Compress PDF

How to compress a PDF

1. Drop a PDF file on the drop zone, or click it to choose one. The tool shows its name, number of pages and size.
2. Choose the Image resolution, from 72 DPI (smallest, screen only) to 300 DPI (print). 150 DPI (recommended) suits reading on screen and ordinary office printing.
3. Set the JPEG quality with the slider (30 to 95; 70 by default). Lower values give smaller files and more visible compression artefacts.
4. Keep Also convert lossless images to JPEG ticked for photos stored without compression loss; untick it for screenshots or diagrams that must stay perfectly sharp.
5. Keep Remove unused objects ticked, and tick Remove metadata (title, author, software…) if you want to strip the document information too.
6. Click Compress PDF. The tool shows the Original size, the Compressed size, how much you Saved and the number of Images recompressed.
7. Click Download the compressed PDF. If the result isn’t smaller than your file, no download is offered: keep your original.

What this tool actually compresses

Compression in the browser works on images only. For each image, the tool finds the largest size at which it is drawn on a page, then lowers its resolution to the chosen DPI and re-encodes it as JPEG. The pixel size it aims for is simply:

pixels = drawn size in inches × DPI

A 3000 × 2000 px photo drawn 5 inches wide on the page needs only 750 px across at 150 DPI, so it shrinks to 750 × 500 px. Images already at or below the target resolution keep their size and are only re-encoded. An image is replaced only when the new version is at least 5 % smaller; otherwise the original stays untouched.

Limits to know

• Text, fonts and vector drawings are not changed. A report made mostly of text, a spreadsheet export or a CAD plan contains few image bytes, so it won’t shrink much. Scanned documents and photo-heavy brochures give the biggest savings.
• Some images are deliberately left alone: very small ones (under about 8 KB), transparency masks, CMYK images, JPEG 2000, black-and-white fax-style images (JBIG2, CCITT) and indexed-colour images.
• Browsers always write colour JPEGs, so grayscale images are saved as RGB JPEG. They still look grey and usually still shrink.
• Remove unused objects deletes data nothing in the document refers to, such as leftovers from earlier edits. Remove metadata deletes the title, author, creating software and XMP data.
• Any change to a digitally signed PDF invalidates its signature; the tool warns you when it detects one.
• Password-protected PDFs must be unlocked first (you need to know the password).

Privacy

The PDF is read, compressed and saved by your browser, on your device. The file is never uploaded to a server.