Compress PDF

The photographs soften; the text does not. Smart re-encodes images at around 150 DPI, which is usually indistinguishable on screen. Strong goes to about 96 DPI and you will see it on detailed images and scans. Neither one touches your text or vector graphics, so headings, body text, tables and line art stay as crisp as they were. Maximum is the exception: it turns each page into a picture, so everything on the page — text included — takes on the quality of that picture.

Use our dedicated Compress PDF to 1MB tool (and its 500KB/2MB siblings) for a target-size workflow — they attempt the target automatically instead of asking you to pick a compression level.

Because there was not much to remove. Compression mostly targets images, and text and vector graphics already take very little space. A PDF made of scanned pages or photographs usually shrinks dramatically; one that is mostly typed text may barely change. A file that has already been compressed once has little headroom left either — running it through again rarely helps.

Not with Smart or Strong. Those two levels only replace the images inside the file: the page content is left exactly as it was, so the text is still real text and can be selected, copied, searched and read by a screen reader afterwards. Links, form fields and bookmarks survive too. One level does destroy it — Maximum renders each page to a picture, and the page warns you before you run it. There is no OCR here, so that change cannot be undone; keep your original if you might need the text back.

Aim comfortably below your provider's cap rather than right at it, because the message body, signature images and encoding overhead all add to the total. Gmail rejects attachments over 25MB and Outlook.com over 20MB, and many corporate mail servers are stricter still — 10MB is a safe target. If you cannot get under the limit, splitting the PDF and sending two messages beats compressing it into illegibility.