Free PDF accessibility checker
Find out in seconds whether a PDF can actually be read by someone using a screen reader — and exactly what needs fixing. No signup, no upload, no cost.
Your files never leave your computer. Every check runs inside your browser. Nothing is uploaded, stored, or transmitted — which means you can safely test documents containing student records, patient information, or anything else confidential.
Choose PDF files to check
Drag PDF files here
or
You can check many files at once.
Results
What this checks
These are the failures that dominate real remediation work, and the ones most likely to make a document unusable for someone relying on assistive technology.
- Tagged structure
- Whether the PDF contains a structure tree. Without one, a screen reader cannot tell a heading from body text, cannot follow reading order, and cannot navigate lists or tables. This is the single largest barrier, and the main thing remediation fixes.
- Document title
-
Whether a real, descriptive title is set — and whether viewers will actually
show it. We also flag placeholder titles like
untitledorMicrosoft Word - draft3.docx, which pass most automated checks but tell a listener nothing. - Language
- Whether the document declares its language. Without it, a screen reader applies the wrong pronunciation rules, which can render correct text incomprehensible.
- Alternative text on images
- Whether figures carry a written description. A chart, map, or signature with no alt text is simply absent for a screen reader user.
- Table headers
- Whether tables identify their header cells. Without them, a screen reader reads disconnected numbers instead of “Q3 revenue: 40,000”.
- Text layer
- Whether the pages contain real text or are scanned images. A scanned document is a picture of words — invisible to screen readers and unsearchable — and needs OCR before anything else can be done.
What this cannot tell you
Automated testing catches a meaningful share of accessibility problems, but not all of them, and we would rather say so plainly than let you assume otherwise.
- Reading order — we can tell you a structure tree exists, but not whether it puts content in a sensible order. That needs a human.
- Alt text quality — we can tell you alt text is present. We cannot tell you whether “image1.jpg” is a useful description of a bar chart.
- Colour contrast and visual design — not assessed here.
- Whether headings are correct — we count them; we cannot judge whether the right text was marked as a heading.
A clean result here means no automated failures were found. It is not a certification, and it is not a legal guarantee of compliance.
Where to go next
- What PDF remediation actually costs — published vendor rates, and what drives the price up.
- How to tag a PDF for accessibility — what tagging is and how to do it in Acrobat or Word.
- Section 508 and WCAG requirements for PDFs — which rules apply to you, and by when.