Skip to content
PdfPress.online
PDF basics

PDF Accessibility Checklist: Make Accessible PDFs Step by Step

Many PDFs fail screen reader and keyboard users in ways sighted mouse users never notice. Work through this checklist before you publish or send.

PdfPress EditorialPublished Updated 6 min read
Share:
Illustration for PDF Accessibility Checklist: Make Accessible PDFs Step by Step
On this page

You've finished the report, exported the PDF, and it looks great on screen. But a screen reader user might hear nothing at all, or hear the sidebar read in the middle of a sentence. A PDF accessibility checker catches many of those problems, and the checklist below covers the rest that software can't judge on its own.

Think of this as a practical pass you can run before publishing or sending a file. It won't replace a formal audit against WCAG or PDF/UA when your organisation needs one, but it will catch the issues that trip up most documents.

Why do so many PDFs fail accessibility checks?

Because the problems are invisible if you read with your eyes and a mouse. A scanned page looks like text but is really a photo. An image looks self-explanatory but has no description for someone who can't see it. A two-column layout reads fine to you, while a screen reader may jump from the top of column one straight into a pull quote.

People reach PDFs in lots of ways: screen readers, magnifiers, keyboards instead of mice, high-contrast settings. A document that works for all of them is also easier to search and reuse for everyone else.

Start in the source document

Most fixes are far cheaper before the PDF exists. If you write in a word processor, use the built-in heading styles instead of just making text bigger and bold. Add alt text as you insert each image. Mark table header rows and set the document language. A proper export carries all of this into the PDF as tags.

Here's a common scenario: a team spends an afternoon trying to patch tags in a finished 40-page PDF, when ten minutes applying heading styles in the original file and re-exporting would have done a better job.

Tip: When exporting from a desktop word processor, look for an option like "document structure tags for accessibility" or "create tagged PDF" and make sure it's switched on.

The checklist: content and structure

  1. The PDF contains real text. Try selecting a single word. If you can't, the page is an image and a screen reader has nothing to read. Run OCR PDF to add a text layer, and see our guide on how to OCR a scanned PDF for the details.
  2. The document is tagged. Tags label headings, paragraphs, lists, tables and figures. Without them, assistive technology has to guess.
  3. The reading order is logical. Columns, sidebars and footnotes should come in the order a person would read them, not the order they happen to be stored in the file.
  4. Headings are real headings. Proper heading tags let screen reader users jump from section to section, the same way you skim bold titles.
  5. Images have alternative text. A chart showing quarterly sales needs a short description of what it shows. A decorative divider line should be marked as decorative so it's skipped.
  6. Tables have header cells. Marked headers let a screen reader announce "March, Revenue" instead of reading a bare number with no context.

The checklist: settings and navigation

  1. The document title is set and shown in the title bar instead of a file name like final_v3_EXPORT.pdf.
  2. The language is specified, so the screen reader uses the right pronunciation. Mark any passages in another language too.
  3. Long documents have bookmarks that mirror the headings. You can add them with Add Bookmarks to PDF.
  4. Links say where they go. Screen reader users often pull up a list of every link on a page, and ten entries that all say "click here" tell them nothing. "Read the annual report" does.

The checklist: visual design and forms

  1. Colour contrast is strong enough, and colour is never the only way information gets across. A chart that separates "on track" from "at risk" using only green and red fails many readers with colour vision differences. Add labels or patterns.
  2. Forms can be completed without a mouse. Each field needs a label or tooltip, and pressing Tab should move through fields in a sensible order. A form that's just lines printed on a page can't be filled in with assistive technology at all.

What can a PDF accessibility checker find on its own?

Automated tools are fast and thorough at spotting things that are missing. They're much weaker at deciding whether what's present actually makes sense.

Check Can automated tools detect it? Needs human judgement?
Real text vs image-only pages Yes Rarely
Tags present Yes Partly (are they correct?)
Reading order Partly Yes
Alt text present Yes Yes (is it meaningful?)
Document title and language Yes Rarely
Colour contrast Partly Yes
Descriptive links No Yes

Alt text is the clearest example. A checker can confirm an image has a description. It can't tell that the description reads "image1.png".

How to run the check with PdfPress

  1. Open PDF Accessibility Checker.
  2. Drag your PDF in or choose it from your device.
  3. Run the check. The analysis happens in your browser, so the document isn't uploaded anywhere.
  4. Read through the findings and note which ones trace back to the source document.
  5. Fix what you can at the source, export again and re-run the check.

Note: A clean report is a good sign, not a guarantee. Follow it with a manual review, and if you can, spend five minutes listening to the document with a screen reader.

Fixing the most common problems

Problem Likely cause Fix
No text found Scanned or image-only pages Run OCR PDF, then re-check
Untagged document Exported without tags, or printed to PDF Re-export from the source with tagging on
No headings detected Manual formatting instead of heading styles Apply heading styles and re-export
Missing alt text Images inserted without descriptions Add alt text in the source document
Missing title Document properties left empty Set the title before exporting

If your source is a Word document, Word to PDF converts it in your browser. Run the checker on the result, since complex layouts with text boxes or floating images sometimes need adjusting.

Warning: "Print to PDF" usually produces an untagged file, even from a well-structured document. Use a proper export or save-as-PDF option instead.

Making accessibility a habit

The teams that get this right don't treat it as a final step. They build templates with heading styles, accessible colours and table settings already in place. They write alt text as images go in. And they re-check every new version, because merging, converting or editing a PDF can quietly drop tags or scramble the reading order.

Be wary of flattening or rasterising public documents, since image-only pages lose their text entirely. And for something really complex, like a 90-page annual report packed with charts, consider offering an HTML version alongside the PDF.

Written by PdfPress Editorial

Our editorial team writes and maintains practical guides on PDF workflows. Articles are reviewed for accuracy and updated when tools or best practices change. Spotted an error? Let us know.

Questions readers ask

What is a tagged PDF?

A tagged PDF contains a hidden structure that labels each part of the page as a heading, paragraph, list, table, figure and so on, in a logical reading order. Screen readers and other assistive technologies rely on these tags to present the content in a meaningful way.

Can an automated checker tell me my PDF is fully accessible?

No. Automated checks are very good at finding missing tags, missing language settings or image-only pages, but they cannot judge whether alt text is meaningful or whether the reading order makes sense. A manual review is always needed.

Is it better to fix accessibility in the PDF or in the source document?

In the source document, whenever possible. Headings, alt text, table headers and the document language set in your word processor usually carry through into the exported PDF, which is far easier than repairing the PDF afterwards.