Drop in the file that has everything mixed together.
Choose how to cut: a keyword that appears on the first page of every document, or the blank sheets your scanner picked up between them.
Download all the pieces at once.
This one needs an upload, and is deleted afterwards
Browsers cannot do this calculation, so the file goes to our server. It is deleted automatically 60 minutes after the job finishes, and the working copy lives in a memory filesystem that is wiped whenever the server restarts.
We do not log file names and do not log file contents — that is a hard rule in our engineering standards, not a setting.
What it does — and what it doesn't
Blankness is judged two ways. Pages with a text layer are judged on their text — even a lone page number counts as content. Only pages with no text layer are judged on pixels. In testing, a creased blank sheet is darker than a page with one line on it, so pixels alone cannot get both right.
Pick a keyword that appears on every first page and nowhere in the middle. "Invoice No." usually works; "Total" usually does not.
The pages it treated as separators are reported back so you can check the cuts.
A scan with no text layer will not match any keyword — run OCR first.
Common questions
It cut in the wrong places.
The report tells you where it cut. Try a different keyword, or switch to blank-page mode. Your original was never modified.
How many pieces can it produce?
No hard limit, but an implausible number stops and asks — that usually means the keyword was wrong, not that there really are 200 documents.
Can the file names include the keyword?
No. Output names are sequential. We do not record or reuse your file names, and putting document content into a file name would contradict that.
