Skip to main content

How to Tag a PDF for Accessibility: Step by Step

Picture of Sidharth Nayyar

Sidharth Nayyar

To tag a PDF, you give it a hidden structure tree. Each heading, paragraph, list, table and image gets a label that tells a screen reader what it is and in what order to read it. Without tags, a PDF is a picture of a page. With tags, it is a document.

This guide shows the steps, in the order that saves the most rework. If you only want the definition first, read what a tagged PDF is.

Before you start: check what you have

Open the file and try to select a single word with the cursor. If you cannot, the page is a scan. A scan has no text for tags to attach to, so it needs OCR (text recognition) first. Tagging a scan without OCR gives you empty tags.

Then run a check so you know the starting point. The free PDF accessibility checker takes an uploaded PDF or a pasted URL and returns a score, the issues it found and a time and cost estimate. Keep the report. You will compare against it at the end.

Step 1: Fix the source file if you can

The cheapest tags come from a well-built source. If the PDF was exported from Word, InDesign or a slide deck, fix it there. Use real heading styles, real list formatting, real table headers, and alt text on images. Then export with the "tagged PDF" or "accessibility" option on. A source with proper styles usually exports with most tags already correct.

If you only have the PDF, go on to step 2.

Step 2: Add tags automatically, then treat the result as a draft

Most PDF editors can auto-tag a document. This is a useful first pass, and it is never the finish. Auto-tagging guesses. It often labels a heading as a paragraph, splits one table into several, or reads a two-column page across the columns.

After the auto-tag run, open the tag tree and read it from the top. You are looking for three things: wrong tag types, wrong order, and content with no tag at all.

Step 3: Set the document basics

Four properties are quick to set and often missed:

  • Marked as tagged. The file must declare that it carries tags.
  • Document title. Set a real title and make the viewer show it instead of the file name (WCAG 2.4.2).
  • Language. Set the main language so a screen reader picks the right voice (WCAG 3.1.1). Mark any passage in another language.
  • Bookmarks. For long documents, add bookmarks that follow the headings.

Step 4: Tag headings, paragraphs and lists

Use one H1 for the document title, then H2 and H3 in order. Do not skip levels to get a smaller look. Body text is P. A list is a list tag (L) with one item tag (LI) per entry, and each item holds a label (Lbl) and the item body (LBody). A row of bullets typed by hand is just text unless you tag it as a list.

Step 5: Fix the reading order

Reading order is the sequence a screen reader follows. It can differ from the visual order. Check these cases:

  • Multi-column pages. The reader should finish column one before it starts column two.
  • Sidebars and pull quotes. Place them where they make sense, or after the main text.
  • Headers, footers and page numbers. These are decoration. Tag them as artifacts so they are not read on every page.

This maps to WCAG 1.3.2, Meaningful Sequence. Test it by listening, not only by reading the tree. Turn on a screen reader and let it read two or three pages.

Step 6: Tag tables

A data table needs a Table tag, row tags (TR), header cells (TH) and data cells (TD). Mark each header cell with its scope, row or column. For a complex table with merged cells, tie data cells to their headers explicitly. Do not use a table for page layout.

Step 7: Tag images and write alt text

Each meaningful image gets a Figure tag with alt text that says what the image conveys, in the context of the page (WCAG 1.1.1). A chart needs the finding stated in words, not "bar chart". An image that only decorates, such as a border or a logo repeated on every page, becomes an artifact with no alt text.

A link needs a Link tag that wraps both its text and its link annotation, and the text must make sense out of context. "Read the permit guide" works. "Click here" does not. Form fields need a label that is tied to the field, and a tooltip that matches the label.

Step 9: Validate

Run the checker again and compare with the first report. Automated checks catch missing tags, missing language, missing alt text and structure errors. They cannot judge whether alt text is good or whether the reading order makes sense, so finish with a short screen reader test. Our guide on how to check a PDF for accessibility covers the checks in more detail, and PDF/UA vs WCAG explains which standard each check comes from.

How long does tagging take?

It depends on the document, not the page count. A clean one-page memo exported from Word can take a few minutes to fix. A 40-page report with multi-column layouts, complex tables and charts can take hours by hand. That is why teams with many PDFs move from tagging one file at a time to a fix process they can repeat.

Why it matters now

Public entities in the United States face a deadline under the Department of Justice web accessibility rule for ADA Title II. The compliance dates are April 26, 2027 and April 26, 2028, depending on the size of the entity. PDFs on a public site count as web content. Section 508 and the European Accessibility Act also reach documents. A tagged, checked PDF is the base requirement in all three.

When to stop tagging by hand

If you have a handful of files, the steps above are enough. If you have hundreds, the work is the same for every file: tags, reading order, alt text, a PDF/UA check. WebAbility's PDF accessibility service fixes the real PDF, not a copy on a separate page. It repairs the tags and reading order, runs a PDF/UA check, and adds an Accessibility Conformance Report in the VPAT 2.5 format. Processing and preview are free. One credit covers up to 10 pages, with a minimum of one credit per document.

For the full picture of what a compliant file needs, see the PDF accessibility checklist and our overview of PDF remediation.

FAQ

Can I tag a PDF for free?

Yes, if you have a PDF editor with a tag panel, or if you fix the source file and export a tagged PDF. You can also check your file for free with the PDF accessibility checker linked above.

Is auto-tagging enough?

No. It gives a first draft. Plan to fix heading levels, reading order, tables and alt text by hand.

What is the difference between a tagged PDF and an accessible PDF?

Tags are one part. An accessible PDF also needs a title, a language, alt text, a sensible reading order, usable contrast and working forms. See what an accessible PDF is.

Do scanned PDFs need OCR before tagging?

Yes. OCR turns the page image into text. After that you can tag the text and check the order.

Quick Questions

Tap to ask AI about this article

Ready to make your website accessible? Engage with our team or start a free trial today.


Related Blogs