Skip to main content
BrowserStay

Command Palette

Search for a command to run...

PDF

Why is my PDF so large?

Almost always the same reason: something scanned paper and stored a photograph of it at print resolution. The fix follows from the diagnosis.

Compress the PDF in the browser

A twenty-page text document should be a few hundred kilobytes. A twenty-page scan of the same document is routinely twenty megabytes, and sometimes eighty. Same paper, same words, forty times the size — because the two files are not the same kind of object.

The four things that make a PDF heavy

  • Scanned images. A scanner photographs the page. At 300 dpi, an A4 page is about 2,480 by 3,508 pixels; stored as a photograph, that is a megabyte or three before compression. A 600 dpi colour scan can be eight. Everything downstream inherits that.
  • Images stored losslessly inside a scan. Some scanning software writes the page as a lossless bitmap instead of JPEG. The result is crisp and enormous — often five to ten times what a JPEG of the same page would take.
  • Embedded fonts. A PDF that will look the same on every machine embeds the font subsets it uses. Text documents with a dozen fonts can add several hundred kilobytes each. Real, but rarely the main cost.
  • Incremental saves. Every “save” in some editors appends a new revision and leaves the old one in the file. Download a document that has been edited eleven times and you may be carrying all twelve versions, of which only the last is visible.

Diagnose before compressing

Open the file and zoom to 400% on ordinary text. If the letters dissolve into grey pixels, the page is a picture and no text-level trick will help — you need to resample the image. If the text stays razor sharp, the file is probably carrying vector text plus a few heavy images or duplicate resources.

The other quick test: does selecting text work? A scanned page has no text layer at all. That single check tells you which family of file you have.

What compression can and cannot do

Compression tools do three legitimate things. They downsample images to a sensible resolution for screen reading. They re-encode those images as JPEG at a chosen quality. They drop unused objects, duplicate resources and stale revisions. On a 40 MB scan, that routinely brings the file to a few megabytes with no visible difference on screen.

What no honest tool can do is make a 600 dpi colour scan into a 300 KB file without throwing away information. If the document has to stay print-quality, it stays large, and the right answer is a smaller page count or a delivery method that does not involve email attachments.

Doing it locally

Compress PDF runs in the tab: it reads the file, re-encodes what can be re-encoded, and gives you a download. Nothing is uploaded, which matters more for the second reason in that list above than the first — a heavy PDF is usually the one with scans of things you would not post on a wall.

The trade-off is memory. A 200 MB scan on a phone may not fit in the tab, and the tool will fail rather than quietly truncate your document. On a laptop it is unremarkable.

Prevention, next time

  • Scan at 200–300 dpi for reading copies, not 600.
  • Scan to JPEG, not to a lossless bitmap, when the page is text plus a photograph.
  • If the source is already digital, print to PDF from the application rather than scanning a printout.
  • After editing, use “Save as” occasionally — it rewrites the file instead of appending to it.