Automatically organize invoices and receipts on Windows

You can automatically organize invoices and receipts on Windows without relying on the file name — by reading the text inside each PDF. That’s the difference between hunting for document (3).pdf at tax time and having every receipt already filed by month, tagged, and ready for a return. This guide shows you how.

The problem: the file arrives with a useless name#

Anyone who freelances or runs a small business knows the scene. The store emails an invoice and the file is called inv_00019283.pdf. The bank sends a statement as stmt_99812.pdf. The vendor portal downloads download (12).pdf. None of those names say what the file is — so no name-based rule can organize them. You open each one, rename it, drag it to the right folder. Every week.

The good news: the document knows what it is. Every order confirmation prints “thank you for your order” or “order confirmation.” Every invoice carries “invoice number” and “amount due.” Those markers live in the text, even when the file name is junk. The app just has to read it.

Content conditions: the app reads the PDF for you#

Elegant File Explorer has conditions that look at the file’s content, not just its name — for PDF, DOCX, and TXT. When a rule uses “Content contains (PDF, DOCX, TXT)”, the app reads the document’s text on its own. And here’s the detail that changes everything: it works even on scanned PDFs, with built-in OCR that runs 100% on your PC. A receipt that came in as a scanned image is read just the same.

The first pass over a full folder can take a few minutes — the app is reading every document once. After that, each new file that arrives is recognized instantly.

A note on accuracy: content matching is by text fragment. A loose word like “invoice” shows up in any contract that mentions “payment against invoice” — so good rules aim at the fixed phrases the real document prints: “thank you for your order”, “order confirmation”, “invoice number”, “amount due”. That precision is what keeps a genuine receipt from being confused with a document that merely talks about one.

Step by step: a receipt that files itself#

This is the tag + move flow. Open Auto-organization, click + New rule, and fill in the wizard:

1. Rule name: Receipts.
2. Monitored folders: click + Add folder, Browse…, and pick Downloads (you can add the Desktop too).
3. When to run: “Automatically, when a file arrives (real time)”.
4. Which files (conditions): add two conditions with + condition, both matching under AND:
   • “Extension is” with the value .pdf (so only PDFs are considered).
   • “Content contains (PDF, DOCX, TXT)” with the value order confirmation.
5. What to do (actions): add, in order, with + Add action:
   1. “Apply tag” → Receipt.
   2. “Move to” → destination Documents\Receipts\{year}\{month-name} (the year/month placeholders sort everything chronologically).

Click “Simulate effect” to see the Preview — nothing is actually changed — and only then “Save rule”. From there, download (12).pdf that carries “order confirmation” becomes Documents\Receipts\2026\July\download (12).pdf, tagged Receipt.

Tax forms and statements, by their content#

The same idea scales to the paperwork you dread. A W-2 always prints “Wage and Tax Statement”; letters and forms from the IRS print “Internal Revenue Service,” “Form 1099,” or “Form 1040.” A rule that combines “Extension is” .pdf with “Content contains (PDF, DOCX, TXT)” on those phrases can file your whole tax season into Documents\Taxes\{year}, tag it, and even create a reminder for the deadline — with “Create reminder (tag)” — no matter what the file was named.

Bank statements work the same way, matching phrases like “account summary” and “statement” inside the PDF.

The sharpest fingerprint: the official number the document carries#

Sometimes even a fixed phrase isn’t enough — a contract might mention an invoice, but only the real invoice actually prints its official invoice number, its purchase-order code, or the tax ID on a form. The app can recognize a file by that number, the fingerprint no lookalike document carries. It’s the safest way to tell the genuine receipt or tax form apart from anything that merely talks about one.

You don’t have to configure anything special: the content recipes already know these markers. For advanced cases, the content conditions also accept custom patterns — but for everyday paperwork, the ready-made recipes cover it. Everything else in the recipe — the tag, the reminder, the “Move to” destination with {year}\{month-name} — works exactly as described above.

The shortcut: the ready-made recipes#

You don’t have to build any of this by hand. In the Recipe Gallery, the content pack ships it ready:

• “USA/UK: receipts by content” — reads each incoming PDF, recognizes “thank you for your order,” “order confirmation,” “invoice number,” and “amount due,” and files receipts into Documents/Receipts by year and month with the Receipt tag.
• “USA: taxes by content (W-2, IRS)” — reads the text (even scanned, all on your PC) and files W-2s and IRS paperwork into Documents/Taxes by year, tagged Tax, with a deadline reminder.
• “Contracts by content (PDF and DOCX)” — recognizes the anatomy of a contract (“between the parties,” “hereinafter”) inside PDFs and DOCX and files them into Documents/Contracts.

Prefer to organize by name (faster, if you already name files well)? The by-name versions exist too: “USA/UK: receipts and order confirmations”, “USA: tax forms (W-2, 1099)”, and “USA/UK: bank statements”. The content recipes complement those — they catch exactly the files that arrive badly named.

All of them run through the simulation first and let you “Undo” afterward. And, as with any automation here, nothing is deleted: moving is moving, and the history logs every run so you can revert in one click.