A PDF that is too large can bounce back from email servers, fail to upload to job portals, or eat up your phone’s storage. The good news: you can compress a PDF online for free and shrink it dramatically without any visible quality loss. Here is how to do it, plus what to try when the file is still too big.
An online compressor is the fastest option and works on any device with a browser.
Mac users can shrink a PDF without any third party tool.
Be aware that Apple’s Reduce File Size filter is aggressive, images can look noticeably softer afterward. If quality matters, an online compressor with a quality slider gives you more control.
Windows has no built in PDF shrinker, so the practical free options are a browser tool or the compress/optimize feature inside a free desktop PDF reader. For a one off file, the online method takes under a minute. If you compress PDFs regularly and prefer working offline, look for a free desktop app with an “Optimize” or “Reduce file size” option, which recompresses images and strips redundant objects from the file.
Open the online PDF compressor in your phone’s browser, upload the file, compress, and download the result. This works on both iPhone and Android with no app install. If the original is a photo heavy document you scanned yourself, you can also shrink the images before scanning to keep the PDF small from the start.
Images are almost always the culprit. A scanned document saved at high resolution, or photos pasted into a report, can push a file to tens of megabytes while the text itself takes up almost nothing. Compression works by recompressing those images at a sensible resolution, removing embedded thumbnails and duplicate font data, and flattening unused layers. Text only PDFs barely shrink because there is nothing redundant to remove, and that is normal.
Scanned documents deserve special attention because every page is a full image. If you control the scanner, choose 150 to 200 DPI for text documents instead of 300+ DPI, the result is perfectly readable and the file starts out much smaller. For black and white text scans, saving pages as black and white rather than color or grayscale cuts the size further. If the scan is already done, a compressor with an image quality slider lets you dial the resolution down after the fact.
Here is a common real world case. You scanned a 24 page signed contract on a home scanner at its default settings and the PDF is 18 MB. The job portal you need to upload it to accepts a maximum of 5 MB, so the upload fails. Here is how the numbers typically play out with the PDF compressor:
Why did the scan start so large? A colour A4 page scanned at 300 DPI is an image of roughly 2,480 × 3,510 pixels. At 150 DPI the same page is about 1,240 × 1,755 pixels, a quarter of the pixel data, and for plain text documents it reads identically. That is the whole trick behind the second pass. And if even the strong setting had not reached 5 MB, the fallback would have been to split the PDF into two halves of about 2.3 MB each and upload them as separate attachments where the portal allows it.
| Setting | Best for | What it does | Watch out for |
|---|---|---|---|
| Light | Files slightly over a limit; documents you will print | Removes unused data, duplicate fonts and thumbnails; barely touches images | May not shrink an image heavy scan enough on its own |
| Balanced | Everyday emailing and uploads | Recompresses images at a moderate resolution while keeping text sharp | Photos lose a little fine detail at high zoom |
| Strong | Strict portal limits; scans that start out huge | Renders page images at a noticeably lower resolution | Fine print and photos can soften, always inspect the result |
| Preview’s Reduce File Size (Mac) | A quick one off shrink with nothing to install | Applies Apple’s fixed Quartz filter | No control over strength; often softer than a balanced online pass |
The working rule: use the lightest setting that gets you under your target size, and treat “strong” as a tool for scanned documents rather than for files full of photographs you care about.
It helps to know why a 24 MB file can fail against a “25 MB limit.” Email does not carry files in their raw form, attachments are encoded as text (a scheme called base64) so they survive the journey between mail servers, and that encoding makes the data roughly a third larger than the original file. A 20 MB PDF can therefore occupy about 27 MB inside the actual email message and bounce off a limit you thought you were safely under. Portals and upload forms do not have this overhead, but they set their own caps, often much lower than email’s.
Two practical consequences follow. First, aim comfortably below any stated limit rather than just under it, for email attachments, targeting two thirds of the limit leaves room for the encoding overhead. Second, when a recipient says “just send the PDF,” the compressed version is a courtesy as much as a necessity: it downloads faster on a phone, opens faster on an old laptop, and does not eat a noticeable bite of someone’s mailbox quota. Compression is rarely about your file. It is about every system and person the file has to pass through on its way to being read.
If you send large documents regularly, build the habit at the source: scan at 150 to 200 DPI in black and white for text, save compressor ready originals separately from the compressed copies you send, and name the copies clearly (contract compressed.pdf) so you never overwrite or confuse the two. That small routine prevents nearly every “file too large” moment before it happens.
Two minutes of diagnosis tells you which fix will actually work. Start with the simplest ratio: file size divided by page count. A 30 page document at 3 MB is carrying about 100 KB a page, ordinary. Three pages at 15 MB means roughly 5 MB a page, which only happens with high resolution images: photographs, or scans saved at a scanner’s highest settings. That kind of file compresses beautifully, because nearly all of its weight is image data the compressor can work on.
The opposite diagnosis matters just as much. If a long file is heavy mainly because it has 400 pages of modest content, compression will shave a little and disappoint you, the honest fixes are removing sections you do not need or splitting it into parts. And if the PDF came from a design program, look for embedded extras: stock photos placed at full size and then shrunk on the page still carry their full data inside the file. Knowing which of these you are holding turns compression from hopeful clicking into a predictable result.
With a sensible compression level, no, text stays sharp and images look the same at normal zoom. Aggressive compression can soften photos, so use the lightest setting that gets you under your size target and always review the result before sending it.
It depends on the content. Image heavy scanned documents often shrink by 50 to 90 percent. Text only PDFs may barely change because there is little redundant data to remove.
Reputable tools process the file and delete it from their servers after a short time. For highly sensitive documents, prefer an offline desktop tool so the file never leaves your computer. Never upload files you are not authorized to share.
You need to unlock it first, which requires the password. See our guide on how to unlock a PDF you can only remove protection from files you are authorized to open.
The file probably contains mostly text or vector graphics, which are already compact. Try a stronger compression setting, or check whether embedded fonts and duplicate images are the issue, a desktop optimizer can strip those more aggressively.