Method 1: Rule-Based (Fixed Pattern or Zone)
The rule-based approach applies a fixed transformation, add today's date, insert a counter, or read text from a fixed rectangle on the page (a "zone") and drop it into the name. For a single vendor whose invoice layout never changes, this can work: the invoice number always sits top-right, so you read that box every time.
The problem is that invoices don't stay identical. A vendor redesigns their template, a field moves, a scan comes in slightly rotated, and the fixed zone reads blank or the wrong text, silently. And a pattern rule can't put a vendor name into a filename that never contained it. So rule-based renaming holds up only for one uniform source and falls apart across the mixed invoices most people actually have.
The failure is quiet, which is what makes it costly. A fixed page zone doesn't announce that it landed on the wrong box; it simply copies whatever characters happen to sit inside that rectangle now. When a supplier moves the invoice number two centimeters to accommodate a new logo, the zone might grab a phone number, a tax line, or nothing at all, and the file is still renamed and filed away looking perfectly valid. You often don't notice until weeks later, when you go looking for a specific invoice and the name no longer matches the document inside. Every vendor you add multiplies this risk, because each layout you support is one more rule you have to write, test, and maintain the day that vendor changes anything.
Method 2: Content-Based (Read Each Invoice)
Content-based renaming reads the invoice the way a person would. renamer.ai runs AI and OCR over each file, understands it's an invoice, and finds the vendor, invoice number, date, and amount wherever they sit, then builds the filename from them. Because it locates fields by what they say, not where they are, a redesigned layout or a rotated scan still reads correctly, and a folder of invoices from twenty different vendors comes out consistently named in one pass.
This is the practical advantage when your invoices come from many suppliers at once. You never have to sort the pile by vendor first, register each template, or maintain a separate rule per source. The same read that identifies a field on one vendor's layout identifies it on the next, so one messy folder of scans, exported PDFs, and forwarded attachments turns into one clean, uniformly named set. When a vendor eventually changes their design, nothing breaks: the tool simply reads the new page and keeps naming it correctly, with no rule for you to rewrite.
- Add your invoices, scans, vendor PDFs, and phone photos together, with no sorting first.
- Let AI and OCR read each one and extract the vendor, invoice number, date, and amount.
- Pick a naming template once, for example {date}_{vendor}_{invoice-number}_{amount}.
- Review the previewed names across the batch, then apply; low-confidence reads are flagged, not guessed.
- Optionally point a watched folder at your invoice inbox so new invoices are named automatically on arrival.
The Fields That Make an Invoice Filename Useful
A good invoice filename carries four things: the date (in a sortable format), the vendor, the invoice number, and often the amount, for example 2024-11-15_AcmeCorp_INV-2847_$3200.pdf. With those in the name, you can sort a folder chronologically, scan for a vendor at a glance, or search an invoice number and land on the right file, without opening anything. Content-based renaming produces exactly this because it reads those fields off the page; a rule-based tool can only include what was already in the filename.
The date's position in the name matters more than it looks. Put the date first in a year, month, day format, and your file manager sorts the whole folder into chronological order by default, so the newest invoices sit together and a month's worth of billing reads top to bottom without any manual arranging. Lead with the vendor instead and the folder groups by supplier, which is handy when you reconcile one account at a time. The point is that a name built from real fields lets you choose the order that fits how you actually look things up, whereas a counter like invoice_014.pdf tells you nothing about when it arrived or who sent it.