accessiBe vs UserWay: Do Accessibility Overlays Work?

A law firm emails you. It says your website blocks screen-reader users, cites the ADA, and names a settlement number. You search for a fix, and the first results promise ADA compliance for $49 a month — one line of JavaScript, done by lunch.

That purchase does not fix your website and it does not end the letter. On January 3, 2025 the FTC announced that accessiBe would pay $1 million to settle allegations that it misrepresented the ability of its AI-powered tool to make any website compliant with WCAG; the Commission approved the order on a final basis in April 2025. This post is the honest version: what overlays actually do to your page, what the evidence says, and what to fix first.

Do accessibility overlays actually make a website accessible?

No. Not one of them gets you to full conformance with any accessibility standard. The overlay factsheet — signed by over 1,000 accessibility practitioners, including WCAG and ARIA spec contributors and people who write code for the NVDA and JAWS screen readers — states it flatly: 'No overlay product on the market can cause a website to become fully compliant with any existing accessibility standard and therefore cannot eliminate legal risk.'

The reason is structural. Conformance to WCAG means satisfying every success criterion — there is no partial credit. An overlay is a script that runs in the visitor's browser after your page loads, guessing at what your markup meant. Guessing works for some things and fails at exactly the things that block people.

Per the factsheet, automated repair is unreliable for all of the following:

• Text alternatives for images — the script does not know what your photo of the crew on a roof is for
• Form field labels, error handling, and focus control — the parts a customer must use to book you
• Keyboard access — whether a user can tab to your quote button and press it
• Component-based UIs (React, Angular, Vue) that change page state after the overlay has run
• Content the overlay cannot see at all: PDFs, video, HTML5 canvas, SVG, Java, Flash

Will an overlay widget stop an ADA demand letter or lawsuit?

No — and there is a case count. UsableNet's 2025 midyear report counted 2,019 digital accessibility lawsuits filed in the first half of 2025, and plaintiffs sued companies that were running an accessibility widget in every single month of that half-year, with 132 filed in the peak month.

The same report found 64% of those lawsuits targeted companies under $25 million in revenue. This is not a Fortune 500 problem. By industry, e-commerce drew 69% of filings, food service 18%, healthcare 4%, and fitness and wellness 3%.

The legal exposure is real even without a regulation to point at. The DOJ's guidance on web accessibility and the ADA says the Department has 'consistently taken the position' that Title III covers the goods and services businesses offer online, while also confirming it has no regulation setting out detailed technical standards for private businesses. Flexibility in how you comply is not permission to skip complying.

We build websites — we are not lawyers, and nothing here is legal advice. If you have a letter in hand, your first call is a lawyer who handles ADA Title III claims, not a vendor.

What do accessiBe, UserWay, and AudioEye actually do to your page?

All three load a third-party JavaScript file that runs after your page renders, then rewrites the DOM in the browser — injecting ARIA attributes, guessing at alt text, relabelling controls — and pins a floating button that toggles contrast, font size, and spacing. accessiBe's accessWidget starts at $49/month on its Micro plan, for up to 5,000 page views a month; UserWay's entry Pro plan is listed at $490 a year, roughly $41 a month, for up to 100,000 page views.

Verdict: none of the three widgets wins, because the widget is not the product that helps. AudioEye is the only one whose lineup sells the thing that changes what a screen reader encounters — human expert audits and custom fixes — and that work is priced and delivered separately from the automation. If the entry-tier script is what you are buying, you are buying a badge.

One more reason to be suspicious of automated ARIA injection: the 2026 WebAIM Million found home pages using ARIA averaged 59.1 detected errors versus 42 on pages without it. ARIA sprayed at a page by a script is not free — done wrong, it invents barriers.

Why does the accessibility community oppose overlays?

Because the people who test with screen readers every day rate them as failures. In WebAIM's 2021 survey of 758 accessibility practitioners, 67% rated overlays, plugins, and widgets 'not at all' or 'not very' effective. Among respondents who have a disability, that rose to 72%, with only 2.4% calling them very effective.

This is not a competitor smear campaign. The factsheet's signatories include contributors and editors of the WCAG, ARIA, and HTML specifications, internal accessibility staff at companies like Google, Microsoft, Apple, and Shopify, lawyers for the disabled, and code contributors to JAWS and NVDA.

Their second complaint is that the visible widget is redundant. A blind user already runs a screen reader. A low-vision user already has OS-level zoom and contrast. If they needed your floating button to use the web, they would need one on every site they visit — which is not how assistive technology works.

What does real WCAG remediation cost and take?

Nobody can quote you a credible flat price, and any vendor who quotes one is selling the same promise that cost accessiBe $1 million. What is knowable: in WebAIM's practitioner survey, only 17.6% of respondents believed automated testing can detect even half of accessibility issues — the most common answer was 20–30%. So a real fix is part tooling, part manual keyboard and screen-reader testing, and part developer time in your own codebase.

The good news for a 20-page service-business site: the barriers are boringly repetitive. The 2026 WebAIM Million found six error types account for 96% of all detected errors across a million home pages.

That list is CSS and HTML, not a rebuild. It is the same work we fold into a conversion-focused website build — accessible markup and fast markup are mostly the same markup.

What should you fix first if you just got a demand letter?

Fix keyboard access and forms first, because that is where both the complaint and your revenue live: the quote form, the call button, the booking flow. Work in this order:

• Call an ADA Title III lawyer. The letter has a clock on it and a vendor cannot answer it for you.
• Unplug your mouse and tab through the pages the letter names. If you cannot reach and submit your own form, neither can the plaintiff.
• Fix the six error types above at the source — contrast, alt text, labels, empty links, empty buttons, page language.
• Run a screen reader through your booking flow and listen. NVDA is free on Windows; VoiceOver ships on every Mac.
• Publish an accessibility statement with a working way to report problems. The DOJ guidance names this as a practice that helps owners find and fix issues.
• If you already bought an overlay, remove it. It is not defending you, and it may be adding barriers of its own.

Everything on that list is permanent. It survives your next redesign, your next CMS, and your next agency — which is the point. You should own the fix, the same way you should own your codebase, your ad accounts, and your content.

Does an overlay slow your site or hurt conversions?

It adds a third-party script to every page, and the overlay factsheet states that overlay repairs 'can either slow down page load times or cause unexpected page changes for assistive technology users.' You are paying a subscription to make the page heavier.

That matters because Google's current Core Web Vitals thresholds are an LCP of 2.5 seconds or less, an INP of 200 milliseconds or less, and a CLS of 0.1 or less, each measured at the 75th percentile of real-user page loads. A widget that repaints the DOM after load spends your interactivity budget for you. If you are chasing those numbers, our guide to hitting perfect Core Web Vitals scores is a better use of the money.

And speed is revenue on a lead-gen site. Portent's 2022 analysis of 20 sites and 5.6 million sessions found B2B sites loading in 1 second converted at roughly 3× the rate of sites loading in 5 seconds. Slowing the page while leaving the barrier in place is losing twice — which is exactly the kind of thing a proper CRO audit surfaces.

Is there any case where an overlay is the right call?

Close to none, and the case gets weaker every year. The strongest honest argument for an overlay is that some automated repair is genuinely possible — the factsheet concedes 'a non-trivial array' of problems can be patched this way. But partial repair is not conformance, the user-facing panel duplicates tools your visitors already have, and since January 2025 the compliance marketing itself carries an FTC order behind it.

If you have already installed one and are mid-way through a real remediation project, removing it is not urgent on day one — but it is not buying you the protection you think it is, and the accessibility community's position is that it should come off. Do not let it become the reason the source-code work never gets scheduled.

The honest summary: an overlay converts a code problem into a subscription. The code problem is still there, plaintiffs know how to look for it, and your customers still cannot book you.

If you would rather fix the site than rent a badge, we build conversion-focused websites with accessible markup as part of the build, month-to-month, no lock-in. Get my free audit and we will tell you what is actually broken — including the parts a widget was never going to reach.