Skip to main content
Technical

PDF/A vs PDF: what archival compliance actually means for your documents

Why “save as PDF” is not the same as preparing a file that must still open in thirty years — and which ISO rules close that gap.

By · Updated May 17, 2026

Regular PDF is a flexible container. PDF/A is a restricted profile of that container, defined so archival systems can render the document without depending on external resources, proprietary plugins, or security policies that expire. If you submit records to a court, national archive, or regulated repository, “PDF” alone may be rejected — they often require PDF/A by name. This article explains the standard, why ordinary PDFs fail preservation tests, how the three conformance families differ, and how to produce and verify compliant files.

What PDF/A is and who created it

PDF/A (“PDF for Archiving”) is standardized as ISO 19005. Work began in the early 2000s with input from national libraries, including the U.S. Library of Congress, and the archival community’s need for a vendor-neutral, self-contained format. The PDF Association and industry partners aligned PDF/A with the existing PDF specification (ISO 32000) but removed features that undermine long-term reproducibility.

PDF/A is not a separate file format users open differently — it is still PDF, with a declared conformance level (for example PDF/A-2b) embedded in XMP metadata and structural constraints enforced by validators.

Why regular PDFs are not reliable for long-term archiving

A typical “Save as PDF” from office software or a browser may produce a valid PDF that fails archival rules:

  • Font subsetting without full embedding — future systems substitute fonts and reflow text.
  • References to external fonts or ICC profiles on the network — links rot.
  • Encryption and rights management — keys and policies become unsupported.
  • Embedded JavaScript or launch actions — security models and interpreters change.
  • Multimedia, 3D, or proprietary attachments — decoders disappear.
  • Transparency and compression modes unsupported in stricter PDF/A levels — renderers disagree.

Archivists need a file that behaves the same when opened in 2035 on an unknown OS. PDF/A trades flexibility for predictability.

PDF/A-1, PDF/A-2, and PDF/A-3 — conformance levels

Each ISO part defines a version of the standard. Within a version, suffix letters describe conformance (commonly b for basic visual preservation and a for accessible structure with tagged PDF).

PDF/A-1 (ISO 19005-1)

Based on PDF 1.4. The strictest baseline: no transparency, no JPEG 2000, no embedded files, no optional content (layers) in level B. Widest support in legacy archival systems. Appropriate when your repository specification names PDF/A-1b explicitly — many government ingest pipelines still do.

PDF/A-2 (ISO 19005-2)

Based on PDF 1.7. Adds JPEG 2000, transparency, layers, and digital signatures with embedded timestamps. PDF/A-2u requires Unicode text. Use when source documents need transparency (soft masks in graphics) or when your authority accepts newer ISO parts.

PDF/A-3 (ISO 19005-3)

Same rendering constraints as PDF/A-2 for the PDF/A representation, but allows embedding arbitrary files (XML, CSV, source data) as associated files alongside the archival view. Common in invoices (ZUGFeRD / Factur-X) where humans read the PDF and machines read the XML attachment.

Choose the part your recipient mandates. When unspecified, PDF/A-2b is a common default for new systems; PDF/A-1b remains the conservative choice for maximum validator compatibility.

What PDF/A requires in practice

  • All fonts used for visible text fully embedded — no reliance on system font substitution.
  • Color spaces defined with embedded ICC profiles where required by the part and level.
  • No encryption or password protection — archival copies must be readable without keys.
  • No external dependencies — no URLs required to fetch fonts, images, or content.
  • No JavaScript, no hidden executable actions.
  • No audio or video — the archival object is a static visual record (attachments differ in PDF/A-3).
  • XMP metadata identifying PDF/A conformance.

Conversion tools remap or reject forbidden features. A file that merely “looks” like a scan is not compliant until it passes machine validation.

Who needs PDF/A

  • Government agencies and national archives — formal submission guidelines often cite ISO 19005.
  • Legal firms and courts — especially in jurisdictions that require archival PDF for e-filing.
  • Healthcare — long-term medical record retention where format stability is audited.
  • Academic institutions — thesis repositories and library digitization projects.
  • Financial services — communications and report archiving under retention rules.

If your obligation is only short-term sharing with a colleague, standard PDF may suffice. If the obligation is retention measured in decades, PDF/A is the engineered answer.

How to convert a regular PDF to PDF/A

Conversion walks the file through normalization: embed fonts, strip or disable forbidden features, assign color profiles, and write conformance metadata. Manual “print to PDF” does not guarantee compliance — it often creates a new PDF that still omits full font embedding or inherits transparency.

Use a dedicated converter such as way2pdf PDF to PDF/A: upload the source PDF, run conversion, and download the output. Our service targets PDF/A-1b via PyMuPDF’s PDF/A export for broad compatibility with archival validators. Files are deleted within about one hour; no signup is required.

If the source is password-protected, remove encryption first with Unlock PDF — PDF/A cannot retain passwords. If the source is a scan without text, run OCR only when you need searchability; OCR does not replace PDF/A conversion for archival conformance.

How to verify genuine PDF/A compliance

Do not trust the filename or a label in the document title. Verification steps:

  1. Open the file in a PDF/A-aware validator (veraPDF, Adobe Preflight, or your repository’s ingest checker).
  2. Confirm the reported part and conformance level (e.g., PDF/A-1b) matches your requirement.
  3. Review the error report — font embedding, transparency, and color space issues are the most common failures.
  4. Visually spot-check critical pages after conversion — validators catch structure, not semantic content errors.

Repositories often reject on a single fatal violation. Keep the pre-conversion source until sign-off.

Common mistakes when creating PDF/A

  • Using fonts with embedding restrictions — the converter may substitute or fail validation.
  • Leaving JavaScript or interactive form actions from the source PDF.
  • Applying password protection after conversion — invalidates PDF/A.
  • Assuming “PDF/A export” in creative tools without running an independent validator.
  • Embedding hyperlinks that depend on live external content for meaning — the visual archive should stand alone.
  • Converting heavily redacted or repaired files without checking that repair did not inject disallowed objects.

After conversion, if file size is problematic for upload limits, compress only after confirming your pipeline accepts re-compressed PDF/A — some archives forbid post-processing that breaks conformance.

PDF/A is insurance on readability

Standard PDF optimizes for authoring convenience. PDF/A optimizes for reproducibility. Knowing the difference keeps submissions accepted and preserves trust that a document opened today will still represent the same record a generation from now.

Convert to PDF/A

Related: PDF security guide · Repair PDF tool

In-depth guides & tools

Step-by-step documentation on way2pdf tools—not just the blog article above.

Tool guide
PDF Form Designer — AcroForms explained

Build fillable PDF forms: field types, PDF vs web forms, download capabilities.
Developer guide
Code formatters in CI/CD & pipelines

JSON, YAML, SQL in GitHub Actions, debugging logs, secure review.
Blog guide
All 15 code formatters overview

JSON, XML, CSS, SQL, YAML, Markdown, Base64, and more.
Directory
Formatters & code tools hub

Open any formatter—runs in your browser, nothing uploaded.

way2pdf exists so you can finish document tasks quickly—without installing desktop suites, creating yet another account, or wondering where your file was stored after you clicked “convert.”

Why PDFs still dominate work and school

A PDF freezes layout: fonts, spacing, and images look the same on a phone, a courtroom laptop, or a print shop PC. That stability is why contracts, syllabi, invoices, and government forms are shared as PDF. The trade-off is that PDF is a presentation format first—editing, extracting tables, or reusing text often requires a deliberate step (conversion, OCR, or merge/split) rather than typing directly into the file.

Our tools are organised around those real steps: turn a scan into searchable text with OCR, shrink an oversized scan for email with compression, combine exhibits with merge, pull out one chapter with split, or move between Word, Excel, and PDF when you need an editable source document again.

If you are unsure which tool fits your situation, start with the FAQ or the PDF glossary for short definitions of terms like vector text, rasterisation, and PDF/A. For longer walkthroughs—redaction, compression settings, cloud imports—see the guides on the blog.

Privacy and retention

Files are processed for your session and removed automatically afterward; we do not use your documents to train models or build marketing profiles. Details vary slightly by tool (for example, AI-assisted features that call an external API)—read the plain-language breakdown on our privacy policy and the about page for how the service is run.

When something goes wrong

Conversion quality depends on the source: heavily compressed scans, unusual fonts, or password-locked PDFs can all affect results. If a specific file fails or looks wrong, use contact and include the file type, roughly how many pages, and what you expected versus what you got—we read those messages and use them to prioritise fixes.

Preparing files before you upload

A few minutes of preparation often saves a failed conversion or a layout surprise. For Office documents, embed unusual fonts where possible and simplify very complex charts before exporting to PDF. For scans, use straight pages, adequate contrast, and at least 300 DPI if you plan to run OCR—thin pencil marks and coloured highlighter streaks are the hardest patterns for any recognition engine.

For password-protected PDFs, you may need to unlock them first using the password the author gave you; we cannot bypass encryption on someone else’s file. For very large files, try compressing or splitting into sections so your browser can upload reliably on slower connections.

If you are comparing several operations (for example, PDF to Word then Word to PDF for a “clean” PDF), download and inspect each intermediate result so you can spot whether the issue comes from the source file or a specific tool chain.

Formats and tools at a glance

Word (DOCX/DOC) is ideal when you need flowing paragraphs and tracked changes; Excel (XLSX) preserves formulas and pivots; PowerPoint (PPTX) keeps slide masters and speaker notes in ways PDF cannot always round-trip. HTML is useful for archiving web pages or sharing read-only snapshots, while images (JPG, PNG) are best when the source is already photographic or when you need one page at a time from a PDF via PDF to JPG.

Beyond conversion, way2pdf includes utilities for day-to-day hygiene: rotate mis-scanned pages, watermark drafts, number long reports, redact sensitive fields, compare two revisions, and repair damaged files when the viewer shows errors.

Developers and analysts often use our code and data formatters—JSON, XML, SQL, YAML, CSV, and more—entirely in the browser for configuration files and API payloads, alongside the PDF tools when documentation needs to ship as a single portable file.

Students & educators

Merge readings, compress hand-ins, OCR lecture scans, and export slides—see PDF tools for students.

Legal teams

Redaction, combine exhibits, page order, and unlock workflows for bundles you control—see PDF tools for lawyers.

Finance & accounting

Tables, statements, and appendices in one package—see PDF tools for accountants.

All conversion tools  ·  Terms & conditions  ·  Support the project