Average detected accessibility errors per home page across 29 sectors, from government and education to retail and sport, measured on one million sites.
Headline finding
The sectors with the fewest detected barriers are the ones that have been legally required to care the longest. Government, non-profit, and education lead; shopping and sport trail by roughly 29 errors per page.
Source: WebAIM Million 2026. Automated results do not establish WCAG conformance.
Executive summary
Government, non-profit, and education sites are not the best-funded on the web, but they sit at the top of this table. Retail, travel, and entertainment sites have far larger budgets and sit at the bottom. The difference between them is that one group has had a legal duty for decades.
Even the leaders are not doing well in absolute terms. The best sector still averaged 42.4 automatically detected errors on its home page.
Government at 42.4 errors, then Non-Profit at 43.0 and Science at 45.0. Education came in at 48.9, all below the 56.1 average.
Sports was worst at 71.4, with Shopping at 71.0 and Style and Fashion at 66.7. These are the sectors built around imagery, carousels, and marketing scripts.
The spread from best to worst is about 29 errors per home page. Every sector average is far above zero, so this is a comparison of degrees of failure.
Site categories were assigned by WebAIM using Amazon Web Services Bedrock AI, then averaged across the home pages in each category. That means a category label is a machine judgement about what a site is for, not a self-declared industry or a regulatory classification. Sector averages also mix organisations of wildly different size and budget, so treat a figure as a benchmark to compare your own audit against, not as a grade for an industry.
Government home pages had the fewest detected errors at 42.4, and Sports the most at 71.4. The gap between the best and worst sector is roughly 29 errors per page, which is smaller than the gap between the best and worst JavaScript library.
24.4% below average
23.3% below average
19.8% below average
19% below average
17.7% below average
17% below average
13% below average
12.8% below average
12.6% below average
6.3% below average
3.5% below average
3.1% below average
1.1% above average
1.3% above average
2.3% above average
2.4% above average
5.1% above average
5.5% above average
7.7% above average
8% above average
8.8% above average
10.2% above average
10.4% above average
12.9% above average
13.4% above average
17.4% above average
18.9% above average
26.6% above average
27.3% above average
Source: WebAIM Million 2026
Interpretation
The sectors at the top of the table are the ones that have been legally obliged to care the longest. The sectors at the bottom sell things.
Government sites averaged 42.4 errors, 24.4% below average, and Non-Profit and Education were close behind. These are the sectors covered by Section 508, ADA Title II, and the EU Web Accessibility Directive. Sustained legal obligation correlates with measurably better markup.
Shopping sat near the bottom of the table, and every ecommerce platform we measured separately also came in above average. Retail is simultaneously the sector with the most detected barriers and the one facing the most digital accessibility litigation.
The spread across languages is far wider than across industries. Home pages in the worst-performing language averaged 136.2 errors, more than double the million-page average, which dwarfs any sector difference in this dataset.
The range here is the widest segmentation in the study, from 46 errors to 136.2. Language is a proxy for regional web conventions, popular platforms, and local regulation rather than anything about the language itself.
18% below average
17.1% below average
16.4% below average
4.2% above average
7.5% above average
8.8% above average
10.8% above average
14.7% above average
14.8% above average
17.9% above average
22.5% above average
23.5% above average
29.2% above average
32.8% above average
39.2% above average
68.7% above average
142.8% above average
Source: WebAIM Million 2026
Automated WAVE evaluation of the home pages for the top one million websites in February 2026. Technology and category detection was performed alongside the accessibility scan.
Sample Size
1,000,000 home pages classified into 29 categories and 17 languages
Date Range
February 2026 sample, published 2026-03-30
Last updated: 2026-08-03
Use the research
Useful as a benchmark in an audit report or a business case. Keep the link to WebAIM when publishing extracts.
The PDF contains the full sector and language tables with the methodology note. The CSV contains one row per sector and language.
Accessibility.build. (2026). Web Accessibility by Industry. https://accessibility.build/research/accessibility-by-industry
Questions about the data
What these sector averages support, and what they do not.
Comprehensive tools, checklists, and guides to help you create inclusive digital experiences