OmniDocPDF/A-1a vs PDF/A-3: Choosing a PDF/A Version
Learn how PDF/A-1a, PDF/A-3 and XML-embedded PDF/A differ, how to pick the right conformance level, and how to check the result.
If you need to convert a PDF to PDF/A-3, you have probably hit a requirement from an archive, a government portal or an e-invoicing system. Those systems often name a specific conformance level, and the names look similar enough to cause mistakes. PDF/A-1a, PDF/A-3 and "PDF/A with embedded XML" are not three equal options. They solve different problems, and choosing the wrong one can get a file rejected later.
This guide explains what each level requires, when the differences matter, and how to check the result. It also covers a limit you should know up front: OmniDoc's PDF editor does not list PDF/A export among its features. You will read below how to handle that gap. The goal is to help you pick the version your receiver expects and confirm that your file meets it.
Which PDF/A version should you pick? The short answer
Choose the version that matches what your receiver asks for. If they name a level, use it. If they don't, decide based on whether you need tagged structure or embedded source files.
| Your situation | What to do | Why |
|---|---|---|
| The receiver asks for PDF/A-1a or accessible archiving | Convert to PDF/A-1a and confirm the file is tagged | 1a requires logical structure, reading order and Unicode text |
| You only need a long-term visual copy of a report | Use PDF/A-1b or PDF/A-2b, which are simpler | You don't need tagging or embedded files |
| You must attach XML or another source file inside the PDF | Use PDF/A-3 and attach the XML as an associated file | PDF/A-3 allows any embedded format; 1a and 2 do not |
| An e-invoice or similar system requires PDF/A-3 with XML | Follow that system's profile for the attached file | The receiver checks the specific relationship and file name rules |
| You are not sure what the receiver wants | Ask for the exact conformance level in writing | Guessing is a common cause of rejected files |
What PDF/A-1a, PDF/A-3 and XML-embedded PDF/A mean
PDF/A is an ISO family of standards for long-term preservation. Each part and level sets rules for what the file may contain. The letter after the number describes how strict the requirements are.
PDF/A-1a
The "a" stands for accessible. PDF/A-1a is built on PDF 1.4 and requires a tagged PDF, which means the document has a logical structure: headings, paragraphs, lists, tables and figure descriptions. Text must map to Unicode, so screen readers and search tools can read it. Fonts must be embedded, and transparency is restricted.
PDF/A-1a is strict. A scanned page with no text layer will not pass, because nothing is tagged. Use it when the receiver needs accessibility, not just visual fidelity.
PDF/A-3
PDF/A-3 is built on PDF 1.7. It keeps the same visual preservation rules as the earlier versions, and it adds one major change: you may embed files of any format inside the PDF. That is the feature that makes PDF/A-3 useful for XML.
PDF/A-3 has its own accessible variant, PDF/A-3a, which adds tagging requirements. Plain PDF/A-3 (often written PDF/A-3b) does not require tags. If the receiver names "PDF/A-3" without a letter, ask whether they mean 3a or 3b.
XML-embedded PDF/A
"PDF/A with embedded XML" is not a separate level. It is a PDF/A-3 file that also contains an XML file as an attachment. The visible pages show the human-readable document, and the XML carries structured data that software can read. Electronic invoice standards commonly use this pattern.
Two rules matter here:
- The XML must be declared as an associated file, with a relationship type that tells the reader what it is (for example, source data or alternative representation).
- The attachment must follow the naming and metadata rules of the target profile. A file that is simply attached without the right declaration may still open, but it can fail the receiver's check.
Common mix-ups
- "PDF/A-3 is newer, so it is better." Newer does not mean suited to your case. PDF/A-1a is the right choice for many accessibility requirements, and PDF/A-3 is only needed for embedded files.
- "Any PDF with an XML file attached is PDF/A-3." Attachment alone does not make a file PDF/A-3. The base file must conform to PDF/A-3 throughout.
- "Renaming the file or changing the extension converts it." It does not. Conformance depends on the file's internal structure.
- "Tagged means accessible." Tagging is required for 1a, but accessibility also depends on good alt text, reading order and language settings.
📊 At a glance
| Version | Based on | Embedded files | Tagging required | Typical use |
|---|---|---|---|---|
| PDF/A-1a | PDF 1.4 | Not allowed | Yes | Accessible long-term archives |
| PDF/A-1b | PDF 1.4 | Not allowed | No | Visual long-term copies |
| PDF/A-3b | PDF 1.7 | Any format | No | Archives that carry source files |
| PDF/A-3a | PDF 1.7 | Any format | Yes | Accessible archives with source files |
Use this table to narrow your options, then confirm against the receiver's written profile. Some systems add their own rules on top of the ISO standard.
Putting it to use
Follow these steps to choose and check a PDF/A file.
- Get the exact requirement. Write down the conformance level (for example, PDF/A-3b), whether tagging is required and whether you must attach any XML.
- Check your source PDF. Open it and select a line of text. If you can select and highlight the words, the file has a text layer. If you can't, it is probably a scanned image. For a scanned PDF, use the OCR feature in the OmniDoc desktop client (download it at https://download.omnidoc.top) to add searchable text before any PDF/A step. Check the OCR result against the original, because recognition errors are common in poor scans.
- Keep the original. Save an untouched copy before any conversion. Conversion can alter fonts, transparency and embedded images.
- Convert with a PDF/A-capable tool. OmniDoc's facts do not list PDF/A export, so use a dedicated PDF/A conversion step or a desktop office suite that offers a PDF/A export option. Pick the exact level your receiver named.
- Attach the XML if required. For PDF/A-3, attach the file with the relationship and name the target profile expects. Then confirm the attachment appears in the file's attachment list.
- Validate the output. Run the result through a PDF/A validator that reports the conformance level. A passing check at the right level is the only reliable sign the file is compliant.
- Review the content. Open the converted file and compare each page with the original. Check tables, formulas, headers, footers and any text that OCR produced.
🎯 Which approach fits you?
- Accessibility or screen-reader requirements: PDF/A-1a or PDF/A-3a, with tagging verified.
- Archive copy with no special requirements: PDF/A-1b or PDF/A-2b.
- Invoice or data package with an XML file: PDF/A-3, with the attachment declared as the receiver specifies.
- Scanned source with no text layer: run OCR first, then convert, and review the text before you archive.
❓ FAQ
1. Can I convert a PDF to PDF/A-3 without an XML file?
Yes. A plain PDF/A-3 file does not need any attachments. Choose PDF/A-3 without an embedded XML file only if your receiver accepts that level, because many receivers pick the level for a reason.
2. Does OmniDoc produce PDF/A files?
OmniDoc's listed features do not include PDF/A export. Its PDF editor can edit text, sign, fill forms, redact, manage pages, run OCR, compare files and export to Word, Excel and PowerPoint. For a PDF/A file, use a conversion tool that lists PDF/A output and then validate the result.
3. Is PDF/A-1a the same as an accessible PDF?
Not exactly. PDF/A-1a requires tagging, Unicode text and other rules that support accessibility. Accessibility also depends on alt text for images, correct reading order and document language, so check those separately.
4. How do I know whether my scanned PDF can become PDF/A-1a?
A scanned PDF has no text layer, so it cannot be tagged as is. Add OCR first, then check that the recognized text matches the page. Recognition can make mistakes, so review important content before you convert.
📝 Summary
PDF/A-1a suits accessible archives, PDF/A-3 suits files that carry embedded sources such as XML, and the right choice always follows the receiver's stated level. Convert the file with a PDF/A-capable tool, keep your original, and validate the result before you send or archive it. Use this approach to convert PDF to PDF/A-3 or any other level with confidence.
