"Compress this PDF" sounds simple, but knowing what compression actually changes is the difference between a great result and a disappointing one. Here is what is really happening under the hood, in plain terms.
Most of the weight is images
In the typical oversized PDF, the text is a tiny fraction of the file. The bulk is images, scans, photos, screenshots, stored at far higher resolution than the page ever displays. Compression works primarily by re-encoding those images: lowering their resolution to what the page actually needs and applying efficient image compression. That is where the dramatic size drops come from.
Lossless vs lossy
- Lossless optimisation tidies the file's internal structure and removes redundant data without changing how anything looks. The savings are modest but completely safe.
- Lossy compression reduces image quality slightly to save much more space. Done sensibly, the loss is invisible; pushed too far, images soften and text-as-image gets fuzzy.
What compression does not shrink
Actual text is already tiny, so compressing a text-only document yields little, there is simply not much to remove. If a mostly-text file is huge, the cause is usually embedded fonts or hidden leftover data, not the text itself. This is also why the same setting can shrink one file by 80% and barely touch another: it depends entirely on what is inside.
The rasterisation trade-off
Some compressors, in strong mode, rebuild each page as a single flattened image. This maximises savings and is perfect for scans, but it converts real, selectable text into a picture. For documents people need to search or copy from, choose a gentler level that preserves the text layer.
The takeaway
Compression is mostly image management. Once you see it that way, the results stop being surprising: image-heavy files shrink a lot, text-only files barely move, and the right setting is simply the gentlest one that gets you under your limit.
