What Is PDF Reading Order and Why Does It Matter?

You've created a polished two-column handout for your students. The headings are clear, the sidebar has a helpful glossary, and the footer shows your department name on every page. It looks great on screen. So why is a student using a screen reader telling you the document is incomprehensible?

The answer is almost always reading order.

Reading Order, Defined

Reading order is the sequence in which a screen reader reads the content of a PDF. It's determined by the document's internal tag structure, not its visual layout. When reading order is wrong, a screen reader might read a sidebar before the main text, jump between columns incorrectly, or read a footer in the middle of a page — making the document confusing or unusable for anyone who relies on assistive technology.

Think of it this way: a sighted reader can glance at a page and instantly understand the layout. They know the sidebar is supplementary, the footer is ignorable, and the two columns should be read left-then-right, top-to-bottom. A screen reader has none of that spatial intuition. It follows the tag order, one element after another, in a single linear stream. If that stream is jumbled, so is the experience.

Why Visual Layout Does Not Equal Reading Order

This is the part that confuses most people. If the document looks correct, why would it read incorrectly?

The reason is architectural. PDFs store content as individually positioned objects on a canvas — a block of text at coordinates (72, 400), an image at (300, 200), a text box at (50, 600). There is no inherent flow. A two-column layout might be stored with all left-column objects first, or with objects from both columns interleaved by vertical position, or in whatever order the authoring tool happened to write them.

When you export from Word or InDesign, the application tries to write tags in a sensible order, but it frequently gets it wrong — especially with complex layouts. The visual rendering engine and the tag-writing engine are solving different problems, and they don't always agree.

Without correct reading order tags, anything goes. And "anything goes" is exactly what a screen reader will deliver.

Common Reading Order Problems

Certain document patterns cause reading order failures more than others. If you recognise any of these in your materials, they're worth checking:

• Multi-column layouts. The most common culprit. Screen readers may read across both columns in a single line instead of down each column sequentially, turning sentences into nonsense.
• Text boxes and floating elements. Word's text boxes are especially problematic. They often get tagged at the end of the document regardless of where they appear visually.
• Sidebars and callout boxes. A "Key Terms" sidebar placed visually beside paragraph three might be read before paragraph one — or after the last paragraph on the page.
• Headers and footers read inline. Instead of being skipped, repeated page headers and footers get read as part of the content flow, interrupting every single page.
• Figures interrupting text. An image placed between paragraphs might cause the text after it to be read before the text before it, depending on how the objects were layered.
• Table captions misplaced. A caption visually above a table might be tagged after the table's data, so the reader hears rows of numbers with no context and the explanation only after.

How to Check Reading Order

There are three reliable ways to check, in order of increasing accuracy:

1. Adobe Acrobat's Reading Order Tool. Open the PDF in Acrobat Pro, go to Accessibility > Reading Order. This highlights tagged regions with numbers showing the order. It's a good quick visual check, but it can be misleading for complex layouts.

2. PAC (PDF Accessibility Checker). PAC's screen reader preview shows you exactly what a screen reader will encounter, in order, as plain text. This is the fastest way to spot jumbled content. It's free and widely used in accessibility work.

3. Test with an actual screen reader. NVDA on Windows or VoiceOver on Mac. There is no substitute for hearing your document read aloud. If you get confused, your students will too. Even testing the first two pages will reveal most reading order problems.

For a broader accessibility check beyond reading order, see our guide on how to check PDF accessibility.

How to Fix Reading Order

Fix in the Source Document

The best approach is to prevent the problem. In your source document (Word, PowerPoint, Google Docs):

• Avoid text boxes when possible. Use the document's built-in layout features instead.
• Use simple, single-column layouts for handouts and course materials. If you need two columns, use your word processor's column feature rather than manual positioning.
• Check the reading order before exporting. In PowerPoint, the Selection Pane (Home > Arrange > Selection Pane) shows the order of objects on each slide. In Word, use the Navigation Pane to verify heading structure.

For a complete walkthrough, see our full guide to making PDFs accessible.

Fix in the PDF Directly

If you're working with an existing PDF, Acrobat Pro's Reading Order tool lets you select content regions and drag them into the correct sequence. You can also reassign tags — marking a paragraph as a heading, or a decorative image as a background artifact.

This works, but it's tedious. A 20-page document with complex layout can take an hour or more to fix manually.

Mark Headers and Footers as Artifacts

One specific fix deserves its own mention: repeated headers and footers — page numbers, document titles in the header, department names in the footer — should be marked as artifacts. Artifacts are elements that screen readers skip entirely.

This is correct behaviour. A screen reader user does not need to hear "University of Melbourne, School of Engineering" repeated on every single page. Marking this content as artifacts keeps the reading experience clean and focused on actual content.

If your headers and footers are being read aloud as part of the document flow, they haven't been properly artifacted. This is one of the most common issues we see in university PDFs and one of the easiest to fix — once you know to look for it.

How Aelira Fixes Reading Order Automatically

Manually reordering tags in Acrobat Pro is effective but doesn't scale. If your department has hundreds or thousands of PDFs to remediate before the April 2027 ADA deadline, manual work alone won't get you there.

Aelira uses a dual strategy to automatically fix reading order:

• Heuristic analysis for standard layouts. For single-column documents, simple two-column layouts, and other common patterns, position-based clustering algorithms can reliably determine the correct reading order. This approach is fast and accurate for the majority of academic documents.
• AI vision analysis for complex layouts. For documents with sidebars, mixed column widths, callout boxes, and irregular arrangements, AI vision models analyse the page as a human would — understanding spatial relationships and content grouping to determine the intended reading sequence.

This dual approach means straightforward documents are processed quickly and efficiently, while complex layouts get the attention they need. Headers and footers are automatically detected and marked as artifacts.

Aelira uses a dual strategy — heuristic analysis for standard layouts and AI vision for complex ones — to automatically fix reading order. See how it works.