← Back to blog

WCAG 2.2 Target Size: Make Analytics Controls Easier to Use

WCAG 2.2 adds a practical Level AA minimum for pointer targets. Apply its 24 CSS pixel rule to filters, menus, and controls without turning dense interfaces into guesswork.

Small controls are a usability problem before they become a conformance issue. WCAG 2.2 Success Criterion 2.5.8 sets a 24 by 24 CSS pixel minimum for pointer targets at Level AA, subject to defined exceptions. It applies to interface controls people activate with a mouse, touch, stylus, or another pointer—not just buttons designed for phones.

That matters in analytics dashboards, where compact filter chips, date arrows, chart legends, and overflow menus often place several small targets close together. A visually tiny icon can still have a sufficiently large clickable area, but a large-looking button can fail if its actual target box is too small or overlaps a neighbor.

Understand the minimum and the spacing exception

The default test is straightforward: the pointer target should contain a 24-by-24 CSS-pixel square. When a target is smaller, the spacing exception can apply if a 24-pixel-diameter circle centered on the target’s bounding box does not intersect another target or the equivalent circle for another undersized target. This checks the usable hit area, not merely the visible glyph.

WCAG 2.2 also defines exceptions, including inline targets in text, equivalent controls that meet the size requirement, user-agent-controlled targets, and targets whose presentation is essential. Apply an exception only when its conditions are met; “the toolbar is crowded” is not itself an exception. The criterion is about reliable activation, not whether a design looks neat.

Separate the AA floor from a stronger target

The 24-pixel threshold belongs to SC 2.5.8, a Level AA criterion. SC 2.5.5 Target Size (Enhanced) is a separate Level AAA criterion with a 44-by-44 CSS-pixel target. The larger value is a useful design goal for frequently used or consequential actions, but it is not the AA minimum.

Do not solve every small icon by enlarging its visible artwork. Add padding to the interactive element while keeping the icon visually compact. Use a consistent minimum block and inline size, and provide enough gap between adjacent controls. A focus outline should reveal the actual interactive boundary so keyboard and pointer users can understand what will activate.

Test rendered targets, not source values

CSS pixels are not physical screen pixels. Responsive layout, zoom, transforms, and neighboring controls can change the effective geometry, so inspect the rendered page at representative viewport widths. Test menus, chart interactions, and controls that appear only on hover or after a filter is selected. A CSS declaration such as width: 24px is not proof if the clickable element is smaller or its target overlaps another one.

Include both automated and manual checks. Automated accessibility tools can flag many undersized targets, but they may not correctly evaluate every spacing exception or dynamic state. Use browser devtools to inspect bounding boxes, then try the control with touch or a coarse pointer. Check that the pointer target remains large after the UI wraps, scales, or opens a popover.

For a dense dashboard, prioritize the controls people use often and those that can trigger an expensive or hard-to-reverse action. Give clear visual grouping to filters and destructive actions, avoid putting adjacent icon-only buttons edge to edge, and preserve a generous target area even when the visual layout is compact. These are small implementation choices that prevent accidental selections.

Keep accessibility data out of analytics payloads

A team can measure whether interface changes improve task completion without collecting pointer coordinates or recording individual interaction trails. Use aggregate usability tests, accessibility audits, and coarse event counts that do not encode sensitive labels or user-entered values. Accessibility instrumentation should not become an excuse to add a behavioral profile.

The target-size criterion is a baseline, not a guarantee that every control is easy to use. Larger targets, clear labels, keyboard support, and visible focus still matter. Treat the hit-area geometry as part of the component contract, so a future icon swap or spacing change cannot silently make a control harder to operate.

Sources

Comments

Loading comments…