What are the WCAG?
The Web Content Accessibility Guidelines (WCAG) are the internationally recognised standard for digital accessibility. Published by the World Wide Web Consortium (W3C), the current version is WCAG 2.2, a W3C Recommendation since October 2023. WCAG explains how to make digital content usable by disabled people, including people who use screen readers, magnification software or a keyboard alone.
WCAG is organised around four principles:
- Perceivable: Information must be available in forms that people can perceive, for example through alt text for images and sufficient contrast.
- Operable: Navigation and interaction must work without a mouse, and users must be able to orient themselves within the document.
- Understandable: Content and operation must be comprehensible. This includes declaring the language of the content programmatically.
- Robust: Assistive technologies must be able to interpret the content reliably, which requires a sound technical structure.
Each principle is supported by specific success criteria assigned to one of three conformance levels:
| Level | Meaning |
|---|---|
| A | Minimum requirements; without them, many users cannot access the content at all. |
| AA | The conformance level commonly required in the EU, including by EN 301 549 and BITV 2.0. |
| AAA | Additional requirements that legislation does not generally require in full. |
In practice, WCAG conformance almost always means level AA: all success criteria at levels A and AA.
Does WCAG apply to PDFs?
Yes. WCAG is technology-neutral and applies to content, not only to websites. The European standard EN 301 549 provides the formal link to documents: Chapter 10, “Non-web documents”, applies the level A and AA success criteria to documents outside the web, including PDFs.
This matters wherever legislation references EN 301 549:
- The German Accessibility Strengthening Act (BFSG) and the European Accessibility Act (EAA) require certain products and services to be accessible. PDFs such as contract documents, product information and invoices form part of those services. See our guide to the BFSG and accessible PDFs.
- For authorities, universities and other public-sector bodies, Section 3 of BITV 2.0 also references EN 301 549. These requirements have applied for years and expressly cover file formats such as PDF.
A PDF provided on a website or customer portal must therefore meet the relevant WCAG criteria just as the website does. Note that the currently harmonised version of EN 301 549 references WCAG 2.1; WCAG 2.2 is the current W3C Recommendation and a sensible benchmark for new documents.
Key WCAG success criteria for PDFs
Not every WCAG success criterion applies to documents; many concern interactive web features. The following criteria are particularly important for typical PDFs:
| Criterion | Level | Requirement for PDFs |
|---|---|---|
| 1.1.1 Non-text Content | A | Images, graphics and diagrams need alt text describing their content. Purely decorative elements are marked as artefacts so that screen readers skip them. |
| 1.3.1 Info and Relationships | A | Headings, paragraphs, lists and tables must use structure tags. Only then can a screen reader distinguish a heading from a table header, for example. |
| 1.3.2 Meaningful Sequence | A | The reading order in the tag tree must follow the logical content sequence. This is a common source of errors in multi-column layouts, information boxes and sidebars. |
| 1.4.3 Contrast (Minimum) | AA | Text needs a contrast ratio of at least 4.5:1 against its background, or 3:1 for large text. |
| 2.4.2 Page Titled | A | The document metadata needs a meaningful title that applications can display instead of the file name. |
| 2.4.5 Multiple Ways (correspondingly) | AA | Long documents need ways to navigate them. In PDFs, this primarily means bookmarks that reflect the document structure. |
| 3.1.1 Language of Page | A | The primary language of the document must be declared programmatically so that text-to-speech software uses the correct pronunciation. |
| 3.1.2 Language of Parts | AA | Passages in another language must be marked accordingly, such as an English quotation in a German document. |
As a basic prerequisite, text must be encoded as actual, machine-readable text. A scanned document consisting only of page images already fails success criterion 1.1.1; to a screen reader, it is empty.
WCAG-conforming or PDF/UA-conforming: what is the difference?
In short: WCAG describes what must be accessible; PDF/UA describes how to implement it correctly in PDF.
- The WCAG is technology-neutral. For example, it requires information and relationships to be programmatically determinable, but does not prescribe which PDF tags to use.
- PDF/UA (ISO 14289, currently PDF/UA-2) is the specialist standard for accessible PDFs. It precisely defines requirements for tag structures, artefacts, metadata and embedded fonts. PDF/UA does not address criteria unrelated to the file format, such as colour contrast.
The two standards complement each other. A document structured correctly in line with PDF/UA meets the structure-related WCAG criteria; full WCAG conformance also requires design considerations such as contrast. In practice, the standards should therefore be considered together. Our guide Understanding PDF/UA.
Common WCAG failures in real-world PDFs
Many PDFs fail WCAG because the tools used to create them do not include a proper structure. The most common failures are:
- Scanned PDFs without a text layer: The document consists only of page images. Screen readers find no text, and search and copy functions do not work, resulting in a failure of criterion 1.1.1, among others.
- Missing or incorrect tags from layout and office applications: Many export processes produce untagged PDFs or mark everything as body text. Headings, lists and tables are then not identifiable as such, resulting in a failure of criterion 1.3.1.
- Tables as images: Tariff or pricing tables are embedded as graphics. The data is visible but unavailable to assistive technologies, resulting in failures of criteria 1.1.1 and 1.3.1.
- Incorrect reading order: In multi-column layouts, the reading sequence jumps between columns and boxes, resulting in a failure of criterion 1.3.2.
- Missing document title or language: These two small metadata fields have a major effect. Without them, text-to-speech software may pronounce German using English rules, and applications may display cryptic file names, resulting in failures of criteria 2.4.2 and 3.1.1.
These errors can accumulate by the thousand across large document collections, especially in the following sectors: Insurance, Banks or Authorities.
How Accessful automatically remediates machine-testable failures
Accessful works at the structural layer where most WCAG failures in PDFs occur. Its AI analyses text, layout and visual content, adds structure tags, corrects the reading order, generates alt text for images and graphics, and sets the document title and language—in more than 110 languages, with results typically available in seconds. The document’s layout and content remain unchanged. Every document comes with an audit-ready report that provides evidence for monitoring bodies and audits.
An honest distinction is important: not every WCAG criterion can be remediated within a finished file. Colour contrast under criterion 1.4.3, for example, is a design decision in the source document. A tool that preserves layout and visual design cannot and should not change it automatically. Accessful identifies such issues in the audit results so that you can address them in your templates. Content questions such as plain language also remain an editorial responsibility.
For details of automated auditing and remediation through the web app, API or an on-premises deployment, see PDF accessibility with AI. To assess the current accessibility of your documents, use scan.accessful.de for a free PDF check.