PDF/A Explained: What It Is, When You Legally Need It, and How to Convert

If you've ever been told a document "must be submitted as PDF/A" — by a court e-filing system, a research repository, a tax authority, or a records office — and you weren't sure what that meant or how to comply, this guide is for you. PDF/A isn't a different file extension or a premium feature. It's a stricter, well-defined version of the PDF you already know, built for one job: staying readable for decades.

What PDF/A Actually Is

PDF/A is an ISO standard (ISO 19005) for the long-term archiving of electronic documents. The "A" stands for archive. It was created because a regular PDF — flexible as it is — can quietly depend on things that won't be there in twenty years.

A normal PDF might reference a font installed on your computer rather than embedding it. It might link to an external color profile, rely on JavaScript, or be locked with a password. Open that file on a 2045 machine and the font could be substituted, the colors could shift, the script could break, and the encryption could lock you out entirely. None of those are problems today. All of them are problems for an archive whose entire purpose is to outlive its creator's hardware.

PDF/A removes that fragility by forbidding anything that makes a document non-self-contained. The result is a file that carries everything it needs to render itself, identically, on any compliant viewer, forever.

How PDF/A Differs From a Regular PDF

The differences come down to a short list of hard rules. A PDF/A file must:

• Embed every font it uses. No relying on the reader's system fonts. If a document uses Times New Roman, the actual glyph data travels inside the file.
• Be fully self-contained. No external references — no linked color profiles, no streamed audio/video, no dependencies on files that live somewhere else.
• Specify color unambiguously. Color is defined through an embedded output intent (an ICC profile) so a red stays the same red on any device.
• Carry standardized metadata. Document properties are written in XMP so cataloguing systems can index it reliably.

And a PDF/A file must not:

• Be encrypted. Password protection and access restrictions are banned outright. An archive you might be locked out of isn't an archive. This is the rule that trips people up most often: if your source PDF has any security applied, conversion will fail or strip it.
• Contain JavaScript or executable content. A document should display, not run code.
• Use audio, video, or LaunchActions. Anything that triggers an external process is gone.
• Use transparency or encodings the standard doesn't allow (the exact list depends on the conformance level — more on that below).

A useful mental model: a regular PDF is a document that might be portable. A PDF/A is a document that is guaranteed to be, because the spec strips out every way it could fail to be.

Conformance Levels, Briefly

You don't need to memorize the matrix, but you should recognize the labels you'll see in submission requirements.

• PDF/A-1 — the original, strictest version. Two flavors: PDF/A-1b ("basic") guarantees the document looks right; PDF/A-1a ("accessible") adds tagged structure so the text is machine-readable and accessible to screen readers.
• PDF/A-2 — relaxes a few constraints and adds modern features like JPEG2000 compression and the ability to embed other PDF/A files.
• PDF/A-3 — like A-2, but allows arbitrary attachments (e.g., the original spreadsheet or XML behind an invoice). This is what electronic-invoicing standards such as ZUGFeRD/Factur-X use.

For everyday "make this an archival PDF" needs, a level-b conformance (visual fidelity) is the common, safe target. If a system demands a specific level, it will say so explicitly — match exactly what they ask for.

Who Requires PDF/A — and Why

PDF/A shows up wherever a document has to be trusted years after it's filed:

• Courts and legal e-filing. Many electronic filing systems mandate PDF/A so the record is tamper-evident in the sense of being self-contained and viewer-independent.
• Government and tax agencies. Records-retention laws often run 7, 10, or 30+ years; archival format is part of compliance.
• Universities and research repositories. ETD (electronic thesis and dissertation) systems frequently require PDF/A so a 2026 dissertation still renders in 2056.
• Libraries, museums, and national archives. Digital preservation programs standardize on it.
• Regulated industries — pharma, finance, engineering — where audit trails and document integrity are legally enforced.

The common thread: these institutions can't control what software future readers will have. PDF/A is how they remove that variable.

How to Convert a PDF to PDF/A on pdf-edit.tech

You can convert directly in the browser with the [PDF to PDF/A converter](https://pdf-edit.tech/pdf-to-pdfa). Upload your file and it produces a PDF/A-conformant output: fonts get embedded, an output intent is set, and disallowed features are normalized.

A few practical things to get right before and after conversion:

1. Remove passwords first

Because PDF/A forbids encryption, a protected source file can't be archived as-is. If your PDF is password-protected and you can already open it, remove the password first with [unlock PDF](https://pdf-edit.tech/unlock-pdf), then convert. (To be clear, that tool removes a password you already know — it does not crack or recover a forgotten one.)

2. Flatten forms and layered content

Interactive form fields and overlapping layers can complicate conversion and aren't ideal for an archive that's meant to be a fixed snapshot. Running [flatten PDF](https://pdf-edit.tech/flatten-pdf) first bakes form field values into the page so what you see is permanently what's stored — no live fields to render differently later.

3. OCR scanned documents before archiving

Here's the catch that surprises people: a scanned PDF is just a picture of a page. Converting that image to PDF/A makes it a self-contained image — still no searchable, selectable text underneath. For most archives, that defeats half the point. The fix is to add a real text layer first with [OCR PDF](https://pdf-edit.tech/ocr-pdf) (powered by Tesseract), then convert the searchable result to PDF/A. Now the archive is both visually faithful and full-text searchable — which is what records systems and accessibility requirements actually want. (The PDF/A conversion itself runs on the same OCR-and-Ghostscript toolchain, so it's purpose-built for exactly this.)

A Concrete Workflow

Say you're submitting a 40-page scanned contract to a court that requires PDF/A:

1. Unlock it if it carries a password you can open it with.
2. OCR it so the text becomes selectable and searchable.
3. Flatten any signature fields or form annotations into the page.
4. Convert to PDF/A as the final step.

Do those in order and you hand over a file that's self-contained, searchable, password-free, and standards-compliant — exactly what the system is checking for.

The Practical Takeaway

PDF/A isn't a fancier PDF — it's a disciplined one. It trades flexibility (encryption, scripts, external links) for a single guarantee: the document will render the same way, on its own, decades from now. When an institution asks for it, they're not being fussy; they're protecting a record they're legally obligated to keep.

If you only remember one thing: convert to PDF/A last, after you've removed passwords, OCR'd any scans, and flattened any forms. Start at the [PDF to PDF/A converter](https://pdf-edit.tech/pdf-to-pdfa) when your file is ready.

---

pdf-edit.tech is a free PDF toolkit by [TakshaGroups](https://takshagroups.in). Files are processed over encrypted HTTPS in server memory and deleted immediately after your output is generated — nothing is stored.

More PDF guides