When a PDF Will Not Compress
What is taking up the space, and what actually works.
The Short Answer
Compression is not failing. There is simply nothing left for it to remove.
Almost all of a large PDF is images, and those images are already compressed. Compressing compressed data does not make it meaningfully smaller — it only makes it worse-looking.
What Is Actually In Your File
Open the file size in your head as a budget and it becomes obvious where it went:
| Content | Typical size | Compresses? |
|---|---|---|
| Text | Tiny | Already minimal |
| Vector graphics | Small | Barely |
| Embedded fonts | 0.1–2 MB | Only by subsetting |
| Photos and scans | Almost everything | Already JPEG |
If your PDF is 40 MB, roughly 39 MB of it is pictures. That is the only thing worth attacking.
Why Re-Compressing Does Nothing
JPEG works by discarding detail your eye is unlikely to miss. Once that pass has happened, the redundancy it targets is gone. A second pass finds very little to remove, so it discards more real detail for a very small size gain.
This is why a scan can look visibly worse after compression while barely shrinking. You paid the quality cost without getting the size benefit.
The Three Things That Genuinely Work
1. Reduce the image resolution. This is the big one. Scanners default to 300 or 600 DPI, which is far more than a screen needs. Halving the DPI removes three quarters of the pixel data. At 150 DPI most documents remain perfectly readable on screen and still print acceptably.
2. Convert colour scans to greyscale. A scan of black text on white paper stored in colour is carrying three channels to describe grey. Dropping to greyscale removes roughly two thirds of the data with no meaningful loss for a text document.
3. Go back to the source. If the PDF was exported from a document, presentation or design file, re-export it with lower image quality settings. Doing it at export time is always better than trying to undo it afterwards, because the original still has the full-quality images to work from.
When The Real Goal Is an Attachment Limit
Most people compressing a PDF are trying to get under an email limit. If the file will not shrink far enough, splitting it is the honest answer — two 8 MB emails beat an hour spent fighting a 40 MB scan.
The PDF Splitter will divide it by page range in your browser. For a stubborn image-heavy document, the other route is extracting the images, running them through the Image Compressor at a lower resolution, and rebuilding with Images to PDF.
Try compressing in your browser →Accepting the Floor
Every PDF has a size below which it cannot go without losing something you care about. A 200-page colour scan is not becoming a 2 MB file while remaining legible. Recognising you have hit that floor is more useful than trying a fifth compression tool — they are all doing the same arithmetic.
Frequently Asked Questions
Why does compressing my PDF barely change the file size?
Because the space is almost certainly taken by images that are already compressed. Photos and scans inside a PDF are usually stored as JPEG data, which has already had its redundancy removed. Running JPEG compression over JPEG data a second time gains a few percent at best while degrading quality.
What does the error file is already compressed mean?
It means the tool analysed the document and concluded there is nothing left it knows how to squeeze. It is a statement about the file, not a fault. A text-only PDF and a well-optimised scan will both produce it.
What actually reduces the size of a scanned PDF?
Lowering the image resolution. A page scanned at 600 DPI holds four times the pixel data of the same page at 300 DPI, and 150 DPI is usually still readable on screen. Converting colour scans to greyscale is the other big win. Both throw away real data, which is why generic compression will not do it for you automatically.
Why is a PDF full of text already tiny?
Text is stored as characters plus a font reference, not as a picture of words. A hundred pages of text is often under a megabyte before any compression. If your text document is large, something else is inside it — usually an embedded font set or a logo image repeated on every page.
If nothing works, what are my options?
Split the document and send it in parts, which is often the real goal when someone is fighting an email attachment limit. Or re-export from the original source with lighter settings, which almost always beats compressing the finished PDF.