Accessible PDFs explained

A PDF can look perfect and still be unusable for someone relying on a screen reader. Accessibility is about the invisible structure underneath the visuals — and it's increasingly a legal requirement, not just a nicety.

Why plain PDFs are a problem

As we cover in how PDFs work, a page is fundamentally a set of positioned glyphs with no inherent notion of headings, paragraphs, or reading order. A screen reader looking at an untagged PDF may read text in the wrong order, skip images, or announce a table as meaningless streams of numbers.

What "tagged" adds

A tagged PDF includes a hidden structure tree that labels content: this is a Heading 1, this is a paragraph, this is a list, this image has alt text, these cells form a table with these headers. That tree gives assistive technology a logical map of the document.

Ingredients of an accessible PDF

• A correct tag structure and logical reading order
• Alt text for every meaningful image
• Real table headers, not just visual formatting
• A document title and language set in the metadata
• Sufficient colour contrast and selectable (not scanned) text

Standards to know

PDF/UA (ISO 14289) is the accessibility standard for PDFs, and WCAG guidelines apply too. Many organisations are legally obliged to meet them. If you start from a well-structured source document and export carefully, you're most of the way there.