Accessibility of US Small Business Websites 2026: A 994-Site Scan
We scanned 994 US small business home pages: 85.6% failed an automated WCAG 2.2 test, 60% had low contrast text, and overlay sites did no better.
Our content is written and reviewed by the Four Cents team. We may recommend our own services. Read our editorial standards.
On this page
Study summary: We scanned the home pages of 994 US small businesses in November 2026. 85.6% had at least one automatically detectable WCAG 2.2 A or AA failure. Low contrast text was the most common problem, on 60% of sites. Of the 121 sites running an accessibility overlay widget, 92.6% still failed.
This is original research by DeveloperOfWeb. We built a sample of 994 real small business websites from Google Maps listings across 9 industries and 11 states, then tested each home page with axe-core, the open source engine behind many accessibility checkers. We publish aggregates only. No business is named. This page is general information, not legal advice.
Summary and key findings
Most US small business websites have accessibility barriers that a free automated tool can find in seconds. In our scan of 994 home pages, 85.6% failed at least one WCAG 2.2 A or AA rule that axe-core can test. The typical site had 7 failing elements on its home page alone. The problems are mostly basic: text that is too faint to read, links and buttons a screen reader cannot name, images with no description.
We scanned the home pages of 994 US small businesses in November 2026. 85.6% had at least one automatically detectable WCAG 2.2 A or AA failure. Low contrast text was the most common problem, on 60% of sites. Of the 121 sites running an accessibility overlay widget, 92.6% still failed.
The table below holds every headline number. Each one is a share of the 994 home pages we could test, unless it says otherwise.
| Finding | Result | Detail (n) |
|---|---|---|
| Home pages scanned | 994 | 1,242 sampled, 248 could not be tested |
| At least one detected WCAG A/AA failure | 85.6% | 851 of 994 |
| No detected failures | 14.4% | 143 of 994 |
| Low contrast text | 60% | 596 sites |
| Links with no accessible name | 40.8% | 406 sites |
| Images with no alt text (any image rule) | 20.3% | axe image-alt, role-img-alt, input-image-alt, svg-img-alt |
| Form fields or dropdowns with no label | 5.2% | axe label, select-name |
| Buttons with no accessible name | 6.1% | 61 sites |
| No page language set | 3.9% | 39 sites |
| At least one "critical" failure | 29.1% | about 289 sites |
| At least one "critical" or "serious" failure | 84.6% | |
| Median failing elements per home page | 7 | mean 13.6 |
| Overlay widget detected | 12.2% | 121 sites |
| Overlay sites with a detected failure | 92.6% | 112 of 121 |
| Link to an accessibility statement or page | 14.8% | 147 sites |
| Served over HTTPS | 95.4% |
One number matters most to an owner. 29.1% of sites had at least one failure that axe-core rates "critical". A critical failure usually means a blind visitor cannot use part of the page at all. It might be an unlabeled booking field, or a button with no name.
How many small business websites pass an accessibility scan?
Not many: about one in 7. Only 143 of 994 home pages (14.4%) had zero failures that axe-core could detect. Passing an automated scan is also a low bar. Automated rules catch only part of what WCAG asks for, so even those 143 sites may still have barriers a person would find. Deque, which makes axe-core, found in a 2021 study of more than 2,000 audits that automated testing covered 57 percent of issues by volume. That leaves a large share that only manual testing finds.
The spread is wide. Most failing sites have a handful of problems, not hundreds. The chart shows how many failing elements we found per home page.
28.7% of sites had between 1 and 5 failing elements. Another 36.7% had 6 to 20. At the far end, 4.4% had more than 50. This is good news for owners. A site with 1 to 20 failing elements is usually a short repair job, often a matter of a few theme colors and a few missing labels. The cost and effort rise when failures come from a page builder or plugin that repeats the same mistake on every page.
The typical site failed 2 different rules. That number is a better guide to repair effort than the element count. One contrast failure in a theme's footer color can show up as 30 failing elements, and it is fixed with one line of CSS.
What "failing" means in this study
We count a site as failing if axe-core reported at least one violation of a rule tagged WCAG 2.0, 2.1 or 2.2 at Level A or AA. We did not count best-practice rules, which go beyond WCAG. We did not count "needs review" results, where axe-core could not decide, for example contrast on text placed over a photo. So our failure figures are a floor, not a ceiling.
What are the most common accessibility failures on small business websites?
Low contrast text is the most common failure by a wide margin, found on 60% of the home pages we scanned. The next most common were links with no name (40.8%) and tap targets too small (18.5%). Together these few rules explain most of what is wrong.
Here is the full top 12, with the WCAG success criterion each rule maps to. The criterion mapping comes from axe-core's own rule tags.
| Rank | Failure (axe-core rule) | Sites failing | Share of sites | Failing elements in total | axe impact | WCAG criterion |
|---|---|---|---|---|---|---|
| 1 | Low contrast text color-contrast | 596 | 60% | 6,345 | serious | 1.4.3 Contrast Minimum (AA) |
| 2 | Links with no name link-name | 406 | 40.8% | 2,105 | serious | 2.4.4 Link Purpose (A); 4.1.2 Name, Role, Value (A) |
| 3 | Tap targets too small target-size | 184 | 18.5% | 1,336 | serious | 2.5.8 Target Size Minimum (AA) |
| 4 | Images with no alt text image-alt | 175 | 17.6% | 1,037 | critical | 1.1.1 Non-text Content (A) |
| 5 | Zoom disabled on mobile meta-viewport | 121 | 12.2% | 121 | moderate | 1.4.4 Resize Text (AA) |
| 6 | Embedded frames with no title frame-title | 110 | 11.1% | 153 | serious | 4.1.2 Name, Role, Value (A) |
| 7 | Links shown by color only link-in-text-block | 70 | 7% | 171 | serious | 1.4.1 Use of Color (A) |
| 8 | Broken list markup list | 61 | 6.1% | 100 | serious | 1.3.1 Info and Relationships (A) |
| 9 | Buttons with no name button-name | 61 | 6.1% | 437 | critical | 4.1.2 Name, Role, Value (A) |
| 10 | Hidden content still focusable aria-hidden-focus | 47 | 4.7% | 158 | serious | 4.1.2 Name, Role, Value (A) |
| 11 | No page language set html-has-lang | 39 | 3.9% | 39 | serious | 3.1.1 Language of Page (A) |
| 12 | ARIA used where not allowed aria-prohibited-attr | 38 | 3.8% | 136 | serious | 4.1.2 Name, Role, Value (A) |
Low contrast text
Contrast failed on 60% of sites and accounted for 6,345 of the 13,487 failing elements we found, or 47%. WCAG 2.2 criterion 1.4.3 asks for a contrast ratio of at least 4.5:1 for normal text and 3:1 for large text. The usual culprits on small business sites are pale gray body text, white text on a light brand color, light text in footers, and placeholder text in forms. Contrast is a design decision made once in the theme, which is why one fix can clear dozens of failing elements.
Links and buttons with no name
40.8% of sites had at least one link a screen reader cannot name, and 6.1% had a nameless button. On small business sites these are almost always icons: a Facebook or Instagram logo, a phone icon, a map pin, a hamburger menu, slider arrows. A sighted visitor knows what the icon means. A screen reader user hears "link" or "button" and nothing else. The fix is a short text label, often added with an aria-label attribute.
Images with no alt text
20.3% of sites had at least one image with no text alternative under any of axe-core's four image rules. For a dentist, that might be the photo of the team. For a restaurant, it is often the menu, posted as an image. A menu image with no text version is a real barrier, because the visitor cannot get the one piece of information they came for.
Form fields with no label
5.2% of sites had a form field or dropdown with no programmatic label. This is lower than we expected, partly because many small business home pages have no form at all, and partly because booking tools often run inside a frame that this scan did not open. Where a form does exist, a missing label usually means placeholder text is doing the job of a label. The placeholder disappears as soon as someone starts typing.
Tap targets too small
18.5% of sites had tap targets smaller than WCAG 2.2 allows. Criterion 2.5.8 Target Size (Minimum) is new in WCAG 2.2 and asks for targets of at least 24 by 24 CSS pixels, or enough space around them. Carousel dots, tiny social icons and cramped footer links are the usual cause. WCAG 2.2 was published in October 2023, and many themes in use today were designed before it.
Missing page language
3.9% of sites did not declare a page language. It is the easiest fix on this list: one attribute on one tag. Without it, a screen reader may read English text with the wrong pronunciation rules.
How serious are the failures?
Most failures are rated "serious" by axe-core, and a large minority are "critical". axe-core gives every rule an impact level: critical, serious, moderate or minor. It is the tool's own rating of how much the problem is likely to block a user, not a legal measure.
| axe-core impact | Share of sites with at least one | Failing elements in total | Typical examples |
|---|---|---|---|
| Critical | 29.1% | 1,901 | Images with no alt text, unlabeled form fields, nameless buttons |
| Serious | 82.8% | 11,465 | Low contrast, nameless links, untitled frames |
| Moderate | 12.2% | 121 | Zoom disabled, some landmark and structure problems |
| Minor | 0% | 0 | Rare in the WCAG rule set we ran |
Contrast is rated serious, and it is everywhere, which is why the serious row is so high. The critical row is the one to act on first. A critical failure often stops a task outright. Think of a customer who wants to book a cleaning, request a quote or read a menu and cannot.
The impact rating is also a good proxy for what a customer notices. An unlabeled field or an unnamed button is not a cosmetic issue for someone using a screen reader: it is a dead end. We cover what the law says further down. The point here is simple. Many of the failures that block people most are also among the cheapest to fix.
Which industries have the least accessible websites?
Medical clinics had the most failing elements per site in our sample, with a median of 14. Restaurants had the fewest, with a median of 2. The differences between industries are real but smaller than the differences between individual sites. Every industry we tested had a failure rate of at least 73%.
| Industry | Sites | Any failure | Critical failure | Low contrast | Median failing elements | Overlay | Statement link |
|---|---|---|---|---|---|---|---|
| Medical clinics | 85 | 84.7% | 35.3% | 65.9% | 14 | 16.5% | 20% |
| Dentists | 146 | 93.8% | 30.8% | 67.8% | 9 | 32.9% | 29.5% |
| Contractors | 133 | 90.2% | 29.3% | 69.2% | 9 | 4.5% | 3% |
| Retail stores | 80 | 86.3% | 37.5% | 46.3% | 9 | 7.5% | 15% |
| Real estate | 125 | 90.4% | 36% | 68.8% | 8 | 8% | 16% |
| Law firms | 136 | 85.3% | 29.4% | 60.3% | 6.5 | 8.1% | 5.1% |
| Salons | 121 | 79.3% | 18.2% | 55.4% | 5 | 1.7% | 5% |
| Hotels | 57 | 82.5% | 43.9% | 40.4% | 5 | 24.6% | 35.1% |
| Restaurants | 111 | 73% | 11.7% | 48.6% | 2 | 9% | 16.2% |
| All sites | 994 | 85.6% | 29.1% | 60% | 7 | 12.2% | 14.8% |
A few patterns stand out.
- Hotels had the highest share of sites with a critical failure, at 43.9%. Retail stores followed at 37.5%.
- Restaurants and retail matter most legally. EcomBack's 2025 report put restaurant, food and drink sites at 1,368 suits (34.65%) and fashion and apparel at 1,025 (25.96%), the two largest groups. In our scan, restaurants failed at 73% and retail stores at 86.3%.
- Healthcare (dentists and medical clinics) often runs booking and intake forms on the home page. Those are the pages a patient with a disability most needs to use. Dentists failed at 93.8% and medical clinics at 84.7%.
- Law firms failed at 85.3%, close to the overall rate of 85.6%.
Industry samples range from 57 to 146 sites. Small gaps between industries, a few percentage points, are within the noise for samples this size. We would not rank one industry above another on a gap of less than about 8 points.
If you run one of these businesses, we have industry guides that cover accessibility alongside design: web design for dentists and restaurant website design.
Does the state make a difference?
Not much. Sites in the five states with the most website accessibility lawsuits were no more accessible than sites elsewhere. Across New York, California, Florida, Illinois and Pennsylvania, 85.9% of 582 sites failed, with a median of 8 failing elements. Across the other six states, 85.2% of 412 failed, with a median of 7.
| State | Metros sampled | Sites | Any failure | Critical failure | Low contrast | Median failing elements | Statement link |
|---|---|---|---|---|---|---|---|
| Florida | Miami, Tampa | 143 | 81.8% | 29.4% | 60.1% | 8 | 15.4% |
| California | Los Angeles, San Diego | 142 | 83.8% | 29.6% | 59.9% | 7 | 19% |
| Pennsylvania | Philadelphia, Pittsburgh | 134 | 88.1% | 29.1% | 59% | 8 | 19.4% |
| Washington | Seattle | 86 | 81.4% | 20.9% | 52.3% | 4 | 15.1% |
| New York | New York | 84 | 94% | 38.1% | 63.1% | 8 | 14.3% |
| Illinois | Chicago | 79 | 84.8% | 31.6% | 60.8% | 7 | 16.5% |
| Arizona | Phoenix | 71 | 85.9% | 22.5% | 64.8% | 7 | 7% |
| Colorado | Denver | 68 | 85.3% | 19.1% | 52.9% | 7 | 16.2% |
| Georgia | Atlanta | 64 | 85.9% | 28.1% | 65.6% | 7 | 10.9% |
| Texas | Houston | 62 | 93.5% | 37.1% | 67.7% | 8.5 | 6.5% |
| North Carolina | Charlotte | 61 | 80.3% | 34.4% | 55.7% | 6 | 11.5% |
We picked those five states because they lead lawsuit counts. Seyfarth Shaw's count of 2025 federal website suits put New York first (1,021), then Florida (961), Illinois (585), Minnesota (162) and Pennsylvania (137). California had only 4 federal website suits. That is not because California is quiet. Seyfarth notes that a significant number of demand letters and suits are filed in state courts, which its federal count leaves out, and EcomBack's count, which includes state courts, puts California third with 787 website suits in 2025.
The lesson for an owner is uncomfortable but clear. Higher legal risk in your state has not, so far, translated into better small business websites. Owners in New York or Florida are not more likely to have fixed the basics than owners in Colorado or Arizona. Each state figure is based on one or two metro areas, so treat it as a description of those cities rather than of the whole state.
Which website platforms do best?
Platform made a visible difference. WordPress ran 35.3% of the sites we scanned, far more than any other platform, and 91.2% of WordPress home pages failed, with a median of 10 failing elements. Squarespace sites did best of the larger groups: 66.4% failed, with a median of 2. Across platforms with at least 15 sites in our sample, Squarespace had the lowest median failing elements (2) and Shopify the highest (11).
| Platform | Sites | Share of sample | Any failure | Critical failure | Low contrast | Median failing elements |
|---|---|---|---|---|---|---|
| WordPress | 351 | 35.3% | 91.2% | 32.8% | 71.5% | 10 |
| Other / custom | 264 | 26.6% | 83.3% | 33% | 54.2% | 6 |
| Squarespace | 116 | 11.7% | 66.4% | 14.7% | 41.4% | 2 |
| Wix | 72 | 7.2% | 84.7% | 5.6% | 66.7% | 4.5 |
| Shopify | 38 | 3.8% | 92.1% | 34.2% | 57.9% | 11 |
| Next.js (custom) | 36 | 3.6% | 75% | 16.7% | 55.6% | 3 |
| GoDaddy Website Builder | 34 | 3.4% | 100% | 64.7% | 47.1% | 7.5 |
| Webflow | 26 | 2.6% | 100% | 19.2% | 73.1% | 8 |
| Duda | 21 | 2.1% | 90.5% | 28.6% | 52.4% | 4 |
| Weebly / Square Online | 15 | 1.5% | 100% | 20% | 60% | 9 |
Platforms with fewer than 15 sites in our sample are left out of the table because the numbers are too small to compare: Dental/medical vendor platform (6), HubSpot CMS (5), Framer (3), Scorpion (3), Drupal (2), BentoBox (1), Joomla (1).
A platform does not make a site accessible or inaccessible by itself. The theme, the page builder, the plugins and the content choices do most of the work. A WordPress site on a well-built theme can pass; a Squarespace site with white text over a photo will fail contrast. Hosted builders such as Wix and Squarespace control more of the code, so their sites tend to be more uniform. That cuts both ways: their common patterns are reasonably good, but an owner has less control when something is wrong.
We detected platform from code signatures on the page, such as wp-content paths for WordPress or cdn.shopify.com for Shopify. "Other or custom" covers hand-built sites, agency frameworks and platforms we could not identify. If you are choosing a platform, our comparisons of Squarespace vs WordPress and WordPress vs Wix cover accessibility among other factors.
Do accessibility overlay widgets fix the problem?
Not in our data. We found an overlay widget on 121 home pages (12.2%). 92.6% of those pages still had at least one detected WCAG failure, higher than the 84.7% we saw on sites without a widget. The median overlay site had 9 failing elements, against 7 for sites without one.
The widgets we detected were: UserWay (65), accessiBe (24), AudioEye (16), EqualWeb (5), One Click Accessibility (WP plugin) (5), Accessibility Adapter / Elfsight (4), Sienna (3), Equally AI (1). The most common failures on overlay sites were low contrast text (62.8%), links with no name (42.1%), tap targets too small (26.4%), images with no alt text (18.2%), embedded frames with no title (16.5%). Those are the same failures we found everywhere else.
We want to be careful about what this does and does not show.
- It is a measurement, not an accusation. We are reporting what axe-core found on pages where a widget's code was present. We are not saying any vendor's product is defective or that any business acted in bad faith.
- We tested the page as a visitor first sees it. Our browser loaded each page, waited for scripts including any widget, and then ran axe-core. Some widgets apply extra changes only when a visitor opens the widget and turns on a profile. We did not do that, because many visitors never open it, and screen reader users rely on their own software.
- The groups are not random. Owners who install a widget may differ from owners who do not. With 121 overlay sites, the gap we found should not be read as widgets causing failures. It shows only that they did not prevent them.
What the data does show is narrow and useful. Having an overlay installed did not mean the page passed an automated WCAG test. That matches the wider record. In April 2025 the Federal Trade Commission approved a final order requiring accessiBe to pay $1 million. The FTC's complaint said the company's accessWidget "did not make all user websites WCAG-compliant", and the order bars it from claiming its automated products can make any website WCAG-compliant without evidence to support the claim. Lawsuit trackers also report suits against sites that had widgets: EcomBack counted 983 such suits in 2025 (24.90% of its total), and UsableNet reported that approximately 20 percent of companies sued in its 2026 data had a widget or overlay installed.
If you already have a widget, you do not need to rip it out today. Just do not treat it as the fix. The money is better spent repairing the theme colors, labels and alt text where the failures live. Our guide to ADA website lawsuits for small businesses explains why an overlay widget is not the fix.
How many small business sites publish an accessibility statement?
Very few. Only 147 of 994 home pages (14.8%) linked to anything we could identify as an accessibility statement or accessibility page. We looked for a link whose text or address mentions accessibility or the ADA, anywhere on the home page, including the footer.
Sites with a statement link were slightly less likely to fail: 80.3% against 86.5%. The median failing elements were 6 with a statement link and 7 without. A statement is a promise, not a repair. Overlay sites were also more likely to have a statement link (39.7% against 11.3%), One possible reason is that some widget setups include a statement link.
No US law requires a private business to post an accessibility statement. It is still a good practice. A useful statement says which standard you aim for (WCAG 2.2 AA is the current version), when the site was last tested, known gaps, and, most important, a phone number and email so a customer who hits a barrier can still book, buy or ask. A statement that overclaims, such as "this site is fully ADA compliant" when it is not, can do more harm than good. Ours is at /accessibility if you want an example of the format.
How does this compare with the WebAIM Million?
Our results point the same way as the WebAIM Million, but our percentages are mostly lower, and the reasons matter. WebAIM's February 2026 analysis found detected WCAG 2 failures on 95.9% of the top one million home pages, with an average of 56.1 errors per page. We found failures on 85.6% of small business home pages, with a mean of 13.6 failing elements.
| Measure | This study (axe-core) | WebAIM Million 2026 (WAVE) |
|---|---|---|
| Sample | 994 US small business home pages from Google Maps listings, outside the Tranco top 1 million | The top 1,000,000 home pages from the Tranco ranking |
| When | November 2026 | February 2026 |
| Pages with a detected WCAG failure | 85.6% | 95.9% |
| Low contrast text | 60% | 83.9% |
| Missing image alt text | 20.3% | 53.1% |
| Missing form labels | 5.2% | 51% |
| Empty links | 40.8% | 46.3% |
| Empty buttons | 6.1% | 30.6% |
| Missing document language | 3.9% | 13.5% |
| Average per page | 13.6 failing elements (median 7) | 56.1 errors |
Four differences likely explain most of the gap.
- Different tools. WebAIM uses WAVE. We used axe-core. axe-core is designed to report only what it can confirm, and it puts uncertain cases into a separate "needs review" list that we did not count. Contrast over images and gradients often lands there. WAVE flags some patterns as errors that axe-core treats as needing review. So the same page can score differently in each tool.
- Different units. WebAIM counts errors. We count failing elements per axe-core rule. The two are close but not identical, so the averages are not directly comparable.
- Different sites. The WebAIM Million draws on the Tranco ranking of the most visited sites. We excluded every domain in the Tranco top one million. Our sites are smaller and simpler. WebAIM reports an average of 1,437 page elements per home page in 2026; more elements mean more chances to fail.
- Different page builders. Small business sites lean heavily on themes and hosted builders. These ship with reasonable defaults for basics such as page language, which helps explain why missing language was rarer in our sample (3.9%) than in WebAIM's (13.5%).
The conclusion is the same in both studies. Most sites have detectable barriers, contrast is the top problem, and the same handful of failure types make up most of what is found. Our study adds a narrower view. Small companies make up most defendants (68% of suits through June 2026 involved companies under $50 million in revenue, per UsableNet), and their sites have the same problems as the big sites, just fewer of them per page. For the broader set of numbers, see our web accessibility statistics for 2026.
What should small business owners fix first?
Fix contrast, link and button names, image alt text and form labels first, then tap target size. Those groups cover most of the failures in our scan, each maps to a WCAG Level A or AA criterion, and most can be fixed in the theme or template in a day or less. The table below ranks the failures we found by how common they were and how much they block people, with the WCAG 2.2 criterion and a realistic effort for each.
| Priority | Failure | Sites in our scan | WCAG 2.2 criterion | Who it blocks | Typical fix effort |
|---|---|---|---|---|---|
| 1 | Low contrast text | 60% | 1.4.3 Contrast (Minimum), AA | Low vision, older visitors, anyone on a phone in sunlight | Low to medium. Usually a few theme colors: body text, light gray captions, white text on brand-color buttons and hero images. |
| 2 | Links with no name | 40.8% | 2.4.4 Link Purpose (A); 4.1.2 Name, Role, Value (A) | Screen reader and voice control users | Low. Add visible text or an aria-label to icon links (social icons, logo, phone and map icons). |
| 3 | Images with no alt text | 17.6% | 1.1.1 Non-text Content (A) | Blind and low vision visitors using screen readers | Low. Write alt text for meaningful images; use alt="" for decorative ones. |
| 4 | Buttons with no name | 6.1% | 4.1.2 Name, Role, Value (A) | Screen reader and voice control users | Low. Name icon buttons such as menu toggles, slider arrows and close buttons. |
| 5 | Form fields with no label | 3.1% | 1.3.1 Info and Relationships (A); 3.3.2 Labels or Instructions (A); 4.1.2 (A) | Screen reader users, people with cognitive disabilities | Low. Give every field a real label element, not just placeholder text. This matters most on booking and contact forms. |
| 6 | No page language set | 3.9% | 3.1.1 Language of Page (A) | Screen reader users (wrong pronunciation) | Minutes. Add lang="en" to the html element. |
| 7 | Zoom disabled on mobile | 12.2% | 1.4.4 Resize Text (AA) | Low vision visitors who pinch to zoom | Minutes. Remove maximum-scale=1 and user-scalable=no. |
| 8 | Embedded frames with no title | 11.1% | 4.1.2 Name, Role, Value (A) | Screen reader users | Minutes. Add a title to embedded maps, videos and booking widgets. |
| 9 | Tap targets too small | 18.5% | 2.5.8 Target Size (Minimum), AA, new in WCAG 2.2 | People with tremors or limited dexterity, all phone users | Medium. Enlarge or space out small tap targets such as carousel dots and footer icons. |
| 10 | Links shown by color only | 7% | 1.4.1 Use of Color (A) | Color blind visitors | Low. Underline links inside paragraphs. |
| 11 | ARIA used where not allowed | 3.8% | 4.1.2 Name, Role, Value (A) | Screen reader users | Medium. Usually a theme or plugin bug; fix the template, not each page. |
| 12 | Scroll area not keyboard reachable | 2.3% | 2.1.1 Keyboard (A) | Keyboard-only users | Low to medium. Make scrolling panels focusable or let them expand. |
Effort ratings are for a typical small business site of 5 to 20 pages on a mainstream theme or builder. They assume the fix is made once in the template, not page by page.
A one-week fix plan
- Day 1: measure. Run an automated scan on your home page and your three most important pages, such as booking, contact and menu or services. Our free accessibility checker does this. Save the results as your baseline.
- Day 2: theme colors. Fix body text, link, button and footer colors to meet 4.5:1. Check white text on photos and brand-color backgrounds.
- Day 3: names and labels. Add accessible names to icon links and buttons. Give every form field a visible label. Add a title to embedded maps, videos and booking frames.
- Day 4: images and content. Write alt text for meaningful images. Put your menu, price list or hours in real text, not only in an image or PDF.
- Day 5: keyboard and phone. Tab through the page with no mouse. Can you reach and use the menu, the booking button and the form? Can you see where focus is? Pinch to zoom on a phone. Then re-run the scan and post an honest accessibility statement with a contact route.
This plan clears most automated failures. It does not make a site conform to WCAG on its own, because a large share of WCAG needs human judgment: whether alt text is meaningful, whether headings make sense, whether a screen reader can complete a booking. That is what a manual audit adds. Our ADA compliance audit page explains what a manual audit covers, our guide to accessibility audit costs compares the options, and our WCAG 2.2 checklist walks through the criteria in plain English.
If you would rather hand the work over, we fix accessibility as part of a build or as a standalone accessibility remediation project. Our published price band for WCAG 2.2 AA build or remediation work on a small site is $500 to $2,000, depending on how many templates are affected. Ask for a fixed quote and we will tell you which band your site falls in after looking at it.
What is the legal context for small business websites in 2026?
Private businesses open to the public are covered by Title III of the Americans with Disabilities Act, and US courts and the Department of Justice have applied it to websites, though no federal regulation sets a technical standard for private business sites. The DOJ's guidance says businesses open to the public must make their websites accessible and points to WCAG as helpful guidance. In 2017 the DOJ withdrew its earlier plans to write a Title III web regulation, so there is still no Title III rule that names a WCAG version.
How many website lawsuits are filed?
Thousands each year, and the number rose again in 2025. Counts differ because trackers count different courts and case types.
| Source | Period | What it counts | Figure |
|---|---|---|---|
| Seyfarth Shaw | 2025 | Federal website accessibility suits | 3,117 (up 27% from 2,452 in 2024) |
| Seyfarth Shaw | 2025 | Top federal states for website suits | New York 1,021; Florida 961; Illinois 585; Minnesota 162; Pennsylvania 137; California 4 |
| EcomBack | 2025 | Website suits, federal and state | 3,948; New York 1,108, Florida 950, California 787 |
| EcomBack | 2025 | Suits against sites with a widget | 983 (24.90%), up from 722 (22.65%) in 2024 |
| UsableNet | January to June 2026 | Share of digital suits against companies under $50M revenue | 68% |
| UsableNet | 2026 projection | Digital accessibility suits, all courts | about 6,176 |
| UsableNet | 2026 | Companies sued that had a widget or overlay installed | approximately 20 percent |
Seyfarth Shaw counts federal cases only. It found 3,117 federal website accessibility suits in 2025, up 27% from 2,452 in 2024. Website cases were 36% of all 8,667 federal ADA Title III suits that year. Its September 2026 mid-year report counted 5,006 federal Title III suits of all types in the first half of 2026, up 9%. EcomBack, which includes state courts, counted 3,948 website suits in 2025. UsableNet, which counts more broadly again, projected about 6,176 digital accessibility suits for 2026.
Small businesses are the main target. UsableNet reported that 68% of digital accessibility suits through June 2026 involved companies with under $50 million in revenue. It also found approximately one quarter of suits were filed against companies that had faced a similar claim before. So fixing a site once and then letting it drift is a real risk.
What about Title II and the 2027 and 2028 deadlines?
Title II is a separate rule for state and local governments, not private businesses, but it sets the first federal technical standard and it is coming. In April 2024 the DOJ published a rule requiring state and local government websites and mobile apps to meet WCAG 2.1 Level AA. An interim final rule published on April 20, 2026 moved the compliance dates. ADA.gov now gives April 26, 2027 for governments serving 50,000 or more people and April 26, 2028 for smaller governments and special district governments. We cover that rule in detail in our ADA Title II web rule guide.
Why should a dentist or a contractor care? Three reasons. The DOJ's own Title III guidance points businesses to WCAG, and the Title II rule makes WCAG 2.1 AA the federal government's stated standard for public entities. Many small businesses sell to local governments, schools and transit agencies, and those buyers now need accessible vendors. And if you sell into Europe, the European Accessibility Act has applied since June 2025; see our guide to the European Accessibility Act for US companies.
None of this is legal advice. If you have received a demand letter, talk to a lawyer who handles ADA defense. What we can say from the data is practical: the failures we found on most small business sites are basic WCAG Level A and AA problems, and most of them are inexpensive to fix.
Methodology
We designed this study to answer one question: what does an automated WCAG test find on the home page of a typical US small business that has a website? Here is exactly how we did it, so anyone can repeat it.
Sample
- Source. Business listings from the DataForSEO Business Data API, which returns Google Maps business profiles, queried on November 1, 2026.
- Industries. 9 primary Google Maps categories: dentist, law firm, restaurant, clothing store (our retail group), medical clinic, real estate agency, hotel, hair salon and general contractor.
- Places. 14 metro areas in 11 states, searched within 20 km of each city center. Ten metros are in the five states that lead lawsuit counts: New York City; Los Angeles and San Diego; Miami and Tampa; Chicago; Philadelphia and Pittsburgh. The rest are Houston, Atlanta, Phoenix, Denver, Seattle and Charlotte.
- Inclusion. The listing had to show a website. We kept one site per domain.
- Exclusions. We removed links to social networks, booking apps and directories (for example a Facebook page or a booking profile), domains that appeared more than once across searches (a sign of a chain or franchise), well-known national brand domains, and every domain in the Tranco top one million list (list ID 7462X). The Tranco exclusion removes big brands and makes our sample separate from the sites the WebAIM Million draws on.
- Result. 1,242 unique domains. 994 could be tested. The rest were excluded for these reasons: bot challenge page (79), domain does not resolve (55), HTTP error status (53), TLS certificate error (24), near-empty page (20), other connection error (9), timeout (8).
Scan
- Browser. Playwright with headless Chromium, desktop viewport of 1280 by 900 pixels, US English locale.
- Engine. axe-core 4.13.0, limited to rules tagged wcag2a, wcag2aa, wcag21a, wcag21aa and wcag22aa. Best-practice and experimental rules were not run.
- Page. The home page only. We loaded it, waited for the load event (up to 12 seconds) and a further 2.5 seconds for scripts and widgets, then ran axe-core on the rendered page.
- Politeness. One page per site, at most three sites at a time, a 30 second navigation timeout, and a browser user agent that identifies us as DeveloperOfWebResearch/1.0 with a link back to this site. We did not submit forms, log in, or follow links.
- Overlay detection. We matched the page's script sources and markup against known signatures for overlay and toolbar vendors, including accessiBe, UserWay, AudioEye, EqualWeb, Equally AI, AllAccessible, Recite Me, Sienna and common WordPress accessibility toolbar plugins.
- Accessibility statement. A link on the home page whose visible text or address contains "accessibility" or "ADA". We did not judge the statement's content.
- Platform. Code signatures, such as wp-content for WordPress, static.wixstatic.com for Wix, static1.squarespace.com for Squarespace and cdn.shopify.com for Shopify.
Definitions
- Failing site: at least one axe-core violation in the WCAG A/AA rule set above.
- Failing elements: the number of page elements (axe-core "nodes") that failed a rule. One element can fail more than one rule.
- Critical, serious, moderate, minor: axe-core's impact rating for each rule.
Ethics
We publish aggregates only. We do not name, rank or contact any business in the sample, and we will not share a list of failing sites. The raw scan data is kept privately. If you run a small business and want to know how your own site scored, run it through the free checker, which runs the same kind of test.
Limitations
An automated scan finds only part of the accessibility problems on a site, so treat every number here as a minimum. These are the limits we know about.
- Automated testing is partial. Deque's 2021 study found automated testing covered 57 percent of issues by volume. WebAIM states that "absence of detected errors does not indicate that a page is accessible or conformant". Keyboard traps, confusing headings, poor alt text, inaccessible PDFs and broken screen reader flows mostly need a human.
- One page per site. Inner pages, booking flows, checkouts and forms inside frames were not tested. Those are often worse than the home page.
- Desktop only. We used one desktop viewport. Mobile layouts can fail differently, for example on target size and reflow.
- Not a random sample of all US businesses. We sampled businesses with a website and a Google Maps listing in large metro areas, in nine industries. Rural businesses, businesses without a website and other industries are not represented. Results by state describe one or two metros each.
- Overlay detection can miss. Some widgets load late, from unusual addresses, or only after consent. A few sites in our "no overlay" group may have one.
- Exclusions shape the sample. Sites that block automated browsers or did not load were left out. Their accessibility may differ.
- A point in time. Sites change. These results describe November 1, 2026.
- Not a legal finding. A failed automated rule is not a finding that a site violates the ADA, and a passed scan is not a finding that it complies.
Cite this study
You are welcome to quote and chart these findings with a link to this page. Please cite it as:
DeveloperOfWeb (2026). Accessibility of US Small Business Websites 2026: A 994-Site Scan. Published November 1, 2026. https://developerofweb.com/blog/small-business-website-accessibility-study-2026
Short form for a sentence: "85.6% of US small business home pages failed at least one automated WCAG 2.2 test in a November 2026 scan of 994 sites by DeveloperOfWeb." Journalists and researchers who want the aggregate tables or the method in more detail can reach us through the contact page. We do not share data that identifies individual businesses.
Next steps for your business
Start with a free scan, fix the four common failures, then decide whether you need a manual audit. If your home page is like most in our study, it has a handful of failures, mostly contrast and missing names, and a few days of work will clear them. If it is one of the 4.4% with more than 50 failing elements, the problem is probably in a theme or page builder, and a template-level fix or an ADA-ready redesign may be cheaper than patching.
- Run the free accessibility checker on your home page and key pages.
- See what a manual ADA compliance audit adds to an automated scan.
- Use our WCAG 2.2 checklist to work through what a scan cannot test.
- Request a free quote for remediation or an accessible rebuild.
Frequently asked questions
No scan can prove ADA compliance, but in our November 2026 scan of 994 US small business home pages, 85.6% had at least one automatically detectable WCAG 2.2 A or AA failure. Only 14.4% had none, and automated tests find only part of the issues, so the true share of fully accessible sites is lower still.
Low contrast text. It failed on 60% of the home pages we scanned and made up 47% of all failing elements. It is usually fixed by changing a few theme colors to meet the WCAG 4.5:1 contrast ratio for normal text.
Our data does not support that. 92.6% of the 121 sites with an overlay widget still had a detected WCAG failure, compared with 84.7% of sites without one. In 2025 the FTC ordered accessiBe to pay $1 million and barred it from claiming its automated products can make any site WCAG-compliant without evidence.
Yes. Courts have applied ADA Title III to business websites, and small companies are the main target: UsableNet reported that 68% of digital accessibility suits through June 2026 involved companies with under $50 million in revenue. Seyfarth Shaw counted 3,117 federal website suits in 2025.
In federal court in 2025, New York (1,021), Florida (961), Illinois (585), Minnesota (162) and Pennsylvania (137) led, according to Seyfarth Shaw. California had only 4 federal website suits because most California claims are filed in state court; EcomBack, which counts state courts, put California third with 787.
No. The ADA Title II web rule applies to state and local governments. ADA.gov gives April 26, 2027 for governments serving 50,000 or more people and April 26, 2028 for smaller ones. Private businesses fall under Title III, which has no technical web rule; the DOJ guidance points businesses to WCAG.
No. It is the right first step and it catches the most common failures, but Deque found automated testing covered about 57 percent of issues by volume. Keyboard use, screen reader flows, meaningful alt text and forms need manual testing.
Our published band for WCAG 2.2 AA remediation or an accessible build on a small site is $500 to $2,000, depending on how many templates are affected. Most failures in our study were contrast, missing names and alt text, which are fixed in the theme rather than page by page.
Sources
Each figure and date above comes from one of these. The retrieved date is when we last confirmed the source still says what we say it says.
- The WebAIM Million: 2026 report on the accessibility of the top 1,000,000 home pages WebAIM, Utah State University. Retrieved 2026-11-01.
- Federal Court Website Accessibility Lawsuit Filings Bounce Back in 2025 (March 25, 2026) Seyfarth Shaw LLP, ADA Title III blog. Retrieved 2026-11-01.
- 2026 Mid-Year Report: ADA Title III Federal Lawsuit Numbers Are Climbing (September 9, 2026) Seyfarth Shaw LLP, ADA Title III blog. Retrieved 2026-11-01.
- 2025 Annual ADA Website Accessibility Lawsuit Report EcomBack. Retrieved 2026-11-01.
- Inside the 2026 midyear numbers: where digital accessibility litigation is going (July 8, 2026) UsableNet. Retrieved 2026-11-01.
- Digital accessibility lawsuits in 2026: five trends companies should know (July 31, 2026) UsableNet. Retrieved 2026-11-01.
- FTC approves final order requiring accessiBe to pay $1 million (April 22, 2025) Federal Trade Commission. Retrieved 2026-11-01.
- Fact Sheet: New Rule on the Accessibility of Web Content and Mobile Apps Provided by State and Local Governments ADA.gov, U.S. Department of Justice. Retrieved 2026-11-01.
- Guidance on Web Accessibility and the ADA ADA.gov, U.S. Department of Justice. Retrieved 2026-11-01.
- Nondiscrimination on the Basis of Disability; Notice of Withdrawal of Four Previously Announced Rulemaking Actions, 82 FR 60932 (December 26, 2017) U.S. Department of Justice, Federal Register. Retrieved 2026-11-01.
- Automated testing study identifies 57 percent of digital accessibility issues (March 10, 2021) Deque Systems. Retrieved 2026-11-01.
- axe-core rule descriptions Deque Systems (open source). Retrieved 2026-11-01.
- Web Content Accessibility Guidelines (WCAG) 2.2 W3C. Retrieved 2026-11-01.
- Tranco: a research-oriented top sites ranking (list 7462X) Tranco (KU Leuven and partners). Retrieved 2026-11-01.
- DataForSEO Business Data API: Business Listings DataForSEO. Retrieved 2026-11-01.
About the author
Founder, Accountant & Project Manager, Four Cents
Hamza Hai is a founder of Four Cents and its accountant and project manager. She keeps projects on track and the numbers clear.
She manages timelines, scope and client communication, so there are no surprises for clients along the way.
She writes the guides on DeveloperOfWeb.com, including its website costing and planning guides. They draw on the quoting and scoping work she does for Four Cents clients every week.
Areas of expertise
- Software and app project management
- Development costing and estimates
- Project scoping
- Budgeting for software projects
All articles by Hamza Hai (106 on DeveloperOfWeb.com)
Reviewed by Zee, Founder & Lead Developer, Four Cents
Zee leads every Four Cents build as the senior full-stack engineer, from the first call to launch and beyond, and technically reviews the articles on this site.
Our editorial standards: who writes and reviews our articles, where our prices come from and how to report a correction.