You try to attach a PDF and the email bounces, or a government portal says the file must be under a certain size. The document itself is fine. It’s just too heavy.
Shrinking it is usually easy, but the right method depends entirely on what’s inside the file. A text PDF exported from Word and a 30-page phone scan need very different treatment, and using the wrong approach either does nothing or ruins the quality. Here’s how to tell them apart and which settings to choose.
Try it: PDF CompressionReduce PDF file size losslessly, or shrink scanned and image-heavy PDFs dramatically by re-encoding pages.
Open tool — freeWhy some PDFs are so big
A PDF is a container. The text, fonts, vector drawings and images inside it all add weight, but not equally.
- Text and vector graphics are tiny. A 50-page report with no photos is often well under a megabyte.
- Embedded fonts add some weight, especially if full font files were included instead of just the characters used.
- Photos and scans are the heavyweights. A scanned page is one big image, and a high-resolution color scan can be several megabytes on its own.
- Leftovers such as duplicate images, metadata, thumbnails and old revisions can bloat files that have been edited many times.
So the first question is simple: is this a document of text, or a stack of pictures of pages? Try selecting a word with your cursor. If you can highlight individual words, it’s text-based. If you can only drag a box around the whole page, it’s a scan.
Lossless compression vs re-encoding pages
There are two broad ways to make a PDF smaller, and our PDF compressor offers both.
Lossless compression
This rewrites the file more efficiently: it tidies the internal structure, compresses data streams better and drops unused objects. Nothing you can see changes. Text stays selectable, links keep working, and vector graphics stay sharp at any zoom.
The catch is that savings are modest, and sometimes close to zero on a file that was already well built. That’s normal, not a failure.
Re-encoding pages (rasterizing)
This renders each page as an image at a resolution you choose, then stores it as a compressed image. For scans and photo-heavy PDFs, the reduction can be dramatic, because you’re replacing oversized images with sensibly sized ones.
The trade-off is real. Every page becomes a picture, so text in a formerly text-based PDF is no longer selectable or searchable, links stop working, and small print can soften. Use this method for scans, or when the file has to get under a limit and nothing else works.
| Your PDF | Best method | What to expect |
|---|---|---|
| Exported from Word, Google Docs or a browser | Lossless | Small savings; file stays fully editable and searchable |
| Phone or scanner scan | Re-encode pages | Large savings; already an image, so little is lost |
| Brochure or report with many photos | Try lossless first, then re-encode | Re-encoding saves more but flattens the text |
| Contract or form you’ll sign or fill | Lossless only | Keeps form fields and text intact |
Choosing dpi and quality settings
When you re-encode, two settings decide the result: resolution (dpi) and image quality. Here’s how I’d think about them.
- 150 dpi is the sweet spot for most people. Text is crisp on a laptop or phone screen and files are a fraction of a typical scan.
- 200 to 300 dpi makes sense if the document will be printed, or if someone will run OCR on it later. Small print and fine lines survive better.
- 100 dpi or lower is for emergencies, when you must get under a strict limit. Check that the smallest text is still readable.
For quality, a medium JPEG setting is usually indistinguishable from high on ordinary documents. Push it lower only after you’ve tried lowering the dpi. And if the document is black text on white paper, grayscale is a cheap win: color scans of plain documents carry data you don’t need.
How to compress a PDF step by step
- Check what kind of PDF you have. Try to select a word. Text-based files go to lossless; scans go to re-encoding.
- Open the PDF compressor and add your file. It’s processed in your browser, so the document never leaves your device.
- Choose the method. Start with lossless for text PDFs. For scans, pick re-encoding and set around 150 dpi.
- Compress and compare. Note the new size against the limit you’re trying to meet.
- Inspect a page at 100 percent zoom. Look at the smallest text, any signatures and any stamps. If they’re fuzzy, go up a step in dpi or quality.
- Download and keep the original. You can’t recover detail from a compressed copy, so never overwrite the source file.
Other ways to get the size down
Compression isn’t the only lever, and sometimes it’s not the best one.
- Export smarter in the first place. Word and many other apps offer a smaller export option for online use. Re-exporting from the source often beats compressing afterwards.
- Remove pages nobody needs. Blank backs of scanned sheets, duplicate pages and long appendices all add weight. Our guide to extracting only the pages you need covers this.
- Scan at a sensible resolution. Many scanner apps default to very high settings. For everyday paperwork, 150 to 200 dpi is enough.
- Compress photos before making the PDF. If you’re building a PDF from phone pictures, see our guide on turning phone photos into a clean PDF.
- Split instead of squeezing. If a portal accepts several uploads, two readable files are better than one blurry one.
On a Mac, Preview has a built-in “Reduce File Size” export filter. It’s quick but can be aggressive, so check the result carefully before relying on it.
Mistakes to avoid
- Rasterizing a contract or form you still need to fill in. You’ll lose the fields.
- Compressing the same file over and over. Each lossy pass degrades it further; go back to the original instead.
- Assuming upload limits are the same everywhere. They vary by email provider and portal, so check the actual limit before you start.
- Sending a file you haven’t opened. A file that meets the size limit but is unreadable will just come back to you.
A sensible default
Try lossless first on any PDF. If it’s a scan or still too big, re-encode at 150 dpi with medium quality, then zoom in and read the smallest text before sending. Keep the original, and if you’re still over the limit, cut unneeded pages rather than lowering quality until the document is a blur.
Frequently asked questions
What dpi should I use to compress a scanned PDF?
For documents that will only be read on screen or attached to an email, 150 dpi is plenty. Use 200 to 300 dpi if the file will be printed or run through OCR later.
Will compressing a PDF make the text blurry?
Lossless compression never touches text. Re-encoding pages as images can soften small text at low resolutions, so check a page at 100 percent zoom before you send the file.
Why did my PDF barely get smaller?
Text-based PDFs exported from Word or a browser are usually already compact, so there is little to remove. Big savings come from scanned or image-heavy pages.
