Nine criteria added, one removed
WCAG 2.2 became a W3C Recommendation in October 2023, and most organisations are still testing against 2.1 out of habit. The update adds nine new success criteria and removes one: 4.1.1 Parsing, now obsolete because modern browsers handle malformed HTML consistently enough that the old rule stopped being meaningful. The total now sits at 87 success criteria across the three conformance levels.
The nine additions target three groups the earlier version underserved: people with low vision, people with cognitive or learning disabilities, and mobile users with limited fine motor control.
The criteria that actually change how you test
Focus Not Obscured means a sticky header or cookie banner can no longer sit on top of the element that currently has keyboard focus. Test this by tabbing through a page with any overlay open, not just with the mouse.
Target Size (Minimum) sets a 24 by 24 pixel floor for clickable targets, unless there's enough spacing around a smaller one. This one catches a lot of icon-only buttons and closely packed mobile navigation that pass a visual review but fail on a real device.
Dragging Movements requires that anything currently operated only by a drag gesture, a slider or a reorderable list, also works with a single click or tap alternative. If your only way to test this is to drag something with a mouse, you haven't tested it.
Consistent Help asks that help mechanisms, a contact link, a chat widget, a support phone number, appear in the same relative place across pages. It sounds minor. It fails constantly on sites built page by page rather than from a shared template.
Redundant Entry and Accessible Authentication (Minimum) both target forms: don't make a user re-enter information the system already has in the same session, and don't make login depend purely on a cognitive test like solving a puzzle, unless there's an alternative.
What this means for a test plan
Most WCAG 2.1 test scripts don't cover any of the above, because none of it existed yet. Automated accessibility scanners catch a handful of the older criteria reliably: colour contrast, missing alt text, heading structure. They catch almost none of the WCAG 2.2 additions well, because most of the new criteria depend on interaction, not markup. Target size and dragging alternatives need a person testing with a keyboard and a touch device, not just a linter running against the DOM.
If your last accessibility audit predates October 2023, it didn't test against these nine criteria, because they didn't exist yet. That's not a failure, but it is a gap worth closing before your next one.