Free PDF Accessibility Checker Online

What is a PDF accessibility checker?

A PDF accessibility checker is a tool that inspects a PDF file for the features people using screen readers and other assistive technology depend on, such as tags, a document title, a language, and alternative text. It reports each problem against PDF/UA (ISO 14289) and the W3C Web Content Accessibility Guidelines (WCAG).

TestMu AI built this PDF accessibility checker for testers, developers, and content teams who need a fast first pass on a document. PDF/UA is described by the PDF Association as a means of making PDF files that conform to WCAG. All processing happens in your browser. No data is uploaded.

What does this PDF accessibility checker test?

The checker runs 12 automated checks. Eight of them map to a PDF technique in the W3C Techniques for WCAG 2.2, and each technique maps to a WCAG success criterion:

How do you check a PDF for accessibility online?

Checking a PDF takes six steps and needs no signup or software:

• Add a PDF: Press Upload PDF, drop a file onto the upload box, or press Untagged sample or Tagged sample to try a built in document.
• Read the score: The ring shows a score out of 100, with counts of failed checks, warnings, passed checks, and checks that could not be run.
• Filter the checks: Use the All, Issues, Warnings, and Passed tabs to narrow the list of 12 checks.
• Open a check: Select any check to read why it matters and how to fix it, or press Expand all to open every check in the current tab.
• Export the report: Press Copy report to copy the results as text, or Download report to save them as a text file.
• Fix and check again: Correct the issues in the source file, export a new PDF, and press Replace to check the new version.

The capture above shows the untagged sample. It scores 28 out of 100 because it has no tags, no title, and no language, so three more checks cannot run at all. To check a locked file, remove the password with Unlock PDF first.

How is the PDF accessibility score calculated?

Each of the 12 checks carries a weight, and the weights add up to 100. A passed check earns its full weight, a warning earns half, and a failed check or a check that could not run earns nothing.

• 25 points: Tagged PDF, because every structural check depends on it.
• 15 points: Real, selectable text.
• 10 points each: Document title, document language, and alternative text on figures.
• 8 points: Heading structure.
• 5 points each: Table header cells and form field labels.
• 2 to 4 points each: Bookmarks (4), title shown in the window bar (3), assistive technology access (3), and the PDF/UA identifier (2).

A worked example: the untagged sample passes selectable text (15), form field labels (5), bookmarks (4), and assistive technology access (3), and gets a warning for the PDF/UA identifier (1). That adds up to 28.

What can an automated PDF accessibility check not tell you?

An automated check confirms that accessibility features exist, not that they are correct. The Matterhorn Protocol from the PDF Association is a list of all the possible ways to fail PDF/UA-1, and this tool covers only part of it. Review these five points by hand:

• Reading order: The tag order must match the visual order of every page.
• Alternative text quality: The text must describe what each image conveys, not just that an image exists.
• Color contrast: Normal text needs a contrast ratio of 4.5 to 1, which you can test with the Color Contrast Checker.
• Link text: Links must make sense out of context.
• Artifacts: Decorative images and repeated headers or footers should be marked as artifacts so they are skipped.

What are the use cases of a PDF accessibility checker?

A PDF accessibility checker fits anywhere a PDF is produced or published. Common uses in QA and development work:

• Testing PDF export features: Check the invoices, statements, or reports your application generates, as part of accessibility testing.
• Comparing export libraries: Run the same document through two PDF libraries and compare the scores to see which one writes tags.
• Pre-publish review: Check a report or brochure before it goes on a website.
• Triage of old documents: Sort a backlog into scanned files that need OCR, untagged files, and files that only need a title or language.
• Choosing an alternative format: When a PDF cannot be fixed, convert it with PDF to HTML and publish the HTML version.

This tool checks PDF files. For web pages and apps, TestMu AI accessibility testing runs automated WCAG scans, and Accessibility DevTools brings the same checks into the browser. Setup steps are in the accessibility testing documentation.

Frequently Asked Questions (FAQs)

What is PDF/UA?

PDF/UA is the ISO 14289 family of standards for accessible PDF files. PDF/UA-1 (ISO 14289-1) builds on PDF 1.7, and PDF/UA-2 (ISO 14289-2:2024) was written for PDF 2.0. A PDF/UA file is tagged, has a title and a language, and gives every meaningful image a text alternative.

What is a tagged PDF?

A tagged PDF is a PDF that carries a hidden structure tree alongside the visible page. The tree labels each piece of content as a heading, paragraph, list, table, or figure and fixes the reading order. Screen readers rely on these tags, so an untagged PDF fails the first check in this tool.

Is your PDF uploaded to a server when you check it?

No, your PDF is not uploaded. The PDF accessibility checker loads the open source pdf.js engine into your browser and reads the file on your own device. The document never leaves your machine and nothing is stored between visits, so contracts, invoices, and unpublished reports can be checked safely.

Can an automated checker prove a PDF is fully accessible?

No, an automated checker cannot prove a PDF is fully accessible. Software can confirm that tags, a title, a language, and alternative text exist, but it cannot judge whether the reading order makes sense or whether the alternative text describes the image. A person still needs to review those points.

Why does a scanned PDF fail the accessibility check?

A scanned PDF fails because each page is a picture of text, not text. Screen readers find nothing to read, and people cannot search, select, or enlarge the words. Run optical character recognition (OCR) on the scan to create real text, then add tags, a title, and a language.

How do you fix an untagged PDF?

The most reliable way to fix an untagged PDF is to return to the source file and export it again with tagging switched on. Word processors and layout tools offer an option to create a tagged PDF. If the source is lost, add tags in a PDF editor, then check the file again.

Does a PDF need bookmarks to be accessible?

Bookmarks matter for long PDF documents. W3C technique PDF2 recommends them for documents that exceed a few pages, because an outline lets people jump between sections. This checker shows a warning when a PDF of 21 pages or more has no bookmarks, and passes shorter documents without them.

What are the file size and page limits of this checker?

This PDF accessibility checker accepts one PDF of up to 50 MB at a time. It scans the first 500 pages and shows a notice when a longer document is cut off. Password protected PDFs cannot be opened, so remove the password before running the check.

What is the difference between WCAG and PDF/UA?

WCAG is the W3C guideline for accessible web content in any format, written as testable success criteria. PDF/UA is an ISO standard that sets technical rules for the PDF file itself, such as tagging and metadata. A PDF that follows PDF/UA has a strong technical base for meeting WCAG.