Why won't my PDF compress? What makes it big, and what to do

A PDF will not compress when there is nothing in it that can be made smaller without ruining it. Size in a PDF usually sits in the images: scans, photos, and logos saved at a far higher resolution than the page shows. A PDF made only of text and vector drawings is already small, and a PDF whose images are already well compressed has little left to give.

Try the Target Size File Compressor first. It tells you in plain words what it found, and that message usually answers the question. The rest of this guide explains how to read it and what to do for each cause.

A PDF is a container. Inside are the pages, and with them whatever the pages need to look right. Four things take up most of the space.

• Images. A scanned page is one big picture. A document with photos holds one picture per photo. This is the usual reason for a file of many megabytes, and the only part the compressor can shrink.
• Fonts. A PDF carries the fonts it uses, so it looks the same on every device. A document that uses many fonts, or a font with a very large character set, is heavier for it.
• Text and drawings. Words and vector shapes take very little room, even over many pages.
• Structure and extras. Page descriptions, form fields, attachments and metadata add a small amount each, and a file with attachments can add a lot.

Compare the file with its page count. A one-page scan can be bigger than a fifty-page contract typed in a word processor. If a text-heavy file is large, the weight is probably fonts or attachments rather than the words.

The compressor re-encodes eligible images inside the PDF as JPEG, lowering quality and, when needed, resolution. It does not rasterize the text or vector content, so text remains selectable and searchable. Some image types are skipped when they cannot be safely decoded or preserved.

You give it a size you need, for example 500 KB or 2 MB. It tries from the best-looking setting down to the lowest one and keeps the best-looking result that fits. Everything runs in your browser, so the file is not uploaded.

It will not hand you a worse answer and call it better. If the result would be larger than the file you gave it, you get the original back, marked as unchanged.

When a PDF does not reach your target, the compressor says why. These are the cases and what each means.

Some images are skipped on purpose, because changing them would damage the document. The compressor counts them and says why, so a file that barely shrank explains itself.

• Formats a browser cannot decode. JPEG 2000, fax-style black-and-white (CCITT, JBIG2) images and similar formats need decoders the browser does not provide. Some scanners and archiving software save pages this way.
• Images with transparency settings the tool cannot preserve. These are skipped to avoid changing how the page looks.
• Stencil masks. These are shapes used for cutting out other images, not pictures.
• Very small images. In a target-size run, images under about 2 KB are skipped. They are not why a file is large.
• Images that would grow. An image that is already compressed well can come out bigger when re-encoded. That one is left as it was.

Work on the source, not on the PDF. A compressor cannot take fonts or attachments out of a finished file, but you can often avoid them when you create it.

1. 1

   Remove pages you do not need

   Use Delete pages from PDF for blank pages, cover sheets or duplicates. Each page you remove takes its images with it.

2. 2

   Split the file if the limit is per upload

   If a portal accepts several files, Split PDF cuts one large file into parts that each fit.

3. 3

   Check the images in the source file

   If you made the PDF from a document or a design, shrink the pictures there before exporting, and choose a lower-quality export option if the program has one. This is the cleanest fix, because the picture is made small once instead of squeezed afterwards.

4. 4

   Rescan at a lower resolution

   A scan at a very high setting makes a very large page picture. If you still have the paper, scanning again at a lower setting gives a smaller file than any compression applied to the first scan.

For a scan that you also need to search or copy from, see scanned vs searchable PDF. Making a scan searchable does not make it smaller, so do the size job first.

If the compressor says the PDF could not be opened, the file may be damaged or password-protected (encrypted). It does not unlock a protected file. Remove the password first with Unlock PDF, if you are allowed to, then compress the unlocked copy.

There is also a size limit: a file over 100 MB is refused, because compressing it in a browser tab could run the device out of memory. Split it into parts first.

• It will not reach every target. If a file is mostly text, fonts or attachments, or its images are already at the lowest usable quality, it will tell you it could not, and give back its smallest result.
• It lowers image quality. Pictures get softer as the target goes down. Check the result on screen, and look closely at signatures, stamps and small print.
• It does not change your words. Text is untouched. It also will not remove fonts, attachments or hidden data from the file.
• It will not turn pages into pictures to save space. That would shrink the file but turn your text into an image you could no longer search or copy.
• It does not unlock files. A password-protected PDF has to be unlocked first.