VSThiran

Why won't my PDF compress below 100KB?

Quick answer

A PDF's size floor is set by what is actually inside it. A page that is a full scanned photograph has real detail that has to be thrown away to shrink further, and there is a point past which doing that makes the page illegible - a good compressor stops there rather than forcing an arbitrary number. A PDF made of real text and vector graphics, by contrast, is already close to its floor and was never going to shrink much regardless of the target.

  1. 1Check whether the PDF is a scan (a photograph of a page) or real text - this alone decides how compressible it is.
  2. 2For a scan: try a lower scan resolution to begin with, or accept the smallest size a compressor can honestly produce.
  3. 3For text: the size is usually fonts/structure, not images - there is little a compressor can safely remove.
  4. 4If a portal genuinely requires an unreachable size, split the document into smaller pieces instead.

Free, no sign-up, and your file is read on your own device rather than uploaded.

A PDF compressor works almost entirely by re-encoding the photographs inside a document at a lower quality. That means how far a PDF can shrink is set by how much of it is actually photographic - and a scanned page, even of plain text, is a photograph as far as a compressor is concerned.

A tool that promises to hit any target size no matter what is either lying about the result (rasterising everything into an unreadable blur) or quietly failing. The honest version tells you plainly when a target genuinely cannot be reached, and gives you the smallest version it could responsibly produce instead.

Step by step

  1. Work out what kind of PDF you actually have

    A scanned document (from a phone camera or scanner) behaves completely differently to one exported from Word or a website - the first is images, the second is real text.

  2. For a scan, expect a real floor

    A one-page scan can often reach 100-200 KB. A five-page scan usually cannot, without the text becoming hard to read.

    Compress PDF to 200KB
  3. For text, look elsewhere for the size

    Embedded fonts, not images, are usually what makes a text PDF larger than expected - compression of images will not help much here.

Tips

  • Scanning at a lower resolution in the first place (150-200 DPI is plenty for a text document) produces a smaller starting file than scanning high and compressing hard afterward.
  • If a portal accepts multiple files, splitting a long scanned document into two or three smaller PDFs is often more reliable than fighting to hit one aggressive size target.

Common problems

A tool claims to have hit the target size, but the result looks blurry or unreadable.

That tool rasterised the page rather than honestly re-encoding it - it hit the number by destroying the content. A compressor that tells you plainly when a target cannot be reached is more trustworthy than one that always "succeeds".

A one-page document should easily fit but doesn't.

Check the scan resolution and colour mode - a full-colour scan at 600 DPI carries far more data than a black-and-white one at 200 DPI, for the same readable content.

Frequently asked questions

What actually determines how small a PDF can get?
How much of it is photographic content (scanned pages, embedded photos). Real text, fonts and vector graphics are already compact and have little left to remove.
Is there a genuine minimum size for a scanned document?
Yes - past a certain point, further compression destroys legibility rather than just file size. A responsible compressor stops before that point and tells you the result, rather than forcing an arbitrary target.
What should I do if a portal requires a size I genuinely cannot reach?
Re-scan at a lower resolution if you have the original document, or split it into smaller files if the portal accepts more than one upload.

Related PDF tools

Related articles

Ready to do it?

Free, no sign-up, and nothing is uploaded to a server.