input labels were found on 51% of the top 1 million home pages in 2026, according to WebAIM.
-
Empty links were found on 46.3% of the top 1 million home pages in 2026, according to WebAIM.
-
The top 1 million home pages averaged 1,437 page elements each in 2026, a 14.3% rise in one year, according to WebAIM.
-
The top 1 million home pages averaged 66.6 images each in 2026, and 16.2% of those images had no alternative text, according to WebAIM.
-
ARIA was used on 82.7% of the top 1 million home pages in 2026, up from 79.4% in 2025, according to WebAIM.
-
Home pages using ARIA had 40.7% more detectable accessibility errors than pages without it in 2026, calculated by StatisticsGlobal from WebAIM data.
How we calculated it: Average errors on home pages with ARIA (59.1) / average errors on pages without ARIA (42) = 1.407, so 40.7% more. WebAIM notes pages with ARIA were also more complex.
-
The share of top home pages with low-contrast text rose 4.8 percentage points between 2025 and 2026, calculated by StatisticsGlobal from WebAIM data.
How we calculated it: Low-contrast text on 83.9% of home pages in 2026 minus 79.1% in 2025 = 4.8 points.
-
The share of top home pages missing image alt text fell 14.9 percentage points between 2019 and 2026, calculated by StatisticsGlobal from WebAIM data.
How we calculated it: Missing alt text on 68% of home pages in 2019 minus 53.1% in 2026 = 14.9 points.
-
WordPress home pages had 60% more detectable accessibility errors than Squarespace home pages in 2026, calculated by StatisticsGlobal from WebAIM data.
How we calculated it: Average errors on WordPress home pages (52.8, 252,302 pages) / Squarespace home pages (33.0, 2,669 pages) = 1.60, so 60% more. WebAIM notes errors cannot always be attributed to the technology.
-
The median Lighthouse accessibility score was 85% in 2025, up from 83% in 2019, according to HTTP Archive.
-
Only 30% of websites passed the Lighthouse colour contrast audit on desktop and mobile in 2025, according to HTTP Archive.
-
The share of websites passing the colour contrast audit rose 8 percentage points between 2019 and 2025, calculated by StatisticsGlobal from HTTP Archive data.
How we calculated it: Sites with sufficient colour contrast: 30% in 2025 minus 22% in 2019 = 8 points.
-
HTTP Archive implies 70% of sites fail colour contrast in 2025, 13.9 points below the 83.9% WebAIM found in 2026, calculated by StatisticsGlobal from HTTP Archive and WebAIM data.
How we calculated it: HTTP Archive: 100% minus 30% passing = 70% failing (2025). WebAIM: 83.9% of home pages with low-contrast text (2026). 83.9 minus 70 = 13.9 points. Different tools, samples and years.
-
Default focus outlines were removed on 67% of websites in 2025, according to HTTP Archive.
-
Accessibility overlay apps were used on about 2% of desktop websites in 2025, and 0.2% of the top 1,000, according to HTTP Archive.
| Failure type | 2019 | 2023 | 2024 | 2025 | 2026 |
|---|---|---|---|---|---|
| Low-contrast text | 85.3% | 83.6% | 81% | 79.1% | 83.9% |
| Missing image alt text | 68% | 58.2% | 54.5% | 55.5% | 53.1% |
| Missing form input labels | 52.8% | 45.9% | 48.6% | 48.2% | 51% |
| Empty links | 58.1% | 50.1% | 44.6% | 45.4% | 46.3% |
| Empty buttons | 25% | 27.5% | 28.2% | 29.6% | 30.6% |
| Missing document language | 33.1% | 18.6% | 17.1% | 15.8% | 13.5% |
Website page weight and speed statistics
HTTP Archive measures the bytes each page downloads in its July crawl. The Web Almanac 2025 reports medians for home pages and inner pages, on desktop and mobile.
Pages keep getting heavier. The median home page grew 7.8% in a year to 2.7 MB. The median mobile home page weighed 2,559 KB in July 2025, about 5.1 times its October 2014 weight. Desktop pages were 303 KB heavier than mobile ones. That small gap suggests most sites send phones nearly the same page as laptops.
Images and JavaScript are the weight. Images were about 35.6% of the median mobile home page, and the two together about 60.3%. JavaScript costs more than its size suggests, because the phone has to run it as well as download it. Home pages carry most of the extra images. The median mobile home page was 44.7% heavier than an inner page.
The median page made 72 requests on mobile and 77 on desktop. Each request adds a round trip, which hurts most on slow mobile networks. For how many sites there are and what they run on, see our website statistics.
Source: HTTP Archive, Web Almanac 2025, Page Weight chapter, Figure 14.5.
-
The median mobile home page weighed 2,559 KB in July 2025, according to HTTP Archive.
-
The median desktop home page weighed 2,862 KB in July 2025, according to HTTP Archive.
-
The median home page grew 7.8% in a year to 2.7 MB in 2025, according to HTTP Archive.
-
The median mobile home page used 911 KB of images in 2025, according to HTTP Archive.
-
The median mobile home page used 632 KB of JavaScript in 2025, according to HTTP Archive.
-
The median web page made 72 requests on mobile and 77 on desktop in 2025, according to HTTP Archive.
-
The median mobile home page was 5.1 times heavier in July 2025 than in October 2014, calculated by StatisticsGlobal from HTTP Archive data.
How we calculated it: Median mobile home page 2,559 KB in July 2025 / 505 KB in October 2014 = 5.07. Desktop: 2,862 KB / 1,208 KB = 2.37, a 136.9% rise. Crawl setup has changed over time.
-
The median desktop home page was 303 KB, or 11.8%, heavier than the median mobile home page in 2025, calculated by StatisticsGlobal from HTTP Archive data.
How we calculated it: 2,862 KB desktop minus 2,559 KB mobile = 303 KB. 2,862 / 2,559 = 1.118, so 11.8% heavier.
-
The median mobile home page was 44.7% heavier than the median mobile inner page in 2025, calculated by StatisticsGlobal from HTTP Archive data.
How we calculated it: Median mobile home page 2,559 KB / median mobile inner page 1,769 KB (July 2025) = 1.447, so 44.7% heavier.
| Resource | Desktop home page (KB) | Mobile home page (KB) |
|---|---|---|
| Images | 1,058 | 911 |
| JavaScript | 697 | 632 |
| Fonts | 139 | 122 |
| CSS | 82 | 77 |
| HTML | 22 | 22 |
| Whole page | 2,862 | 2,559 |
Web design statistics for mobile
Google's Core Web Vitals measure loading (LCP), responsiveness (INP) and visual stability (CLS) for real Chrome users. HTTP Archive reports the share of sites that pass all three.
Mobile is catching up, slowly. The share of mobile sites with good Core Web Vitals rose by half between 2021 and 2025, from 32% to 48%. Desktop stalled at 56% after a big jump in 2024. The gap is now 8 points. Still, more than half of mobile sites fail at least one test.
Responsiveness is the biggest mobile weakness. Only 77% of mobile sites had a good Interaction to Next Paint, against 97% on desktop. Loading is the next. 62% of mobile pages had a good Largest Contentful Paint, against 74% on desktop. The largest element is usually an image, and about 16% of pages still lazy-load it, which delays the main content.
Small screens need zoom. Yet 19% of mobile sites restricted it in 2025, which fails users with low vision. For how traffic splits between phones and computers, see our mobile vs desktop usage page.
- Mobile
- Desktop
Source: HTTP Archive, Web Almanac 2025, Performance chapter, Figure on good Core Web Vitals over time.
-
Good Core Web Vitals were reached by 48% of mobile websites in 2025, up from 44% in 2024, according to HTTP Archive.
-
Good Core Web Vitals were reached by 56% of desktop websites in 2025, up from 55% in 2024, according to HTTP Archive.
-
Desktop websites led mobile on good Core Web Vitals by 8 percentage points in 2025, calculated by StatisticsGlobal from HTTP Archive data.
How we calculated it: 56% of desktop sites minus 48% of mobile sites with good Core Web Vitals in 2025 = 8 points. The gap was 9 points in 2021 (41% minus 32%).
-
The share of mobile websites with good Core Web Vitals rose by half between 2021 and 2025, calculated by StatisticsGlobal from HTTP Archive data.
How we calculated it: 48% of mobile sites with good Core Web Vitals in 2025 / 32% in 2021 = 1.50, a 50% relative rise (16 points).
-
A good Largest Contentful Paint was reached by 62% of mobile pages in 2025, against 74% of desktop pages, according to HTTP Archive.
-
A good Interaction to Next Paint was reached by 77% of mobile websites in 2025, against 97% on desktop, according to HTTP Archive.
-
The largest element on screen was an image on 76% of mobile pages and 85.3% of desktop pages in 2025, according to HTTP Archive.
-
About 16% of web pages lazy-loaded their largest image in 2025, delaying the main content, according to HTTP Archive.
-
Zooming was restricted on 19% of mobile websites and 21% of desktop websites in 2025, according to HTTP Archive.
| Measure | Mobile | Desktop |
|---|---|---|
| Good Core Web Vitals | 48% | 56% |
| Good Largest Contentful Paint | 62% | 74% |
| Good Interaction to Next Paint | 77% | 97% |
| Largest element is an image | 76% | 85.3% |
| Zoom restricted | 19% | 21% |
| Median home page weight | 2,559 KB | 2,862 KB |
Ecommerce UX and checkout statistics
Baymard Institute tests checkout usability and compiles abandonment studies. WebAIM's scan shows how online shops compare with other kinds of site on accessibility.
Design choices cost sales. Baymard's average documented cart abandonment rate is 70.22%. In its US survey, 17% of shoppers who left a checkout blamed a long or complicated process, and another 17% blamed site errors or crashes. Both are design problems a shop can fix. Our cart abandonment statistics page covers every reason in detail.
Checkouts ask for too much. The average US checkout showed 23.48 form elements, 9.48 more than the top of Baymard's ideal range. Baymard estimates better checkout design could lift a large site's conversion rate by 35.26%.
Shops are also among the least accessible sites. Shopping home pages averaged 71.0 errors in WebAIM's 2026 scan, 67.5% more than government home pages. Shopify and Magento home pages each averaged about 75 errors. Every barrier on a product or checkout page can block a sale.
Source: WebAIM, The WebAIM Million 2026.
-
The average documented online shopping cart abandonment rate was 70.22% in 2025, across 50 studies, according to Baymard Institute.
-
A too long or complicated checkout led 17% of US shoppers who abandoned a checkout to leave in 2025, according to Baymard Institute.
-
Website errors or crashes led 17% of US shoppers who abandoned a checkout to leave in 2025, according to Baymard Institute.
-
The average US checkout showed 23.48 form elements by default in 2025, according to Baymard Institute.
-
The average US checkout showed 9.48 more form elements in 2025 than the 14 at the top of Baymard's ideal range, calculated by StatisticsGlobal from Baymard Institute data.
How we calculated it: 23.48 average default form elements minus 14, the upper end of the 12 to 14 elements Baymard says an ideal checkout needs = 9.48.
-
Better checkout design could raise the average large ecommerce site's conversion rate by 35.26% in 2025, according to Baymard Institute.
-
Shopping home pages averaged 71.0 detectable accessibility errors in 2026, against 42.4 for government home pages, according to WebAIM.
-
Shopping home pages had 67.5% more detectable accessibility errors than government home pages in 2026, calculated by StatisticsGlobal from WebAIM data.
How we calculated it: Average errors on Shopping home pages (71.0, 95,039 pages) / Government home pages (42.4, 22,345 pages) = 1.675, so 67.5% more. Categories were assigned by WebAIM using AI.
| Category | Home pages | Average errors | Difference from 56.1 average |
|---|---|---|---|
| Government | 22,345 | 42.4 | −24.4% |
| Non-Profit/Charity | 52,884 | 43.0 | −23.3% |
| Technology & Computing | 300,245 | 49.0 | −12.6% |
| Business | 107,509 | 52.6 | −6.3% |
| Travel | 128,711 | 61.9 | +10.4% |
| Style and Fashion | 21,184 | 66.7 | +18.9% |
| Shopping | 95,039 | 71.0 | +26.6% |
| Sports | 15,640 | 71.4 | +27.3% |
Web designer salary and employment statistics
The US Bureau of Labor Statistics groups web developers with web and digital interface designers. Pay comes from its May 2025 wage survey, and projections run from 2025 to 2035.
Designers out-earn developers in this group. US web and digital interface designers earned a median $104,000 in May 2025, $11,350 more than web developers. They also hold most of the jobs, 60.3% of 220,100. The whole group earned 1.95 times the median wage for all US workers.
Pay varies a lot. The top 10% of designers earned at least 3.75 times the bottom 10%. Many work for themselves. Self-employed workers held 18% of web developer jobs and 13% of designer jobs.
Growth is steady but not fast. BLS projects 5% growth from 2025 to 2035, against 3% for all jobs. About 13,600 openings a year are expected, equal to about 6.2% of today's jobs. Many come from the need to replace people who leave the field. These figures cover the US only.
Source: US Bureau of Labor Statistics, Occupational Outlook Handbook.
-
Web developers and digital designers in the US earned a median $99,520 a year in May 2025, according to the US Bureau of Labor Statistics.
-
Web and digital interface designers in the US earned a median $104,000 a year in May 2025, according to the US Bureau of Labor Statistics.
-
Web developers in the US earned a median $92,650 a year in May 2025, according to the US Bureau of Labor Statistics.
-
US web developers and digital designers earned 1.95 times the median wage for all workers in May 2025, calculated by StatisticsGlobal from US Bureau of Labor Statistics data.
How we calculated it: Median annual wage $99,520 for web developers and digital designers / $50,980 for all occupations = 1.952.
-
The top 10% of US web and digital interface designers earned at least 3.75 times the bottom 10% in May 2025, calculated by StatisticsGlobal from US Bureau of Labor Statistics data.
How we calculated it: Highest 10% earned more than $201,550; lowest 10% earned less than $53,750. 201,550 / 53,750 = 3.75.
-
The US had 220,100 web developer and digital designer jobs in 2025, according to the US Bureau of Labor Statistics.
-
Web and digital interface designers held about 132,700 US jobs in 2025, according to the US Bureau of Labor Statistics.
-
Self-employed workers held 18% of US web developer jobs and 13% of web and digital interface designer jobs in 2025, according to the US Bureau of Labor Statistics.
-
US employment of web developers and digital designers is projected to grow 5% from 2025 to 2035, according to the US Bureau of Labor Statistics.
-
About 13,600 US openings for web developers and digital designers are projected each year from 2025 to 2035, according to the US Bureau of Labor Statistics.
| Occupation | Median pay, May 2025 | Jobs, 2025 | Projected jobs, 2035 | Growth, 2025 to 2035 |
|---|---|---|---|---|
| Web and digital interface designers | $104,000 | 132,700 | 140,800 | 6% |
| Web developers | $92,650 | 87,400 | 90,600 | 4% |
| Web developers and digital designers (total) | $99,520 | 220,100 | 231,400 | 5% |
| All occupations | $50,980 | n/a | n/a | 3% |
How WebAIM, HTTP Archive, Baymard and BLS measure web design
WebAIM runs its WAVE engine on the rendered home pages of the top 1 million sites in the Tranco ranking. The latest scan was in February 2026. It counts errors that are very likely WCAG 2.2 A or AA failures. Automated tools miss many problems, so a page with no detected errors is not proof of accessibility. Site categories were assigned with AI.
HTTP Archive crawls millions of sites from the Chrome UX Report each July, on both a desktop and a mobile setup. Its Web Almanac 2025 uses the July 2025 crawl and was published in January 2026. Page weights are medians, and each resource type has its own median, so they do not add up. Core Web Vitals come from real Chrome users. The next Almanac will replace these figures.
Baymard's 70.22% abandonment rate is an average of 50 studies by other parties. Its reasons for abandoning come from its own US survey, and its form counts from its checkout benchmark. BLS figures cover the US only. Its Occupational Outlook Handbook groups designers and developers together. Where sources disagree, as on colour contrast, we show both and say why.
Frequently asked questions
What share of websites are accessible in 2026?
Very few pass automated checks. WebAIM found detectable WCAG 2 failures on 95.9% of the top 1 million home pages in February 2026, so fewer than 4.1% could be fully conformant. The average page had 56.1 errors.
Is web accessibility getting better or worse than last year?
Worse on WebAIM's measure. Pages with detected failures rose from 94.8% in 2025 to 95.9% in 2026, and errors per page rose from 51 to 56.1. HTTP Archive's median Lighthouse accessibility score edged up to 85% in 2025.
How big is the average web page?
The median mobile home page weighed 2,559 KB and the median desktop home page 2,862 KB in July 2025, according to HTTP Archive. The median home page grew 7.8% in a year.
Why are websites slow on mobile?
Pages are heavy and phones are slower. Images were about 35.6% of the median mobile home page in 2025, calculated by StatisticsGlobal from HTTP Archive data. Only 48% of mobile sites had good Core Web Vitals, against 56% on desktop.
How much do web designers earn in the US?
US web and digital interface designers earned a median $104,000 a year in May 2025, according to the US Bureau of Labor Statistics. Web developers earned $92,650.
Is demand for web designers growing?
BLS projects US employment of web developers and digital designers to grow 5% from 2025 to 2035, faster than the 3% average. About 13,600 openings are expected each year.
Sources
- WebAIM, The WebAIM Million: The 2026 report on the accessibility of the top 1,000,000 home pages, March 2026 (last updated 30 March 2026), accessed 7 November 2026. Link · Saved data copy
- HTTP Archive, Web Almanac 2025, Chapter 14: Page Weight, 15 January 2026, accessed 7 November 2026. Link · Saved data copy · Licence: Apache 2.0
- HTTP Archive, Web Almanac 2025, Chapter 11: Performance, 15 January 2026 (updated 5 May 2026), accessed 7 November 2026. Link · Saved data copy · Licence: Apache 2.0
- US Bureau of Labor Statistics, Occupational Outlook Handbook: Web Developers and Digital Designers, 2026 edition (page modified 27 August 2026), accessed 7 November 2026. Link · Saved data copy
- HTTP Archive, Web Almanac 2025, Chapter 6: Accessibility, 15 January 2026 (updated 3 November 2026), accessed 7 November 2026. Link · Saved data copy · Licence: Apache 2.0
- Baymard Institute, Cart Abandonment Rate Statistics, 22 September 2025. Link · Saved data copy
Changes to this page
- : Page published with WebAIM Million 2026 data, HTTP Archive Web Almanac 2025 data, Baymard Institute checkout research and BLS data for 2025.