Lossy vs. lossless compression: what it is

In 30 seconds

Compressing means storing the same information with less data.
Lossless gets you the exact file back; lossy throws away barely noticeable detail and saves far more.
In a PDF, images are usually the heaviest part, not the text.
The best approach is to compress a copy and not repeat the process on the same file.

1What is compression and what is it for?

Compressing means representing the same information with less data. It makes a file take up less disk space, take less time to send and fit within the limits of web forms and email attachments. In everyday life, almost everything you download is compressed: photos, music, videos, web pages and documents.

Compression is not a single trick but a set of techniques. Some are fully reversible and others accept losing something in exchange for much bigger savings. Understanding the difference lets you choose based on what matters most in each case: fidelity or file size.

2How does compression work under the hood?

Lossless compression takes advantage of repetition. If a file contains the same sequence many times, instead of repeating it, it is stored once with a note of how many times it appears. When decompressed, exactly what was there is rebuilt. This is what the ZIP and PNG formats do, and it is also part of PDF's internal compression.

Lossy compression goes further: it exploits what the eye or ear does not distinguish well. In a photo, for example, small color variations in a uniform area are simplified. The harder you compress, the more information is discarded and the sooner visible flaws appear, such as blocks or halos next to the edges of text.

3What types of compression are applied to a PDF?

A PDF combines several things. The most important is image compression: you can reduce their quality, their resolution or both, and this is where the biggest savings come from. Another is structural: removing duplicate data, simplifying fonts and stripping out information that is not displayed. And the third is switching color to grayscale when the document does not need color.

Image formats vary too: JPG uses lossy compression and is ideal for photos; PNG is lossless and works better for screenshots, logos and flat-color drawings. Choosing the right format before inserting the image into a document already avoids half the problem.

4When should you compress, and when not?

Compress when you have a size limit, when you will share by messaging or email, or when the document will only be viewed on screen. A moderate level of compression usually goes almost unnoticed in normal reading.

Do not compress if the document will be printed in high quality, if the images are working material with fine detail such as blueprints, technical photos or very small type, or if the file is already light. A PDF made up mostly of text will barely shrink, and compressing a JPG or a ZIP again saves almost nothing.

5Common mistakes and myths

One myth is that all compression keeps the file intact. Only lossless does. Another mistake is believing that compressing the same file several times shrinks it more each time: every lossy pass accumulates flaws and the extra savings are small. It is also thought that PDF text degrades when compressed, when in fact the text is vector-based and stays almost intact; it is the images that suffer.

Zipping a folder does not help much with PDF or JPG files either, because they are already compressed. To really save space you have to work on the images inside the document itself.

6Common problems and how to fix them

If the PDF barely shrinks, it was probably already optimized or is made up mostly of text. If it shrinks but looks blurry, you tried a level that was too strong: go back to the original and use gentler compression. If it is heavy because of a scan, the root of the problem is resolution, and scanning at 200 or 300 DPI in grayscale solves it better than any later compression.

In rare cases, the resulting file can even be larger than the original. This happens when the original was already highly optimized. The solution is to keep the original. And remember that lossy compression is irreversible: there is no way to recover the discarded detail.

7Compression and privacy in Rayo PDF GO

Compress PDF and Compress image in Rayo PDF GO work on your device, so documents with personal data are not uploaded to any server to reduce their size. That is a meaningful difference, because many online compressors require you to send the entire file.

A practical guideline: keep the original, compress a copy, and check the result by zooming in on a page with images before sending it. If the size is still not where you want it, Convert to grayscale or shrinking the images with Resize image before building the PDF usually delivers the missing savings.

Related tools

Frequently asked questions

Is compression noticeable in the text of a PDF?

Usually not. The text is vector-based and takes up almost no space; what mainly gets compressed is the images.

What is the difference between lossy and lossless compression?

With lossless compression, the file is recovered exactly the same. With lossy compression, barely noticeable information is discarded to save more space, and that information cannot be recovered.

Can I compress a PDF several times?

You can, but it is not a good idea. Each lossy pass adds flaws and the extra savings keep shrinking. It is better to always start from the original.

Why does my PDF not get smaller?

It may be made up mostly of text or may already be optimized. In those cases there is not much left to reduce.

Does compressing a PDF make it unreadable?

Not if you use a moderate level. If the result looks blurry, the level was too strong: go back to the original and try a gentler one.

Keep learning

See the whole knowledge base ›