How to Compress a Scanned PDF Without Making It Unreadable

Scanned PDFs get huge for one simple reason: most “scans” are high‑resolution photos saved inside a PDF. If your upload portal says “file too large” (common for HR, school, government, and banking sites), a basic PDF compressor sometimes helps—but scans often need a different approach.

This guide shows a reliable workflow using Dogufy tools to shrink a scanned PDF while keeping text readable.

Quick answer (featured snippet)

To compress a scanned PDF without making it unreadable:

Make sure it’s a scan by trying to select text. If you can’t, it’s image‑based.
Fix page orientation first with Rotate PDF.
If you only need certain pages, extract them with Split PDF.
Try Compress PDF once (fastest option).
If it’s still too large, convert pages to images with PDF to JPG.
Reduce the image file sizes with Image Compressor (and optionally Image Resizer if dimensions are excessive).
Rebuild a smaller PDF from the optimized images with JPG to PDF.
If you need copy/paste and search, run OCR after (see How to Make a Scanned PDF Searchable (OCR)).

When this workflow is the right move (and when it isn’t)

Use the “convert → optimize → rebuild” workflow when:

Your PDF came from a scanner or phone camera
You can’t select text in a PDF viewer
The file size is driven by image pages (not vector text)

Avoid this workflow when:

Your PDF contains selectable text, charts, or vector artwork you need to keep crisp
You need hyperlinks, bookmarks, or form fields preserved (image-based PDFs won’t keep these the same way)

If you’re not sure, check: try selecting a sentence. If selection is impossible or weird, treat it as a scan.

Why scanned PDFs are so large

A typical scan app saves each page as a big image (often far larger than necessary for screen viewing). That means:

File size scales with page count (10 pages can become 50–200 MB depending on settings)
Over-resolution is common (great for archiving, overkill for uploads)
Many “compressors” can only squeeze so much without changing the images

So the most effective fix is often to optimize the underlying images.

Step-by-step: Compress a scanned PDF by rebuilding it from optimized images

Step 1: Rotate pages so text is upright

OCR and image optimization work better when pages aren’t sideways.

Use: Rotate PDF

Tip: Rotate first, then continue. If you convert sideways pages to JPG and fix rotation later, you may waste time and get inconsistent results.

Step 2: Split out only the pages you actually need (optional but recommended)

Most “upload failed” situations don’t require your entire scan.

Use: Split PDF

Examples:

Keep only the signature page + ID page
Extract the 3 pages a portal asks for
Remove blank pages before you optimize anything

Step 3: Try standard compression once (fast path)

Before doing an image rebuild, run the PDF through:

Compress PDF

If it hits your target size, you’re done. If the result barely changes (common with scans), keep going.

Step 4: Convert the PDF pages to JPG (the “scan-friendly” format)

JPG is usually the best format for scanned pages because it compresses photo-like content efficiently.

Use: PDF to JPG

If you’re deciding between JPG and PNG for scans, this quick rule works well:

Scans/photos of paper: choose JPG
Crisp UI screenshots, line art, or logos: PNG might look cleaner (but usually larger)

Step 5: Compress the JPGs (and resize if necessary)

Now optimize the page images:

Use: Image Compressor

If the images are massive in dimensions (for example, pages exported at extremely high pixel sizes), resizing often reduces file size more than “compression strength” alone:

Use: Image Resizer

Practical “don’t ruin readability” guideline:

Prefer light compression + moderate resizing over extreme compression.
After any big resize/compress change, zoom in on a paragraph and check that letters don’t smear together.

Step 6: Rebuild a smaller PDF from the optimized images

Once your images are optimized, convert them back into a single PDF:

Use: JPG to PDF

Order matters. If your pages are out of order, use the move up/down controls in the tool before converting.

Step 7: Optional finishing steps (depending on what the upload wants)

If you still need to shave off a bit more size, try Compress PDF again on the rebuilt PDF.
If you need to add a signature after rebuilding, use Sign PDF.
If you need to type extra text (like a date or name), use Edit PDF.

Troubleshooting (the common “why is it still big?” cases)

“I compressed the JPGs but the PDF is still too large”

Most of the time this means the images are still too large in dimensions, not just in file size. Resize the pages (moderately) and rebuild again:

“Now it’s small, but the text looks blurry”

Back off in this order:

Undo aggressive resizing (keep more pixels).
Use lighter image compression.
If you only need certain pages, split and upload fewer pages instead of over-compressing:
- Split PDF

“I need the PDF to be searchable / copyable”

Rebuilding from JPGs produces an image-based PDF. To make it searchable, you’ll need OCR:

How to Make a Scanned PDF Searchable (OCR)

Tip: If you plan to OCR, avoid extreme downscaling—OCR accuracy depends on readable letter shapes.

FAQ

Will this preserve selectable text and links?

If your original file is a true scan (image-only), there’s no selectable text to preserve. Rebuilding from images keeps it readable but image-based. If you need selectable text, run OCR after.

If your original PDF contains real text and links, avoid the image rebuild and use Compress PDF instead.

Should I use JPG or PNG for scanned pages?

For paper scans, JPG is almost always the best balance of readability and size. Use PDF to JPG, then Image Compressor.

If the document contains lots of sharp-edged graphics (like UI screenshots), PNG can look cleaner but may be larger. Compare both if quality is critical.

What’s the fastest way to meet a strict upload limit?

Do this in order:

Split out only what’s required: Split PDF
Compress: Compress PDF
If it’s a scan and still too big, rebuild from optimized JPGs:
- PDF to JPG → Image Compressor → JPG to PDF