Does my file get uploaded?

No. The conversion runs entirely in your browser, on your own device, and the file never reaches a server. This matters more here than for most tools: the documents people convert to PDF/A are theses, court filings, patient records and grant submissions. We measured one competitor directly — iLovePDF — and its conversion uploads the document to its servers.

Which level should I pick?

PDF/A-2b unless you have been told otherwise — it is the usual requirement and allows transparency and modern compression. Pick 1b if an archive asks for it by name; it is the oldest and strictest, and some repositories still specify it. Pick 3b only if your document has files attached to it, because that is the only level that permits attachments. The tool checks your document first and greys out any level it cannot actually reach, with the reason.

Why does it say my fonts will be replaced?

PDF/A requires every font to be stored inside the file. Many PDFs — especially older ones — only name a font like Helvetica or Times and rely on your reader to supply it, so there is nothing to store. We embed a metrically matched open-licence substitute instead: Liberation Sans for Helvetica and Arial, Liberation Serif for Times, Liberation Mono for Courier. We name each replacement on screen before you convert, which is the part other converters skip.

No. We measured this: text positions shift by 0.0 pt, because the substitutes have the same letter widths as the fonts they replace, so every line breaks exactly where it does now. The letterforms themselves are drawn slightly differently — an "a" has a marginally different shape — and the ink on a line can extend up to about 3 pt further. That is an appearance difference, not a layout one. If a document uses a font we have no metric match for, we say so and tell you the lines may re-break.

Can you tell me my file is valid PDF/A?

Not honestly, and we would rather say so. Validating PDF/A properly means running a validator like veraPDF, which is a Java program — nothing equivalent runs in a browser, so your individual file cannot be independently checked on your device. What we can tell you is three things, all of them measured: what level the output declares; what we confirmed by reading the result back (that every font is embedded, and — for the fonts whose program and encoding we can read — that the character widths they declare match the font actually embedded, which is the requirement most converters get wrong); and that this converter is run against the veraPDF reference validator on every build, over a set of test documents, at every level we offer. We show you how many fonts were width-checked rather than implying it was all of them.

Why was my file refused?

Because a file that declares PDF/A and fails validation is worse than no file — you would only find out when an archive rejected it. We refuse when a document uses CMYK colour (the sRGB profile we attach does not cover it), when it uses a symbol font like Symbol or ZapfDingbats that no text substitute can draw, when it uses a large-script font such as Chinese or Japanese that is not embedded, or when it is password-protected. Each refusal names the specific reason and what would fix it.

Why did the file get bigger?

Because the fonts are now inside it. A PDF that merely names Helvetica carries no font data at all; a PDF/A file must carry the whole typeface. We embed each face once, no matter how many times the document refers to it, and cut each one down to just the characters it can address — but a few tens of kilobytes per typeface is the unavoidable cost of a file that will still render correctly in fifty years.

My PDF is a scan. Will this make it searchable?

No, and it will tell you so. A scanned PDF is a picture of a page with no text in it. Converting it to PDF/A is legal and useful for archiving, but it adds no searchable text — so the tool says plainly that the pages are images. Run OCR PDF first if you need the text.

PDF Tools

Convert PDF to PDF/A

Convert a PDF to PDF/A-1b, 2b or 3b entirely in your browser: every font embedded, an sRGB profile attached, and every font we have to replace named before you convert. The file is never uploaded. No account, no email.

Frequently asked questions

Does my file get uploaded?: No. The conversion runs entirely in your browser, on your own device, and the file never reaches a server. This matters more here than for most tools: the documents people convert to PDF/A are theses, court filings, patient records and grant submissions. We measured one competitor directly — iLovePDF — and its conversion uploads the document to its servers.
Which level should I pick?: PDF/A-2b unless you have been told otherwise — it is the usual requirement and allows transparency and modern compression. Pick 1b if an archive asks for it by name; it is the oldest and strictest, and some repositories still specify it. Pick 3b only if your document has files attached to it, because that is the only level that permits attachments. The tool checks your document first and greys out any level it cannot actually reach, with the reason.
Why does it say my fonts will be replaced?: PDF/A requires every font to be stored inside the file. Many PDFs — especially older ones — only name a font like Helvetica or Times and rely on your reader to supply it, so there is nothing to store. We embed a metrically matched open-licence substitute instead: Liberation Sans for Helvetica and Arial, Liberation Serif for Times, Liberation Mono for Courier. We name each replacement on screen before you convert, which is the part other converters skip.
Will my text move?: No. We measured this: text positions shift by 0.0 pt, because the substitutes have the same letter widths as the fonts they replace, so every line breaks exactly where it does now. The letterforms themselves are drawn slightly differently — an "a" has a marginally different shape — and the ink on a line can extend up to about 3 pt further. That is an appearance difference, not a layout one. If a document uses a font we have no metric match for, we say so and tell you the lines may re-break.
Can you tell me my file is valid PDF/A?: Not honestly, and we would rather say so. Validating PDF/A properly means running a validator like veraPDF, which is a Java program — nothing equivalent runs in a browser, so your individual file cannot be independently checked on your device. What we can tell you is three things, all of them measured: what level the output declares; what we confirmed by reading the result back (that every font is embedded, and — for the fonts whose program and encoding we can read — that the character widths they declare match the font actually embedded, which is the requirement most converters get wrong); and that this converter is run against the veraPDF reference validator on every build, over a set of test documents, at every level we offer. We show you how many fonts were width-checked rather than implying it was all of them.
Why was my file refused?: Because a file that declares PDF/A and fails validation is worse than no file — you would only find out when an archive rejected it. We refuse when a document uses CMYK colour (the sRGB profile we attach does not cover it), when it uses a symbol font like Symbol or ZapfDingbats that no text substitute can draw, when it uses a large-script font such as Chinese or Japanese that is not embedded, or when it is password-protected. Each refusal names the specific reason and what would fix it.
Why did the file get bigger?: Because the fonts are now inside it. A PDF that merely names Helvetica carries no font data at all; a PDF/A file must carry the whole typeface. We embed each face once, no matter how many times the document refers to it, and cut each one down to just the characters it can address — but a few tens of kilobytes per typeface is the unavoidable cost of a file that will still render correctly in fifty years.
My PDF is a scan. Will this make it searchable?: No, and it will tell you so. A scanned PDF is a picture of a page with no text in it. Converting it to PDF/A is legal and useful for archiving, but it adds no searchable text — so the tool says plainly that the pages are images. Run OCR PDF first if you need the text.

Guide

What is PDF/A and how to convert your PDF to it

PDF/A is the ISO standard for archiving PDFs in a self-contained format. This guide explains what it is, when you need it, and how to convert your PDF to PDF/A without uploading to a third-party service.

Read the guide →

Frequently asked questions

What is PDF/A and how to convert your PDF to it

Related tools