PDF Compression — How Small Can You Go Without Losing Quality?
July 14, 2026
I once emailed a 200-megabyte PDF to a client. You can probably guess how that went. Their mail server bounced it, my outbox got stuck, and I had to spend twenty minutes uploading the thing to a file-sharing service while feeling like a complete amateur. The worst part? That PDF was just a proposal with a few product photos and some text. It did not need to be that big.
PDF compression is one of those skills you don't realize you need until you're in a situation where you desperately need it. And then suddenly it's all you can think about. So let me save you the trouble I went through and break down what actually matters when you're trying to shrink a PDF.
The number one thing that determines your file size is images. I cannot stress this enough. A single high-resolution photo from a modern smartphone can be 5 to 10 megabytes all by itself. If your document has ten of those, you're already at 50 to 100 MB before you add a single word of text. The PDF wrapper adds very little overhead — it's the images that are killing you. So when you're trying to compress a PDF, the first question you should ask is: can these images be smaller?
Most compression tools give you a few different levels to choose from. Low compression keeps the images almost intact and gives you a modest file size reduction. Medium compression starts to slim things down more aggressively, usually with very little visible quality loss. High compression can cut the file size in half or more, but you'll start to see artifacts if you look closely. The sweet spot for most documents is somewhere in the medium range. That's where you get the biggest savings without anyone noticing the difference.
There's a technical distinction worth understanding here. Lossless compression means the file gets smaller but every single pixel stays exactly the same. It's like zipping a folder on your computer. Lossy compression actually discards some of the image data to save space, and it can achieve much higher compression ratios as a result. The trick is finding the point where the data you're discarding isn't something a human would notice. JPEG compression is lossy by nature, and it's very good at this. A JPEG compressed to 80 percent quality often looks identical to the original to the naked eye, but it takes up a fraction of the space.
Beyond images, there are a few other things that add weight to your PDFs. Embedded fonts can be surprisingly heavy, especially if you're using a font with a large character set. Some PDF creators embed the entire font file when they only need a handful of characters. Metadata and annotations also add up over time. And if your PDF contains vector graphics or CAD drawings, those can be complex in their own way.
The workflow I use now is pretty simple. I check the image resolution first. If I'm including photos, I make sure they're scaled appropriately before I generate the PDF. A 4000-pixel-wide photo doesn't need to be in a document that will only ever be viewed on screen. Then I run the finished PDF through a compression tool at medium settings. If the result is still too big, I bump it up to high compression and inspect the output closely. Nine times out of ten, medium does the job and nobody ever knows the difference.
These days I don't think twice about sending PDFs through email. A few megabytes, a quick attachment, done. No file-sharing services, no bouncebacks, no embarrassment. It took me one bad experience to learn the lesson, but I'm glad I did.