Automations/Finance
Split a Scanned Stack Into Separate Documents
You put eleven invoices through the sheet feeder and got one PDF back. The scanner treated a stack of separate documents as a single document, because as far as it knows a stack is what it is.
The short answer
A sheet feeder produces one file per pass, not one file per document, so a stack of eleven invoices arrives as an eleven-document PDF. The Split Document step separates it into individual documents, and each one is then treated on its own terms — the Read Details from File step reads it, the Rename step names it from its own date and party, and the Move step files it. Splitting first is what makes the rest possible; a merged stack cannot be named, because it is not one thing.
Steps this uses
Before and after
What actually changes
As they arrive
- Scan_0051.pdf (11 invoices, 14 pages)
After the workflow
- 2026-08-01-invoice-northwind-4821.pdf
- 2026-08-01-invoice-acme-1180.pdf
- 2026-08-03-invoice-dataflow-0092.pdf
- … 8 more
One sheet-feeder pass becomes eleven documents, each named from its own page.
Setting it up
Described once, then it runs
This is the sentence. Send it to the builder and the steps below appear on a canvas, wired and named, for you to change before anything runs.
When a scanned stack lands in my scans folder, split it into separate documents, read the date and the vendor off each one, rename them date-first, and file each under its vendor.
Scan the stack as usual
No change to how you scan. One pass, one file, into a folder the workflow watches.
Split Document separates them
The step divides the scan into individual documents rather than individual pages, so a three-page contract stays whole.
Each document read on its own
The Read Details from File step runs per document, which is why each one can be named for its own date and party.
Named and filed individually
The Rename and Move steps then act on each document separately, in place in your cloud storage.
Documents, not pages
The distinction that matters is between splitting by page and splitting by document. Splitting a fourteen-page scan into fourteen files is usually wrong, because a three-page contract is one document. The Split Document step separates documents, so multi-page items stay intact.
Why splitting has to happen first
A file containing eleven invoices has no single date, no single party, and no correct name. Any attempt to name it is a choice between eleven answers. Splitting resolves that before the naming question is asked, which is why the order of the steps is not arbitrary.
Stacks that arrive by email
Batches sent by a bookkeeper or a scanning service arrive as attachments. The Save a Copy step gives the attachment a home in cloud storage first, and the split proceeds from there exactly as it would for a scan from your own machine.
Then build
Variations worth building next
When a scanned stack lands in my scans folder, split it into separate invoices, read the vendor, invoice number, date and total off each one, file it under the vendor, and send those fields to our accounting webhook.
Gets each invoice's figures into the accounting system without anyone retyping them.
When a scanned stack lands in my scans folder, split it into separate documents, rename each one from its date and vendor, and ask me to approve only when the vendor is unclear before filing it under that vendor.
Keeps routine invoices moving and stops a misread vendor from filing to the wrong place.
When a scanned stack lands in my scans folder, split it into separate documents, rename and file each one under its vendor, then move the original combined scan to the trash.
Leaves one copy of each invoice instead of the stack plus its parts. Trash stays recoverable.
Limits
What to know before you build it
It runs on new files, not your backlog
This workflow fires when a file arrives. It will not go back over the documents already sitting in the folder, so switching it on leaves the existing pile exactly where it is. Working through what is already there is a scheduled run instead, which is a different workflow.
Organize the files you already haveWhat you need to run it
Pointed at a folder in The Drive AI, this works on every plan including the free one, because the files are already here. Pointing it at a Google Drive, OneDrive, SharePoint or Dropbox folder instead means connecting that account, which is on the Max plan or any team workspace.
FAQ
Common questions
How do I split a scanned PDF containing several invoices?
The Split Document step separates a multi-document scan into individual documents. Each one is then read, named from its own date and party, and filed separately.
Does it split by page or by document?
By document. A three-page contract inside a scanned stack stays as one file rather than becoming three, which is the distinction that makes splitting useful rather than destructive.
Why not just name the whole scan?
Because a file holding eleven invoices has eleven dates and eleven parties. There is no correct single name for it, which is why splitting comes before naming.
Can it split scans that arrive as email attachments?
Yes. The Save a Copy step files the attachment into cloud storage first, since an attachment has no home of its own, and the split runs from there.
One pass in, eleven documents out.
Split, named and filed, each on its own terms.
5 GB free · No credit card required