What makes a PDF inaccessible, and why it matters

A PDF is inaccessible when someone using a screen reader, magnification software, or keyboard navigation cannot read it properly. This happens because the PDF was created from a scanned image with no text layer, or because text and images lack descriptions, or because the document has no logical reading order. When a PDF is inaccessible, a blind or low-vision reader gets nothing; a deaf reader misses captions on embedded video; someone with motor difficulties cannot navigate using only a keyboard.

The most common cause is scanning a paper document directly to PDF without running optical character recognition (OCR) to convert the image into readable text. The second most common cause is exporting from Microsoft Word or Google Docs without tagging headings, lists, and images so assistive technology knows what each element is. A third cause is using PDFs created from presentations or design software where text exists only as a visual element, not as actual text data.

Making a PDF accessible takes time but does not require special software beyond what you may already have. The steps differ depending on whether you are starting with a new document, converting an existing one, or fixing a PDF that already exists.

Key Takeaways

  • Scanned PDFs need optical character recognition (OCR) run on them before they become readable to screen readers; most PDF readers and online tools can do this.
  • PDFs created from Word, Google Docs, or PowerPoint need proper heading and list formatting in the original document before export, not after.
  • All images and charts must have alt text describing what they show, and decorative images should be marked as decorative so screen readers skip them.
  • The reading order must flow logically from top to bottom, which you can check and fix using the Tags panel in Adobe Acrobat or free alternatives like LibreOffice.
  • Testing with a screen reader like NVDA (free, Windows) or VoiceOver (built into Mac) reveals problems that visual inspection cannot catch.

Converting a scanned document to accessible text

If you have a PDF made from scanning paper, the first step is running OCR (optical character recognition) to convert the image into actual text. Without this step, a screen reader sees only a picture and cannot read anything.

Adobe Acrobat Pro includes OCR built in: open the PDF, go to Tools, select Recognize Text, choose Recognize Text in This File, and let it process. The free alternative is to upload the PDF to Google Drive, right-click it, select Open With, and choose Google Docs. Google Docs runs OCR automatically and converts the result to a document you can then export as a PDF. Another free option is Smallpdf.com or ILovePDF.com, both of which have OCR features; upload your file, run OCR, and download the result.

After OCR runs, open the PDF and read through it to check that the text converted correctly. OCR is not perfect, especially with poor-quality scans, handwriting, or unusual fonts. Correct obvious errors before moving to the next steps, because a screen reader will read the text exactly as it appears in the file.

Adding structure with headings and lists

A screen reader user navigates a document by jumping between headings the way a sighted person scans a page. If your PDF has no headings, or if headings are just larger text with no structural tag, the screen reader cannot help the user find sections.

The best time to add structure is before you create the PDF. In Microsoft Word, use the Styles panel to mark text as Heading 1, Heading 2, and so on, rather than just making text bold or larger. In Google Docs, use the Format menu to apply heading styles. In PowerPoint, use the slide title and content placeholders rather than inserting text boxes. Then export to PDF: Word and Google Docs have Export as PDF options in the File menu, and both preserve heading structure automatically.

If you already have a PDF and cannot go back to the original document, you can add structure using Adobe Acrobat Pro. Open the PDF, go to Tools, select Accessibility, and choose Add Tags to Document. Acrobat will attempt to identify headings and lists automatically, though you will need to review and correct its work. For free alternatives, LibreOffice Draw can open a PDF, and you can then export it back out with corrected structure, though this is more manual work.

Writing alt text for images and charts

Alt text is a short description of what an image shows. A screen reader reads this text aloud so the user knows what they are missing. Decorative images (like a colored line or background graphic) should be marked as decorative so the screen reader skips them entirely.

Write alt text that describes the content and purpose, not just "image" or "chart". For a photograph of three people in an office, write "Three office workers reviewing documents at a conference table" rather than "photo". For a bar chart showing sales by quarter, write "Bar chart showing Q1 sales at $50,000, Q2 at $75,000, Q3 at $90,000, and Q4 at $110,000" rather than "sales chart". Keep it under 125 words; if you need more detail, add a longer description in the document text itself.

In Word or Google Docs, right-click an image and select Alt Text (Word) or Alt Text (Google Docs), then type your description. When you export to PDF, this alt text travels with the image. In an existing PDF, use Adobe Acrobat Pro: open the Tags panel on the left, find the image tag, right-click it, select Properties, and add the alt text in the Alternate Text field. In free tools like LibreOffice, the process is similar: right-click the image and look for an Alt Text or Description option.

Checking and fixing the reading order

Reading order is the sequence in which a screen reader encounters content. If your PDF has columns, sidebars, or images mixed with text, the reading order might jump around in a confusing way. A sighted person sees the layout and understands it; a screen reader user hears content in whatever order the file specifies.

To check reading order in Adobe Acrobat Pro, open the Tags panel (View menu, then Tags Panel), and you will see a tree of all tagged content. Read through this tree from top to bottom and confirm it matches the logical order you want. If a heading appears after a paragraph that should follow it, you can drag and drop tags to reorder them. If content is missing from the Tags panel, it means it was not tagged, and you will need to tag it manually.

For free tools, LibreOffice Draw shows the reading order visually: open the PDF, go to View, select Navigator, and you will see the structure. Reordering is less intuitive than in Acrobat, but you can delete and re-add tags to fix the sequence. After any changes, export the file back to PDF.

Testing with a screen reader

The only way to know if your PDF is truly accessible is to test it with the assistive technology an actual user would employ. NVDA (NonVisual Desktop Access) is free and runs on Windows; download it from nvaccess.org, install it, and open your PDF in any reader. NVDA will read the content aloud and you will immediately hear if headings are tagged, if alt text is present, and if the reading order makes sense.

On Mac, VoiceOver is built in: press Command + F5 to turn it on, then open your PDF. Use the rotor (press VO + U, where VO is Control + Option) to jump between headings and images and confirm they are all tagged correctly. On both systems, try navigating using only the keyboard: press Tab to move forward through links and form fields, Shift + Tab to move backward, and arrow keys to read line by line. If you get stuck or cannot reach certain content, your PDF needs more work.

After testing, go back and fix what you found. Common problems include missing alt text on images, headings that are not actually tagged as headings, and reading order that jumps around. Each fix makes the document usable for more people.

Choosing the right tool for your situation

Your choice of tool depends on what you are starting with. If you have a Word document or Google Doc, export it directly to PDF with proper heading and list formatting already in place—this is the fastest route. If you have a scanned document, run OCR first, then add structure and alt text. If you have an existing PDF that needs fixing, Adobe Acrobat Pro is the most powerful option, but it costs money; LibreOffice Draw is free but slower for large documents.

For one-off documents, free online tools like Smallpdf or ILovePDF can handle OCR and basic fixes. For documents you create regularly, invest time in learning your authoring tool's accessibility features—Word, Google Docs, and PowerPoint all have them built in, and using them from the start is faster than fixing a PDF later. For organizations that produce many PDFs, consider a workflow where accessibility is checked before the document is finalized, rather than after.

Frequently Asked Questions

Can I make a PDF accessible without Adobe Acrobat Pro?

Yes. Google Docs can run OCR on scanned PDFs for free. LibreOffice Draw can open and re-export PDFs with corrected structure. NVDA is a free screen reader for testing. The process is slower and more manual than with Acrobat Pro, but it is possible. For simple documents with few images, free tools are often enough.

What if my PDF has text but it is not selectable?

That means the text is part of an image layer and OCR has not been run. Use Google Docs, Smallpdf, or Adobe Acrobat Pro to run OCR. After OCR, the text becomes selectable and readable to screen readers.

Do I need to add alt text to every image?

Every image that conveys information needs alt text. Decorative images—like a colored border or background graphic—should be marked as decorative so screen readers skip them. If you are unsure, ask yourself: would a sighted person miss important information if this image disappeared? If yes, it needs alt text.

How long does it take to make a PDF accessible?

A simple document with a few headings and images might take 15 to 30 minutes. A complex document with many images, charts, and columns might take an hour or more. The fastest approach is to build accessibility into the original document before exporting to PDF, rather than fixing a PDF after the fact.

What if I do not know what alt text to write?

Describe what the image shows and why it matters in the document. For a chart, include the data values. For a photograph, describe the people, objects, and action. Read your alt text aloud to yourself—if it makes sense without seeing the image, it is good enough. If you are stuck, ask a colleague or someone who uses a screen reader what they would want to know.