Target Size

What Is Target Size?

Target size is how big the active area of a button, link or control actually is, and how much room sits between it and the next one. It is easiest to recognize when it goes wrong: the row of tiny social icons in a footer where you tap Facebook and open Instagram, or the close button on a popup that takes three attempts to hit.

Visual Size and Active Size Are Different Things

A control can look large and still have a small target, and the reverse is also true. An icon drawn at sixteen pixels can carry a comfortable tap area if padding is included in the clickable element, while a large-looking button can have a tiny target if only the text inside the button is actually clickable. This is why target size problems survive design review: on screen the layout looks generous.

What the Requirement Asks For

WCAG 2.2 introduced a minimum for target size, with exceptions for controls placed inline within a sentence, for targets that have an equivalent elsewhere on the page, and for cases where the presentation is essential. Spacing counts as well as size, so small controls separated by enough space can still pass where the same controls packed together would not.

How It Affects Real Users

Anyone with a tremor, limited fine motor control, or a hand injury feels this immediately, and so does anyone using a touchscreen one-handed on a moving bus. The consequence is rarely that the task is impossible. It is that the wrong thing gets tapped: the wrong quantity, the wrong date, delete instead of edit. On a checkout or a booking flow, a mis-tap costs a sale rather than just patience.

Where Target Size Shows Up in an Accessibility Review

Reviewers look hardest at dense interfaces: pagination, calendar pickers, data table row actions, icon toolbars, cookie banners and anything crowded into a mobile viewport. Automated tools can measure the dimensions of a control, but deciding whether an exception applies still needs a person, which is why this finding usually arrives with a judgment attached rather than a raw count.