What does that producer string in your PDF mean?

What the producer field is

Two fields describe a PDF's origin. /Creator names the application the content came from — Word, a layout tool, a reporting system. /Producer names the software that wrote the PDF bytes. When a document is exported once, they often disagree in a perfectly ordinary way: created in Word, produced by a print driver.

Both are ordinary strings in the document's Info dictionary or XMP packet. Any tool can write them, edit them or remove them entirely, in seconds. So a producer string is evidence, never proof — worth reading, never worth deciding on alone.

The distinction that matters: did it create the file, or consume one?

The useful question is not whether a tool is "consumer" or "professional" software. It is whether the file's existence begins with that tool, or whether the tool took a PDF that already existed and wrote a new one.

• Authoring, print drivers, design tools, browsers — Word, Canva, Illustrator, Print to PDF, Skia/PDF, Google Docs. The document starts here. Naming one says nothing about whether anything changed afterwards, because there was no "afterwards" to change.
• PDF editors and re-distillers — iLovePDF, Sejda, PDF-XChange, Acrobat's interactive editor, Ghostscript. These take an existing PDF as input. That is a claim about the file's history — and still not proof of anything wrong, because merging, compressing, signing and filling in a form are things honest people do to honest PDFs every day.

Getting this backwards is a mistake we made and published: our producer-fingerprint family used to treat any consumer tool as a signal, and on a random sample of the Brazilian web that meant flagging 206 Microsoft Word documents. The write-up is here, numbers included.

How to read the last column

The rightmost column is our own false-positive rate, not a fraud rate. Every document in the sample was published by the organisation that made it, so each is presumed genuine, and every score above low counts against us. A producer with a high number is a toolchain whose ordinary output our engine finds unusual — useful to know when you are reading a report on a file that came out of it.

Sample: 985 PDFs from the Library of Congress "1000 .gov" random set plus 743 fetched from a Common Crawl sample of the Brazilian web, scored on engine 1.22.0. Producers appearing fewer than eight times are not listed — 432 distinct strings appeared in total, 252 of them exactly once, which is itself the most useful fact on this page.

The producer strings, and what each one is

Filter

Reading a file of your own

The metadata viewer shows every field a PDF carries, including both origin strings, with nothing stored. The revision history answers the question a producer string cannot: how many times the file was saved, and which save changed the page. And the metadata forensics guide puts both in context.