Why is my PDF so large?

A two-page document should not be eight megabytes. When a PDF balloons far past what its visible content seems to justify, one of a small number of things is responsible, and knowing which one changes what you should do about it.

Suspect #1: images at camera resolution

This is the cause in the overwhelming majority of oversized PDFs. A modern phone photo is 3–8 MB and about 4,000 pixels wide; a scanned page at 600 DPI is similar. When a photo or scan is placed in a document, most software embeds the original file untouched, even if it's displayed at a width of 600 pixels. One headshot on a résumé, one pasted screenshot, and your 60 KB document is suddenly 4 MB.

The fix is downsampling: re-encoding each embedded image at the resolution the page uses. Done right, it's visually indistinguishable on screen and in print, and it's how a 4 MB résumé becomes 180 KB. This is what a good PDF compressor does, and it's also why compression barely helps text-only files: there are no images to re-encode.

Suspect #2: why scanned documents are so large

Scanners make two expensive decisions on your behalf. They scan in full color even for black-and-white text, and they scan at print-shop resolutions for documents that will only ever be read on screens. A 300 DPI color scan of a text page can be ten times the size of what the content needs. If you control the scanner, scanning at 150–200 DPI in grayscale prevents the problem; if you're stuck with the file, compression with a size target recovers most of it.

Suspect #3: embedded fonts (usually fine, occasionally not)

PDFs embed the fonts they use so the document renders identically everywhere. That normally costs a few hundred kilobytes at most. The pathological case is documents with dozens of fonts (often from merged files or CJK typefaces embedded whole), which can add several megabytes. You can see what a PDF carries with the metadata inspector; font weight is a rebuild-the-source problem more than a compression problem.

Suspect #4: dead weight from editing history

Some editors append changes to the end of the file rather than rewriting it (it's how PDF's "incremental save" works), so a document edited many times carries every previous version inside it. Deleted pages, replaced images, and old form data are all still in the file. Rebuilding the document discards all of it. That's the reason running a file through a compressor sometimes shrinks it dramatically before any image work happens: the rebuild alone dropped the dead weight.

What doesn't help

Zipping a PDF rarely saves more than a few percent, because the images inside are already compressed and ZIP can't improve them. Printing to PDF again sometimes helps (it rebuilds the file) but often rasterizes text and makes things worse. And "Save as reduced size" in various tools is just compression with the settings hidden.

Rules of thumb

• A text-only PDF over 1 MB means editing dead weight or font bloat; a rebuild fixes it.
• For any PDF with photos or scans, suspect image resolution; downsampling recovers 80–95%.
• If it needs to hit an exact limit, use a target-size mode (50 KB, 100 KB, 200 KB, 500 KB, 1 MB) rather than guessing with a quality slider.