How to Compress a Scanned PDF
By PDFOak · Published September 9, 2026 · Updated October 1, 2026
A scan stores each page as an image, so meaningful compression usually changes resolution or image quality. Keep the original and inspect critical details after compression.
Choose the byte limit the recipient actually sets. Going lower than that only costs readability.
How it works
- Open the scanned PDF and check whether any text is selectable.
- Choose the target size the recipient requires.
- Accept rasterization only if losing an OCR text layer is safe; image-only scans are rasterized to reach the target.
- Compress the file in your browser.
- Inspect names, dates, fine print, stamps, signatures and barcodes before you submit it.
Why scans are large
Every page can hold millions of pixels. Colour, high resolution, paper texture, shadows and camera noise all add detail to describe. Clean grayscale scans often compress better than phone photos.
If colour proves a seal or an annotation, keep it even when grayscale would be smaller.
What happens to OCR text
Some scans already carry an invisible OCR text layer. Exact-size compression first tries to keep it by recompressing the images in place. A scan that is mostly picture may still be rasterized to reach the target, and that removes the text layer. If search matters, try lossless mode first.
To get text back after compressing, run PDFOak’s OCR tool on the compressed file. OCR uploads the file.
Frequently asked questions
Should I scan in colour or grayscale?
Use grayscale when colour carries no evidence or meaning, since it is often smaller. If colour proves something, keep it.