Skip to main content
Accessibility Pro
Startscanning

Guides · Choosing a tool

Accessibility overlays versus scanners

Both are sold as ways to deal with website accessibility, and both involve software looking at your pages. They do opposite things. An overlay changes the page in the visitor's browser. A scanner reports what is wrong so your team can change the page itself. Here is what each can and cannot do, and what public bodies have said about them.

Overlay

  • A script, usually JavaScript, added to every page.
  • Adds a toolbar and tries to repair the page in the visitor's browser.
  • Your source code is unchanged.
  • Stops working if the script fails to load or is blocked.

Scanner

  • Loads pages and tests them against WCAG rules.
  • Reports each failure with the element and the criterion.
  • Your developers change the code, and every visitor gets the fix.
  • Adds nothing to the page people load.

What public bodies say about overlays

  • The European Commission's web accessibility policy page states that overlays are not an appropriate solution, and that it is best to fix accessibility issues at their source.
  • Germany's federal monitoring body for accessible information technology, BFIT-Bund, concluded in March 2025 that adding an overlay tool does not necessarily make an inaccessible site accessible under the law.
  • The Overlay Fact Sheet, with more than 1,000 signatories as of September 2026, states that full compliance cannot be achieved with an overlay.

What scanners cannot do

Scanners have limits too, and an honest one says so. W3C's guidance on evaluation tools puts it plainly: tools cannot check all accessibility aspects automatically, human judgement is required, and tools can produce false or misleading results. Two studies measure how much automation finds, in different ways:

  • By number of issues: Deque, which makes axe-core, reports that its automated tests identified 57.38% of the issues in a sample of first-time audits. It also notes the automated issues fell under only 16 of the 50 WCAG 2.1 Level AA success criteria.
  • On a deliberately broken page: the UK Government Digital Service built a page with 143 barriers in 2017; the tools it tested found 71% between them, the best single tool found 41% when its prompts for manual checks are counted, and 42 barriers were missed by every tool.

That is why Accessibility Pro reports which criteria it could not evaluate as not tested, never as passed, and publishes its own precision and recall.

Choosing

If your goal is a site that works for people using assistive technology and holds up in an audit, the fixes have to be in the code, whichever tool finds them. A scanner makes that work faster and keeps it from regressing, especially when it runs on every pull request. It does not replace testing with people, and neither does an overlay. Start with the fix guides for the most common failures, or compare specific products: accessiBe and Siteimprove.

Sources

Checked on 2026-09-13. The Overlay Fact Sheet signatory count changes; the figure is the count on that date.