The short answer
OCR (optical character recognition), usually combined today with AI-based field extraction, can read a supplier invoice, create a draft purchase invoice in ERPNext or Prometheus, match it to the purchase order and receipt, and route it for approval. In a well-designed set-up, clean invoices that match need only a quick confirmation, and a person's time goes into the mismatches, duplicates and unreadable scans. It is one of the most reliable automation wins in an ERP, but only if you design the review step, the matching rules and the tax fields carefully. Capture that posts straight to the ledger without review is not automation; it is a new source of errors.
How it works in an ERP
A typical flow in ERPNext or Prometheus looks like this:
- Intake. Invoices arrive at a dedicated accounts payable mailbox, a scanner folder or an upload screen.
- Extraction. The OCR or extraction service reads the supplier, invoice number, date, line items, subtotal, sales tax and total.
- Draft record. A draft purchase invoice is created, with the original file attached.
- Matching. The draft is matched to the purchase order and purchase receipt; quantity, price and tax differences are flagged.
- Review and approval. Clean matches go to the accounts payable clerk to confirm. Mismatches, new suppliers and invoices over a threshold go to an approver.
- Posting. A person submits the invoice, and it is scheduled for payment.
Example: three invoices on a Monday morning
- Invoice A from a regular packaging supplier matches its purchase order and receipt exactly. The clerk checks the highlighted fields and submits it.
- Invoice B from a freight company has no purchase order. It is routed to the operations manager to approve and code.
- Invoice C has the same supplier and invoice number as one entered last month. The system blocks it as a likely duplicate and flags it for review.
The US details that matter
Generic OCR tools handle a clean invoice well, but US accounts payable has a few specifics your capture needs to get right.
Sales and use tax. There is no federal VAT in the US. Sales and use tax is set by states and local jurisdictions, so one invoice may show state, county and city tax, a single combined line, or no tax at all. Your extraction should capture each tax line where it appears and flag invoices with no tax charged on taxable purchases, because your ERP's tax templates and any use tax accrual depend on it. Tax-exempt purchases also need the supplier to hold your exemption certificate on file.
Vendor tax information. Collect an IRS Form W-9 from vendors before you pay them, and store the taxpayer identification number, legal name and entity type on the supplier record. That information feeds year-end information returns such as Form 1099, and missing details are far easier to catch at onboarding than in January. Ask your accountant which payments need reporting.
Payment details. Invoices increasingly arrive with changed bank details, a common fraud pattern. Flag any change to ACH or wire instructions for call-back verification using a number already on file, not one printed on the invoice.
Invoices in other languages and formats. Overseas suppliers may send invoices in other languages, with other currencies and with day-first dates (04/05 can mean April 5 or May 4). Test extraction on those documents, including accented supplier names.
Keeping the original. Attach the original file to the ERP record, keep it for the retention period your accountant and counsel advise (the IRS expects records that support the figures on your returns, and state rules vary), and include it in backups. If you scan paper invoices, keep clear, complete images.
An audit trail. Auditors, lenders and tax authorities expect a trail from source document to financial records, and public companies have internal control reporting duties on top of that. Your set-up should show who confirmed or changed each extracted value. See the IRS for current record-keeping guidance.
This is general information, not tax advice; confirm your requirements with your accountant.
Rather talk it through? Send us a handful of real supplier invoices (with personal details removed) and we will show you how they would be captured, matched and approved in your ERP. Talk to a Promatics specialist
What usually goes wrong
- Automating before cleaning supplier data. Duplicate supplier records and missing tax numbers make matching unreliable. Clean the supplier list first.
- No purchase orders to match. If most spending has no PO, OCR saves typing but cannot validate much. Decide which spending categories should require a PO.
- Trusting confidence scores blindly. Extraction tools report how sure they are, but a confident misread of a total is still wrong. Key fields such as total, tax and invoice number should always be visible to the reviewer.
- Ignoring the long tail. Handwritten invoices, poor scans and multi-page statements will never extract cleanly. Give them a manual-entry route rather than forcing them through.
- Sending data somewhere you have not approved. Many extraction services are cloud-based. Know where documents are processed and stored, and whether that fits your privacy and contractual obligations.
- We know our monthly invoice volume and how many arrive as PDF versus paper.
- Our supplier list is de-duplicated, with W-9 details and tax-exempt status recorded.
- We have decided which spending requires a purchase order.
- Approval thresholds and approvers are agreed.
- Key fields (total, tax, invoice number, supplier) are always shown to the reviewer.
- Duplicate detection is tested.
- Changes to supplier bank details trigger a call-back check.
- We know where the extraction service processes and stores documents.
- Originals are attached to the ERP record and included in backups.
How we build it
We start by collecting a sample of your real invoices and agreeing what "done" looks like. We then choose an extraction approach that suits your volume, your languages and your data-location requirements, and build the capture, matching and approval flow in a separate custom app, so it survives ERPNext upgrades. Every build is tested against your own documents, including the awkward ones, and failures alert a named person rather than failing silently. The same approach extends to receipts, expense claims and other forms; see ERPNext and Prometheus customization.
When to bring in help
If you receive a few dozen invoices a month from a handful of suppliers, a standard add-on or careful manual entry may be enough. Bring in help when volumes are high, when invoices must be matched to purchase orders and receipts, when you need multi-jurisdiction sales and use tax handled correctly, or when an auditor will ask how you know each invoice was checked. That is where design, testing and accountability matter more than the OCR engine itself.
Sources and further reading
Product capabilities and guidance change. These are the primary sources this article relies on, checked on the review date above.
- Tax information for businesses, Internal Revenue Service
- Purchase Invoice, ERPNext documentation (Frappe)
This article is general information, not legal, accounting or security advice for your specific situation. Examples are hypothetical unless stated otherwise.