Making a PDF accessible means adding a machine-readable structural layer: tags, a correct reading order and alt text. There are three approaches—creating an accessible source document, manual remediation and automated remediation. The right choice depends primarily on whether you are producing new documents or remediating an existing collection.
What makes a PDF accessible?
To a sighted reader, an accessible PDF looks like any other. The difference lies in its invisible structural layer, which enables screen readers and other assistive technologies to present the document meaningfully. An accessible PDF needs:
- tags that identify headings, paragraphs, lists and tables semantically,
- a logical reading order, including in multi-column layouts,
- alt text for images, graphics and diagrams,
- properly marked-up table structures with header cells and cell associations,
- document metadata, including a title and primary language.
The principal technical standard is PDF/UA-2 (ISO 14289-2), complemented by WCAG 2.2 and the European standard EN 301 549. For the detailed requirements, see Understanding PDF/UA.
The German Accessibility Strengthening Act (BFSG) has made the subject a legal requirement for many businesses since 28 June 2025 where their products or services fall within Section 1 BFSG, including consumer banking and e-commerce services. BITV 2.0 additionally applies to public-sector bodies. For details, see BFSG and accessible PDFs.
Approach 1: create accessible source documents
The cleanest approach for new documents is to build accessibility into the source application. Microsoft Word and Adobe InDesign can export tagged PDFs when the source consistently uses styles, heading levels, alt text and a declared document language.
Limitations of this approach
- Suitable only for new documents. Your existing collection, including forms, reports and brochures from previous years, remains unchanged.
- Quality varies. An export produces usable tags only when everyone uses the templates consistently. In practice, documents are often created under time pressure or by different authors, and small deviations can result in a faulty structure.
- Complex layouts remain difficult. Multi-column pages, nested tables and design-heavy publications often need remediation even after export.
Creating accessible source documents is therefore a sound foundation for future documents, but it does not address existing PDFs. The exported PDF still needs a final accessibility check.
Approach 2: manual remediation
Specialists can remediate existing PDFs in tools such as Adobe Acrobat Pro by auditing and correcting the tag tree, setting the reading order, adding alt text and marking up tables—page by page.
Limitations of this approach
- Specialist knowledge is essential. Correct tagging requires a command of PDF/UA, tag semantics and the tools themselves. Most operational departments do not have this expertise, and it cannot realistically be acquired as a side task.
- Effort grows with every document. Every page is edited individually, and length and layout complexity determine the processing time. This may be reasonable for a few long-lived documents, but becomes a bottleneck for hundreds.
- Difficult to scale consistently. Different specialists make different decisions. Maintaining consistent quality across a large collection requires additional review cycles.
Manual remediation, whether in-house or outsourced, remains an option for small collections and exceptional cases. It is not suitable as the standard process for a continuous flow of new documents.
Approach 3: automated remediation with AI
The third approach also starts with the finished PDF but replaces manual work with software. Accessful uses AI to understand the document’s text, layout and visual content, then automatically adds the missing structural layer—tags, reading order and alt text in more than 110 languages. The layout and content remain unchanged; remediation affects only the structural layer.
Results are typically available in seconds. Every document comes with an audit-ready report as evidence for audits and internal records. The AI reliably identifies common barriers, while the report flags anything that cannot be assessed conclusively by automated means.
There are three ways to use Accessful:
- Web app with drag and drop and live status, requiring no installation and suitable for operational teams,
- REST API for integration with editorial and output-management systems, including batch processing of thousands of documents and webhook notifications,
- On-Premises for deployment entirely within your infrastructure, optionally air-gapped.
See the complete process, from document submission to the audit report, on How it works.
Which approach suits which situation?
The three approaches are not mutually exclusive; most organisations combine them. Use this table as a guide:
| Situation | Recommended approach | Why |
|---|---|---|
| New documents, a small number of templates and trained authors | Accessible source documents plus automated final checks | Accessibility is built in at source, while the final check safeguards export quality. |
| Small collection of individual, long-lived documents | Manual or automated remediation | Manual effort may be manageable for a few documents, although automation is usually faster. |
| Large collection or a continuous flow of new documents | Automated remediation through batch processing or the API | Manual remediation does not scale. Automation processes thousands of documents in a single run and integrates with existing workflows. |
An automated process is particularly important in document-intensive sectors such as Banks, Insurance or in Authorities, where output-management and editorial systems continually generate new PDFs.
Checklist: five steps to an accessible document collection
- Catalogue the collection. Identify which PDFs are publicly available or form part of your services, which are produced continuously and which are archival material.
- Check current status. Test representative documents against PDF/UA-2. For example, use the free PDF check from Accessful, which returns results in seconds.
- Prioritise. Start with actively used documents covered by the BFSG or, for public-sector bodies, BITV 2.0: forms, contract documents and product information. Address rarely accessed archive material later.
- Remediate. Choose the appropriate approach for each document group using the table above. For large volumes, use automated correction with batch processing; the audit reports document every result.
- Embed the process. Prevent new barriers by using accessible source templates, automated auditing and remediation before publication—for example through the API directly within the publishing workflow—and storing audit reports systematically.
This turns a one-off clean-up into an ongoing process: the existing collection becomes accessible step by step, and new documents remain accessible from the outset. For packages and terms, see Pricing.