Gmail caps attachments at 25MB. Most portals for job applications, tenders, or government forms cap it lower — often 2MB or 5MB. So when a "quick PDF" turns out to be 40MB, you go looking for a way to shrink it. Here's what actually happens when you compress a PDF, and — just as important — when it won't help.
Why PDFs get big in the first place
Three things usually account for most of a PDF's size:
- Embedded fonts. A PDF that embeds a full font family (rather than just the characters it actually uses) can add hundreds of kilobytes per typeface, especially with multiple weights.
- High-resolution images. A photo captured at 12 or 48 megapixels doesn't need to be that large to look sharp on a page — but if it's pasted in at full resolution, the PDF carries that full weight.
- Scanned pages. A "scanned" PDF is usually just a sequence of full-page images, one per page, with no actual text underneath. These are almost always the biggest files, because every page is a photograph, not text.
What "compressing" a PDF actually does
There are two very different things people mean by "PDF compression," and it's worth knowing which one you're getting:
- Re-serializing the content streams. A PDF's internal structure (text positioning, drawing instructions, metadata) can usually be re-encoded more efficiently without touching a single pixel of any image inside it. This is lossless — nothing about how the document looks changes.
- Re-encoding the images. A more aggressive tool will also open every embedded image and re-save it at a lower resolution or a higher JPEG compression ratio. This is where the big size reductions come from on scanned or photo-heavy PDFs — but it's also where quality can visibly drop if pushed too far.
Our Compress PDF tool deliberately does only the first kind. It re-serializes the document's content streams, which is completely safe — you'll never get a blurrier or lower-resolution result — but it also means a PDF that's already full of well-compressed JPEG scans may not shrink much, because there's nothing left to squeeze out of the structure itself. If your original file comes back the same size, that's not a bug — it means it was already about as small as it can safely get without touching the images.
If your PDF is mostly scanned pages or photos
For that case, the real savings are in the images themselves, before they ever become a PDF. Two options:
- If you're building the PDF from images you control, run them through an image compressor first, then combine them with Image to PDF. You control exactly how much quality you're trading for size.
- If you already have the oversized PDF and can't rebuild it from the source images, a stream-only compressor like ours is still worth trying first — it's free, instant, and occasionally the file does carry recoverable overhead (unused embedded fonts, redundant metadata) even when it looks image-heavy.
Using the tool
Upload one PDF (up to 20MB), and you'll get a download link along with a before/after size comparison. If there's no reduction, the original file is returned unchanged — you'll never end up with a larger file than you started with.
Try it here: Compress PDF.