Accessibility finding
PDF files that are untagged, scanned, or fail the document checker
Accessibility Error Severity: High
Part of the audit PDF accessibility. Finding id: accessibility.pdfDocuments.
What it costs you
The file has tags, but they are incomplete or wrong at these places. Screen reader users miss the image, table, heading or form field the checker names.
What to do
Fix what the checker reports in the source document and export the PDF again. Start with the files that most pages link to. Best is to publish the content as a web page.
What it checks
- PDF has no structure tags
- PDF is only a scanned image
- PDF title or language missing
- PDF headings missing or wrong
- PDF table problems
- PDF image descriptions missing
- PDF form fields not labelled
How to fix it
Export a tagged PDF from the source file
- Open the source document, not the PDF.
- Use real heading styles, table header rows and alternative texts for images there.
- Set the document title and language, then export with the option that keeps tags or document structure.
- For a scan, run text recognition first, or replace it with the original document.
- Replace the file at the same URL, so existing links keep working.
How to check: Recrawl: the file is no longer reported. Open it with a screen reader and move by heading.
A real case
UK monitoring of public sector websites in 2020 and 2021 found PDFs generally less accessible than web pages, often without the information assistive technology needs. In the simplified tests there was normally an issue with the PDF that was tested.