ADA compliance: what it asks of a digital product

The Americans with Disabilities Act names no web standard. Here is what an auditor tests, and where the risk really sits.

What ADA compliance means online

The Americans with Disabilities Act (ADA) is a United States civil rights law. Two of its titles reach digital products. Title II covers state and local government services, including their websites and apps. Title III covers businesses open to the public, and courts have read that to include selling online.

Neither title names a web standard. That gap is why every accessibility conversation ends at WCAG, the Web Content Accessibility Guidelines, which the World Wide Web Consortium maintains. It is a technical standard: it says what has to be true of the code, not who is liable when it is not.

Where the standard is written down

One rule is explicit about it. The US Department of Justice published a final rule on 24 April 2024 setting WCAG 2.1 Level AA as the technical standard for state and local government web content and mobile apps.

Those compliance dates have moved since. An Interim Final Rule from the Department of Justice, published on 20 April 2026, extended the date for public entities with a population of 50,000 or more to 26 April 2027, and for entities under 50,000 and for special district governments to 26 April 2028. The technical standard did not change. Only the clock did.

Private sector work has no equivalent rule. Claims are decided case by case, and a settlement usually names WCAG 2.1 Level AA as the benchmark. We keep the dated changes on Accessibility news rather than guess at them.

What an ADA compliance audit tests

An ADA compliance audit is a WCAG audit framed for legal risk. The testing is the same work as any accessibility audit. What changes is the order of the findings and the language of the report.

  • A scope in writing. Which pages, flows, apps and documents were tested, at which viewport and text sizes, and what was left out.
  • Manual passes with assistive technology. Screen readers, keyboard only, screen magnification and voice control, on current versions of the browsers and devices your users have.
  • Every finding tied to a criterion. Severity, the success criterion, steps to reproduce and the fix, in that order.
  • A retest. The same cases run again after your fixes, so the second report says what changed and what did not.

The words "we tested" are cheap. A report that names its method, its tool versions and its gaps is the only kind your counsel can use.

Where products fail first

Across the audits we run, the same five areas come up before anything exotic does.

  • Forms. Fields announced with no name, errors shown in red alone, and a success message that never reaches the reader.
  • Keyboard reach. Menus, modals and date pickers that a keyboard cannot open, close or escape.
  • Images that carry meaning. Charts, product photos and icons with no text alternative, so the information is simply absent.
  • Documents. PDFs that pass a tag check and still read in the wrong order, because no one listened to one.
  • Third party code. Chat widgets, payment frames and consent banners that arrive with their own barriers.

None of these need a specialist to find once you know the shape of them. The ADA compliance checker on this site runs a first pass on markup, and the WCAG 2.2 checklist lists what to do by hand.

What an audit cannot decide

It cannot tell you whether you are complying with the law. That is a question for your lawyers, and the answer depends on who your users are, what you sell and where. We test the product and write down what we found, in a form your counsel can read.

It cannot make an inaccessible product accessible either. An audit produces evidence and a work list. The fixes are yours, or ours if you want them under consulting and support.

The order to work in

Test the product as it is, then fix blockers before polish, then prove the fixes with a retest. Test the flows that carry money, personal data or a legal obligation first, because that is where a barrier costs the most.

If you are starting from nothing, start with one page. Our free one-page audit tests a page by hand and sends back the findings with a proposed scope for the rest.

Test the page that matters

One free page, tested by hand, with the findings written out.