Accessibility overlays vs code fixes: what the FTC case and lawsuit data show

What an overlay is, and what it cannot change

An overlay runs after your page loads. It can add a toolbar (bigger text, contrast modes, a reading mask) and it can try to patch the page automatically, for example by guessing alt text or adding ARIA attributes.

What it cannot do is change your HTML, CSS and components at the source. A product grid built from unlabelled divs, a cart drawer that loses keyboard focus, a checkout error that is never announced: those live in your theme code and are still there for every user whose assistive technology reads the real page. The European Disability Forum and IAAP put it plainly in 2023: overlays "do not make the website accessible or compliant with European accessibility legislation" and "do not constitute an acceptable alternative or a substitute for fixing the website itself" (EDF and IAAP).

Toolbar features also duplicate settings people already have in their browser, operating system or screen reader, and can conflict with them.

The FTC accessiBe case

In January 2025 the US Federal Trade Commission announced a complaint and proposed order against accessiBe, the best-known overlay vendor. On 22 April 2025 it approved the final order, which requires accessiBe to pay USD 1 million (FTC).

The FTC alleged two things (FTC case page, complaint and order):

• accessiBe claimed its accessWidget product could make any website compliant with WCAG and keep it compliant, and in the FTC's words "accessWidget did not make all user websites WCAG-compliant and these claims were false, misleading, or unsubstantiated";
• it presented articles and reviews as independent when they were connected to the company.

The order bars accessiBe from claiming that its automated products make a site WCAG compliant, or keep it compliant over time, unless it has the evidence to support the claim, and from passing off paid or connected endorsements as independent (FTC).

The case was about advertising claims, not about whether overlays are lawful to sell. It is still the clearest regulator finding on the core promise of the category.

What the lawsuit data shows

The best long-running count comes from UsableNet, which tracks US digital accessibility lawsuits (2025 year-end report):

• 5,114 ADA digital accessibility lawsuits were filed in 2025: 3,195 in federal courts and 1,919 in New York and California state courts.
• E-commerce was the target in 70% of them.
• The report's monthly chart counts lawsuits against companies already using an accessibility widget: between 95 and 155 a month, 1,416 across the year. That is about 28% of all filings. UsableNet's conclusion: "Accessibility widgets did not slow litigation in 2025" (report, summary).

Customers have turned on vendors too. In June 2024 a New York dermatology practice that had bought accessiBe filed a proposed class action against it, alleging the product was sold as making its site compliant and that the practice was sued anyway (complaint). Those are allegations, not findings.

These are US figures. EU enforcement works differently, through regulators, associations and in Germany competitors, but the same question decides a complaint in both systems: can a disabled customer use the site? A widget does not change the answer if the code underneath fails.

Where the accessibility community stands

• European Disability Forum and IAAP (2023): overlays do not make a site accessible or compliant with European law and are no substitute for fixing the site (statement).
• Overlay Fact Sheet: 1,048 signatories on 1 October 2026, including accessibility specialists from Google, Microsoft, Apple and Shopify, commit to "advocate for the remediation of accessibility issues at the source of the original error" and call on site owners to remove overlays (overlayfactsheet.com).
• National Federation of the Blind (US): its 2021 convention resolutions criticised overlay vendors' claims (NFB), and it banned accessiBe from its national convention that year (Forbes, 26 June 2021).

What a code fix looks like

The failures that show up most often are mundane and fixable in source. The WebAIM Million 2026 scan of one million home pages found detectable WCAG failures on 95.9% of them. The six most common failure types were:

• low contrast text, on 83.9% of home pages: change the colour tokens;
• missing image alt text, 53.1%: add meaningful alt text in the product and media templates;
• missing form input labels, 51%: add real labels to search, newsletter and checkout fields;
• empty links, 46.3%: give icon links an accessible name;
• empty buttons, 30.6%: name the cart, menu and close buttons;
• missing page language, 13.5%: set the lang attribute per locale.

Each of these is a change to your theme or components. Once it is made, it stays made, and a re-scan proves it. Beyond the scanner's reach, a person has to check keyboard flow, focus handling in drawers and modals, and whether error messages are announced.

What each option costs

Entry-tier overlay plans cost around USD 490 a year, every year, for as long as the script stays on (accessiBe, UserWay). The code underneath does not improve.

A code fix is a one-time project, then upkeep when the site changes. A CodeBetter Fix Sprint starts at EUR 1,500, fixes the in-scope issues in your own code, has a person review every change, and ends with a before and after re-scan as evidence. Start with the free scan to see what you are dealing with.