Accessibility scans detect approximately 25% of accessibility issues. The remaining 75% requires human evaluation by an accessibility professional. This accessibility scan detection rate has remained consistent across scan categories, including browser-based scanners, API-based scanners, and open source command-line tools.
| Key Point | What It Means |
|---|---|
| Detection Rate | Scans flag approximately 25% of accessibility issues on a given page or application |
| What Scans Evaluate | HTML structure, CSS properties, and ARIA attributes that can be checked programmatically |
| What Scans Miss | Keyboard operability, screen reader behavior, logical reading order, and context-dependent content |
| Implication for Conformance | A clean scan result does not indicate Web Content Accessibility Guidelines (WCAG) conformance |
What the 25% Includes
Scans work by loading a web page and running automated checks against WCAG success criteria. They evaluate the underlying code: HTML elements, CSS values, and ARIA roles and properties.
The types of issues scans reliably identify include missing alternative text attributes, incorrect heading hierarchy, missing form labels, duplicate element IDs, and language attribute omissions. These are issues where the presence or absence of a specific code pattern gives a clear pass or fail result.
Why 75% Requires Human Evaluation
Most WCAG success criteria depend on context, intent, or user experience rather than the presence of a single code attribute. A scan can confirm that an image has an alt attribute, but it cannot determine whether the text in that attribute accurately describes the image content.
Screen reader testing, keyboard testing, and visual inspection are needed to evaluate how content behaves in practice. Whether a modal traps keyboard focus correctly, whether a dynamic content update is announced to assistive technology users, or whether a custom component conveys its state are all determinations that require a trained evaluator.
How the Detection Rate Affects Automated Accessibility Strategy
The 25% detection rate means scans are one component of an accessibility evaluation strategy, not a substitute for one. Organizations that rely on scans alone operate with visibility into roughly a quarter of potential issues on their site.
Scans are most effective as a first pass before a full audit and as ongoing monitoring between audits. Scheduled scans, conducted daily, weekly, or monthly, catch regressions in the issues they can detect. New code deployments that break heading structure or remove form labels surface quickly through recurring scans.
A Clean Scan Does Not Mean Conformance
A scan that returns zero issues means the page passed automated checks, which cover approximately 25% of WCAG criteria. The remaining 75% is unexamined. Presenting a clean scan as evidence of WCAG conformance misrepresents the scope of the evaluation.
An audit conducted by an accessibility professional evaluates against the full set of applicable WCAG success criteria at a specified conformance level, typically 2.1 AA or 2.2 AA. That evaluation identifies issues across the complete range of criteria, including the 75% that scans cannot reach.
The accessibility scan detection rate of 25% is a useful benchmark for understanding where automated tools fit within a broader conformance program and where human expertise becomes necessary.