Compressing a scanned PDF without making it unreadable

A scanned PDF is a stack of images with no text layer. That is why compressing one behaves differently from compressing a word-processed document: nearly all of the file size lives in the images, so the compression level decides whether the result is readable.

Which level to choose

• Screen — smallest. Text stays sharp, but photos and shading turn blocky. Good for a file you only need to read on screen, or for getting under an upload limit.
• Ebook — the sensible default. Typically 50–70% smaller with text still crisp and photos acceptable.
• Print — keeps the highest image resolution. Use it when the document will be printed or archived.

The trick that saves the most

Duplicate detection: scanners often store the same background texture repeatedly. Ghostscript's duplicate image detection removes the repeats, which is where a lot of the size in scanned sets actually lives. All three levels on this site use it.

What to check afterwards

1. Open the compressed file and zoom to 200% on a page of small text.
2. Check the page count is unchanged.
3. Check none of the pages came out blank — the one failure mode worth watching for on old or damaged scans.

If a page came out blank, run the file again at Print level. That almost always means the scan itself is unusual rather than the compression being wrong.

---

Ready to try it? Open the tool.