My scans are named scan0001, scan0002 — can they still be sorted?

Yes, in two ways. The “Scanned documents by month” recipe recognizes them by that very naming (scan, scanner, CamScanner) and dates them. And content routing ignores the name entirely — it reads the text inside the scan, so a file called scan0043.pdf still gets filed as the invoice or contract it actually is.

Does reading the content send my documents anywhere?

No. The OCR runs 100% on your own PC — no upload, no account, no cloud. Reading a scanned page to recognize its text happens entirely on your machine, which is exactly what you want for sensitive paperwork.

Why is the first scan-folder pass slow?

Because the recognition is done for real, one file at a time, at a gentle pace so it never fights you for the computer. It’s a once-per-file cost: after a scan is read, its text is remembered, and later runs are instant. It only re-reads a file if the file changes.

Can I split invoices, contracts and medical results into separate folders?

Yes — that’s the point of content routing. A “Content contains” rule (or a ready recipe from the “Inside the file” pack) recognizes each type by words inside it and sends it to its own folder, tagged, even when every scan shares the same anonymous name.

What if a scan is recognized as the wrong type?

Simulate first and the preview shows exactly where each file would go, so you catch it before anything moves — and you can tune the words the rule looks for. After a run, “Undo” reverses it, and nothing is ever deleted. Available now on the Microsoft Store.

Organize scanned documents: from scanner to the right folder

Every scanner does the same thing. You feed it a page, it hands you back scan0001.pdf. Feed it another: scan0002.pdf. A month later your scan folder is a wall of scan#### files — an invoice, a signed lease, a lab result, a warranty, all wearing the same anonymous name, all in the same heap. When you actually need the electric bill from March, you’re opening files one by one like it’s a memory game.

The mistake isn’t scanning. It’s treating that scan folder as a destination when it’s really an inbox — a place things arrive, not a place they should live. This guide is about the flow that empties it: scan lands, gets recognized, gets filed, gets tagged, on its own. Not “where do scans go to die,” but “how does a scan reach the right folder without me.”

Treat the scanner like an inbox#

An inbox has one job: hold things briefly until they’re sorted out. Your email inbox works because rules and habits move messages onward. Your scan folder fails because nothing ever moves out of it — so it grows forever.

The fix is to point automation at that folder (or at Downloads, if that’s where your phone-scanner app drops files) and let two layers do the sorting: a first pass that clears the inbox into a dated archive, and a smarter second pass that reads what’s inside each scan and routes it by type. Here’s the whole flow at a glance:

Scan the pageLands in the inboxContent readFiled by typeTagged

Layer 1: empty the inbox into a dated shelf#

The first, simplest pass gets scans out of the inbox and onto a timeline. The ready-made recipe “Scanned documents by month” recognizes scans by name — scan, scanner, digitized, CamScanner — and files them into Documents\Scanned, split by year and month, tagged “Scanned.” That pile of scan0001 becomes a dated, browsable archive: Documents\Scanned\2026\July, and so on.

Turning it on takes under a minute:

Open Auto-organization and click + New rule — the wizard offers to start from a ready-made recipe.
In the Recipe Gallery, pick “Scanned documents by month.”
Point its Monitored folders at wherever your scans land — your scan folder or Downloads.
Click “Simulate effect” for the preview — nothing is actually changed — then “Save rule.”

That alone ends the “wall of scan0001” problem. But a folder-per-month still mixes an invoice with a contract with a medical result. For that, you read what’s inside.

Layer 2: route by what the scan actually says#

Here’s the part a scanner can’t do and a name-based rule can’t either: recognizing a document by its content. Elegant File Explorer can read the text inside your PDFs with built-in OCR that runs 100% on your PC — no upload, no account, nothing sent anywhere — even when the PDF is a photo of a page, which every scan is. It focuses on the opening pages, where the text that identifies a document lives: the invoice number, the parties’ names, the header of a lab report.

Once it can read the page, a rule can act on those words. The “Content contains” condition matches the text inside the file — scanned PDFs included — so you can send every contract to its own folder, every invoice to another, every medical result to a third, no matter what the file is named. The “Inside the file” pack has ready recipes for exactly this: “Contracts by content (PDF and DOCX)” recognizes a contract by words like hereinafter and between the parties and files it to Documents\Contracts; “USA/UK: receipts by content” catches order confirmations and invoices by their fixed phrases. For anything else — say, lab results — a rule you build yourself does the job: a “Content contains” condition on reference values sends every scanned test into Documents\Health, even the one named scan0043.pdf. (Patient records specifically get a dedicated walkthrough in organize patient files locally.)

Two honest notes about the OCR, because real recognition isn’t magic. Reading content is opt-in — you turn it on once with “Turn on deep search” in the search footer, since reading text costs memory and the choice should be yours. And the first pass over a folder full of scans takes a while: the app reads them one at a time, unhurried, so it never fights you for the machine. After that, each scan’s text is remembered, so it’s only ever read once — new scans are handled the moment they arrive.

Build the routing rule yourself#

Recipes are a starting point; your paperwork is specific. A rule that sends every scanned invoice to its folder looks like this:

Open Auto-organization and click + New rule.
Rule name: Scanned invoices.
Monitored folders: your scan folder (or Downloads).
Which files (conditions): add “Content contains” with a word that appears inside your invoices — invoice number, amount due.
What to do (actions): “Apply tag” → Invoice, then “Move to” → Documents\Invoices.

Simulate, save, and — when you trust it — promote “When to run” to real time so it happens the moment a scan lands. Layer it under the “Scanned documents by month” recipe and you get the best of both: content routing catches what it recognizes, the dated archive holds the rest.

How this differs from two neighbors#

This post is about the incoming flow — emptying the scanner inbox. Two related jobs are covered elsewhere, and pointing you to them keeps this focused. If your goal is the whole household’s paper life — rent, manuals, IDs, travel — sorted into home categories, that’s organize household documents. And if you’ve already filed your scans and now need to find a word buried inside one, that’s the search side: find text in a scanned PDF, 100% on your PC. Same OCR, different job — one files, the other finds.

Through all of it, the safety net holds: “Simulate effect” shows every move before it happens, “Undo” reverses any run, and moving is moving — never deleting. A scan filed to the wrong folder is one click from home.