Measure the activation target rather than the visible icon

Measure the active shape rather than its picture

W3C’s WCAG 2.2 guidance for Success Criterion 2.5.8, checked 7 November 2026, uses a 24-by-24 CSS-pixel square wholly inside the target, not merely its bounding box. A spacing alternative uses 24-pixel circles centered on target bounding boxes, subject to the documented conditions and exceptions. These are criterion-specific rules, not a universal legal standard or complete accessibility assessment.

Ask the implementer to identify the actual active region for each disputed control in a supplied test page. Record its shape, dimensions and position in CSS coordinates. Keep the visible icon outline separate. A decorative outline or transparent padding does not establish what will respond to input. Inspect the agreed implementation evidence rather than guessing the hit area from an enlarged screenshot.

Use independent geometry to distinguish the cases

Create a synthetic pair of controls: a circular target with a diameter of 24 CSS pixels and a square target measuring 24 on each side. Both have equally sized bounding boxes, but their interior shapes differ. In the circle, the proposed square’s corners lie outside the target: their distance from the center is about 16.97, while the circle’s radius is 12. The square target contains the proposed square exactly. This arithmetic is an original fixture, not a copied source example.

Keep this shape test separate from the alternative spacing analysis. Draw the relevant assessment circles and neighboring active regions on a coordinate sheet using the actual layout. Identify intersections explicitly instead of saying the controls look comfortably separated. A control that lacks the required interior square is not automatically a failed page; the qualified reviewer must consider the applicable alternative and exceptions rather than applying a shortcut based on appearance.

Exercise the layout that users actually receive

Specify representative narrow and wide layouts, expanded menus and any control that appears after an error or selection. Request measurements for these states from the implementer. A generous desktop arrangement can become crowded when a text label wraps or an adjacent action appears. Keep the actual responsive states identifiable so the geometry result cannot be mistaken for approval of an untested arrangement.

For a reported exception, record the exact reason and evidence the reviewer uses. Do not stretch an exception to cover an inconvenient layout simply because redesign would take time. Ask whether a different control provides the equivalent action where that exception is relevant, and confirm the reviewer’s interpretation. Keep numerical geometry, exception judgment and a possible design improvement in separate columns so a discretionary preference does not masquerade as a requirement.

Choose a remedy that preserves the intended action

Compare feasible remedies with the designer and implementer: change the active shape, increase separation or reorganize competing controls. Test whether a larger hit area overlaps another action or captures input intended for nearby content. Making one control easier to activate is not a successful repair if it causes accidental activation of its neighbor. Preserve the user’s ability to distinguish the actions.

Request a supplied post-change coordinate record and an interaction check for the relevant states. Keep this pointer review alongside, not instead of, keyboard, focus and screen-reader acceptance. Record the criterion scope, tested variants and unresolved exceptions. This exercise changes no real interface or settings and claims no full conformance; it produces a geometric decision that the authorized product and accessibility owners can evaluate.

Make the reasoning concrete.

A fictional compact product selector uses two round icons whose visible bounds each measure 24 by 24 CSS pixels. A reviewer assumes the dimensions alone settle the question, while the implementer has not yet described the actual active shapes or their separation.

The reviewer requests the coordinate evidence and keeps any conformance claim bounded to the assessed criterion and states. No live controls are modified.

Take the questions into your next brief.

Use these prompts in your own notes. Nothing is submitted from this page.

1. Region

   What shape actually activates each action?

2. Interior

   Can the required axis-aligned square fit wholly inside that shape?

3. Neighbors

   What does the spacing analysis show about nearby active regions?

4. States

   Which responsive or dynamic arrangement is least favorable?

5. Decision

   What evidence supports the chosen remedy or documented exception?

Watch for the wrong conclusion.

• Equating a bounding box with an interior square.
• Measuring the icon instead of the active region.
• Expanding hit areas without checking conflicting neighboring actions.

Sources & scope

Workflows, fixtures, worksheets and examples are original editorial analysis. Sources support the narrow attributed statements, not measured business results. Platform guidance is scoped to the documented implementation and date checked. Synthetic offline planning only; no external account, file publication, permissions or production configuration is changed.

• Understanding SC 2.5.8: Target Size (Minimum) (Level AA) →Primary platform documentation or published technical standard · Supports: Supports the interior-square, spacing and exception distinctions; the numerical circle comparison is original.