Accessibility widget for your website.

Title III of the Americans with Disabilities Act applies to public accommodations, and US courts have consistently held that this includes commercial websites. Plaintiffs filed roughly 4,000 ADA Title III web lawsuits in 2024, with ecommerce, hospitality, and education sites the most-targeted. The Department of Justice published a final rule in 2024 confirming WCAG 2.1 AA as the applicable standard for state and local government sites under Title II, and most defense lawyers now treat WCAG 2.1 AA as the de facto standard for Title III as well. A widget alone has not held up as a defense in litigation. Several reported cases involved sites that had deployed an accessibility overlay and were sued anyway, with courts finding that the underlying HTML, alt text, and form labeling were still non-compliant. The widget reduces friction for many visitors and signals good-faith effort, but the underlying code has to do its share.

The European Accessibility Act (EAA, Directive 2019/882) became enforceable across all 27 EU member states on 28 June 2025. It applies to ecommerce, banking, electronic communications, e-books, transport ticketing, and media services that serve consumers in the EU. The technical standard is EN 301 549, which incorporates WCAG 2.1 AA at minimum and WCAG 2.2 in some categories. Penalties vary by member state but include fines, removal from the market, and reputational damage. Non-EU companies are in scope if they offer services to EU consumers. The widget is helpful for visitor-side controls, but semantic HTML, real alt text, keyboard reachability, screen reader compatibility, accessibility statements, and feedback channels remain site-owner responsibilities outside this widget.

WCAG 2.1 Level AA contains 50 success criteria across four principles: perceivable, operable, understandable, and robust. WCAG 2.2, finalized in October 2023, adds nine new criteria including focus appearance, target size minimum, dragging movements, and consistent help. Most regulators reference 2.1 AA as the floor; 2.2 is becoming the working standard. A widget can help visitors adjust presentation through controls such as font scaling, contrast modes, spacing, animation pause, and reading guides. It does not fix missing alt text, incorrect heading order, broken form labels, keyboard traps, focus management, or ARIA mistakes in your underlying code.

Be honest about this. The disability community has been openly critical of AI-only accessibility overlays for years. Public statements from disability advocates, accessibility consultants, and screen reader users have documented cases where overlays interfered with assistive technology, made navigation harder, or provided a false sense of compliance to site owners. The Overlay Fact Sheet, signed by hundreds of accessibility professionals and disabled people, recommends against marketing overlays as a complete solution. ADA lawsuits have proceeded against sites that deployed overlays. Poper takes the position that a widget is a useful complement: it gives visitors immediate user-facing controls (font sizing, contrast, dyslexic font, animation pause) that they would otherwise have to dig into browser settings to find. It is not a substitute for writing accessible code, conducting manual audits, testing with real assistive technology, and fixing the issues those tests reveal.

Industry-standard accessibility programs combine four things. First, manual code remediation: fixing alt text, heading order, form labels, focus management, and keyboard reachability in the source HTML. Second, automated testing in CI using axe-core or Pa11y to catch regressions before they ship. Third, manual testing with real users of assistive technology (screen readers, voice control, switch access) and with disability consultants who can identify barriers automated tools miss. Fourth, user-facing controls so visitors can adapt presentation to their needs without leaving your site. Poper covers the fourth pillar honestly. Your team still owns the statement, feedback process, testing, and source-code remediation.