This video covers what accessibility scans are able to detect and where they fall short when evaluating a website for WCAG conformance.
Automated scans work by reviewing the HTML, CSS, and ARIA attributes of a web page against known WCAG success criteria. They are effective at flagging issues like missing alternative text on images, missing form labels, and incorrect heading order. These are patterns a machine can reliably identify because they follow predictable code structures.
However, scans only detect approximately 25% of accessibility issues. The remaining 75% requires human evaluation. A scan cannot determine whether alternative text is meaningful, whether a custom component is operable with a keyboard, or whether content makes sense to a screen reader user in context.
These assessments depend on human judgment. Scans are a useful starting point within a broader evaluation strategy, but they are not a substitute for a full audit conducted by an accessibility professional. Understanding what falls inside and outside automated coverage helps organizations plan their evaluation approach more effectively.