Why some PDFs won't compress (and what to do)

You run a PDF through a compressor, wait for the progress bar, and the result is the same size. Or worse, it's bigger. This is frustrating and common, and it almost always has a specific, knowable cause. A PDF doesn't refuse to compress out of spite. It refuses because there's nothing left to squeeze, or because the thing taking up space isn't the kind of data compression helps. Here's how to tell which it is.

Reason 1: the file is already mostly text

Text in a PDF is tiny. A 50-page report with no images might be a few hundred kilobytes, and that's already close to the floor. The bytes are characters and font references, not pixels, and there's very little redundancy for a compressor to remove. If your file is small to begin with and full of selectable text, there's no meaningful compression to be had. That's not a failure. It's a file that was already efficient.

Reason 2: the images are already compressed

The biggest savings in most PDFs come from images. But if those images are already stored as well-compressed JPEGs at a reasonable resolution, re-compressing them does almost nothing. Worse, running an already-JPEG image through another lossy pass can add artifacts while saving only a few percent. A good compressor recognizes this and leaves them alone, which is why you sometimes see "0% reduction" on a file that looks image-heavy. The images were optimized before you ever touched the file.

Reason 3: the content is vector, not raster

Charts, diagrams, logos, and CAD exports are often vector graphics: instructions to draw lines and shapes, not grids of pixels. Vectors are already compact and scale infinitely without quality loss. Image downsampling does nothing to them because there's no resolution to lower. A 10 MB PDF of complex technical drawings might be 10 MB of pure vector math, and that's simply how much the drawing costs to describe.

Reason 4: it's a scan saved the wrong way

Here's the case where a file looks compressible but fights you. A scanned document is a stack of images. If the scanner saved each page as a lossless PNG or a maximum-quality JPEG at 600 DPI, the file is huge. It can compress, a lot, but only if you downsample the resolution and re-encode. If your compressor is set to "lossless" or conservative defaults, it won't touch the resolution and you'll see no change. The fix is to allow downsampling, not to compress harder on the same settings.

Reason 5: the file is bloated with hidden data

Some PDFs carry weight you can't see: embedded full fonts, document revision history (every edit ever saved into the file), embedded thumbnails, attached files, or leftover objects from the software that made it. Editing tools sometimes append changes instead of rewriting the file, so a document that's been opened and saved twenty times can be far larger than its content justifies. Here the fix isn't image compression at all. It's rewriting the PDF cleanly, which drops the orphaned data and flattens the revision history.

Reason 6: it's encrypted or password-protected

A locked PDF can't be modified until it's unlocked, and compression is a modification. Most tools will either refuse outright or ask for the password first. If your file won't compress and you remember setting a password, that's the wall you've hit. Unlock it, compress, and re-protect if you still need the password.

How to diagnose your specific file

1. Can you select and copy the text? If yes, the file is text-based and probably already small. Don't expect much.
2. Is it a scan (text is part of the image, not selectable)? If yes, enable downsampling and set a screen-appropriate DPI.
3. Are the visuals crisp at any zoom level? That's vector. It won't shrink through image settings.
4. Has the file been edited and saved many times? Try a "rewrite" or "optimize" pass to clear hidden bloat.
5. Does the tool ask for a password? Unlock first.

When the file genuinely can't get smaller

Sometimes the honest answer is that the file is already as small as it can be without throwing away content you need. A dense, well-optimized document has no slack. If you need it smaller anyway, the only options left involve real trade-offs: lower the image quality past the point of invisibility, drop the resolution below what the screen needs, or split the document so each piece is lighter. None of those are "free" the way removing waste is.

Doing the diagnosis privately

You shouldn't have to upload a confidential file just to find out whether it'll compress. PDFShore runs in the browser, so you can drop the file, attempt compression, and see the result without the document ever leaving your device. If it doesn't shrink, you've learned that locally, with no server copy sitting somewhere waiting on a deletion timer. The diagnosis costs you nothing but a few seconds.