ADA compliance audit services compared
An "ADA compliance audit" of a website is a test against WCAG. The ADA sets no detailed web standard for businesses, DOJ points to WCAG as helpful guidance, and no one certifies a site. What you buy is a person who works through your pages by keyboard and screen reader against WCAG 2.1 or 2.2 Level AA and writes down, criterion by criterion, what fails. Here are five accessibility audit services that do that work by hand, compared on what their own pages promise. For the law behind it, start with what the ADA asks of a business website.
The verdict by use case
Best overall
Its audit page names the WCAG version and level, keyboard and screen reader testing, the success criterion on every issue, and a check of your fixes afterwards.
Best for a small WordPress site
It audits and then fixes WordPress sites itself, in the source code, with blind and visually impaired testers checking the result.
Best for a one-time audit before a deadline
Its page promises a report in 1 to 2 weeks and answers the demand-letter case directly; ask in writing whether a person retests your fixes.
Side by side
Scroll sideways to see all vendors.
| Requirement | Accessible.org | Equalize Digital | Knowbility | WebAIM | Deque |
|---|---|---|---|---|---|
| States WCAG 2.1 AA (or 2.2) as the target | Yes | Check with the vendor | Yes | Not stated | Not stated |
| Manual testing by a human auditor, not scan-only | Yes | Yes | Yes | Yes | Yes |
| Keyboard and screen-reader testing named | Yes | Yes | Check with the vendor | Not stated | Not stated |
| Report cites success criteria by number | Yes | Not stated | Check with the vendor | Not stated | Not stated |
| Fixes aimed at your source code, not an overlay | Yes | Yes | Yes | Not stated | Yes |
| Retest after fixes | Yes | Yes | Yes | Yes | Check with the vendor |
| Writes an ACR (a completed VPAT®) for your product | Yes | Yes | Not stated | Yes | Yes |
| Works with your CMS | Not stated | Yes | Not stated | Not stated | Not stated |
| PDF remediation | Yes | Not stated | Not stated | Not stated | Not stated |
| Captioning | Not stated | Not stated | Not stated | Not stated | Not stated |
The vendors
Accessible.org
- What it is
- An audit firm that tests by hand and hands back a spreadsheet of issues. Its page says every auditor is a certified DHS Trusted Tester, and that it makes no fixes during the audit.
- Who it fits
- An owner with a developer who wants a fixed-scope report, with the WCAG version and level written into the order.
- WCAG 2.1 AA or 2.2 AA named as the targetYes
- Keyboard and screen reader testing namedYes
- Each issue cites its success criterionYes
- Your fixes checked again afterwardsYes
- Works with your CMSNot stated
Equalize Digital
- What it is
- A WordPress accessibility agency, maker of the Accessibility Checker plugin, that audits a site and then fixes it as it goes. Its page says it can make the fixes as long as they are in WordPress.
- Who it fits
- A small WordPress site that wants one team to find and fix the problems in the theme, plugins and content.
- WCAG 2.1 AA or 2.2 AA named as the targetCheck with the vendor
- Keyboard and screen reader testing namedYes
- Fixes made in your source codeYes
- Your fixes checked again afterwardsYes
- Works with your CMSYes
Knowbility
- What it is
- An Austin, Texas accessibility organisation whose experts test a sample of screens by hand and, after a clean retest, issue a letter of validation for the screens they examined.
- Who it fits
- An owner who wants the fixes explained on a call and a written validation letter at the end.
- WCAG 2.1 AA or 2.2 AA named as the targetYes
- Keyboard and screen reader testing namedCheck with the vendor
- Fix advice down to the codeYes
- Your fixes checked again afterwardsYes
- Writes an ACR (a completed VPAT®)Not stated
WebAIM
- What it is
- The Utah State University team behind the WAVE checker. Its audit is a deep look at a representative sample, usually around 8 to 15 pages or views, written to teach your team the patterns.
- Who it fits
- A team that wants to learn to fix things itself, and a public conformance statement once the barriers are verified fixed.
- WCAG 2.1 AA or 2.2 AA named as the targetNot stated
- Manual testing by a personYes
- Your fixes checked again afterwardsYes
- Writes an ACR (a completed VPAT®)Yes
- Keyboard and screen reader testing namedNot stated
Deque
- What it is
- A large accessibility firm, maker of the axe testing tools, whose audits combine automated and manual testing. Its page says reports arrive in 1 to 2 weeks.
- Who it fits
- A business on a short clock, or one that will also need an ACR for a product it sells.
- Manual testing by a personYes
- Fixes made in your source codeYes
- Your fixes checked again afterwardsCheck with the vendor
- Writes an ACR (a completed VPAT®)Yes
- WCAG 2.1 AA or 2.2 AA named as the targetNot stated
A scan, an overlay and an audit are three different things
A scan is software reading your code. It finds an image with no alt attribute in seconds, but it cannot tell you that the alt text on your logo says "image123", or that the booking calendar swallows the Tab key. An overlay is a third-party widget or toolbar script sold as making a site conform without changing its source. An audit is a person working through your pages with a keyboard, a screen reader and the source, and writing down every place a WCAG success criterion fails.
What the Department of Justice says
A "clean" report does not necessarily mean everything is accessible. Also, a report that includes a few errors does not necessarily mean there are accessibility barriers. Pairing a manual check of a website with the use of automated checkers can give you a better sense of the accessibility of your website.
From DOJ's guidance on web accessibility and the ADA, on ada.gov. The guidance is informal and not legally binding.
Most of the firms above also run a scanner as a second check, and that is fine. What you are paying for is the person. If a quote describes the work as a scan, it is a scan, whatever it is called. The free checkers are compared in accessibility testing tools compared, and why a widget does not change your code is in our guide to accessibility overlays. If a widget is coming off your site, accessiBe alternatives that fix your code sets out the four steps that replace it.
What one finding in a good report looks like
A long list of errors with no page, no criterion and no code is hard to act on. A useful finding tells your developer where the barrier is, who it stops, which success criterion it fails and what to change, so the fix takes minutes rather than a meeting. This is the shape to ask for, with a real fail a small site often has.
| Field | What the auditor wrote |
|---|---|
| Issue | The date picker on the booking form traps the keyboard. Tab moves into the calendar and never leaves it; Escape does nothing. |
| Where | /book, step 2, the date field's calendar |
| Tested with | Keyboard alone, then NVDA with Chrome on Windows 11 |
| Success criterion | 2.1.2 No Keyboard Trap (Level A) |
| Code | The calendar's keydown handler calls preventDefault() on every Tab press |
| Evidence | A 20-second screen recording of the focus stuck on the 14th |
| Fix | Let Tab and Shift+Tab leave the calendar, close it on Escape, and send focus back to the date field |
| Retest | Pass, same keyboard and screen reader set-up, after the fix shipped |
What WCAG asks
If keyboard focus can be moved to a component of the page using a keyboard interface, then focus can be moved away from that component using only a keyboard interface, and, if it requires more than unmodified arrow or tab keys or other standard exit methods, the user is advised of the method for moving focus away.
Success criterion 2.1.2 No Keyboard Trap, Level A, WCAG 2.2.
You can run the same keyboard walk on your own pages before an auditor does with the focus order checklist, and the screen reader part with screen reader testing for owners.
How an auditor chooses the pages
Nobody tests every page of a large site; they test a sample that stands in for it. W3C published the second edition of its method for this, WCAG-EM 2.0, as a Group Note on 23 July 2026. It asks for a structured sample of common pages, key tasks and each type of content, plus a random set worth 10% of that sample on top, and every step of any process a visitor has to complete. It also notes that Level AA is the generally accepted target.
| Page or view | Why it is in the sample |
|---|---|
| Home page | A common view every visitor sees |
| About the team | A common view, and the page with the most photos |
| One treatment page | Stands in for the 12 pages built from the same template |
| One blog post | Stands in for every post |
| Contact page with its form | Essential functionality |
| Booking, step 1: pick a treatment | Part of a complete process, so every step is in |
| Booking, step 2: pick a date | Part of a complete process |
| Booking, step 3: your details and confirm | Part of a complete process |
| New patient forms page with PDF links | A different sample type: documents |
| Location page with the embedded map | A technology the site relies on: a third-party embed |
| One page picked at random | The random set: 10% of the 10 pages above, so 1 page |
The booking steps are the easiest pages to cut from a quote, and they are where real customers get stuck. Keep in mind that WCAG conformance is claimed for full pages: an audit reports on the pages in its sample, and the template fixes it leads to are what carry the result to the rest of the site.
Check the scope line
Before you sign, ask for the sample as a list of URLs. A quote that says "up to 10 pages" with no list leaves the choice of pages, and the result, to the firm.
Nine questions to send before you sign
Each question matches a row of the table. Send them with your page count; a firm that answers all nine in writing has told you most of what the comparison could not. For what drives the quote itself, read what ADA website compliance costs.
- Which WCAG version and level will you test against (WCAG 2.1 AA or 2.2 AA), and will you write it into the order?
- Which pages and views are in the sample, does every step of our booking or checkout count, and do you add pages picked at random?
- Do you test by keyboard alone, and with which screen reader and browser pairs?
- Does every finding name its success criterion by number, the page, the element and the code involved?
- Who makes the fixes, and do they go into our source code rather than a script or overlay?
- Is a retest of our fixes included, and is it done by a person or by a scan?
- What do we hold at the end: the report, a conformance statement, an ACR?
- Are our PDFs and videos in scope, or are they separate work?
- On what date will the report arrive?
How we compared
Vendor pages read on
We listed firms whose own audit page says a person tests the site against WCAG, and which sell no overlay widget. Every status word comes from the firm's own pages as read that day. Yes means the page says it. Not stated means the page does not say it, which is not the same as no. Check with the vendor means the page says part of it.
Prices are left out on purpose; ask each firm to quote on your list of pages. Official lists of tools and certified testers are on our evaluation tools and auditor lists page, and what a finished ACR contains is in what is a VPAT.
Verdicts are editorial: no vendor can buy a verdict position or displace another vendor.
No overlay widget is listed. By an overlay we mean a third-party widget or toolbar script sold as making a site conform without changing its source. In April 2025 the FTC approved a final order that bars one overlay vendor from claiming, without evidence to support it, that its automated products can make any website WCAG-compliant. Read the FTC's announcement of the order.
Nothing here is legal advice. For a specific case, ask a lawyer.
Questions
Is there such a thing as an ADA certified audit?
No. DOJ has no regulation setting detailed web standards for businesses open to the public; its position is that the ADA's general rules still apply to websites. No federal body certifies a site or an auditor. Some firms issue their own paper at the end, such as WebAIM's conformance statement or Knowbility's letter of validation. Those are the firm's word on the pages it tested, which is worth having, but they are not a government certificate.
Should the audit use WCAG 2.1 or WCAG 2.2?
Ask for WCAG 2.2 Level AA if the firm offers it. Content that conforms to WCAG 2.2 also conforms to WCAG 2.0 and 2.1. WCAG 2.2 added 9 success criteria to WCAG 2.1: 2 at Level A, 4 at AA, 3 at AAA, so an AA audit to 2.2 adds six new ones to test. DOJ's Title II rule for state and local governments names WCAG 2.1 Level AA, so a supplier to a city or county may be asked for that version by name. The differences are in what changed in WCAG 2.2.
Can I run a free checker instead of paying for an audit?
Run one first; it is free and it clears the easy errors before the auditor's clock starts. It will not replace the audit: DOJ says automated checkers need to be used carefully and a clean report does not mean a site is accessible. Our WCAG 2.2 AA checklist lists every Level A and AA criterion as a plain test you can run between audits.
Does an audit cover our PDFs and videos?
None of the five audit pages we read says the audit covers them. PDFs are usually separate work (Accessible.org, for one, sells PDF remediation as its own service), and captions are a separate service again. Ask whether they are in scope, and compare PDF remediation services and captioning services if they are not.
We have received a demand letter. Is an audit the first step?
An audit shows what is actually wrong with the site, which you and your lawyer will both need. Speak to the lawyer before you commission it, and keep a dated record of every step. The demand letter response checklist lists them in order.
This is general information, not legal advice. For your own situation, ask a lawyer.
Related
- ADA compliant website: what the law asks10 min read
- What ADA website compliance costs12 min read
- VPAT and ACR writing services comparedFirms that test a software product and write its accessibility conformance report.
- Accessible alternatives to accessiBeAudit services and testing tools instead of an overlay widget, and why overlays are left out.