Figure tag

A Figure tag is the structure tag that marks an image, chart, map, diagram or other graphic as meaningful content in a PDF. PDF/UA, the ISO standard for accessible PDFs, requires every Figure tag to have alternate text, and PAC, the free PDF Accessibility Checker, also checks that each figure records its position on the page. When PAC flags a Figure tag, one of these requirements is missing, or the tag is being used for something it shouldn't be.

What this means

A correct Figure tag has all of the following:

1. Alternate text (the Alt entry) that says what the image means in context. PDF/UA requires it on every Figure tag, and machine checks fail when it's missing.
2. A bounding box (BBox) that records where the image sits, when the figure is on a single page. Recent versions of PAC check this too. See Figure has bbox.
3. Only meaningful images. Decorative images, such as borders, background swirls and spacer graphics, should be artifacts (content screen readers skip), not figures. A figure with alt text like "decorative line" just adds noise.
4. One tag for one picture. A chart drawn as dozens of small shapes belongs in a single Figure tag with one description, not in many tiny figures. Checkers can't detect this, so it needs a person's eye.
5. No real text hidden inside. Alt text replaces everything inside its tag, so text placed in a Figure tag is hidden from screen readers. Keep real text, including captions, in their own tags next to the figure.
6. No links hidden inside. If an image is also a link, don't put alt text on a tag that wraps the link, or the link becomes unreachable. Microsoft's Word export, for example, places the Link tag beside the Figure tag rather than around it.
7. The right place in reading order, next to the text that refers to it.

People often search for this topic as "Figure tag PDF/UA tagging".

Why it matters

Screen reader users get only what the tag provides: a useful description, a confusing one, or nothing but the word "graphic". If the image carries information, such as a map of a road closure or a chart of rising costs, a missing or misplaced description means missing information. That's WCAG 2.1 success criterion 1.1.1 Non-text Content. The bounding box matters for people who use reflow or magnification (1.4.10 Reflow).

How to fix it in Acrobat Pro and re-check in PAC

To work through every figure in order, Acrobat's alternate text tool shows each one with a box for the description and a checkbox to mark it as decorative. It's called Set Alternate Text in older versions and Add alternate text in newer ones.

Fix it in the source file

Figures exported from the source come out right more often than figures fixed in the PDF:

• Word and PowerPoint: add alt text in the Alt Text pane, or select Mark as decorative. Group the parts of a diagram into one object before describing it. In Word, set pictures to In Line with Text so they stay in reading order.
• InDesign: set alt text in Object > Object Export Options > Alt Text, and set decorative frames to export as artifacts.
• Charts: give a short alt text that states the main point, and put the detailed values in the body text or a data table.

Fix it automatically with Includoc

Includoc adds a bounding box to every figure automatically and moves misplaced alt text to the right tag, so descriptions don't hide text or links.

Alt text itself needs a person. Our AI writes a description for each figure, using the image, the text around it and the document's language, and suggests marking decorative images as artifacts. You approve or edit every one on the review screen. Machines can't judge whether a description says what an image means in context, so nothing is applied without your approval.

Standards

Related errors

• Figure has bbox: the bounding box check in more detail.
• Figures alternate text – Failed: Acrobat's check for figures with no alt text.

For help writing descriptions, read How to write alt text, or check a PDF free to find every figure that needs attention.