PDF/A vs PDF: What's the Difference?
Side-by-side comparison
The short version: PDF/A trades flexibility for permanence. A standard PDF can do more today; a PDF/A is engineered to still open and look the same long after today's software is gone. The table below contrasts the two across the properties that matter for storage and compliance.
| Property | Standard PDF | PDF/A (ISO 19005) |
|---|---|---|
| Primary purpose | Sharing, printing, interactive forms, everyday distribution | Long-term preservation and archiving |
| Embedded fonts | Optional, may reference system fonts | Mandatory, all fonts fully embedded |
| External dependencies | Allowed (linked fonts, remote resources) | Forbidden, the file must be self-contained |
| JavaScript & executable content | Allowed | Banned entirely |
| Audio, video, 3D, embedded files | Allowed | Disallowed (varies by version) |
| Encryption / password protection | Allowed | Not permitted |
| Colour handling | Device-dependent colour is fine | Requires device-independent colour (embedded ICC profiles) |
| Metadata | Optional | Standardized XMP metadata required |
| Future reliability | No guarantee, rendering may drift over time | Designed to render identically for decades |
What PDF/A actually restricts
PDF/A is a profile of PDF that removes anything which could make a file render differently, or fail to render, in the future. The rule of thumb: if a feature depends on something outside the file, PDF/A either embeds it or bans it.
Everything is embedded
Fonts must be fully embedded so text never falls back to a substitute typeface that shifts layout. Colour must be device-independent, with ICC profiles embedded so a teal stays the same teal on any screen or printer. Metadata is standardized as XMP so cataloguing systems can read it consistently. The goal is a file that carries its entire rendering context inside itself.
Risky features are removed
JavaScript and any executable content are banned outright, both for security and because behaviour that depends on a script engine cannot be guaranteed to work in fifty years. Encryption is not permitted, since a password or DRM scheme could lock the file out of a future reader. Audio, video, and 3D objects are disallowed because they rely on external codecs and players. What remains is static, predictable, and verifiable.
PDF/A versions and conformance levels
PDF/A is not a single thing. It comes in versions (PDF/A-1, -2, -3, -4) and conformance levels (b, a, and u) that trade off compatibility, accessibility, and modern features. Pick the combination that matches your retention and accessibility needs.
The versions
- PDF/A-1 is the original, based on PDF 1.4. It is the strictest and most universally supported, but allows no JPEG2000, layers, or transparency.
- PDF/A-2 adds modern compression (JPEG2000), transparency, layers, and the ability to embed other PDF/A files. It is a common default today.
- PDF/A-3 is like A-2 but allows embedding arbitrary files (for example, the source spreadsheet behind an invoice). It is used for e-invoicing formats such as ZUGFeRD/Factur-X.
- PDF/A-4 is the newest revision, based on PDF 2.0, simplifying and modernising the earlier conformance model.
The conformance levels
- Level b (basic) guarantees reliable visual reproduction. It is the easiest level to produce and the most common.
- Level a (accessible) adds full logical structure and tagging for accessibility and reflow. It is harder to author and validate.
- Level u (unicode) adds reliable Unicode text mapping for search and extraction, sitting between b and a.
When you need PDF/A
Use PDF/A whenever a document must remain readable, faithful, and self-contained for years, typically because a law, regulator, or records policy requires it. For everyday sharing, a standard PDF is fine and more flexible.
Clear cases for PDF/A
- Legal and regulatory records. Court filings, contracts, and compliance documents that must be preserved verbatim and admissible later.
- Government and institutional archives. Many national archives and records-retention rules mandate PDF/A for digital deposit.
- Financial and tax documents. Invoices, statements, and audit trails kept for multi-year retention periods.
- Healthcare, academic, and library archives. Patient records, theses, journals, and any material meant to outlive its source software.
When a standard PDF is the better choice
If a document is interactive (fillable forms with calculations), contains media, needs password protection, or is just being emailed and discarded, a regular PDF is simpler and capable. Converting it to PDF/A would strip the very features that make it useful. Archive a PDF/A copy only once the document is final and you actually need long-term integrity.
Verdict: which should you use?
There is no universal winner. The two formats answer different questions. Choose by the lifespan and obligations of the document, not by habit.
- Need it to survive and stay faithful for years: use PDF/A, and validate the output against your chosen level.
- Bound by a records, legal, or e-invoicing mandate: use PDF/A, and check exactly which version the rule names.
- Everyday sharing, forms, media, or protected files: a standard PDF is more flexible and perfectly fine.
- Unsure which PDF/A flavour: default to PDF/A-2b unless you need full accessibility (an 'a' level) or embedded source files (use PDF/A-3).
In practice, many teams keep the working document as a normal PDF and generate a validated PDF/A version at the point of archiving. That gives you flexibility while you work and durability once the record is closed.
Frequently asked questions
Is PDF/A the same as a regular PDF?
No. PDF/A is a constrained subset of PDF defined by ISO 19005 for archiving. Every PDF/A file is a valid PDF, but a regular PDF is not automatically PDF/A. It may use embedded video, JavaScript, external font references, or encryption that the archival standard forbids.
Can a PDF/A file contain hyperlinks or JavaScript?
PDF/A bans JavaScript and any executable content entirely, because they create dependencies and security risks that break long-term reliability. Hyperlinks are allowed, but external dependencies are discouraged. The file must render the same decades from now without relying on anything outside itself.
How do I convert a regular PDF to PDF/A?
Use Save As or Export in Acrobat, a print-to-PDF/A driver, or a server library like Ghostscript or a PDF SDK. Conversion embeds all fonts, flattens disallowed features, and tags the file. Always validate the output against the chosen conformance level afterward rather than trusting the export blindly.
Which PDF/A version and level should I use?
For most archives, PDF/A-2b is a safe default: it allows modern compression and transparency while staying broadly compatible. Choose an 'a' level (PDF/A-1a or 2a) only when you need full tagged accessibility and logical structure, since it is harder to produce and validate.
Last updated: 14 June 2026