PDF Accessibility: The Complete Practical Guide

Updated · 6 min read

On this page
  1. What does an accessible PDF contain?
  2. Which standard applies to your PDFs?
  3. How do I check whether a PDF is accessible?
  4. How do I make a PDF accessible?
  5. What a tool can fix, and what still needs a person
  6. What are the most common PDF accessibility failures?
  7. Every guide on this site

PDF accessibility means a PDF that assistive technology can actually read: one whose structure tree names every heading, paragraph, list and table, whose reading order matches the way a person reads the page, and whose language, title and image alternatives are set. Most PDFs fail that test because they were exported or scanned as a picture of a layout, with no tags at all. This guide covers what the standards ask for, how to test a file, and how to fix the ones you already have.

What does an accessible PDF contain?

Accessibility is not one property you switch on. A PDF is usable by a screen reader only when several things are true at once:

  • A tag structure (the structure tree) that identifies each element: heading levels, paragraphs, list items, table headers, and figures.
  • A reading order that follows the logical flow, not the order objects happen to sit in the file. Two-column layouts are where untagged files fail hardest.
  • Alternative text for images that carry meaning, and an explicit decision to mark decorative images as artifacts so they are skipped.
  • A document language (/Lang) so the screen reader pronounces the text correctly, and a document title that shows in the window and tab rather than the filename.
  • Contrast and colour that do not rely on colour alone, and labels for any form fields.

The practical difference between a tagged and an untagged file is stark. An untagged file is a stack of glyphs at coordinates; the screen reader has to infer structure from position and will often read a table cell before its header, or jump to a footer mid-sentence. A tagged file tells the screen reader what each piece is, so navigation by heading actually works.

Which standard applies to your PDFs?

Three frameworks show up in practice, and a well-tagged PDF serves all of them:

  • WCAG 2.1 AA, published by the W3C at the WCAG 2.1 specification. It is written for web content but applies to documents you publish on the web, PDFs included.
  • PDF/UA (ISO 14289), the PDF-specific standard. See the PDF/UA guide for what it adds.
  • Section 508, which requires US federal agencies to publish accessible electronic documents; its Revised standards incorporate WCAG 2.0 AA. See Section508.gov and the Section 508 PDF guide.

The deadline driving most 2026 work is the ADA Title II web rule. The DOJ published an interim final rule in April 2026 extending the compliance dates to 26 April 2027 for public entities serving 50,000 or more people, and 26 April 2028 for smaller entities and special district governments, with WCAG 2.1 AA covering the documents they publish. The DOJ’s own resource is at the DOJ web rule. In the EU, the European Accessibility Act has applied since 28 June 2025 to services in scope.

How do I check whether a PDF is accessible?

You cannot tell by looking. A file that renders perfectly can have no tags at all. The checkers that do this well are free, and they are covered in detail in the PDF accessibility checker guide:

  • PAC 2024, a free Windows desktop checker from the PDF/UA Foundation.
  • veraPDF, a free open-source validator for PDF/UA and PDF/A.
  • Adobe’s free online accessibility checker, which reports problems and then points you to Acrobat Pro to fix them.

All three report. None of them writes tags into a PDF you did not author. That gap is why the checking half of this work is easy and the fixing half is where most teams stall.

How do I make a PDF accessible?

There are two routes, and which one you take depends entirely on whether you still have the source document.

  1. You have the source. Export it with tags on: in Word choose Save as PDF and tick “Document structure tags for accessibility”; in Google Docs use File > Download > PDF after setting a title and real heading styles; in LibreOffice tick “Tagged PDF” in the export dialog. This is the correct fix, because the tags are generated from real structure. The step-by-step version is in how to make a PDF accessible.
  2. You only have the PDF. Scans, vendor brochures, inherited board packets and files from other departments have no source to go back to. Here you remediate the PDF itself, either by hand in Acrobat Pro or with a hosted tool. The honest breakdown of what is free and what is not is in make a PDF accessible for free.

A worked example of the second route, on a 24-page board packet a coordinator received as a single PDF:

Language detected: en-US Document title: “FY2026 Budget Adoption Packet” Heading outline inferred: H1 “FY2026 Budget” → H2 “Summary” → H2 “Revenue” → H3 “Property tax” → H2 “Appropriations” Flagged for a human: 6 images need alt text; page 18 is a scanned attachment with no text layer and could not be tagged.

That is what a remediation run produces: a tagged file plus a plain list of what it could not do.

What a tool can fix, and what still needs a person

TaskAutomatableNeeds a person
Headings, paragraphs, list itemsYesVerify levels make sense
Reading order on single-column pagesYesMulti-column and sidebars
Document language and titleYesConfirm the title is meaningful
Alt text for meaningful imagesNo, flags onlyAlways
Complex tables (merged cells, headers)PartialUsually
Scanned pages with no text layerNoOCR or retype first
Interactive form field labelsNoAlways

TagRight sits in the automatable column: it returns a remediated PDF with a real structure tree and a report of everything it could not fix, so a coordinator can triage 300 files instead of opening each one. It is a remediation aid, not a PDF/UA certification, and it says so. You can see the free tier on the pricing page and start with two documents, no account, at tag a PDF free.

If you need a certified result for a binding filing, a human remediation service is the better buy; that trade-off is covered in PDF remediation services.

What are the most common PDF accessibility failures?

Across thousands of files, the same handful of failures recur:

  • No tags at all. The file was exported or scanned without structure, so “Tagged PDF: Failed” is the headline result and everything else follows from it.
  • Fake headings. Bold, larger text that looks like a heading but carries no heading tag, so heading navigation finds nothing to move between.
  • Images with no alt text. Logos, charts and diagrams announced as “image” or skipped entirely, taking their information with them.
  • Tables without headers. Cells read with no context about which column or row they belong to.
  • The wrong reading order. Common in two-column layouts, sidebars and text boxes, where the file order and the visual order disagree.
  • A filename for a title. The document announces itself as “report_final_v4.pdf” instead of a real title.

Most of these trace back to one cause: the file was built as a picture of a layout rather than a structured document. Fixing the source, or remediating the PDF, addresses all six at once. When you cannot tell which failures a file has, run it through a free checker first, as described in the PDF accessibility checker guide.

Every guide on this site

Frequently asked questions

What makes a PDF accessible?

A PDF is accessible when it carries a tag structure that names each element (headings, paragraphs, lists, tables, figures), sets a reading order, declares the document language and title, and gives images text alternatives. Screen readers follow those tags instead of guessing from the visual layout.

Which accessibility standard applies to PDFs?

Most public-sector work is judged against WCAG 2.1 AA, which the DOJ references for documents under the ADA Title II web rule and which Section 508 incorporates. PDF/UA, or ISO 14289, is the PDF-specific standard. A correctly tagged file moves you toward both.

Can a PDF be made accessible without Acrobat Pro?

Yes, if you still have the source document: export from Word, Google Docs or LibreOffice with tags turned on. If you only have the PDF, a hosted tool such as TagRight can write tags into it, and free checkers such as PAC 2024 or veraPDF can verify the result.

Is a tagged PDF the same as an accessible PDF?

No. Tagging is the foundation, but accessibility also needs meaningful alt text, a correct reading order, adequate contrast and a real document title. Tags are necessary and not sufficient.

Does tagging a PDF make it legally compliant?

Tagging moves a file a long way toward WCAG 2.1 AA, but it is not a certificate. Compliance is judged on the published document set, and a binding or high-stakes file should be verified with PAC 2024 or veraPDF, or remediated by a human service.

Tag a PDF free