How to Compress a PDF Without Losing Quality

A 40-page PDF that started life as a Word document is usually a few hundred kilobytes. The same 40 pages, scanned on an office photocopier, can easily land at 80 MB. Both are "one PDF," and that gap is the whole story of why compression works dramatically on some files and does almost nothing on others.

What actually takes up the space

A PDF is a container, not a single format. Inside it you'll typically find some mix of:

• Text and fonts — usually tiny. A page of body text, even embedded with its own font, rarely adds more than a few kilobytes.
• Vector graphics — logos, diagrams, chart lines. Also small, since they're stored as drawing instructions rather than pixels.
• Raster images — scanned pages, photos, screenshots. This is almost always where the size comes from, because a single full-page color scan at print resolution can be several megabytes on its own, and a 40-page scanned document is 40 of those.

That last category is why "compress my PDF" and "compress the images in my PDF" are, in practice, the same task. A compressor that only strips a bit of unused metadata might shave off a few percent. One that actually re-encodes the embedded images is doing the real work.

Why the result varies so much file to file

This is the part that trips people up: run the same compressor on two different PDFs and get wildly different results. The reason is almost always what's inside them, not the tool.

A document exported straight from a word processor or design tool has no raster images to shrink, so a compressor genuinely has little to do. A document scanned from paper is the opposite case: it's effectively a stack of photographs with text-shaped content, and those photographs compress well because scans usually carry far more resolution than needed for on-screen reading.

Compress a PDF works exactly on this principle: it recompresses the JPEG-encoded images inside your file and strips unused objects, then shows you the size difference before you download anything, so you're never guessing whether it actually helped.

Picking a compression target

Most compressors offer some version of light, balanced, or aggressive, and the tradeoff is exactly what it sounds like: a bigger drop in file size costs a bit more image sharpness. A few rules of thumb:

• Light is right for anything that might get printed, or where fine text inside an image (a scanned form, a signature) needs to stay crisp.
• Balanced is the sensible default for emailing a document, uploading it to a portal with a size cap, or general archiving.
• Aggressive makes sense for a bulk archive of scans you'll skim on screen but never zoom into.

Try the target, look at the preview, and back off a level if text inside an image starts to look soft. Since the original file is never modified, there's no risk in trying a stronger setting first and dialing it back.

When compression isn't actually the right tool

Two situations that look like a "compress this PDF" problem usually aren't:

You only need part of the document. If a 200-page report only matters for its first 10 pages, splitting out just those pages will shrink the file far more than compression ever could, because you're removing content instead of re-encoding it.

You're about to combine several files anyway. If the end goal is one document assembled from several PDFs, merge them first, then compress the result once. Compressing each piece separately and merging afterward can leave you re-encoding images twice, with no size benefit over doing it once at the end.

The short version

Most of a PDF's file size lives in its embedded images, not its text. A compressor's job is to re-encode those images more efficiently, which is why the same tool can do almost nothing to a text-heavy document and cut a scanned one by 80% or more. Start with a balanced target, check the size estimate before downloading, and reach for split or merge instead when the actual problem is "this file has content I don't need" rather than "these images are bigger than they need to be."