Scanned PDF Too Big to Send? How to Shrink It Down

A scanned contract at 30+ MB hits email's 20 MB cap and WeChat's file limits—and re-scanning is a pain. The root cause isn't "too many pages" but that scanned PDFs are essentially a stack of high-resolution images. Once you see that, you know what to compress, how small you can go, and when compression is pointless.

Because scanned PDFs store a bitmap photo per page, not text. A scanner or phone captures the paper as an image and puts it in a PDF container—so size is driven by resolution and color depth, not word count. A blank page and a dense contract scan to roughly the same size.

The math is straightforward: one A4 page at 300 DPI color scan is about 2480×3508 pixels—nearly 25 MB uncompressed. Even with basic JPEG compression from the scanner, a single page is often 1–3 MB; a dozen pages easily reach tens of MB. That's unlike Word-exported PDFs, which store text and vectors—a 50-page report may be only a few hundred KB.

So the first step is telling text-based from image/scanned PDFs. The former has little room to shrink; the latter is where compression pays off.

PDF compression mainly does two things: remove redundant structure and re-encode images. For scans, the second step is what saves volume.

• Lossless optimization: cleans duplicate objects, unused resources, and redundant metadata—tightens internal structure without changing any pixels. For pure-image PDFs, this saves little.
• Image re-encoding: resamples page bitmaps to a target resolution, then compresses with a lossy codec like JPEG. That's how a 30 MB scan can drop to a few MB—halving resolution roughly quarters the data.

A solid approach is two-stage compression: lossless cleanup first, then per-page image re-encoding. That avoids blindly destroying quality while getting most gains from the image layer. The PDF compression tool follows this pattern—scanned and image PDFs often shrink to 10%–30% of the original.

Key takeaway first: 150–200 DPI is enough for screen reading and normal printing; below 100 DPI, blur becomes obvious. Match resolution to use:

In practice, dropping a 300 DPI color scan to 150 DPI with moderate JPEG often cuts size to about 1/5 of the original—barely noticeable on phones and laptops. Black-and-white text scans can also be grayscale for extra savings.

If tuning resolution still exceeds the recipient's limit, try splitting or converting:

• Too many pages: use the PDF split/merge tool to split by page range and send several smaller files.
• They only need to view, not keep PDF: use PDF to images to export needed pages as JPG—often smaller and easier to open.
• The scan was already blurry: low resolution at capture can't be fixed by compression—re-scan instead.

Compression isn't universal. These cases yield little or nothing:

• Text/vector PDFs: Word/web exports are already small with no large images—maybe a few percent smaller.
• Already compressed once: re-compressing degrades quality further with minimal size gain.
• Heavy vector graphics/tables: already efficiently encoded; image resampling doesn't apply.

Simple rule: big PDFs are almost always image-based—compress those; small text PDFs aren't worth the effort.

Scanned PDFs are large because they're stacks of high-resolution images—size follows resolution and color depth, not word count. Compression means lossless cleanup first, then resample images to 150–200 DPI for your use case and re-encode—screen viewing often reaches 10%–30% of original size with little visible loss. When that fails, try splitting or converting to images; pure text PDFs are already small. Processing runs locally in the browser—contracts and ID scans never need to be uploaded.