Say you spent a year scanning 1,200 real estate contracts for your office archive. You saved them as PDFs, backed them up to a hard drive, and closed the project. Five years later, someone opens one and finds the typography has subtly shifted, an embedded video no longer plays, and a signature field displays a blank square. Did the file break? No — the original PDF simply was not built for long-term archiving. That is the problem PDF/A was designed to solve, and it is why you are about to see the format pop up in legal, government, medical, and research contexts for decades to come.
This guide explains what PDF/A is, how it differs from a regular PDF, which variant you actually need in 2026, and how to create an archive-ready file without uploading anything to a third-party server.
What is PDF/A?
PDF/A is a version of PDF designed specifically for long-term archiving. It is a subset of the PDF specification defined by the international standard ISO 19005, and its goal is simple: make sure your document looks and reads the same in thirty years as it does today, regardless of what software or operating system opens it.
Where a regular PDF gives you maximum flexibility — you can embed videos, link to external content, run JavaScript, and encrypt the file — PDF/A strips away every feature that might break archival reliability. It requires all fonts to be embedded in the file, prohibits references to external resources, and disallows encryption that could lock out a future reader. In trade, you get a self-contained, predictable, long-lasting document.
Why the “A” matters
The A in PDF/A stands for archive. It signals that the file meets a conformance level designed for records retention, government filings, medical documents, and any other context where the document might need to be opened, read, and trusted many years after it was created. When a court clerk or a records office says they need a PDF/A, they are not asking for a cosmetically different file — they are asking for a file that passed a specific validation test.
The PDF/A family, 2005 to 2026
PDF/A has grown over the years to keep pace with evolving PDF capabilities and cryptography. Here is the short version of how the variants fit together.
| Part | Published | Based on | Key additions |
|---|---|---|---|
| PDF/A-1 | 2005 | PDF 1.4 | Baseline archival subset |
| PDF/A-2 | 2011 | PDF 1.7 | JPEG 2000, layers, PDF attachments |
| PDF/A-3 | 2012 | PDF 1.7 | Attachments of any file type (e.g. XML invoices) |
| PDF/A-4 | 2020 | PDF 2.0 | Modern Unicode, PAdES signatures, AES-256 |
PDF/A-1 was the original, and PDF/A-4 — published as part of ISO 19005-4:2020 — is the current modern baseline, built on PDF 2.0. The big shift in PDF/A-4 is that it drops the older A/B/U conformance-level system and replaces it with three variants (PDF/A-4, PDF/A-4e, PDF/A-4f). Unicode mappings are mandatory, so text is always searchable out of the box.
Most organizations today still store archives in PDF/A-1b or PDF/A-2b because tools and readers have had fifteen-plus years to catch up, but PDF/A-4 is increasingly recommended in new 2026 compliance guidance.
PDF vs PDF/A: the actual differences
Both formats use the .pdf extension and open in the same readers — the difference is what is allowed inside the file.
| Feature | Regular PDF | PDF/A |
|---|---|---|
| JavaScript | Allowed | Forbidden |
| External links & references | Allowed | Forbidden |
| Embedded audio / video | Allowed | Forbidden (PDF/A-1 & -2) |
| Fonts | Can be linked | Must be fully embedded |
| Encryption | Allowed | Forbidden in PDF/A-1/-2/-3; AES-256 allowed in PDF/A-4 |
| Color profile | Optional | Device-independent profile required |
| Transparency | Allowed | PDF/A-2 and later only |
| Metadata | Optional XMP | XMP required |
| File attachments | Any | PDF only (A-2); any type (A-3 and A-4) |
If you read that table and thought “PDF/A sounds restrictive,” that is exactly the point. Archival integrity comes from ruling out every feature that could break in the future. A video codec that stops shipping with operating systems in 2040 will not silently break your contract.
Do you actually need PDF/A?
For most everyday PDFs — a travel itinerary, a receipt, a school handout — you do not need PDF/A. A regular PDF is fine. Use PDF/A specifically when any of the following apply.
- Legal filings. Federal courts in the United States (PACER/CM-ECF) and many state courts require or recommend PDF/A for case documents.
- Government records. Agencies such as the U.S. National Archives (NARA) and many European state archives explicitly list PDF/A as an accepted preservation format.
- Medical records and clinical trials. FDA submissions and ICH e-CTD (electronic common technical document) workflows accept PDF/A for long-term record integrity.
- Academic theses and dissertations. ProQuest and most university libraries require PDF/A for final deposits.
- Corporate retention policies. Any document with a seven-plus year retention requirement — contracts, board minutes, tax records — benefits from PDF/A.
- E-invoicing. Germany’s ZUGFeRD and the EU’s Factur-X standards use PDF/A-3 so the human-readable invoice and the machine-readable XML can travel in a single file.
If a checklist, portal, or clerk asks for “PDF/A,” “PDF/A-1b,” “PDF/A-2b,” or “archive-grade PDF,” that is your cue.
The technical rules that matter
You do not need to memorize the spec, but a few rules explain why your files will behave differently.
Fonts are embedded, not linked
Regular PDFs can reference a font by name and expect the viewer to find it on the reader’s machine. That works today, but fonts change names, licensing terms, and file paths over time. PDF/A requires the full subset of every font used to live inside the file itself. The cost is a slightly larger file. The benefit is that your document always renders identically, whether it is opened in 2026 or 2056.
Color is self-describing
A JPEG or a logo color inside a regular PDF can look subtly different on a calibrated press, a laptop screen, and a phone. PDF/A requires an embedded ICC color profile (or an OutputIntent declaration) so the file carries enough information to reproduce colors faithfully decades later.
No JavaScript, no surprises
Regular PDFs can carry JavaScript, which is useful for fillable forms but is a sizable archival liability. A form that auto-populates a date field in 2026 might throw an error in a sandboxed 2040 reader. PDF/A disallows JavaScript entirely, which keeps the document static and predictable.
Security without lockout
PDF/A-1, -2, and -3 forbid encryption because a lost or expired password would make the archive unreadable. PDF/A-4 softens this — it allows AES-256 encryption when needed for confidentiality, but still rules out weak ciphers like 40-bit RC4. If you need both archiving and a password, PDF/A-4 plus a documented key-escrow process is the modern answer.
PDF 2.0 and PDF/A-4 in 2026
ISO 32000-2:2020, the dated revision of PDF 2.0, is the base specification PDF/A-4 builds on. The main 2026-relevant improvements flowing through to PDF/A-4 include several features that make archiving more predictable and more secure.
- AES-256 only. Weaker encryption schemes are formally deprecated. If you still have PDFs encrypted with 40-bit RC4, 2026 is the year to re-encrypt them.
- PAdES digital signatures. Long-term validation for digital signatures is built into the spec so a signature applied in 2026 can still be verified in 2050.
- Mandatory Unicode mappings. Every glyph in the file must declare its Unicode value, which means reliable text search and copy-paste — no more PDFs that paste as gibberish.
- Richer logical structure. Improved tagging support helps screen readers and aligns PDF/A with PDF/UA (Universal Accessibility) expectations.
If you are setting up a records workflow in 2026, PDF/A-4 is the forward-looking choice. If you are handing a file to a 2010-era validator, PDF/A-1b is still the safest bet.
How to convert a regular PDF to archive-ready (step by step)
You can prepare a document for archiving in Tiny PDF Tools without installing a desktop app. Everything runs in your browser — the PDF is opened, processed, and saved back to your device in the same tab, with nothing ever uploaded to our servers. For PDF/A-specific conformance certification you will still want a validator like the open-source veraPDF, but for the preparation steps a client-side toolkit handles the heavy lifting.
- Start with a clean source file. Open your PDF and confirm the fonts and images look correct. If it started life as a Word document, make sure you exported with the “Embed all fonts” option turned on.
- Flatten any form fields or annotations. Fillable forms, sticky notes, and stamps should be baked into the page so no interactive scripting is left behind. Use the Flatten PDF tool to do this in one click.
- Remove JavaScript and external references. Strip any embedded JavaScript and make sure no links point to network resources that might disappear. A hyperlink to an archived document on an internal wiki will rot.
- Combine multi-part documents. If your archive record spans several files, use Merge PDF to produce a single document with a clear page order.
- Shrink the file sensibly. PDF/A files can get large because fonts and color profiles are embedded. Use Compress PDF to reduce size without aggressive down-sampling. Avoid JPEG-2000 compression if you are aiming for PDF/A-1, which does not support it.
- Check the metadata. Confirm that the title, author, and subject fields are accurate — these are the XMP metadata fields PDF/A requires. Overwrite any generic “Microsoft Word — Untitled.docx” entries.
- Validate conformance. Before you hand the file to a regulator, run it through a free validator like veraPDF. This is the authoritative “did I get it right?” check and will tell you the exact conformance level the file meets.
If the validator flags an issue, the most common fixes are embedding a missing font, removing a transparency effect (for PDF/A-1), or replacing an encrypted file with an unencrypted one.
Keep your archive private
One of the reasons archival PDFs exist at all is that the documents inside them tend to be sensitive: contracts, medical records, tax returns, board minutes. Uploading them to a cloud converter to “make it PDF/A” defeats the privacy guarantee you likely promised to auditors or clients. Tiny PDF Tools processes every file entirely inside your browser using WebAssembly and JavaScript — the bytes never leave your device, there are no accounts to create, and nothing is logged. For deeper background on why that matters, see our post on why client-side PDF tools are safer than cloud editors.
Storage hygiene beyond the format
PDF/A alone does not make an archive bulletproof. The format guarantees visual fidelity, but your storage strategy guarantees the file actually still exists years from now. A practical checklist for a durable archive layers several things together: store at least three copies on two different media with one copy off-site, compute a SHA-256 checksum at ingest and verify it on a schedule, and plan a migration every five to seven years so the files move forward as operating systems evolve. Pair PDF/A with that storage discipline and you have a records system that will outlive most of the hardware it sits on.
Frequently Asked Questions
Can I open a PDF/A file with a regular PDF reader?
Yes. Every PDF/A file is still a valid PDF and opens in Adobe Acrobat, Apple Preview, Chrome, Firefox, and Edge without any special reader. The archive-only rules apply at creation and validation time, not at read time.
Is PDF/A the same as a flattened PDF?
No, but there is overlap. A flattened PDF merges interactive elements such as form fields and annotations into the page content. PDF/A goes further by also requiring embedded fonts, embedded color profiles, XMP metadata, and a long list of forbidden features. Flattening is often the first step in producing a PDF/A. See our guide to flattened PDFs for the full picture.
Which PDF/A version should I pick?
If you have no guidance, pick PDF/A-2b — it is broadly supported and allows modern features like transparency and JPEG-2000 images. Pick PDF/A-1b if a validator or portal explicitly asks for it. Pick PDF/A-3 if you need to attach a machine-readable file like an XML invoice. Pick PDF/A-4 for new workflows where PDF 2.0 support is available.
Can I password-protect a PDF/A file?
Only in PDF/A-4. PDF/A-1, -2, and -3 forbid encryption. If you are using an older PDF/A variant and need confidentiality, protect the file at the storage layer — an encrypted disk or access-controlled share — instead. See our PDF password protection guide for AES-256 options on regular PDFs.
Does PDF/A guarantee my file will be readable in 100 years?
Nothing guarantees 100 years, but PDF/A is the closest public standard we have. The format was explicitly designed with long-term preservation in mind, and national libraries and archives worldwide list it as a preferred format. Pair it with good storage hygiene — multiple backups, checksums, and a migration plan — and you will be in excellent shape.
Is PDF/A the same as PDF 2.0?
No. PDF 2.0 (ISO 32000-2) is the base PDF specification published in 2017 and updated in 2020. PDF/A-4 is a subset of PDF 2.0 tuned for archiving. Think of PDF 2.0 as the whole buffet and PDF/A-4 as the dietician-approved plate.
How can I tell if a PDF is already PDF/A?
Open it in Adobe Acrobat — a blue status bar at the top will say “PDF/A view” if the file conforms. You can also check the XMP metadata for a pdfaid:part property. Free validators like veraPDF will confirm it along with the specific conformance level.
The bottom line
If you are reading this because a compliance checklist asked for a PDF/A file, the short answer is that yes, PDF/A is different, yes you probably do need it, and no, you do not have to pay for Acrobat Pro to produce one. A clean source document, a flattened layout, embedded fonts, and a quick validator pass are most of the work. Everything else — merging drafts, compressing for storage, removing sensitive annotations — can happen in your browser in minutes. Start by flattening interactive elements with our free Flatten PDF tool and work from there.
This article is for general informational purposes only. PDF behavior can vary between viewers, operating systems, and PDF versions. Tiny PDF Tools processes your files entirely in your browser — nothing is uploaded to our servers.