How to Reduce Scanned PDF File Size

You scanned a 15-page agreement to email to a colleague, and the file is 48 MB. Your mail server rejects it. You try a different document — a scanned receipt book — and it’s 60 MB. Meanwhile, a 15-page report you exported straight from your word processor is barely 200 KB. Same page count, wildly different sizes. What’s going on, and how do you get those scans down to something sendable?

The answer comes down to what a scanned PDF actually contains, and once you understand that, shrinking it becomes straightforward. This guide explains why scans are so large, walks through the right way to compress them (it’s different from compressing an ordinary PDF), shows how to hit a specific target size, and covers how to do it all in your browser without uploading sensitive paperwork.

If your only goal is to clear an email attachment limit on any kind of PDF, we have a quick ordered method for that in getting a PDF under the email limit. This guide is specifically about scanned PDFs — the big ones — and why they need their own approach.

Why scanned PDFs are so big: it’s all images

Here’s the core fact. A PDF exported from a word processor stores its pages as text — a few kilobytes of letters and font data. A scanned PDF stores each page as an image — a photograph of the page captured by your scanner or phone camera.

That single difference explains the size gap entirely:

• Each page is a picture. A page scanned at 300 DPI (dots per inch) is a high-resolution image that can be several megabytes on its own — before you’ve added a single other page.
• Resolution multiplies fast. DPI measures how many dots make up each inch of the page. Higher DPI means sharper text but far more pixels to store. Going from 150 to 300 DPI roughly quadruples the pixel count.
• Color makes it heavier. A color scan stores red, green, and blue information for every pixel. A black-and-white scan stores far less per pixel. Many scanners default to color even for plain black-text documents, inflating the file for no benefit.
• Page count multiplies everything. A 3 MB scanned page times 20 pages is 60 MB.

Notice what’s not a factor: the amount of text. A one-line scanned memo and a dense scanned contract take roughly the same space, because you’re storing the picture of the page, not the words on it. This is the opposite of a text PDF, where more content means a bigger file.

This is why scanned PDFs need a different size-reduction strategy. Compressing a text PDF squeezes fonts and structure. Compressing a scan means dealing with the images — because the images are the file. For the general mechanics of how PDF compression works across both types, see how does PDF compression work.

The key technique: downsampling the images

Since the images are essentially the entire file, the way to shrink a scan is to downsample them — reduce the resolution (DPI) of each page and re-encode it more efficiently.

Think of it like resizing a photo. A 4000-pixel-wide photo and a 1500-pixel-wide version of the same photo look nearly identical on a screen, but the smaller one is a fraction of the size. Downsampling a scanned page does the same thing: it lowers the DPI to a level that still looks clean for your purpose, discarding pixels you don’t need.

The results are dramatic. Because scanned pages start at such high resolution, downsampling can shrink a scan by 80–95% while keeping it perfectly readable. A 48 MB scanned agreement can routinely come down to 3–5 MB — or even less — with no meaningful loss for on-screen reading.

The question is how far to downsample, and that depends on what you’re going to do with the file.

Choosing the right target resolution

There’s no single “best” DPI — only the right DPI for the destination. Here’s a practical guide:

Use case	Target DPI	Result
On-screen reading, email, web upload	~150 DPI	Clean and clearly readable, smallest size
General sharing where it might be printed	~200 DPI	Good balance of size and print quality
High-quality printing, archival, fine detail	300 DPI	Largest of the three, print-sharp

The mistake people make is reaching for the strongest compression setting on every file. If you’re going to print the document, crushing it to 150 DPI will look soft on paper. If you’re only going to read it on a screen or email it, keeping it at 300 DPI wastes megabytes for sharpness no monitor will show.

Match the level to the destination. For the most common need — emailing or uploading a scan someone will read on a screen — a level targeting around 150 DPI is the sweet spot: it shrinks the file enormously and still looks clean.

A few more tips for the best quality at any target:

• Start from the original scan, not a copy that’s already been compressed once. Compression that throws away detail can’t be undone, so re-compressing a compressed file degrades it further without saving proportionally more.
• Scan in black-and-white for plain text documents if your scanner offers it — it’s much smaller than color and perfectly readable for black text.
• Check the result. Compression that’s too aggressive softens text edges and blurs small print. After compressing, open the file and look at the smallest text on the most detailed page before you rely on it.

How to compress a scanned PDF (step by step)

Here’s the full process using the free in-browser Compress PDF tool. Nothing installs, nothing uploads.

1. Open the tool. Go to Compress PDF in any modern browser — desktop, Chromebook, phone, or tablet.
2. Add your scanned PDF. Drag the file in or click to browse. It’s read locally; nothing is sent anywhere.
3. Pick a compression level. Choose based on the destination — a moderate or strong level (targeting ~150 DPI) for email and screen reading, a gentler level for documents you’ll print. The tool downsamples the page images and re-encodes them.
4. Compress and check the size. The tool shows the new file size. For a scan, expect a large drop.
5. Open the result and verify legibility. Look at the smallest text on the busiest page. If it’s crisp, you’re done. If it’s too soft, step back to a gentler level and re-compress from the original.
6. Download. Save your slimmed-down scan.

That’s the whole process for a single pass. Because Compress PDF targets the images directly, a scan that started at tens of megabytes typically lands in single-digit megabytes.

Hitting a specific target size

Often you have a hard limit — “under 5 MB for the portal,” “under 10 MB for email.” The reliable method is to step and check rather than guess one perfect setting:

1. Compress at a moderate level with Compress PDF and note the resulting size.
2. If it’s still over your target, increase the strength (lower the DPI further) and compress again from the original.
3. Because the images dominate the file, each step down in resolution produces a big drop in size — so you usually reach a target like 5 MB within one or two attempts.
4. Once you’re under the limit, open the file and confirm the smallest text is still legible.

The one rule that makes this work: always compress from the original scan each time, not from an already-compressed version. Re-compressing a compressed file loses quality faster without saving proportionally more space. If you find you’ve over-shrunk, go back to the original and use a gentler level rather than trying to “fix” the degraded file.

For a comparison of compressors and what to expect from each, our best free PDF compressor roundup covers the landscape.

Compression and OCR: get the order right

Many people want a scan to be both searchable and small. Both are achievable, but the order matters.

If you’ve already run OCR to add a searchable text layer (see what is a searchable PDF for what that means), compressing the file preserves that text layer — a good compressor only re-encodes the page images, and the recognized text is stored separately. So you can compress an OCR’d scan and still search and select its text afterward.

The recommended sequence when a document needs both:

1. OCR first with OCR PDF, so recognition runs on the full-resolution images for the best accuracy and adds the searchable text layer. The step-by-step is in how to OCR a scanned PDF.
2. Compress second with Compress PDF, which slims the result while keeping the text layer intact.

Doing it the other way around — compressing aggressively first, then OCR’ing — can hurt recognition accuracy, because OCR is reading degraded, lower-resolution images. OCR first, compress second.

When the file is still too big

If you’ve compressed hard and a scan is still over your limit, a few options remain:

• Split it. A long scanned document can be broken into part 1 and part 2 and sent as two files. This is the simplest fallback for a genuinely enormous multi-page scan.
• Drop unneeded pages. If half the scan is blank backs or pages the recipient doesn’t need, remove them before compressing.
• Extract pages as images if that’s all you need. If the recipient only needs a couple of pages as pictures rather than a full document, you can convert those pages with PDF to JPG and send compact image files instead of a heavy PDF. This is handy when someone just needs to see a page (a signed signature page, a single form) rather than receive the whole document.
• Re-scan in black-and-white if the original was scanned in color unnecessarily — sometimes the cleanest fix is a better scan, not more compression.

Why doing this in your browser protects your documents

Here’s the part most “compress scanned PDF online” sites don’t mention: many upload your file to a server to shrink it. Think about what scans usually are — the reason you scanned a paper document is almost always that it was an official or personal record:

• Signed contracts and agreements
• Tax forms and financial statements
• Medical records and insurance letters
• IDs, passports, certificates
• Bank statements and legal filings

Those are among the most sensitive documents you own, and uploading one to an unvetted server to save a few megabytes is a poor trade. The in-browser approach removes the risk: Compress PDF does the downsampling and re-encoding on your own machine, in JavaScript and WebAssembly inside your browser tab. The file is read from your disk, processed locally, and offered for download — it never travels over the network, never lands on a server, and is gone when you close the tab. No account, no watermark. You can shrink a scanned contract or a scanned ID for email or upload without ever handing the original to anyone.

Conclusion

Scanned PDFs are huge for one simple reason: every page is a high-resolution image, not text, so the file size comes from pixels and DPI rather than from words. The fix is to downsample the images to a resolution that matches how you’ll use the document — around 150 DPI for screen and email, higher for printing — which routinely shrinks a scan by 80–95%. Work down in steps to hit a target size, always compress from the original, OCR before you compress if you need searchable text, and check the smallest text before relying on the result.

Do it with the free, in-browser Compress PDF tool and your scanned documents never leave your device. Ready to slim down that 48 MB scan? Open Compress PDF and shrink it in seconds — no upload required.

Frequently asked questions

Related articles