We do not sell an accessibility overlay
A widget that promises compliance for a monthly fee is the most attractive offer in this market and the one with the least evidence behind it. Practitioners, a disability advocacy organisation, a regulator and even a company that owns an overlay have all said so, in their own words.
Scan a public URL free · no credit card required
- Of practitioners rated overlays not at all or not very effective
- 67%
- Among respondents with disabilities — 2.4% rated them very effective
- 72%
- Payment required by the FTC's final order, approved April 2025
- $1 million
- Products listed on the Overlay Fact Sheet
- 40+
Every figure here belongs to somebody else and was read on 11 September 2026: WebAIM's third Survey of Web Accessibility Practitioners (2021, 758 valid responses); the United States Federal Trade Commission's final order of April 2025; and the Overlay Fact Sheet, signed by over a thousand practitioners, developers and people with disabilities at the time we read it.
What the record says, with names attached
We are an interested party, so almost none of this is our claim. Each card is what a named third party published, in its own words — every source read on 11 September 2026. Check the primary sources rather than this summary of them.
Federal Trade Commission · order approved April 2025
“A line of script makes a site WCAG compliant”
The FTC barred accessiBe from representing that its automated products can make any website WCAG compliant, or keep it so, without evidence — after alleging in January 2025 that the claim was false, misleading or unsubstantiated, and separately alleging undisclosed connections to third-party reviewers. The final order required a $1 million payment.
WebAIM · 3rd Survey of Web Accessibility Practitioners, 2021
“Practitioners find overlays effective”
Of 758 valid responses, 67% rated overlays not at all or not very effective. Among respondents with disabilities the figure was 72%, with 2.4% rating them very effective.
National Federation of the Blind · 2021
“Disability organisations support them”
The NFB passed a resolution condemning the practices of overlay providers, in its own words that accessiBe “overstates the effectiveness of automated testing” and promotes “false claims of attaining Web Content Accessibility Guidelines compliance through the integration of one line of code”. Its board revoked accessiBe's convention sponsorship that June.
UsableNet · 2025 mid-year lawsuit report
“Installing one settles the legal question”
Lawsuits were filed against companies using accessibility widgets in every month of the first half of 2025, with monthly counts running from the mid-eighties into the low hundreds. We draw no legal conclusion from that, and nothing here is legal advice.
Level Access · which acquired the overlay vendor UserWay
“It's a complete solution on its own”
Level Access published a position describing overlay technology as something that can be part of an overall solution but which, on a standalone basis, isn't sufficient to substantiate compliance with most accessibility requirements.
What the category actually is
An overlay is third-party code, usually JavaScript, added to the front end of a site. Most products combine two distinct things, and the two halves attract different criticism.
Automated remediation
The script inspects the page as it loads and attempts repairs — adding attributes, guessing labels, adjusting roles in the accessibility tree. This is where the compliance claim comes from, and where the evidence is least kind.
A preferences toolbar
Larger text, higher contrast, a reading mode. Largely redundant — the people who need these already have them in their operating system, browser or assistive technology, set the way they prefer across every site rather than one.
Who is in the category
accessiBe, AudioEye, UserWay, EqualWeb, Recite Me and Equally.ai among more than forty products named on the Overlay Fact Sheet, a document signed by over a thousand practitioners, developers and people with disabilities.
Why a script cannot finish the job
The limit is not effort or cleverness. Most of the standard asks a question about meaning — whether an alt attribute describes the image or merely exists, whether focus order follows the reading order — and code inspecting a page has no access to meaning. Automated rules reach 23 of the 55 WCAG 2.2 criteria at A and AA, and that is the count with any rule at all, not full coverage.
Recorded by the Overlay Fact Sheet:
- Content an overlay cannot repair — PDFs, Canvas, SVG, media and embedded technologies
- Toolbar features — duplicate controls users already have in their OS, browser or assistive technology
- Detecting assistive-technology use — raises a privacy concern — it discloses something a visitor did not choose to share
- Full WCAG conformance — not achievable through an overlay alone
What we sell instead
Nothing we sell touches your site. That is slower and more expensive than a script, and it is the honest version of the offer.
A scanner, not a script on your page
The scanner opens your pages in a real desktop browser on our own machines and gives you a report. Every change is made by your developers, in your code — the only place a barrier can be removed.
Reporting built to make the work small
The plan is ranked by how many issues each change clears rather than by severity, with a running total on every row, so two hundred findings usually turns out to be four changes and a long tail.
A repair told apart from a coincidence
Each run is compared with the one before it — fixed, introduced and unchanged — so you can tell a repair from a coincidence.
A report that says what it did not measure
Criteria with no automated rule behind them are marked not checked rather than passed — which makes our output look worse than a scanner that reports silence as success, and makes it safe to forward to someone who will act on it.
What this page is careful not to claim
- No legal advice, and no legal conclusions.
- We describe public records and attributed positions. We make no assertion about the lawfulness of any product or about any company's liability, and nothing here is a substitute for advice from a qualified lawyer.
- We are not neutral.
- We sell an accessibility scanner, so we have an interest in this argument. That is why the claims above are attributed to named third parties you can go and read.
- Our tool does not fix anything either.
- It reports. Every repair is a change your developers make in your code, and automated rules reach fewer than half of the WCAG 2.2 AA criteria.
- No certification, no VPAT, no crawler.
- Nothing we produce is a conformance claim; a scan takes up to ten addresses you choose, results are kept ninety days, and positions and orders cited here are dated — check the primary sources.
Frequently asked questions
A piece of third-party code, usually JavaScript, added to a website's front end. Most products do two things: they attempt automated repairs of the page as it loads, and they add a toolbar of user preferences such as larger text or higher contrast. The Overlay Fact Sheet, maintained by accessibility practitioners, lists more than forty such products.
The Overlay Fact Sheet states that full compliance with WCAG cannot be achieved through an overlay alone, and Level Access — which acquired the overlay vendor UserWay — has published that overlay technology on a standalone basis is not sufficient to substantiate compliance. In an order approved in April 2025 the FTC barred accessiBe from representing that its automated products can make any website WCAG compliant, or keep it so, without evidence.
We are not lawyers and this is not legal advice. What is on the public record is that UsableNet's 2025 mid-year report records lawsuits filed against companies using accessibility widgets in every month of the first half of that year, from the mid-eighties into the low hundreds. Take advice from someone qualified about your own obligations.
Nothing to your site at all. We run a scanner in a real desktop browser on our own machines, report the findings against the WCAG version and level you selected, rank the fixes by how many issues each one clears, and compare each run with the previous one. The changes are made by your developers in your code — the only place they can actually be made.
No. There is no overlay in the product, no partner overlay, and nothing we sell injects itself into your pages at runtime. The nearest thing we have to code on your side is a Chrome extension a developer installs on their own machine to work a page in DevTools.
Related
- TestKase vs Sauce LabsReal and virtual devices, per-thread pricing, and one workspace
- TestKase vs LambdaTestHow we compare with LambdaTest, now TestMu AI
- TestKase vs BrowserStackA BrowserStack alternative for real-device testing
- All of Why TestKaseHow we compare with the tools you are probably already paying for
- PricingPer-seat test management and per-thread device testing, priced on the page
The slower answer, which is the only one
Scan a public address, read the ranked plan, and see how much of it four changes clear. Nothing is installed on your site.