How to Compress Scanned PDFs

Why scans are so large

Scanners measure resolution in dots per inch (DPI). An A4 page scanned at 300 DPI is roughly 2480 × 3508 pixels, about 8.7 million pixels. Stored uncompressed in colour, that is around 26 MB per page. Scanners and phone apps compress the image, but often with conservative settings, and some save every page in full colour even when the paper is black and white.

Three settings decide the size of a scan:

• Resolution: halving the DPI cuts the number of pixels by three quarters.
• Colour mode: greyscale stores one value per pixel instead of three; pure black-and-white stores far less again.
• Compression: how strongly the page images are compressed, usually with JPEG.

Best results: get the scan right at the source

If you can scan again, adjusting the scanner is the most effective fix. For most text documents, 150–200 DPI in greyscale is readable on screen and in print, and 300 DPI is a common recommendation when the text will be processed with OCR (optical character recognition). Use colour only when colour matters, such as for photos, stamps or highlighted text.

Option 1: Compress the images inside the PDF

If you only have the PDF, SlimMyFile’s normal levels can help. In Balanced mode, large page images are re-encoded as JPEG at quality 72 and limited to 2400 pixels on the longest side; High mode uses quality 55 and 1600 pixels. That is often a large reduction for high-resolution colour scans, and anything else in the file, such as an existing OCR text layer, bookmarks or links, is kept.

Some scanned PDFs use special formats for their page images, such as JBIG2 for black-and-white pages, JPEG 2000, or CMYK colour. SlimMyFile leaves those images as they are, so these files may not shrink in the normal modes.

Option 2: Scanned document mode (rasterising pages)

Scanned document mode takes a different approach: it renders every page as a single image at a resolution you choose (72, 100, 150 or 200 DPI) and quality you set, then builds a new PDF from those images. Because it rebuilds each page from scratch, it works even when the original page images use formats the normal modes cannot recompress.

The trade-off is significant. The new PDF contains only pictures of pages, so it permanently loses:

• Searchable and selectable text, including any OCR layer the scan had
• Hyperlinks and bookmarks
• Interactive form fields
• Accessibility tags used by screen readers

For a pure scan, which never had real text, the visible loss is only the resolution and quality you chose. For a PDF with real text, this mode is usually a bad idea, which is why SlimMyFile requires you to confirm it explicitly.

Step by step

1. Keep the original file.
2. Open the PDF compressor and select the scan. If the first pages contain no text, SlimMyFile points out that the file looks like a scan.
3. Try Balanced first. If the result is small enough, you keep any text layer the file may have.
4. If not, enable Scanned document mode, read the warning, confirm, and choose 150 DPI with quality around 70 as a starting point.
5. Download the result and zoom in on small print, signatures and stamps to check that they are still legible. If they are not, try 200 DPI or a higher quality.

Common mistakes

• Choosing 72 DPI for documents with small print. Footnotes and fine handwriting can become unreadable.
• Rasterising a PDF that has an OCR text layer and then wondering why search stopped working.
• Sending a heavily compressed scan of a legal or official document when the recipient needs a faithful copy. Ask what quality they require.
• Rasterising a very long document on a phone. Rendering every page takes memory and time on your device; a computer copes better.

Frequently asked questions