Invoice & AP · OCR

Rename PDF invoices based on their content

Most bulk renamers work from the outside of the file: they stamp on the date you saved it, or number the files in the order you dropped them into a folder. That produces tidy-looking names that tell you nothing about the invoice inside. Renamer.ai works the other way around. It reads the content of each invoice PDF with OCR, finds the vendor, the invoice number, the date, and the total, and builds a descriptive filename out of those facts - so Scan047.pdf becomes 2024-11-15_AcmeCorp_INV-2847_$3200.pdf. It is filename generation driven by what the document actually says, applied to one invoice or a backlog of hundreds in a single pass.

The invoice fields Renamer.ai reads from the content

Renaming by content starts with reading the content. Renamer.ai extracts the fields below from the invoice itself - text PDF or scanned image - and any of them can be dropped into the filename. It reads only these filename-relevant fields; it is not a line-item capture or accounting-export tool.

FieldExample
Vendor / supplier nameAcmeCorp
Invoice numberINV-2847
Invoice date2024-11-15
Amount / total$3,200.00
Due date2024-12-15
PO numberPO-11842
CurrencyUSD
Tax / VAT amount$256.00
Bill-to nameNorthwind Ltd
Account numberACCT-55210

Before and after: three real invoice PDFs

The input names carry no meaning - a scanner counter, a browser download, a phone-photo counter. Reading the content turns each into a name you can search, sort, and reconcile.

Text PDF straight off the scanner
Scan047.pdf2024-11-15_AcmeCorp_INV-2847_$3200.pdf
Invoice downloaded from a supplier portal
download.pdf2025-01-09_Northwind_INV-5521_$780.pdf
Scanned image-only invoice (OCR)
img_0002.pdf2025-02-02_Meridian_INV-118_$1450.pdf

How content-based invoice renaming works

From a folder of meaningless filenames to invoices named after what is inside them, in four steps.

  1. 1

    Add your invoice PDFs

    Point Renamer.ai at your AP inbox, scanner output, or Downloads folder. Drop in one invoice or an entire backlog of hundreds - it processes the batch in one pass.

  2. 2

    Renamer.ai reads each invoice

    For text PDFs it reads the embedded text; for scanned or image-only PDFs it runs OCR to reconstruct the text first. Either way it extracts the vendor, invoice number, date, total, and the other fields from the content itself.

  3. 3

    Pick or build a naming template

    Choose a date-first or vendor-first template, or compose your own from variables like {vendor}, {date}, {invoice-number}, and {amount}. The template decides which extracted fields land in the name and in what order.

  4. 4

    Files are renamed on disk

    Each invoice is renamed in place with a descriptive, consistent filename. Nothing is uploaded, and the file stays exactly where your folders, backups, and accounting tools expect it.

Two ready-to-use invoice naming templates

Copy one as-is or build your own from variables such as {vendor}, {date}, {invoice-number}, and {amount}.

Date-first (chronological filing)

{date}_{vendor}_{invoice-number}_{amount}
Result:2024-11-15_AcmeCorp_INV-2847_$3200.pdf

Teams that close by period and want every invoice to sort by date on sight.

Vendor-folder (per-supplier archive)

{vendor}/{date}_{invoice-number}
Result:AcmeCorp/2024-11-15_INV-2847.pdf

Bookkeepers who reconcile one supplier at a time and keep a folder per vendor.

Content-based renaming vs. rename-by-date and rename-by-order

Most bulk-rename utilities never open the invoice. A rename-by-date tool reads the file's modified or created timestamp and stamps it on the name, which tells you when the file touched your disk, not when the invoice was issued - and those two dates are often weeks apart. A rename-by-upload-order tool is worse for invoices: it simply counts the files as they arrive and produces invoice_001.pdf, invoice_002.pdf, and so on, a sequence that is meaningless the moment the folder is re-sorted or a new batch is added. Both approaches move the mess around instead of clearing it, because the one thing that identifies an invoice - its vendor, number, and total - is sitting inside the document where a filename-only tool never looks.

Renaming based on content flips that. Renamer.ai reads the invoice the way a person would: it locates the supplier's name on the letterhead, finds the invoice number the supplier assigned, reads the issue date printed on the page, and picks out the grand total from among the subtotals and tax lines. Those extracted facts, not the clock or the upload counter, become the filename. The result is a name that is correct no matter when you saved the file or what order you processed it in, and that stays correct if you ever re-import or re-sort the folder.

The practical payoff shows up the first time you search. A folder named by date-of-save or by upload order forces you to open files one by one to find the Acme invoice for November. A folder renamed by content lets you type AcmeCorp or INV-2847 and land on the exact file, because the fact you are searching for is the filename. That is the difference between names that merely look organized and names that actually make an invoice findable.

Text PDFs and scanned image invoices, handled the same way

Invoice folders are rarely uniform. Some files are born-digital PDFs exported from a billing system with a clean embedded text layer; others are scans, faxes, or phone photos saved as image-only PDFs with no text inside at all. A renamer that only reads embedded text quietly fails on the second kind, leaving your worst files - the ones that cost the most time to sort by hand - exactly as anonymous as it found them.

Renamer.ai treats both the same. When a PDF has a text layer it reads it directly; when it does not, it runs OCR to reconstruct the text from the image first, then extracts the same vendor, number, date, and total fields it would from any other invoice. So a crisp portal download and a crumpled scan of the same invoice both come out named 2025-01-09_Northwind_INV-5521_$780.pdf. You do not have to sort your folder into 'readable' and 'scanned' piles before you start, and you do not have to retype anything the OCR reads.

Because the OCR and extraction run in the Renamer.ai desktop app on your own machine, the invoice content never needs to be uploaded to a web service just to be read and renamed. For documents that carry supplier identities and payment amounts, keeping that reading local is often the point.

Renaming a whole backlog of invoices in one pass

The hardest version of this job is not one invoice - it is the folder of hundreds that has been accumulating under names like download.pdf, download(1).pdf, Scan047.pdf, and img_0002.pdf. Renaming those by hand means opening each file, reading the vendor and number, and typing a new name, which is exactly the tedium that makes the backlog grow in the first place. Renamer.ai is built for that batch: point it at the folder, choose your template once, and it reads and renames every invoice in the folder in a single pass, mixing text PDFs and scanned images without you having to separate them.

That backlog run is where content-based renaming earns its keep most obviously. Every file comes out in the same convention - the vendor-first or date-first pattern you picked - so a folder that was unsearchable in the morning is fully searchable by the afternoon, with no field retyped by hand. And once the backlog is clear, a Magic Folder can keep it that way: set it on your AP inbox or scanner output and each new invoice is read and renamed the moment it lands, so the pile never rebuilds.

Throughout, the boundary stays clear. Renamer.ai reads the invoice to build a descriptive filename; it does not capture line items, post entries, approve or pay invoices, or store them in a portal. The output is simply your own files, renamed in place, sitting where your accounting tools already expect them. For content-based invoice OCR in the broader sense, see the invoice OCR software hub, or the related workflows for OCR invoice processing and rename invoices for QuickBooks.

Rename-by-content invoice FAQ

How is this different from a rename-by-date or rename-by-upload-order tool?

A rename-by-date tool reads the file's timestamp and a rename-by-upload-order tool just counts files as they arrive - neither one opens the invoice, so the name never reflects the actual vendor, number, or total. Renamer.ai reads the content of the invoice itself and builds the filename from what the document says, not from the clock or the order you processed the files in. That is why the name stays correct even if you re-import or re-sort the folder.

What if the invoice is a scanned image, not a text PDF?

Renamer.ai runs OCR on scanned and image-only PDFs to reconstruct the text before it extracts the fields, so an image invoice is read and renamed just as reliably as a born-digital one. A scanned img_0002.pdf comes out as 2025-02-02_Meridian_INV-118_$1450.pdf, the same as if it had arrived as clean text. You do not need to separate scanned invoices from text ones first.

Can I set my own filename pattern (vendor-first vs date-first)?

Yes. You build the pattern from variables such as {vendor}, {date}, {invoice-number}, and {amount} and arrange them however you like. Use {date}_{vendor}_{invoice-number}_{amount} for a chronological archive, or {vendor}/{date}_{invoice-number} to file each invoice in a folder per supplier. You decide which extracted fields appear and in what order.

Does it read the invoice content or guess from the original filename?

It reads the content. The original filename - Scan047.pdf, download.pdf, img_0002.pdf - carries no useful information, so Renamer.ai ignores it and reads the vendor, invoice number, date, and total from inside the document. The new name is built entirely from what the OCR and extraction find on the page, not from the name the file already had.

Does it store or process payment data?

No. Renamer.ai extracts only the filename-relevant fields - vendor, invoice number, date, total, and the like - to build a descriptive name. It does not process, approve, or store payment information, capture line items, or export to accounting systems. The reading happens in the desktop app on your own machine, and the only output is your own files renamed in place.

Can it batch-rename an existing backlog of invoices?

Yes, that is a core use case. Point Renamer.ai at a folder of hundreds of invoice PDFs, pick your template once, and it reads and renames the whole batch in a single pass, handling text PDFs and scanned images together. A Magic Folder can then keep the folder tidy going forward by renaming each new invoice as it arrives.