10 min read 18 October 2026

PDF Accessibility Remediation: A Practical Guide to Fixing Inaccessible PDFs

PDF accessibility remediation is the process of adding or correcting tagged structure, reading order, alt text, and other accessibility properties in existing PDFs. This guide covers the full remediation workflow — auditing, prioritising, fixing in Acrobat Pro, using automated tools, and validating against PDF/UA — with practical advice for large document libraries.

PDF accessibility remediation is the process of transforming existing, inaccessible PDFs into documents that meet PDF/UA (ISO 14289) requirements and can be used by people with disabilities — screen reader users, keyboard-only users, people with low vision. Remediation is necessary for the vast majority of PDFs that were not authored with accessibility in mind: scanned documents, legacy PDFs, design-heavy brochures, and any PDF generated without a tagged-PDF workflow. Remediation is not a quick fix. It requires understanding both the source document's structure and the PDF accessibility model deeply enough to construct a correct semantic tag tree, logical reading order, and accessible annotations. This guide covers the full remediation workflow: audit, triage, tools, and validation.

Auditing before remediating: what needs to be fixed

Before remediating any PDF, run an automated audit to understand the scope and types of failures. This determines the remediation strategy and effort estimate. Automated audit tools: - PAC (PDF Accessibility Checker) — Windows desktop tool from Access For All. Runs Matterhorn Protocol checkpoint tests and produces a visual report with pass/fail by category (document, pages, fonts, content, embedded files, natural language, structure elements, role mapping, alternative descriptions, metadata, annotations, form fields, actions, XFA). - veraPDF — Open-source command-line and GUI validator. Batch-capable, scriptable, and produces machine-readable JSON/XML output. The definitive tool for large-scale PDF/UA validation. Run: verapdf --flavour ua1 document.pdf. - Adobe Acrobat Pro — Tools > Accessibility > Full Check. Less detailed than PAC/veraPDF but integrated into the Acrobat remediation workflow. Common audit failure categories: 1. Untagged document — No tag structure at all. The most severe failure; full tagging from scratch is required. 2. Missing document title — /Title not present in XMP metadata. Quick fix. 3. Language not set — /Lang absent from document catalogue. Quick fix. 4. Figures without alt text — Every non-decorative image needs an /Alt attribute on its /Figure structure element. 5. Incorrect reading order — Content in the structure tree is not in logical reading order. Affects multi-column layouts, sidebars, and complex page designs. 6. Table structure failures — Missing /TH header cells, missing /Scope attributes, incorrect nesting. See tables guide. 7. Missing font ToUnicode maps — Text cannot be extracted. See font embedding. 8. Artifacts not marked as artifacts — Decorative elements (page numbers, running headers, divider lines) cluttering the reading experience if read by a screen reader.

Remediation workflow in Adobe Acrobat Pro

For individual documents or small batches, Adobe Acrobat Pro is the primary remediation tool. The workflow: 1. Set document metadata — File > Properties > Description. Set Title (required for PDF/UA), Author, Subject. Set document language: File > Properties > Advanced > Reading Options > Language. 2. Run the Accessibility Checker — Tools > Accessibility > Full Check. Review the results panel — each failure links to the problematic element in the document. 3. Fix reading order with the Reading Order tool — Tools > Accessibility > Reading Order. This tool shows numbered regions indicating the current reading order. You can: drag regions to reorder, merge regions, split regions, mark content as figure/text/table/form field/artifact. For complex multi-column layouts, the Reading Order tool is the primary instrument. 4. Add alt text to figures — In the Tags panel, find each /Figure element. Right-click > Properties > set Alternate Text. For decorative images, right-click the tag > Change Tag to Artifact (removes it from the reading experience). 5. Fix table structure — In the Tags panel, navigate to each table. Verify /THead, /TBody, /TR, /TH, /TD nesting. For header cells, right-click /TH > Properties > set Scope to Column, Row, or Both. For merged cells, set ColSpan/RowSpan in Object Properties. 6. Fix form fields — Tools > Prepare Form. Each field must have a tooltip (right-click field > Properties > General > Tooltip) and correct tab order (Fields panel > Tab Order view). 7. Validate — Run PAC or veraPDF after remediation to confirm fixes are complete.

Automated and AI-assisted remediation tools

For large document libraries, manual Acrobat Pro remediation is economically impractical. Several automated and semi-automated approaches exist: CommonLook PDF — A specialised PDF accessibility remediation tool with a guided wizard interface. More efficient than Acrobat Pro for systematic remediation, with better table and list handling. Includes a GlobalAccessible remediation standard checker. commonlook.com Axes4 axesPDF — An Acrobat Pro plug-in (Windows) that adds accessibility-specific workflows: smart alt text suggestions, quick role mapping, batch tag fixes, and guided remediation checklists. Faster than Acrobat's native tools for experienced remediators. axes4.com AI-assisted auto-tagging — Machine learning models that analyse rendered page images and generate tag structure proposals. Tools include Adobe Acrobat's AI tagging (File > Accessibility > Auto-tag Document, improved in 2024–2025 versions), Equidox, and services from Texthelp and Level Access. AI auto-tagging accuracy varies: good for simple single-column documents, problematic for tables, multi-column layouts, and complex graphics. AI output always requires human review. Source document workflows — The most scalable remediation is to fix the upstream authoring process. If documents are generated from templates (Word, InDesign, a CMS), updating the template to produce accessible output eliminates the need for per-document remediation. See tables and fonts guides for authoring-level improvements. Prioritisation for large libraries — For organisations with thousands of inaccessible PDFs, full remediation of every document is often not economically viable. Triage by: (1) current usage frequency (active documents first); (2) regulatory risk (public-facing documents subject to web accessibility regulations first); (3) complexity (simple text-heavy documents are cheap to remediate; design-heavy brochures are expensive).

Validating remediated PDFs

Remediation is only complete when the document passes validation. The validation stack: Automated validation — veraPDF (verapdf --flavour ua1) and PAC both run the full Matterhorn Protocol checkpoint set. A remediated document should pass both. Note that inter-validator differences exist for some edge cases — resolving a veraPDF failure that PAC does not report, or vice versa, may require reading the Matterhorn Protocol specification directly. Screen reader testing — Automated validators catch structural failures but cannot assess whether the document makes sense when read aloud. Test with: - NVDA (free, Windows) + Adobe Reader: the standard combination for PDF screen reader testing. Navigate using NVDA's virtual cursor (arrow keys) and Acrobat's accessibility navigation features. - VoiceOver + Preview (macOS): built-in, no installation required. Less comprehensive PDF navigation than NVDA+Reader. - JAWS (commercial, Windows) + Adobe Reader: used in enterprise/legal environments. Important for government and financial services compliance testing. Keyboard navigation test — Tab through all interactive elements (links, form fields, buttons). Confirm: focus is visible, tab order matches visual order, all interactive elements are reachable without a mouse. Zoom test — View the document at 200% and 400% zoom in a PDF viewer. Confirm text remains readable and layout does not break (important for low-vision users who rely on zoom rather than screen readers).

Frequently asked questions

How much does PDF accessibility remediation cost?

Remediation cost depends heavily on document complexity. Simple text-heavy documents (reports, memos, letters) with good underlying text: USD 15–50 per document for professional remediation. Moderately complex documents (forms, multi-section reports with tables and figures): USD 50–200. Design-heavy documents (brochures, annual reports, PDFs with complex graphics): USD 200–1,000+. Scanned documents requiring OCR before tagging: add OCR cost on top of remediation. These are professional remediator rates — DIY remediation in Acrobat Pro costs only time but is significantly slower. For large libraries, automated pre-processing (auto-tagging + human QA) can reduce per-document cost to USD 5–20 for simpler documents at scale. Budget planning should use a representative sample audit to estimate the distribution of complexity in the full library.

Can I remediating a PDF without Adobe Acrobat Pro?

Fully featured PDF remediation without Acrobat Pro is difficult. The free alternatives: (1) axesPDF QuickFix (free version of Axes4 tool) handles some common fixes. (2) CommonLook has a free trial. (3) Programmatic remediation with PyMuPDF or iText can fix specific issues (add document title/language, fix metadata, add simple alt text to figures) but full structure tree manipulation is complex and not well-supported in most PDF libraries. The practical minimum for serious remediation work is Acrobat Pro (currently subscription-based via Adobe Acrobat). For organisations remediating at scale, specialist tools (CommonLook, axesPDF) are cost-effective despite licensing costs due to improved efficiency over Acrobat Pro.

What is the difference between PDF/UA-1 and PDF/UA-2 remediation targets?

PDF/UA-1 (ISO 14289-1:2012) is based on PDF 1.7 and is the current widely-supported accessibility standard. All validators (PAC, veraPDF), authoring tools, and assistive technologies have mature PDF/UA-1 support. PDF/UA-2 (ISO 14289-2:2024) is based on PDF 2.0 and adds new structure elements (/Aside, /FENote) and MathML support. As of 2026, PDF/UA-1 remains the practical remediation target for most workflows because: the majority of PDF viewers have better PDF/UA-1 support than PDF/UA-2, validator consensus for PDF/UA-2 is still maturing, and most authoring tools produce PDF 1.7 output. Target PDF/UA-2 when: your workflow produces PDF 2.0 output, you have mathematical content requiring accessible MathML embedding, or your regulatory mandate specifically references ISO 14289-2.

Try it yourself

Everything in this article can be done directly in the free PDFTheGuide editor — no account required.

Edit PDF Add text to PDF Fill PDF form Open editor