A PDF can look perfect and still be unusable for someone who relies on a screen reader. The good news is that most failures come from a handful of causes, and you can spot the biggest ones yourself in about a minute, before running any tool at all.
Want to skip the reading? The tool this guide covers is free and needs no sign up.
PDF Accessibility CheckerThree tests you can do right now
Open the PDF in any viewer and try these. If any of them fails, the file is not accessible, and you already know where to start.
- 1Try to select a sentence of text. If you cannot, or the whole page highlights as one block, the page is an image. A screen reader sees nothing there. This is what a scanned document looks like from the inside.
- 2Look at the title bar or browser tab. If it shows the file name, something like "scan_0042.pdf", instead of a proper title, the document has no title set.
- 3Open File > Properties (Ctrl+D in Acrobat Reader). Under "Tagged PDF", it should say Yes. If it says No, there is no structure for a screen reader to follow: no headings, no lists, no tables.
What an automated check adds
Those three tests catch the worst problems, but not the many smaller ones underneath. An automated checker reads the file's internals: whether every image has alternate text, whether the language is set, whether headings skip levels, whether table header cells exist, whether form fields have names, whether fonts are embedded.
Here is an example of why that matters. We ran a 65-page employee handbook exported from Word through our checker. It passed all three quick tests: real text, tagged, and the tags were good, with over 3,000 of them, proper headings, lists and tables. It still failed on 1,548 images without alternate text and on a missing title. Word tags every small decorative icon as a figure unless you mark it as decorative, and nobody had.
An automated check can prove a problem exists. It cannot prove a file is fully accessible, because some things, such as whether alt text is actually useful, need a person to judge.
Fixing the common failures
Fix things in the original document, not the PDF. Repairs made in the PDF are lost the moment someone edits the Word or PowerPoint file and exports again.
- 1Use real heading styles (Heading 1, Heading 2) instead of making text bigger and bold.
- 2Add alt text to every meaningful image: right-click it in Word and choose Edit Alt Text. For purely decorative images, tick "Mark as decorative".
- 3Mark the first row of each table as a header row: Table Properties > Row > "Repeat as header row".
- 4Set a title in File > Info > Properties.
- 5Export with File > Save As > PDF > Options > "Document structure tags for accessibility" ticked. Never use "Print to PDF", which throws the structure away.
- 6For scanned paper, run OCR so the pages contain real text. No other fix helps until that is done.
The checks only a person can do
Once the automated report is clean, spend ten minutes on the rest. Listen to a few pages with NVDA (free for Windows) or Acrobat's Read Out Loud and check that the reading order makes sense. Read each alt text without looking at its image and ask whether it tells you what you need. Check that light grey text and text over photos is still easy to read.
If you need to record that work, our Document Accessibility Test lets you tick off each manual check and includes it in the downloaded report.
Frequently asked questions
Is a tagged PDF automatically accessible?
No. Tagging is necessary but not enough. A file can be fully tagged and still have images without alt text, headings in the wrong order or a reading order that jumps between columns. Tagging gives a screen reader something to work with; the tags still have to be right.
Can I make a scanned PDF accessible?
Yes, in two steps. Run OCR so each page gets a real text layer, then add tags, either by exporting again from a tool that tags or with Acrobat Pro's autotag feature. Check the result, because OCR makes mistakes on poor scans.
Which standard should my PDF meet?
For most public websites, WCAG 2.1 Level AA. US federal work uses Section 508, which points to WCAG 2.0 AA. PDF/UA (ISO 14289) is the PDF-specific standard, and meeting it usually takes care of the WCAG requirements that apply to a document.
Tools mentioned in this guide
See which WCAG success criteria your PDF passes or fails - alt text, titles, language, headings, tables, links and forms.
Check a PDF against PDF/UA-1 rules: tags, alt text, language, fonts, Unicode and tab order, mapped to Matterhorn checkpoints.
Run the automated tests, then tick off the manual checks, for a complete accessibility test you can save and share.
Last updated 7 October 2026. Back to all guides