CompressaPDFto100KB

Somebody has given you a number — an upload field that stops at 100 KB, or a form that rejects anything larger. Whether you can hit it depends almost entirely on what kind of PDF you are holding, and it is worth thirty seconds to work out which.

or drop it here — nothing is uploaded

What 100 KB realistically means

100 KB is a hard target. A one or two page text document will already be near it, and a short scan can reach it on the low quality setting — but a ten page colour scan will not, and no tool can make it, because there is a floor below which a page stops being readable.

How to get there

Work down rather than straight to the bottom:

  • Try medium first and look at the size. A great many files clear the limit here and stay perfectly legible.
  • Drop to low. This is the setting for exactly this situation — a hard cap where readability is all that matters.
  • Turn on greyscale as well. On a scan of black text this is usually the single largest saving available.
  • Delete the pages you do not need. Two pages at 100 KB is a different problem from ten.

If the requirement is a photo of an ID or a single certificate, one page at low quality and greyscale will normally land well under the limit.

If it still will not fit

There are three moves left, in order of how much they cost you:

  • Greyscale. A colour scan of black text carries three channels describing grey paper — dropping them is often a third of the size, at no cost to legibility.
  • Crop. A scanner's black border and wide margins are bytes spent describing nothing. Trim them, then compress.
  • Split. If the limit is per file rather than per submission, send the document in two parts. This is almost always allowed and almost never tried.

What does not help is compressing twice. The second pass re-renders an already-rendered page: you lose quality and save almost nothing.

Why the file was so large to begin with

A PDF from a word processor is a few kilobytes a page, because it stores text as instructions. A scan stores a photograph — an A4 sheet at 300 DPI is roughly 8.7 million pixels — and a scanner that saves those without much compression produces several megabytes per page.

That is the whole story behind almost every oversized PDF. It is not corruption and nothing has gone wrong; it is eleven photographs in a wrapper.

Compress PDF at a glance

Is anything uploaded?
No — the work happens in your browser
Account needed?
No
Daily limit?
None
Watermark on the output?
None
What you get back
a PDF
Text layer
Not preserved — pages are re-rendered
Works offline?
Yes, once the page has loaded
Price
Free

Questions people ask

No, and any tool promising it is guessing. Scans compress dramatically; a document that is already only text has almost nothing to give up, because it was never large for the reason you think.