Finance asked you to hand over the contracts for archiving, and it turned out a regular PDF would not do: they need PDF/A. You open an invoice from five years ago, and instead of the original typeface you see a clumsy substitute. The machine that made the document had Calibri installed, the archive system does not, and the text has shifted. PDF/A exists to prevent exactly that kind of surprise. It is built so a file looks the same today and ten or fifteen years from now, on any computer and in any program.
The core idea behind PDF/A is self-sufficiency. A regular PDF can reference fonts from the operating system, pull in color profiles, hold scripts, or link to outside websites. All of that will break one day: the font gets removed, the site goes down, the plugin stops working. PDF/A bans those dependencies. Fonts are embedded straight into the file, colors are described through an embedded profile, and there is no active content. So when you convert, the file is not simply renamed. It is genuinely repacked, and it helps to know what is happening to it.
How to convert PDF to PDF/A, step by step
1. Open PDF to PDF/A and upload the source document, the one that should go into the archive unchanged. 2. Pick the conformance level. If no regulation dictates a specific one, go with PDF/A-2b. It is a sensible modern default. PDF/A-1b is only for strict intake requirements. 3. Make sure the document has nothing extra in it. Remove blank pages and draft sheets ahead of time, because editing the file after conversion is best avoided. 4. Run the conversion and wait for the finished PDF/A. The tool embeds fonts and normalizes colors on its own. 5. Open the result and page through the whole thing, especially pages with unusual fonts, stamps, and signatures. 6. Save the file under a clear name with a date so it is easy to find in the archive later.
What can go wrong
- **A font did not embed.** Sometimes the source contains a font with no embedding rights (a license restriction baked into the font file itself). The converter then swaps it for a close match. The text stays readable, but the lettering changes. Check those pages by eye and, if needed, re-save the source with fonts that allow embedding.
- **A scan with no text layer.** If you convert a scanned document, you get a PDF/A, but you cannot search it, because it is a set of images. For a searchable archive, recognize the text first and convert afterward.
- **Transparency and layers in PDF/A-1.** Level PDF/A-1 does not support transparency, so semi-transparent stamps or watermarks get flattened or produce artifacts. If your document has elements like that, use PDF/A-2.
- **The size jumped several times over.** Embedded fonts and profiles make the file heavier. That is expected, but if the growth is excessive, compress the document with Compress PDF before converting, not after.
- **A broken source will not convert.** A corrupted PDF throws an error. Fix it first with Repair PDF, then try the PDF/A conversion again.
- **Colors drifted.** If the document used a nonstandard color space, the shades can shift slightly during normalization. For plain text archives it is often simpler to convert the file to grayscale first with PDF to Grayscale.
What to check before handing it to the archive
- All fonts display correctly, with no substitutions or empty boxes in place of letters.
- Stamps, signatures, and logos are visible and have not turned into black blobs.
- Text search and selection work across the document (unless it is a scan).
- The file name carries a date and a clear identifier for the catalog.
- The PDF/A level matches what the receiving side asks for.
Where to go from here
Tidy up the document before you convert. Drop the extra sheets with Remove blank pages, and if an archive unit is assembled from several parts, combine them into one file with Merge PDF. When you need the opposite, breaking a large document into separate cases, Split PDF handles it. After that, convert the finished file to PDF/A and send it off to long-term storage.