Under the hood the compressor resamples embedded images to a DPI target — 150 for Balanced, 96 for Email & web — re-encodes them with the browser’s JPEG encoder, and keeps a new image only when it is at least 8% smaller than the original. Every level also rewrites the file with object streams and stores duplicated resources once, which is how a merged document with the same logo on every page loses weight without any visible change. Text is left as text. Nothing is uploaded: the document stays in the browser tab from drop to download, and the original file on your disk is not modified.
How it works
- Drop the PDF and open Compress
- Compare the three computed sizes with the 1 MB limit
- Download the largest level that fits; keep the original
Questions people ask
Which level should I try first for a 1 MB cap?
Balanced. It is the recommended level, keeps images at 150 DPI — indistinguishable from the original at everyday sizes — and lands most short and medium documents under 1 MB. Only drop to Email & web if the Balanced card shows a number above the cap.
Will a compressed file still be searchable?
Yes. Compression touches embedded images only; the text layer, links, bookmarks and form fields are carried over unchanged, so a document you could search before is just as searchable afterwards. A scan that was OCR’d before compression keeps its invisible text layer, and a scan that has none can be OCR’d after compressing.
Does it work on very large files?
Compression needs the whole file in memory, so the practical limit is your device rather than a cap of Bindery’s. Files above 200 MB switch to big-file mode, which streams Split and Extract but not Compress — split such a file first, then compress the parts.