You run a PDF through a compressor, wait, and get back a file that's... almost exactly the same size. Or worse, slightly bigger. It feels like the tool is broken. It usually isn't — the file is simply telling you that whatever is making it large isn't the thing compression removes.
This guide covers why PDFs resist compression, how to work out what's actually taking up the space in yours, and what to do when compressing harder isn't the answer.
What compression actually does
This is the key to the whole problem. A PDF compressor mainly works on images — it re-encodes them at a lower quality or resolution. That's where the bytes are in most large PDFs, so most of the time it works brilliantly.
But it means compression can only take back what images are holding. If your file is big for a different reason, squeezing the images harder achieves nothing — there's nothing there to squeeze.
The five reasons your PDF won't shrink
1. It's already compressed. The most common cause. If the file was exported by software that already optimised it — or you compressed it once already — a second pass has very little left to take. Compressing twice does not halve the size; the second run is mostly wasted.
2. It's mostly text, not images. A 400-page text document might be large simply because it's 400 pages of text and embedded fonts. Text is already extremely compact — there's no fat to trim. The size is the content.
3. Embedded fonts. A document using many font families (each embedded in full) carries real weight that image compression won't touch.
4. The images are already low-resolution. If the pictures are already small and heavily compressed, asking for more just degrades them without meaningfully shrinking the file.
5. It's an unusually long document. Page count multiplies everything. Fifty pages of anything is heavier than five, and no compression setting changes that arithmetic.
Diagnose it: text or images?
Before you do anything else, work out which kind of PDF you have — it completely changes the fix:
- Can you select the text with your cursor? Then it's a text-based (born-digital) PDF. Compression will do little. Your wins are in page count and fonts.
- Is the text an image you can't select? Then it's a scan, and compression should work very well. If it isn't, see our scanned PDF compression guide — the usual fix is to grayscale it first.
That single test tells you whether to keep compressing or change tactics.
What to do instead
When compression has hit its floor, these are the levers that still work:
- Remove pages you don't need. The most reliable size reduction there is — half the pages, roughly half the file. Extract just the pages you need or delete the rest.
- Split it into parts. If it's genuinely a big document, Split PDF turns one oversized file into two sendable ones.
- Drop the colour. For a scan or a colour-heavy document that doesn't need colour, grayscale it, then compress. This is often the single biggest win on a stubborn file.
- Don't bother zipping it. A PDF is already a compressed container, so a ZIP typically saves a few percent at best — and adds friction for whoever receives it.
Common questions
Does compressing a PDF twice make it smaller?
Barely. The first pass takes the available savings; a second pass finds little left and mostly just degrades image quality further. If one pass didn't get you there, change tactics rather than repeating it.
My compressed PDF came out bigger than the original.
That can happen when the original was already well-optimised — the re-encoding process adds its own overhead without finding savings. It's a sign the file was already about as small as it usefully gets. Keep the original.
Why does the "Extreme" preset barely help?
Because Extreme still only pushes harder on the images. If images aren't what's making your file big (see the five reasons above), Extreme has nothing extra to work on — it just risks quality for no gain.
My PDF is 400 pages of text and it's still huge.
That's expected, and compression won't fix it. Your options are to split it or extract the sections people actually need. The size is the content.
Could the file be corrupted?
Occasionally a damaged PDF behaves strangely — refusing to process or producing odd output. If a file misbehaves across several tools, try repairing it first, then compress.
Wrap-up
If your PDF won't compress, it's almost always because the thing making it big isn't an image:
- Test it — if you can select the text, it's a text PDF and compression has little to take.
- Take the real wins — remove pages, split the file, or drop colour on a scan.
- Don't just compress again — a second pass costs quality and gains almost nothing.
For how compression works and what each preset actually does, see our full compression guide.