Compress PDF
Make a PDF smaller. Start with the safe clean-up, which changes nothing you can see. Only move to rasterising if you need a much smaller file and can live without the text layer.
Runs entirely in your browser. Your files are never uploaded.
How to use Compress PDF
- 1Pick a PDF, or drag it onto the box above.
- 2Run the safe clean-up first and see what it saves.
- 3If that's not enough, switch to replacing pages with images and pick a resolution.
- 4Check the before and after sizes, then download.
Why PDF compressors disappoint people
A PDF is a container. Its size is almost entirely whatever got put inside it, which means embedded fonts and, overwhelmingly, images. There's no general squeeze that makes a PDF smaller without deciding what to give up.
So when a tool promises to halve your file with one click, it has made that decision for you. Usually it re-encoded your images, or replaced your pages with pictures of pages, and you find out when you try to search the document three weeks later. This tool offers both routes and names them.
The safe clean-up
The document gets parsed and written out again using cross-reference and object streams, which plenty of producers don't bother with, especially older office software and some scanner drivers. Objects nothing points at any more get dropped along the way.
Nothing visible changes. Text stays selectable, images keep the quality they had, page sizes don't move. The saving tends to be a few per cent. Occasionally it's much more, on a file from a particularly verbose generator, and sometimes it's zero because the file was already tidy. It costs you nothing to find out, which is why it runs first.
Replacing pages with images
Every page gets rendered at the resolution you choose and stored as a JPEG. On a scanned document this is often a big win. A scan is already pictures of paper, frequently at a far higher resolution than anyone needs, and there was never a text layer to lose.
On a document that came out of a word processor it's the wrong tool entirely. Text stores very efficiently as text and very inefficiently as a photograph of text, so the file may come back the same size or bigger, and now you can't search it. We've measured both outcomes on this page: the same source file came out 18% larger at 1600 pixels and 53% smaller at 1000 pixels with more compression. The tool warns you before it runs and tells you the real before-and-after size afterwards, including when that number goes the wrong way.
When neither one is enough
If the problem is an email limit, splitting the file often beats compressing it. Three 10 MB attachments go through where one 30 MB attachment bounces.
If the bulk is a handful of large photographs, go back to whatever produced the PDF, replace those images with smaller ones and export again. Fixing it at the source always beats fixing it afterwards.
Limits
- 100 MB per file.
- Password-protected PDFs need unlocking first.
- The safe mode can't re-compress images that are already compressed.
- Rasterising removes the text layer, links and form fields.
- Rasterising a text document can easily make it bigger.
Frequently asked questions
- Why did the safe clean-up barely save anything?
- Because the file was already efficiently written and its size is its images. The clean-up reorganises the container. It can't re-compress picture data without changing that data.
- Will my text still be searchable afterwards?
- After the safe clean-up, yes, exactly as before. After rasterising, no. The pages have become images and the text layer is gone.
- Which mode should I use on a scan?
- Rasterising, in most cases. A scan is already images, so there's no text layer to lose, and the saving is normally large. Try 1600 pixels at quality 70 and see where that lands.
- Can you compress a PDF without losing anything at all?
- That's what the safe mode does, and its ceiling is honest: a few per cent, sometimes nothing. If someone promises a big reduction with no loss, they're changing something they haven't told you about.