How to Reduce PDF File Size for Email (Without Wrecking It)

"Message too large." A 40-page scanned report, a deadline, and a mail server that will not budge. Reducing PDF size is simple once you understand where the weight actually is — and frustrating if you attack the wrong thing.

Know the limit you are fighting

Attachment caps are lower than most people assume, and they apply to the encoded message, which is roughly a third larger than the raw file:

• Gmail — 25 MB per message
• Outlook / Microsoft 365 — 20 MB by default, often reduced by administrators
• Corporate mail servers — frequently 10 MB, sometimes 5 MB

A practical rule: aim below 8 MB if you do not know the recipient's limit. For a 10 MB cap, a 10 MB file will still bounce because of encoding overhead.

Where the weight actually lives

This is the key insight: images are almost always the problem. Text is astonishingly compact — a hundred pages of pure text is often under a megabyte. A single full-page scan at 600 DPI can be larger than that on its own.

So the useful question is not "how do I compress this PDF" but "what images are in it, and do they need to be that big?"

Common causes of bloated files:

• Scans at unnecessarily high resolution. 600 DPI is for archival reproduction. 200–300 DPI is ample for reading and OCR.
• Colour scans of black-and-white documents. Scanning a typed letter in full colour can triple the size for no benefit.
• Embedded photographs at camera resolution. A 12-megapixel photo placed in a report is carrying far more detail than the page can display.
• Embedded fonts. Usually small, but a document using many fonts can carry several megabytes of them.

The two kinds of compression

Tools offer what are really two different operations, and conflating them causes disappointment.

Structural compression rewrites the file's internal object tables more efficiently. It is completely lossless, keeps text selectable, and typically saves only a few percent. Many "compress PDF" tools do only this, which is why people often report that compression "did nothing."

Image recompression re-renders the pages at a lower resolution and re-encodes them. This is where the real savings are — frequently 70–90% on scanned documents. The trade is genuine: once a page is re-rendered as an image, its text is no longer selectable.

Our Compress PDF tool exposes both, so you can choose deliberately rather than guess. It also reports the before and after sizes, and refuses to hand you a file that came out larger.

Choosing a level

• Keep text selectable? Use lossless. Accept modest savings.
• Sending a scan for someone to read? Balanced. Around 110 DPI reads perfectly well on screen and cuts size dramatically.
• Fighting a hard limit on a document that only needs to be legible? Strong. Visibly softer, much smaller.

One caution: if the recipient needs to search the document, or you will later convert it to Word, keep the text layer. Compressing to images and then needing the text means running OCR to get back what you deleted.

Things to try before compressing

• Remove pages nobody needs. Removing pages or extracting just the relevant section often solves the problem outright.
• Rescan at a sensible setting. If you control the scanner, 300 DPI greyscale is the sweet spot for text documents.
• Send a link instead. For genuinely large files, cloud storage is the right answer — though note that this does transmit the document to a third party, which may matter.

Privacy while compressing

Compression is often applied to exactly the documents that should not be casually uploaded: scanned contracts, medical letters, financial records. Most free compressors send your file to a server, where both the original and the compressed copy sit on disk for some hours.

Processing in the browser avoids that entirely, and has a pleasant side effect: because there is no upload, there is no file size cap. Server-based free tiers typically stop at 25–100 MB — which is exactly the point at which you most needed help.