Compress PDF for email

Why this works

PDFRun's compressor strips redundant resources, downsamples images intelligently and re-encodes streams — typically reducing files by 50–80% without visible quality loss. The defaults below are tuned specifically for email-safe sizes.

The exact attachment limits vary by provider and are worth knowing precisely. Gmail allows attachments up to 25 MB; above that, the message is auto-converted to a Google Drive link — which some corporate firewalls and mailing lists reject. Outlook.com allows 20 MB inline, with a 33 MB hard ceiling for some account types. Yahoo Mail caps at 25 MB. Apple iCloud Mail caps at 20 MB. Most corporate Exchange and Microsoft 365 deployments enforce stricter caps — 10 MB is the most common policy, with some at 5 MB — set by the recipient's IT admin, not the sender's. If you don't know the recipient's cap, target 10 MB as a safe default: it clears virtually every modern email system without bouncing.

What actually drives PDF size. The largest contributor is almost always embedded images — a single high-resolution photograph at 300 DPI can be 5–10 MB on its own. PDFs from design tools (InDesign, Affinity, Figma exports) often embed photos at print resolution even when the final document is viewed on screen. Scanned pages are similar: each page is a full-resolution image (often 200–300 DPI in colour) so a 10-page colour scan is routinely 30–60 MB. Embedded fonts add 100–500 KB per font face but rarely dominate. Form fields, comments, and revision history add modest overhead but can build up in documents that have been edited many times.

What compression actually does. The Recommended setting downsamples images to 150 DPI — indistinguishable from 300 DPI on a typical desktop or laptop screen, but produces a 75% smaller image file. Fonts are subsetted (only the characters actually used are embedded, not the entire font's glyph table). Redundant resources (duplicate images embedded across multiple pages, unused form-field definitions, orphan annotations) are removed. Stream compression is applied to text and vector content. The net effect on a typical scanned report is 60–80% size reduction with no visible loss at normal viewing distances.

If the Recommended setting doesn't get you under your target, escalate to Maximum compression. It pushes images down to roughly 96 DPI — still readable, but you'll see softening on photographic content if you zoom in. For text-only documents Maximum makes almost no visual difference; for image-heavy decks it's a meaningful trade. If even Maximum doesn't clear the cap, the document genuinely contains more high-resolution image data than will fit — at that point your options are: split the document into smaller PDFs and send as a series, host on Drive/Dropbox and send a link, or convert colour scans to grayscale first (cuts size roughly in half) before re-compressing.

A few practical workflow notes. Don't re-compress an already-compressed PDF — you'll get diminishing returns and may introduce visible degradation. If you've compressed once and the result is still too big, escalate the compression level on the original rather than running compression twice. Test the result before sending — open the compressed PDF to verify text is still readable and images aren't visibly degraded; some compression settings can over-process scanned text. Preserve a copy of the original if the document is important; compression is not perfectly reversible.

How it works