What Is Compliance with PDF/A Standard? A Complete Expert Guide

What Is PDF/A Compliance and Why It Matters

PDF/A compliance—governed by ISO 19005—is the technical and legal benchmark for creating PDF documents that remain fully self-contained, reproducible, and accessible over multi-decade archival periods. Organizations subject to records retention mandates, litigation holds, or accessibility regulations cannot rely on standard PDF files, because standard PDFs can reference external fonts, embedded scripts, or encryption layers that degrade or become unreadable as software environments change.

The standard was first published as ISO 19005-1 in 2005 and has expanded through three active parts covering different archival use cases. Understanding which part and conformance level applies to a given workflow determines everything from font-embedding rules to tag-tree requirements.

What Does PDF/A Compliant Mean?

A PDF/A-compliant document is one that satisfies all mandatory provisions of the applicable ISO 19005 part and conformance level, as confirmed by a validating tool or process. Compliance is not a visual property—two documents can appear identical on screen while only one is ISO-conforming.

Four architectural requirements define compliance across all PDF/A parts:

• Font embedding: Every font used in the document—including subset fonts—must be fully embedded within the file. No reliance on system fonts or font substitution is permitted.
• Device-independent color: All color data must be accompanied by an ICC color profile or use PDF color spaces that do not depend on the output device's color rendering.
• No prohibited content: JavaScript, audio and video content, external content references, encryption, and transparent objects (in PDF/A-1) are not permitted.
• XMP metadata declaration: The file's XMP metadata block must explicitly declare the PDF/A part number and conformance level using the pdfaid:part and pdfaid:conformance schema fields.

Compliance is validated programmatically—tools parse the file's internal structure against the ISO specification and report violations at the object level. Adobe Acrobat includes a preflight profile for PDF/A validation; purpose-built validators such as veraPDF provide more granular reporting. When using RemeDocs, validation is integrated into the remediation pipeline, so conformance is confirmed before any file is delivered.

A common compliance example: a legal brief exported from Microsoft Word contains a proprietary font not embedded in the PDF. That file fails PDF/A-1b validation at the font-embedding rule, even if it renders correctly on the author's machine. Remediation requires re-embedding the font and re-exporting or processing through a PDF/A converter.

What Is PDF/A-1a Compliant?

PDF/A-1a is the highest conformance level within ISO 19005-1 and is the only conformance level that addresses accessibility alongside archival permanence. A PDF/A-1a–compliant file must satisfy every requirement of PDF/A-1b—visual reproducibility—plus two additional structural mandates: tagged PDF structure and logical reading order with Unicode character mapping.

Tagged PDF Structure

The document must contain a complete tag tree—a hierarchical semantic structure that marks up every content element with a role: paragraphs (P), headings (H1 through H6), tables (Table, TR, TD), lists (L, LI), figures (Figure with alt text), and artifacts (page numbers, headers, decorative elements marked as non-content). This structure is what assistive technology—screen readers, braille displays—uses to navigate and interpret the document.

Unicode Character Mapping

Every glyph in the document must map to a Unicode code point. This ensures that text can be reliably extracted, searched, copied, and reflowed regardless of the rendering environment. Documents using custom symbol fonts, ligatures, or glyphs without Unicode mappings fail PDF/A-1a validation even if they pass PDF/A-1b.

PDF/A-1a vs. PDF/A-1b: The Practical Distinction

PDF/A-1b—the basic conformance level—requires only that the document's visual appearance be reproducible. It does not require tagging, reading order, or Unicode mapping. A scanned document with no text layer can theoretically conform to PDF/A-1b but will always fail PDF/A-1a. For organizations subject to accessibility regulations—including those governed by ADA Title II (WCAG 2.1 Level AA) or the European Accessibility Act—PDF/A-1a is the appropriate archival target because it produces documents that are both permanently reproducible and accessible to users of assistive technology.

RemeDocs' remediation workflow targets PDF/A-1a by default for documents entering accessibility-regulated pipelines, building a compliant tag tree, verifying reading order, and confirming Unicode mappings before final validation.

PDF/A vs. PDF: The Architectural Difference

Standard PDF (ISO 32000) and PDF/A (ISO 19005) share the same underlying file format but differ in what they permit. Standard PDF is a general-purpose format optimized for display flexibility and interactivity; PDF/A is a constrained subset optimized for long-term, device-independent preservation.

What Standard PDF Allows That PDF/A Prohibits

• External font references: A standard PDF can reference fonts installed on the viewing system. If those fonts are unavailable, the renderer substitutes an approximation, altering the document's appearance.
• JavaScript and interactive forms: Standard PDFs support embedded scripts, which can alter content dynamically. PDF/A prohibits all JavaScript.
• Encryption and digital rights management: Standard PDFs can be password-protected or encrypted. PDF/A prohibits encryption because it could prevent future access.
• Transparency (PDF/A-1): The PDF transparency model—used by design tools for drop shadows and alpha blending—is prohibited in PDF/A-1. PDF/A-2 and PDF/A-3 permit transparency.
• External content streams: Standard PDFs can reference content hosted outside the file. PDF/A requires all content to be embedded.

What PDF/A Adds That Standard PDF Does Not Require

• Mandatory XMP metadata with conformance declaration
• ICC color profiles for all color data
• Complete font embedding for every glyph used
• Tagged structure and Unicode mapping at the -1a conformance level

The practical implication: converting a standard PDF to PDF/A is not simply a re-save operation. It requires resolving every constraint violation—typically font embedding, color space conversion, transparency flattening, and metadata injection—before the file can pass ISO 19005 validation. A PDF/A converter that only changes the file extension without structural remediation will produce a non-conforming file that fails validation.

PDF/A Conformance Levels Across All Parts

ISO 19005 has three published parts, each targeting a specific archival era and technology capability. Each part defines its own conformance levels, and later parts are not supersets of earlier ones—they address different use cases.

ISO 19005-1: PDF/A-1 (Based on PDF 1.4)

• PDF/A-1a: Full compliance—tagging, reading order, Unicode mapping, visual reproducibility. Required for accessibility-compliant archival.
• PDF/A-1b: Basic compliance—visual reproducibility only. Suitable for scanned documents or image-heavy files where semantic structure is not feasible.

ISO 19005-2: PDF/A-2 (Based on PDF 1.7 / ISO 32000-1)

• PDF/A-2a: Equivalent to -1a requirements plus PDF 1.7 capabilities.
• PDF/A-2b: Basic visual reproducibility under PDF 1.7.
• PDF/A-2u: Visual reproducibility plus Unicode mapping, without full tagging requirement.

PDF/A-2 permits JPEG 2000 compression, transparency, and embedded PDF/A files—capabilities absent from PDF/A-1. It also permits OpenType fonts, which are increasingly common in modern document workflows.

ISO 19005-3: PDF/A-3 (Based on PDF 1.7)

PDF/A-3 is structurally identical to PDF/A-2 but adds one critical capability: arbitrary file attachments. A PDF/A-3 document can embed non-PDF files—XML invoices, spreadsheets, original source files—as attachments while maintaining the outer document's archival conformance. This is the basis for hybrid invoice formats such as ZUGFeRD and Factur-X, which embed machine-readable XML inside a human-readable PDF/A-3 file.

Choosing the Right Part and Level

• Court filings and government records requiring accessibility: PDF/A-1a or PDF/A-2a
• Scanned document archives without text layer: PDF/A-1b or PDF/A-2b
• Financial and invoicing workflows with embedded data: PDF/A-3
• Modern document workflows using OpenType or transparency: PDF/A-2 or PDF/A-3

PDF/A vs. PDF/X, PDF/E, and Other ISO PDF Subsets

PDF/A is one of several ISO-standardized PDF subsets, each engineered for a specific professional domain. Selecting the wrong subset for a workflow produces files that may be technically valid but functionally inappropriate for the intended use.

PDF/X — Print Production

PDF/X (ISO 15930) is the prepress and print production standard. It enforces color management, font embedding, and bleed/trim box definitions required by commercial print workflows. PDF/X prohibits interactive elements for the same reason PDF/A does, but its color requirements are oriented toward CMYK print reproduction rather than screen-independent display. PDF/X and PDF/A serve fundamentally different master purposes: a PDF/X file optimized for offset printing is not suitable as an archival record, and a PDF/A file lacks the print-specific metadata that prepress workflows require.

PDF/E — Engineering Documentation

PDF/E (ISO 24517) targets engineering and technical documentation—CAD drawings, 3D models, manufacturing specifications. It permits 3D artwork and interactive measurement tools that PDF/A prohibits. Organizations managing engineering records archives sometimes mistakenly apply PDF/A requirements to PDF/E documents; the two standards are not interchangeable.

PDF/UA — Universal Accessibility

PDF/UA-1 (ISO 14289-1:2014) is the dedicated PDF accessibility standard. Where PDF/A-1a requires tagging as a component of archival compliance, PDF/UA specifies the full technical requirements for accessible PDFs: tag role mapping, reading order, alt text for figures, table headers, and artifact marking. A document can be PDF/A-1a compliant without being PDF/UA compliant if its tag tree is structurally present but semantically incomplete. For maximum compliance depth—particularly under accessibility regulations—PDF/A-1a and PDF/UA together represent the complete target. RemeDocs produces documents that satisfy both standards simultaneously where the source content permits.

Linearized PDF — Fast Web View

Linearized PDF is not an ISO subset standard in the same sense as PDF/A or PDF/X—it is a file organization optimization. A linearized PDF is structured so that the first page can be rendered by a browser before the entire file downloads. Linearization has no bearing on PDF/A conformance; a file can be both linearized and PDF/A compliant, or either alone.

How to Create a PDF/A File: Technical Requirements

Creating a genuinely conforming PDF/A file requires addressing structural requirements at the point of export or through post-processing remediation—not simply applying a save-as preset and assuming conformance.

Authoring Tool Export

Microsoft Word (via the PDF export dialog, selecting the ISO 19005-1 option), LibreOffice, and Adobe InDesign can produce PDF/A-1b or PDF/A-2b output at export time. These exports embed fonts and inject XMP metadata but do not guarantee full tag-tree completeness for PDF/A-1a. Documents with complex tables, figures without alt text, or custom heading styles frequently fail PDF/A-1a validation despite using the ISO export option.

Adobe Acrobat Preflight

Adobe Acrobat's Preflight tool includes PDF/A conversion profiles. It can fix fixable violations—font embedding, color conversion, metadata—automatically, and reports unfixable violations for manual remediation. This workflow is appropriate for single documents but does not scale to high-volume production pipelines.

PDF/A Converter Tools and Pipelines

Dedicated PDF/A converter tools—server-side applications and APIs—process documents programmatically and can be integrated into document management systems, print workflows, or records management platforms. When evaluating any converter, confirm it produces veraPDF-validated output at the target conformance level, not merely files with a PDF/A metadata declaration.

Remediation for Accessibility Conformance

Reaching PDF/A-1a requires document remediation beyond format conversion: building a complete tag tree, assigning correct role mappings, setting reading order, adding alt text to all figures, and mapping every glyph to Unicode. This is the distinction between format conversion and accessibility remediation. RemeDocs handles both simultaneously—converting to PDF/A while constructing a compliant tag structure and validating against ISO 19005-1 Level A before delivery.

Validation

Validation must be performed by an independent validator, not the tool that created the file. veraPDF is the industry-standard open-source validator for PDF/A-1 through PDF/A-3. It reports violations at the clause level of the ISO specification, enabling targeted remediation. Adobe Acrobat's preflight reports are also valid but use different violation nomenclature.

PDF/A Compliance in Regulatory Contexts

PDF/A is referenced directly or by implication in multiple regulatory frameworks. Understanding which standard applies to which obligation prevents both over-engineering and under-compliance.

ADA Title II and Web/Document Accessibility

The DOJ's ADA Title II rule—published April 24, 2024—mandates WCAG 2.1 Level AA as the technical standard for web and digital content accessibility for public entities. For public entities serving a population of 50,000 or more, the compliance deadline is April 26, 2027; for public entities serving a population under 50,000 and all special district governments regardless of population, the deadline is April 26, 2028, as established by the DOJ interim final rule that extended these dates. PDF documents published on covered entities' websites are within scope. A PDF/A-1a file with a complete, semantically correct tag tree satisfies most of the structural requirements that align with WCAG 2.1 AA for documents, though full WCAG conformance for PDFs also requires PDF/UA-level tag quality.

European Accessibility Act

The European Accessibility Act (Directive (EU) 2019/882) has been in force since June 28, 2025. Conformity is commonly demonstrated against EN 301 549 V3.2.1, which incorporates WCAG 2.1 Level AA. PDF documents distributed as part of in-scope services must meet these requirements. PDF/A-1a with PDF/UA-quality tagging is the appropriate archival and accessibility target for EU-regulated document workflows.

Records Management and Court Filing

Federal and state court e-filing systems increasingly require PDF/A-1a or PDF/A-1b for submitted documents. The U.S. Courts' CM/ECF system specifies PDF/A compliance for certain document types. National archives and records management standards in multiple jurisdictions reference ISO 19005 as the preservation format of record.

PDF/A Compliance Checklist

The following checklist covers the mandatory validation points for PDF/A-1a compliance—the level required for both archival permanence and accessibility conformance.

Font and Character Requirements

• All fonts fully embedded (no system font references)
• All glyphs mapped to Unicode code points (ToUnicode entries present)
• No font substitution warnings in validator output

Color and Visual Requirements

• All color data accompanied by ICC color profile
• No device-dependent color spaces (DeviceRGB, DeviceCMYK without ICC profile)
• No transparency (PDF/A-1; permitted in PDF/A-2 and -3)

Content and Structure Requirements

• Complete tag tree present—every content element tagged with correct role
• Reading order matches visual presentation
• All figures include Alt attribute with descriptive text
• Tables use TH (header cell) and TD (data cell) tags with scope attributes
• Artifacts (page numbers, running headers, decorative elements) marked with Artifact tag
• Document language declared in XMP metadata and tag root

Metadata Requirements

• XMP metadata block present and valid
• pdfaid:part set to correct part number (1, 2, or 3)
• pdfaid:conformance set to correct level (A, B, or U)
• Document title present in XMP dc:title

Prohibited Content

• No JavaScript
• No encryption or password protection
• No external content references
• No embedded audio or video
• No live digital signatures that modify content after signing (signature fields are permitted if properly structured)

Validation

• File validated against target conformance level using veraPDF or equivalent ISO-certified validator
• Zero Level A violations reported
• Validation report retained as part of the document record

Frequently Asked Questions About PDF/A Compliance

What is the difference between PDF and PDF/A?

Standard PDF (ISO 32000) is a general-purpose format that permits JavaScript, encryption, external font references, and interactive content. PDF/A (ISO 19005) is a constrained subset that prohibits all of those features and adds mandatory requirements—font embedding, ICC color profiles, XMP conformance metadata—to ensure the document remains fully readable and reproducible without external dependencies over an indefinite archival period.

What is a PDF/A compliance example in practice?

A county court's e-filing system requires PDF/A-1a for all submitted briefs. An attorney exports a brief from Microsoft Word using the ISO 19005-1 export option, then validates the file with veraPDF. The validator reports a Unicode mapping failure on a symbol font used in footnotes. The attorney replaces the symbol font with a Unicode-mapped equivalent, re-exports, and re-validates. The file now passes PDF/A-1a validation and is accepted by the filing system.

What is PDF/A-1b and how does it differ from PDF/A-1a?

PDF/A-1b is the basic conformance level of ISO 19005-1, requiring only that the document's visual appearance be reproducible—font embedding, color profiles, and no prohibited content. It does not require tagging, reading order, or Unicode mapping. PDF/A-1a adds those three requirements, making the document both permanently reproducible and accessible to assistive technology.

Can I create a PDF/A file using Adobe Acrobat?

Adobe Acrobat's Preflight tool includes PDF/A conversion profiles that can fix common violations—font embedding, color conversion, metadata injection. It is appropriate for individual documents. For high-volume or accessibility-regulated workflows, a dedicated remediation solution such as RemeDocs provides validated PDF/A-1a output with a complete tag tree at scale.

What is PDF/X and how does it relate to PDF/A?

PDF/X (ISO 15930) is the print production standard, governing color management, bleed definitions, and font embedding for commercial printing workflows. PDF/A governs archival preservation. The two standards address different use cases and are not interchangeable—a PDF/X file is not an archival record, and a PDF/A file is not a print-ready file.

What is a linearized PDF?

A linearized PDF is organized so that the first page renders before the entire file downloads—a performance optimization for web delivery. Linearization is independent of PDF/A conformance; a file can be both linearized and PDF/A compliant simultaneously.

Does PDF/A-1a satisfy WCAG 2.1 Level AA for documents?

PDF/A-1a's tagging requirement provides the structural foundation for document accessibility, but full WCAG 2.1 Level AA conformance for PDFs also requires the tag quality specified in PDF/UA-1 (ISO 14289-1:2014)—including correct heading hierarchy, table header associations, and comprehensive alt text. PDF/A-1a plus PDF/UA-1 together represent the complete target for accessibility-regulated document archives.

Immediate Next Steps for PDF/A Compliance

Organizations that need to bring document archives or publishing workflows into PDF/A conformance can act on the following sequence today:

1. Audit your current document output. Run a representative sample of 20–30 documents through veraPDF against your target conformance level (PDF/A-1a for accessibility-regulated workflows). The violation report identifies which rules are failing across your document population—font embedding failures, missing tags, color space issues—and establishes the remediation scope.
2. Identify the correct conformance target. Determine whether your workflow requires PDF/A-1a (accessibility plus archival), PDF/A-1b (visual reproducibility only), or PDF/A-3 (archival with embedded file attachments). Applying the wrong level wastes remediation effort or leaves accessibility obligations unmet.
3. Fix authoring-side issues before remediation. Font selection, heading styles, table structure, and figure alt text are best addressed in the source document (Word, InDesign, etc.) before PDF export. Remediating structural deficiencies in the PDF itself is more labor-intensive than correcting them upstream.
4. Engage a validated remediation pipeline. For documents that cannot be regenerated from source, or for high-volume archives, RemeDocs' PDF remediation process produces PDF/A-1a–conforming output with a complete tag tree, verified Unicode mappings, and a veraPDF validation report for every delivered file.
5. Establish validation as a gate, not an afterthought. Integrate veraPDF or an equivalent validator as a mandatory checkpoint in your document publishing or filing workflow. A conformance declaration in XMP metadata is not a substitute for passing validation—only independent validation confirms that a file actually meets the ISO 19005 specification.