Skip to content
Toolspike

PDF Accessibility Checker

Check PDF accessibility in your browser: tags, title, language, alternate text, headings, tables, lists, fonts, links, form fields and scanned pages are tested against PDF/UA-1 and WCAG, with fixes and an HTML or CSV report.

Runs in your browser — nothing is uploaded

Free trial: 1 run of PDF Accessibility Checker every day. After that, 2 credits per run.

Check a PDF for accessibility problems and see how to fix each one

Add a PDF and get 28 checks based on PDF/UA-1 and WCAG: whether the file is tagged, has a title and a language, gives images alternate text, nests its headings, structures its tables and lists, embeds its fonts, labels its links and form fields, and lets screen readers in. Each result says what was found, on which pages, and what to change.

The file is checked in your browser and never uploaded, which matters for documents that are not public yet. Three checks, reading order, colour contrast and the quality of descriptions, need a person, and the report lists them so they are not forgotten.

What PDF Accessibility Checker gives you

Reads the tag tree

Figures without alternate text, headings that skip a level, tables without header cells or with uneven rows, list items outside a list and custom tags with no role mapping are each found and counted.

Finds untagged and scanned pages

The text of each page, up to 500 pages, is read to find pages with no text layer, which are probably scans, and text that is neither tagged nor marked as page furniture.

Forms and links

Form fields without a tooltip, links without a description, pages whose tab order ignores the structure and dynamic XFA forms are reported, along with links and fields left out of the tags.

A fix for every problem

Each failed check explains the cause in plain words and gives the step that fixes it, usually in Acrobat or in the program the PDF was exported from, with the WCAG and PDF/UA references.

Reports to share

Download the results as an HTML report to send to whoever made the PDF, or as a CSV file to track problems across many documents.

How to use PDF Accessibility Checker

  1. Add the PDF

    Choose a file of up to 100 MB. For a password-protected file, type the password to check its text as well.

  2. Read the results

    Results are grouped by topic, with the checks that need a person at the end. Turn on Show only failures and warnings to hide what passed.

  3. Fix and check again

    Apply the fixes in the source document or in Acrobat, export again and add the new file.

Other names for this tool

People also look for this as PDF/UA checker, PDF accessibility test, check PDF for WCAG, PDF accessibility validator and accessible PDF checker. It is the same tool on this page.

Limitations

  • Covers the main machine-testable PDF/UA-1 and WCAG requirements, not every clause; for a formal conformance claim, also use a full PDF/UA validator and a screen reader.
  • Text is read from the first 500 pages, in files up to 100 MB.

Questions about PDF Accessibility Checker

Is my PDF uploaded?

No. The file is opened and checked by your browser, and the HTML and CSV reports are made there too, so unpublished documents stay on your device.

Does passing every check mean the PDF is compliant?

Not by itself. The automatic checks cover the main machine-testable parts of PDF/UA and WCAG; reading order, contrast and the quality of descriptions still need a person.

Why does my PDF fail the tagged check?

It has no structure tree, so a screen reader gets one stream of text with no headings, lists or tables. Export again from Word, PowerPoint or InDesign with tags turned on, or add tags in Acrobat and then check the reading order.

What is the difference between PDF/UA and WCAG?

WCAG sets out what accessible content has to achieve, for web pages and documents alike. PDF/UA, ISO 14289, describes how a PDF file must be built to achieve it, from tags to fonts. Each result here lists the clauses of both that it relates to.

Can it check a scanned PDF?

It can tell you that it is scanned: pages without a text layer fail the text check and are listed by number. A scan has nothing for a screen reader to read, so run text recognition on it and tag the result before checking again.

How does it compare with PAC or the Acrobat checker?

Those desktop tools run more of the PDF/UA rules, and PAC is the usual reference for a formal check. This one needs no installation, works on any device with a modern browser and explains each problem in plain words, so it suits a first pass and quick checks before publishing.