How to compress a PDF
- Drop your PDF onto the box above, or press Browse files to select it.
- Set the target resolution, 120 DPI for screen reading, 150 if it will be printed.
- Choose an image quality. 70% is a good starting point for scanned documents.
- Press Compress PDF and wait while each page is rebuilt.
- Check the reported saving, open the result to confirm it is still readable, then download.
What this compressor does
- Rebuilds every page at a resolution you control, rather than guessing on your behalf.
- Routinely reduces scanned documents by 80 to 95 per cent.
- Preserves the original page dimensions, so print size and layout are unchanged.
- Reports the before and after size in bytes and as a percentage.
- Warns you plainly when the result came out larger than the original.
- Never uploads the document, which matters for the contracts and records people compress.
Why PDFs get too big
A PDF's size is almost always dictated by its images. A page of text is a few kilobytes, characters and font references take almost no space. A scanned page is a photograph, and a scanner set to 600 DPI in colour produces something close to twenty megabytes for a single sheet. Twenty pages of that is a document nobody can email.
Embedded photographs behave the same way. A report with thirty product images at full camera resolution carries all thirty at full weight, even though each one occupies a few centimetres on the page. The visible size on paper has nothing to do with the pixels stored behind it.
What this tool actually does
It renders each page at your chosen resolution and rebuilds the document from those renders as compressed JPEG images, keeping the original page dimensions. That is a blunt instrument, and it is deliberately blunt, because it is the only approach a browser can take without a full PDF object rewriter.
The consequence is worth being clear about: the output is a picture of each page. Selectable text becomes an image of text. Links stop working. Form fields become pictures of form fields. Searching the document finds nothing. In exchange, a 40 MB scan becomes 2 MB and finally fits the upload box.
When it is a good trade and when it is not
- Excellent: scanned contracts, receipts, forms and anything photographed with a phone, these had no text layer to begin with.
- Good: image-heavy brochures and reports where the pictures dominate the size.
- Poor: text documents exported from a word processor, which are already small and would lose searchability for almost no saving.
- Avoid: forms that need filling in, and any document where the text must remain selectable.
If a text-based PDF is larger than it should be, the cause is usually a handful of oversized images or embedded fonts rather than the text. Compressing those images before building the PDF, with the image compressor, solves it without rasterising anything.
Choosing resolution and quality
Resolution is the bigger lever, because halving it removes three quarters of the pixels. 96 to 120 DPI reads comfortably on screen. 150 DPI survives printing on an office laser printer. 200 and above is rarely worth it here, since a document needing that fidelity probably should not be rasterised at all.
Quality then controls the JPEG compression on top. Seventy per cent suits scanned text; go lower and the characters pick up a visible haze. If the result is unreadable, raise the resolution before raising the quality, soft text is usually a resolution problem, not a compression one.
If the file gets bigger
This happens, and the tool says so rather than pretending otherwise. A PDF that was already well compressed. A text document, or a scan someone else has already reduced, will grow when it is rendered and re-embedded as images. That is a signal to keep your original, not to try harder.
Other routes to a smaller file are worth trying first: remove pages you do not need with the page deleter, or extract just the section you are sending with the page extractor. Both are lossless. Everything here runs in your browser, so nothing is uploaded at any stage.