What DPI Should I Use for a Scanned PDF?

DPI describes sample density relative to physical size. Doubling both width and height approximately quadruples the pixel count before compression, which can sharply increase file size.

A clean typed page may remain readable at a lower resolution than a faded receipt or handwritten note. Color photographs and security features have different requirements from black text on white paper.

Follow the recipient's specification when one exists. A scan that looks clear on your monitor may still fail OCR or print poorly.

In practice, most submission portals are satisfied by 200 DPI in grayscale, which typically produces well under 1 MB per page. Around 300 DPI is the usual default for OCR accuracy and for anything that will be printed at full size. Above roughly 300 DPI, extra pixels rarely change whether a person can read the text, but they increase file size substantially — doubling DPI in both dimensions quadruples the pixel count.

Too low is worse than too high in one specific way: it is irreversible. OCR accuracy drops sharply on low-resolution scans, small punctuation and superscript characters blur together, and security features such as microprinting or fine line patterns can disappear entirely. If you are unsure, scan higher, confirm legibility, then reduce a copy.

A workable default for portals with tight caps: scan at 200 DPI in grayscale, check the smallest text at 200% zoom, and only increase the setting if that check fails. IRCC, for example, suggests 150 DPI or less when saving PDFs to stay under its upload limits.

Bit depth matters as much as DPI. A 1-bit monochrome scan of typed text at 300 DPI is often smaller than a grayscale scan at 200 DPI, because each pixel costs a single bit instead of eight. It destroys photographs, stamps, and colored security backgrounds, so it suits clean typed documents only.