How PDF Compression Works (and Why Some PDFs Barely Shrink)

Updated: 29 August 2026

Compress a scanned contract and it might drop from 8 MB to under 1 MB. Compress a text-only invoice and it might barely move. That difference isn't a bug — it comes down to what's actually inside the file. This guide explains where PDF weight comes from, what a compressor can and can't do about it, and how to get a genuinely smaller file.

What actually makes a PDF big

A PDF is a container. Inside it, the usual contents from heaviest to lightest:

• Images and scans. By far the biggest factor. A page scanned at 300 DPI in colour is a large photograph, and a 20-page scan is twenty large photographs stapled together. A document photographed with a phone and saved as PDF behaves exactly the same way.
• Embedded fonts. To look identical on every device, a PDF carries copies of the fonts it uses. Most tools embed only the characters actually used, but fonts can still add hundreds of kilobytes.
• Text and drawing instructions. Actual text is tiny — a full page of it usually costs only a few kilobytes.
• Metadata and leftovers. Author details, editing history, embedded thumbnails — small, but free to remove.

Lossless vs lossy: the only two options

There are exactly two ways to make any file smaller. Lossless compression re-packs the same information more efficiently — nothing is discarded, and the original can be reconstructed bit for bit. PDFs already apply lossless compression to their text and drawing instructions internally. Lossy compression — what JPEG does to photographs — permanently discards fine detail the eye is unlikely to miss. It's the only technique that produces dramatic size reductions, and it only applies to images.

Why a text-only PDF barely shrinks

Because it's already compressed. The text streams inside a normal PDF are stored losslessly compressed from the moment the file is created, so there is very little slack left for any compressor to squeeze. If your 300 KB report only drops to 280 KB, the tool isn't failing — the file simply contains almost no image data, and image data is where the real savings live.

What the AllToolsHub compressor does

The Compress PDF tool finds the images embedded in your PDF and re-encodes them as JPEGs at a lower quality setting. Text, fonts and vector graphics pass through untouched, so the words on the page never change. The result depends entirely on how much image data the file contains: scans and photo-heavy documents shrink a lot, text-only documents shrink a little. Like every tool on this site, it runs in your browser — the file is never uploaded anywhere.

Five ways to get a smaller PDF

1. Compress the images — the tool above; the biggest win for scans and photos.
2. Scan smarter next time. For text documents, 150 DPI in greyscale is usually perfectly readable at a fraction of the size of 300 DPI colour.
3. Delete pages you don't need. In a scanned file, a blank page costs as much as any other.
4. Split the file when a size limit applies per attachment rather than per document.
5. Strip the metadata — a small saving, and worthwhile anyway when privacy is part of the goal.

When compression is the wrong tool

If a form says "maximum 200 KB" and your file is a 40-page colour scan, no honest compressor can get there without destroying readability. The realistic options are cutting the page count, re-scanning at a lower resolution, or checking whether the limit applies to each file rather than the whole set. Knowing when to stop compressing is part of getting a usable document out the other end.