If you’ve ever opened Adobe Acrobat’s Accessibility Checker, gotten a reassuring collection of green checkmarks, and thought, “Great! This PDF is accessible,” the U.S. Environmental Protection Agency has a resource you may want to bookmark.
The EPA recently updated its guide, 508 Compliance for Existing PDFs: Identifying & Resolving Common Accessibility Issues. It’s a practical walkthrough for finding and fixing common accessibility problems in existing PDF documents using Adobe Acrobat Pro.
One of the most useful things about the guide is what it makes clear almost immediately: making a PDF accessible involves a lot more than running an automated checker.
First: Does It Need to Be a PDF?
Before getting into remediation, the EPA asks an excellent question: Does this content need to be a PDF at all?
For its own web content, EPA recommends defaulting to accessible webpages and reserving PDFs for longer documents. It also recommends addressing accessibility as early as possible by running the accessibility checker while creating source documents in Word or PowerPoint, rather than waiting until after conversion to PDF.
That’s an important accessibility principle well beyond the EPA: accessibility is much easier to build in than bolt on later.
If a document will eventually become a PDF, start with an accessible source document. Use proper heading styles. Build tables correctly. Add meaningful alternative text. Check color contrast. Identify the document language and other properties.
You’ll still need to check the resulting PDF, but you’re starting from a much better place.
What Makes a PDF Accessible?
PDF accessibility depends heavily on the document’s underlying structure.
A sighted reader can look at a page and immediately recognize a heading, paragraph, list, table, photograph, or link. Assistive technologies need that information to be represented programmatically.
That’s where PDF tags come in. Tags identify the structure and purpose of content so technologies such as screen readers can interpret it correctly. The EPA guide also points out that some accessibility information lives outside the visible page itself, including document metadata and properties.
The guide walks through several areas that commonly require attention.
Document properties
Accessible PDFs need appropriate metadata and document settings, including a meaningful title and the correct document language. EPA also recommends configuring the PDF so that its document title, rather than its filename, is displayed when opened.
Logical reading order
Content needs to be presented to assistive technology in a meaningful sequence.
This is one of the places where automated testing falls short. Acrobat flags logical reading order as requiring a manual check. Someone actually has to inspect the document and determine whether the content will be presented in the correct order.
Many documents are presented in logical reading order, but some are not. (I’m looking at you, trifold brochure.)
Alternative text
Informative images (generally tagged as Figures) need alternative text that communicates their purpose in context. Purely decorative elements shouldn’t be tagged, as that will unnecessarily interrupt the experience of someone using a screen reader. Instead, convert them to artifacts, thereby removing them from the tag tree entirely.
The EPA guide includes instructions for finding figures without alternative text and correcting improperly structured figures.
Tables
Tables can be particularly challenging.
The underlying tag structure needs to correctly distinguish table rows, header cells, and data cells. The EPA guide walks through several common table failures and shows how to inspect and repair the tag structure in Acrobat.
Lists and headings
A document that looks like it contains a list or heading isn’t necessarily structured as one.
The guide explains how to correct list tags and heading structure so that assistive technology can understand the relationships between different pieces of content.
Color contrast
Yep, color contrast matters in PDFs too.
The EPA includes contrast among the items that require manual evaluation and points users toward additional tools when needed.
The Accessibility Checker Is a Starting Point
This may be the most important takeaway from the entire resource.
Adobe Acrobat’s Accessibility Checker is useful. I use it. You should use it.
But passing an automated accessibility check does not prove that a PDF is accessible.
Some accessibility requirements require human judgment. Is the reading order logical? Does the alternative text communicate what the image means in context? Does the heading structure accurately represent the organization of the document? Is the table understandable and correctly associated with its headers?
Software can identify certain technical conditions. It cannot reliably answer all of those questions for you.
The EPA’s own process reflects that distinction: run the checker, investigate the results, manually evaluate issues that require human review, and remediate the document’s underlying structure.
What If a PDF Can’t Be Made Accessible?
EPA also addresses the rare situation in which a PDF can’t be made compliant. Its guidance calls for placing contact information prominently in the document so a person with a disability can request assistance or an accommodation.
That’s a fallback, though—not a substitute for making documents accessible whenever possible.
A Useful Resource for Anyone Responsible for Digital Documents
If your organization publishes PDFs online, the EPA guide is worth bookmarking. It’s detailed enough to help someone understand common remediation tasks without pretending to be a comprehensive PDF accessibility standard or training course.
More importantly, it illustrates something organizations preparing for digital accessibility requirements need to understand:
Document accessibility is a process, not a button.
Automated tools are tremendously useful for identifying problems and checking technical requirements. But accessible documents still require thoughtful creation, testing, and—in many cases—manual remediation by someone who understands both the document and how people with disabilities will interact with it.
