How to Compress a PDF Without Losing Quality (and When You Cannot)

First, find out what is inside your PDF

A PDF is a container. Some PDFs hold real text, fonts and vector shapes, such as a report exported from a word processor. Others hold pictures of pages, such as a scan, or a photo of a document turned into a PDF. These behave very differently when you try to make them smaller.

A quick test: open the file and try to select a word with the mouse. If you can select and copy it, the page has real text. If the whole page behaves like one picture, it is a scan.

• Text based PDFs are often small already, even with many pages, because text takes very little space.
• Scanned PDFs are mostly images, and their size depends on the resolution, the colour and the number of pages.
• Some PDFs are a mix, with real text and large photos on the same pages.

What makes a PDF heavy

Nearly all of the weight of a heavy PDF comes from images. A single colour photo at high resolution can take more space than a hundred pages of plain text.

• Resolution: more pixels per page means more data. A scan made at 600 DPI is many times bigger than one made at 200 DPI.
• Colour: a colour scan is bigger than a grayscale one, and a grayscale scan is bigger than a black and white one.
• The number of pages, since every page with an image adds its own weight.
• Images that were stored with very little compression, or pasted into the document at a much larger size than they are shown.
• Embedded fonts and repeated content, which matter far less than images.

Methods that keep the text sharp

These do not blur anything, because they remove content or reduce waste instead of redrawing the pages.

• Delete the pages you do not need, such as a blank back page, a cover sheet or a duplicate. Fewer pages is the one change that costs no quality at all.
• Split a long document into parts when the form or the email accepts more than one file.
• Export again from the original program. If you still have the word processor file or the slide deck, create the PDF again and look for an option about the quality of images, or a minimum size choice. The wording depends on the program.
• Scan again with better settings. For pages that are mostly text, a resolution of about 150 to 300 DPI is often enough to read comfortably, and grayscale or black and white takes much less space than colour.

Methods that trade quality for size

When a scan is already as small as its settings allow, the file can only get smaller by storing the pictures with less detail. There are two common ways to do it.

• Lower the quality of the images inside the PDF. They become smaller and a little softer. If a tool touches only the images, real text on the same pages stays sharp.
• Turn each page into one picture at a chosen quality and build a new PDF from those pictures. This can shrink scans a lot, but the text becomes part of the picture, so you can no longer select or search it, and links and form fields are lost.

A warning about text based PDFs

A page of real text is tiny, and a picture of that page is not. This means that the second method above can make a text based PDF larger than it was. If your file is mostly text and already small, compressing it by turning pages into pictures is the wrong tool, and the best choice is often to leave it as it is, or to delete pages.

How to judge the result

Do not judge the quality from the size number. Open the new file and look at it.

1. Zoom to 100 percent and then to 200 percent, and read the smallest text on the page.
2. Check signatures, stamps, thin lines and tables, because they suffer first.
3. Put one page next to the same page of the original if you are unsure.
4. Check the size on your own device, because that is the number an upload form will see.
5. Keep the original file in a safe place.

Compress from the original, and only once

Lowering the quality of images is a lossy step. Doing it again to a file that was already compressed removes more detail each time and usually saves very little more. If the result is too soft, or still too big, go back to the original and try other settings, instead of compressing the compressed file.

A realistic plan for a scanned document

1. Remove the pages that are not needed.
2. If you can scan again, use a moderate resolution, and grayscale for pages of text.
3. Compress the file, then check the first page, the last page and any page with small print.
4. If it is still too big, use a stronger setting or split the file, and check again.
5. Keep the original, and upload the compressed copy only after you have read it.

What the Compress PDF tool on this site does

The tool draws every page as a picture at the quality you choose and builds a new PDF from the pictures. It works best on scanned and photo based files. For the same reason, the text in the result cannot be selected, and links and form fields are not kept.

If the new file would not be smaller than your original, which can happen with a text based PDF, you are told and you keep your original. The work happens in your browser, so the file is not uploaded to a server.