A PDF that looks perfectly formatted on screen can be completely unreadable to someone using a screen reader. Screen readers โ software used by people with visual impairments to have documents read aloud โ rely on the underlying structure of a PDF, not its visual appearance. A document that looks organized visually can have no meaningful structure at all from the software's perspective. Here's what makes a PDF accessible and how to get there.
Why Most PDFs Aren't Accessible by Default
When you export a Word document or design file to PDF, the visual output looks right but the structural information is often missing or wrong. Headings that look bold and large on screen may have no heading tag in the underlying PDF structure. Columns that appear side-by-side visually may be read left-to-right across both columns rather than down each one. Images have no description unless you specifically add one.
Screen readers follow the document's tag structure, not its visual layout. If that structure is absent or incorrect, the reading order becomes unpredictable โ content from sidebars gets mixed into the main text, tables are read cell by cell in the wrong order, and headings that should help navigation are indistinguishable from body text.
Start With the Source Document, Not the PDF
Accessibility is much easier to build in at the source than to retrofit into a finished PDF. If you're working in Word, the structural decisions you make before exporting determine most of the accessibility of the resulting PDF.
Use real heading styles
Don't create the appearance of a heading by making text bold and large โ use Word's built-in Heading 1, Heading 2, Heading 3 styles. These export to PDF as properly tagged headings that screen readers can navigate. A document with real heading structure lets a screen reader user jump directly to any section, the same way sighted users can scan visually.
Add alt text to every image
Every image, chart, diagram, and graphic in your document needs descriptive alt text โ a brief description of what the image shows and why it's there. In Word, right-click any image and select "Edit Alt Text." A chart showing quarterly revenue growth might have alt text like: "Bar chart showing Q1-Q4 revenue for 2024, with Q3 showing the highest value at $2.4M." Without this, a screen reader announces "image" and moves on โ the content is lost.
Build tables with proper structure
Tables are a common accessibility problem. A table without header rows is just a grid of cells to a screen reader โ there's no way to know which column a value belongs to. In Word, mark the top row of any table as a header row (Table Properties > Row > Repeat as header row). Avoid merging cells where possible, and don't use tables purely for visual layout โ screen readers read table cells in order, and a layout table produces a reading experience that makes no sense.
Use real lists, not manual formatting
If you create a list by typing a dash or number at the start of each line, that's not a list to a screen reader โ it's a series of separate paragraphs that happen to start with a symbol. Use Word's built-in list formatting (the bulleted list or numbered list buttons) to create lists that export as properly tagged list elements.
Export Settings That Affect Accessibility
When exporting from Word to PDF, the method you use matters. The built-in Word PDF export (File > Save As > PDF) preserves more structural information than printing to a PDF printer. In the export dialog, look for options related to accessibility or document structure tags โ these should be enabled.
Specifically in Word's PDF export options: enable "Document structure tags for accessibility" if the option is available. This carries your heading structure, list tags, and table headers through to the PDF. Without this option, even a well-structured Word document can export to a PDF that has no tag structure at all.
Scanned PDFs Are Never Accessible Without OCR
A scanned PDF is an image. There is no text, no structure, and nothing for a screen reader to work with. To a screen reader, a scanned PDF is completely blank โ the software announces the document is open and then has nothing to read.
Running a scanned document through an OCR PDF tool is the first step to making it accessible. OCR converts the image to real text, which screen readers can then work with. WukongPDF's OCR tool at www.wukongpdf.com handles this โ upload the scanned PDF, process it, and download a version with real, readable text. Note that OCR alone doesn't create a fully accessible tagged PDF; for full compliance, additional tagging work in Adobe Acrobat Pro is needed. But OCR is the essential first step without which nothing else is possible.
How to Check Whether Your PDF Is Accessible
The simplest test: try to select and read the text in the PDF. If you can highlight words and the reading order makes sense as you select from top to bottom, the document has real text in a reasonable order. If selection jumps around or nothing selects, you have a problem.
For a more thorough check, Adobe Acrobat Pro has a built-in accessibility checker (Tools > Accessibility > Full Check) that identifies specific issues. Free online tools like PAC (PDF Accessibility Checker) provide similar analysis without requiring Acrobat Pro.
The most reliable test is to open the document in a screen reader and listen to how it reads. NVDA (free, Windows) and VoiceOver (built into Mac and iOS) let you experience the document as a screen reader user would. If the reading order is logical and the headings help navigation, you're in good shape. If it sounds like a random collection of text fragments, there's structural work to do.