How PDF-to-EPUB3 Conversion Satisfies the European Accessibility Act
The European Accessibility Act (EAA) has applied since 28 June 2025 and has been enforced for over a year. For organizations distributing digital documents across the EU, this is not a distant compliance concern — it is a live legal requirement. Documents must meet WCAG 2.2 AA. PDFs, almost universally, do not.
EPUB3 does.
This article explains what the EAA requires, why PDF fails to meet those requirements, and how converting PDF to EPUB3 using toolkit.bot/pdf2epub satisfies EAA digital accessibility obligations in a 30-second workflow.
What the EAA Requires for Digital Documents
The European Accessibility Act applies to digital products and services sold or distributed in the EU. For digital documents and publications, the key requirement is conformance with WCAG 2.2 AA — the same standard used in EN 301 549, the harmonized European accessibility standard.
WCAG 2.2 AA requirements that directly affect document formats include:
- 1.3.1 Info and Relationships: Structure (headings, lists, tables) must be programmatically determinable
- 1.3.2 Meaningful Sequence: Reading order must be correct and determinable by software
- 1.4.10 Reflow: Content must reflow to 400% zoom without loss of content or functionality
- 2.4.1 Bypass Blocks: Navigation landmarks must be provided
- 3.1.1 Language of Page: The language of the document must be programmatically specified
Organizations distributing documents — whether internal reports, publications, government filings, or e-learning materials — must ensure those documents meet these criteria.
Why PDF Fails EAA Accessibility Requirements
PDF was designed as a print format. Its structure is fundamentally at odds with the reflowable, adaptive requirements of WCAG 2.2 AA:
- Reflow failure (WCAG 1.4.10): Most PDFs use fixed page dimensions. At 400% zoom on a standard screen, content overflows horizontally and becomes unreadable. A "tagged PDF" improves this slightly, but multi-column layouts — the norm in academic, legal, and government documents — almost always fail reflow requirements.
- Reading order ambiguity (WCAG 1.3.2): PDF reading order depends on the order of content objects in the file, which rarely matches visual reading order for complex layouts. Screen readers frequently read columns out of sequence, mix footnotes into body text, and misinterpret sidebars.
- Navigation structure (WCAG 2.4.1): PDF bookmarks and tagged heading structure require manual creation. Most PDFs in circulation lack proper heading tags and provide no navigation landmarks accessible to AT users.
- Language (WCAG 3.1.1): PDF language metadata is frequently absent or incorrect in documents generated by common office tools.
Even a carefully hand-tagged PDF is a brittle solution — it breaks on resize, breaks on re-export, and requires specialist tooling to produce correctly.
Why PDF/UA Remediation Is Not Enough
Most compliance tooling targets websites; documents are found failing at audit. The usual response is to remediate to PDF/UA (ISO 14289) with Adobe Acrobat Pro, axesPDF or CommonLook. The output is tagged, has alt text and correct reading order, and passes PAC 2024. It still fails WCAG 1.4.10, because page dimensions are embedded in the file and no tagging changes them. A professionally remediated PDF/UA document can therefore still fail an EAA audit.
| PDF/UA remediation | EPUB3 conversion | |
|---|---|---|
| Screen reader support | ✓ | ✓ |
| WCAG 1.4.10 Reflow | ✗ for any complex layout | ✓ by design |
| Per-document effort | Specialist tagging, hours per complex file | Automated, seconds per file |
| Validation | PAC 2024, PDFix | ACE by DAISY, EPUBCheck |
| EAA result | Partial | Full, when ACE passes |
Remediation still has a place for simple single-column PDFs and for documents whose format is mandated (court filings, regulatory submissions). For anything with columns, tables or design-heavy pages, Reflow is unreachable in PDF.
Why EPUB3 Satisfies EAA Requirements
EPUB3 is built on HTML5, CSS, and XML — the same technologies that power accessible web content. WCAG 2.2 applies directly to EPUB3 content, and the EPUB Accessibility 1.1 specification (a W3C standard) formalizes how to produce conformant EPUB3 files.
- Reflow is inherent: EPUB3 is reflowable by design. Text wraps at any viewport width or zoom level. WCAG 1.4.10 passes by default.
- Reading order is explicit: HTML source order defines reading order in EPUB3. A correctly structured EPUB3 always presents content in the right sequence to screen readers.
- Navigation is mandatory: EPUB3 requires a navigation document (NCX/nav.xhtml) providing a table of contents. Heading structure in HTML maps directly to ARIA landmark navigation.
- Language is required: EPUB3 package metadata requires a language declaration. Every compliant EPUB3 file satisfies WCAG 3.1.1.
- EPUB Accessibility 1.1 metadata: Files produced with conformance metadata allow automated checkers and procurement tools to verify accessibility before distribution.
What the EPUB Itself Must Satisfy: EPUB Accessibility 1.1
EPUB Accessibility 1.1 incorporates WCAG 2.2 by reference and defines three conformance levels: WCAG AA (the minimum, and what the EAA needs), WCAG AAA, and Optimized (AA plus page list and extended descriptions). The success criteria map onto concrete EPUB implementation:
| SC | Requirement | EPUB implementation |
|---|---|---|
| 1.1.1 | Non-text content has alt text | alt on every img; empty alt for decorative |
| 1.3.1 | Structure conveyed in markup | Semantic HTML: h1–h6, ul/ol, table with thead/th, figure |
| 1.3.2 | Meaningful sequence | Document order matches reading order; no CSS-only reordering |
| 1.4.3 | Contrast 4.5:1 | Minimum contrast in embedded CSS |
| 1.4.4 | Text resizable to 200% | em/rem units, not px, for font-size |
| 2.4.1 | Bypass blocks | TOC and landmarks in nav.xhtml |
| 2.4.2 | Page titled | title in each XHTML head |
| 2.4.6 | Descriptive headings | Meaningful heading text, not just “Chapter 1” |
| 3.1.1 | Language of page | xml:lang on the html element of each file |
| 4.1.1 | Valid markup | Passes EPUBCheck |
| 4.1.2 | Name, role, value | ARIA roles only where semantic HTML is insufficient |
Beyond WCAG, the specification requires spine order to match the intended reading sequence, an EPUB 3 navigation document covering all major sections, and accessibility metadata in the package document:
<meta property="schema:accessMode">textual</meta>
<meta property="schema:accessMode">visual</meta>
<meta property="schema:accessModeSufficient">textual</meta>
<meta property="schema:accessibilityFeature">structuralNavigation</meta>
<meta property="schema:accessibilityFeature">alternativeText</meta>
<meta property="schema:accessibilityFeature">tableOfContents</meta>
<meta property="schema:accessibilityHazard">none</meta>
<meta property="schema:accessibilitySummary">Structured navigation, alt text for informative images, semantic headings.</meta>
<meta property="dcterms:conformsTo">EPUB Accessibility 1.1 - WCAG 2.2 Level AA</meta>
The failures most often seen in converted EPUBs are images without alt text (charts and figures from the PDF), tables without th cells, a missing language on html, a missing title, skipped heading levels (H1 to H3) and non-descriptive link text. ACE by DAISY catches all of these; see the EPUB accessibility checker guide.
How to Convert a PDF to EAA-Compliant EPUB3 in 30 Seconds
toolkit.bot produces EPUB3 files with EPUB Accessibility 1.1 metadata from any uploaded PDF. The workflow:
- Go to toolkit.bot/pdf2epub
- Upload your PDF (max 50 MB). No account or email required.
- Wait approximately 30 seconds. The converter extracts text, detects heading structure, maps tables, and builds EPUB3 structure with semantic HTML5.
- Download the EPUB3 file. Open it in any EPUB reading system — Apple Books, Kobo, Adobe Digital Editions, or validate it directly with ACE by DAISY.
The output includes: semantic H1–H6 heading structure, explicit language declaration, EPUB3 navigation document with table of contents, EPUB Accessibility 1.1 conformance metadata, and alt text for images where present in the source PDF.
What EAA Compliance Requires Beyond Conversion
Converting to EPUB3 satisfies the format requirements. For full EAA conformance, organizations should also verify:
- Source document quality: PDFs with missing structure (no headings, images with no alt text) will produce EPUB3 with the same gaps. High-priority documents may need post-conversion review in an EPUB editor (Sigil is free).
- ACE validation: Run the converted EPUB through ACE by DAISY to generate an accessibility report. ACE identifies specific WCAG failures and what metadata claims are substantiated.
- Screen reader testing: For high-stakes documents, test with NVDA + Firefox or VoiceOver + Safari before distribution.
Who This Applies To
The EAA scope includes companies with 10+ employees or €2M+ annual revenue selling digital products in the EU, wherever they are headquartered. Article 2 names e-commerce, banking and financial services (banks, insurers, investment platforms), transport (airlines, rail, ferries, ticketing), media and audiovisual services and telecoms directly, and covers e-books and digital publications as products. Documents delivered as part of any covered service — terms and conditions, product guides, financial disclosures, booking documents — inherit the obligation. Public sector organizations are already subject to the Web Accessibility Directive; the EAA extends equivalent obligations to private sector digital products. Sector-specific document lists are in EAA 2026 by sector.
If your organization distributes PDFs as part of any digital product or service in the EU, those documents are in scope.
Convert a PDF to EAA-compliant EPUB3 — free, no install, no account required.
Convert a PDF →Related guides
- EAA 2026: what publishers must do about PDFs
- PDF to accessible EPUB: guide for libraries and accessibility teams
- EPUB accessibility checker: ACE vs Pagina vs EPUBCheck
- EPUB 3 vs EPUB 2: what changed and why it matters
EAA enforcement is now active. See what organizations need to do immediately →