HowtoreducePDFsizewithoutlosingquality

The savings that are actually free

• Removing pages you do not need. The largest saving available, and it costs nothing at all.
• Cropping a scanner border. Bytes describing the edge of a scanner lid buy you nothing.
• Deleting blank backs from a duplex scan. Half the pages of many scanned documents are empty.
• Greyscale on a black-and-white document. There is no colour to lose.

Do all four before compressing. People habitually reach for the compressor first and then wonder why the result looks soft.

The saving that costs you something

Re-rendering pages as optimised images is what produces the dramatic reductions on scans — and it replaces the text layer with a picture of text. The page looks the same; it is no longer searchable, and text cannot be copied out of it.

For a scan that never had a text layer, this costs nothing real. For a report you may need to search later, keep the original and treat the compressed copy as the one you send.

Find out why it is large first

Divide the file size by the page count. A few kilobytes a page is a text document; a megabyte or more a page is a scan or an image-heavy design.

That number tells you which problem you have. Compression is transformative on the second and nearly useless on the first — where a large file usually means embedded fonts, a full-page background, or a print-resolution logo repeated on every page.

When compression is the wrong tool

• The file is a text document already — look at what is embedded instead.
• The document has to stay searchable — split it, or send it as a link.
• You have already compressed it once — a second pass loses clarity and saves almost nothing.
• The limit is per file and the document is genuinely long — split it and send it in parts.