All articles
File Compression6 min read

PDF Compression Explained: What Actually Shrinks a File

Almost every oversized PDF is oversized for the same reason: it stores far more image data than any screen will ever display. Understanding where those megabytes hide makes the difference between a document that shrinks by 70% and one that barely moves.

Where the megabytes actually live

A PDF is a container. Inside it sit page descriptions, embedded fonts, vector drawings and — usually the culprit — bitmap images. Text and vectors are mathematical instructions and cost almost nothing: a 200-page contract can easily be under 1 MB. A single page scanned at 600 DPI in a lossless format can be 8 MB on its own.

That is why file size correlates with how a document was produced, not how long it is. Scanner apps, print-ready exports and screenshots pasted into Word are the three most common sources of bloat.

Resolution: the single biggest lever

Resolution is measured in dots per inch. Doubling DPI quadruples the pixel count, so the relationship between resolution and file size is steep. The practical question is not "what is the highest quality?" but "what will this document be used for?"

  • About 150 DPI is enough for on-screen reading and web portals.
  • About 200 DPI is a comfortable middle ground that still prints acceptably on an office printer.
  • About 250–300 DPI is what you want for professional printing or documents containing fine legal small print.

Lossy versus lossless encoding

PNG-style lossless compression stores every pixel exactly and works well for line art and screenshots of text. JPEG-style lossy compression discards detail your eye is least sensitive to and is dramatically more efficient for photographs and scans.

Most real-world compression gains come from re-encoding page images as JPEG at a sensible quality setting — typically somewhere between 45% and 85%, depending on how much detail you can afford to lose.

A workflow that consistently works

Order of operations matters more than people expect. Doing the same steps in a different sequence can leave you with a file twice the size.

  • Scan or export once at the highest quality and keep that as your archive copy.
  • Do all editing — merging, page removal, rotation — before compressing.
  • Compress at the end, once, choosing the level that matches the delivery channel.
  • Check page one and the page with the smallest text before you send the file.

When compression will not help

If a PDF is already 400 KB and mostly text, there is nothing meaningful left to remove. Likewise, running a file through a compressor repeatedly compounds JPEG artefacts while saving very little. If a document refuses to shrink, the better fix is usually to split it and send only the pages the recipient needs.

Try it yourself — everything runs privately in your browser.

Compress a PDF now