Three-Way Match: POs, Receipts and Invoices
Short answer
Upload your open purchase orders, receiving log and vendor invoices to Claude and ask it to match each invoice line on price against the PO and on quantity against what was received. Ask for a matched list and an exceptions table, and review both before any bill is approved.
What the source describes
Microsoft Learn describes three-way matching as checking the price on a vendor invoice against the purchase order and the quantity on the invoice against the product receipts, with any difference outside a set tolerance flagged as a matching discrepancy. Its example is an invoice billed at 1.10 per unit against a PO price of 1.00 under a 5 percent price tolerance, which fails the match. The Claude Help Center says Claude can run code on uploaded CSV and other data files and create Excel spreadsheets with working formulas that you download from the conversation.
Result or outcome
Microsoft Learn says three-way matching flags invoices whose price or quantity falls outside your tolerance before they are paid, and the Claude Help Center says Claude can return the comparison as a downloadable Excel workbook with formulas you can inspect.
How to run it
- Export open purchase orders, the receiving log and unpaid vendor invoices as CSV, and swap vendor names for codes if your policy requires it.
- Set your price and quantity tolerances before you run the prompt, so the same rules apply to every vendor.
- Upload the three files to Claude with code execution and file creation turned on, then paste the prompt.
- Open the Exceptions tab and check each line against the PO and the receiving record before you hold or approve the bill.
- Send price disputes back to the vendor and file the workbook with the period's AP support. For the wider process, see How to use AI for accounts payable and the Invoices and AP hub.
Try it yourself
Copyable prompt
I have uploaded three CSV files for [company or client code]: purchase_orders.csv (PO number, line, item, quantity ordered, unit price), receipts.csv (PO number, line, item, quantity received, date received) and invoices.csv (vendor, invoice number, invoice date, PO number, line, item, quantity billed, unit price billed). Do a three-way match for every invoice line. Rules: unit price may differ from the PO by up to [2] percent or [$5] per line, whichever is smaller; quantity billed must not exceed quantity received to date. Return an Excel workbook with three tabs: Matched (lines that pass), Exceptions (lines that fail, with the reason: price over tolerance, billed more than received, no receipt yet, PO not found, or duplicate invoice number for the vendor) and Summary (count and total value of each exception type by vendor). Use formulas, not pasted values, for the variance columns so I can check them. Do not change any amounts from the files, and list any row you could not read.
Replace the [bracketed] placeholders. Remove client names, account numbers, and personal data before pasting into any AI tool.
What to review
Spot-check at least five matched lines by hand, including one partial receipt. Confirm the variance formulas point to the right columns, and treat any line Claude could not read as unmatched until you check it yourself.
Related free download: 50 AI Prompts for Bookkeepers (Prompt pack (PDF))
Source: Microsoft Learn (Accounts payable invoice matching overview); also Claude Help Center (Create and edit files with Claude); also Microsoft Learn (Three-way matching policies)
See also
Not professional advice. This is general information, not accounting, tax, legal or financial advice. Check AI output yourself and consult a licensed professional for your situation. See our Terms of Use.
This entry summarizes a public source. Features, plans, and availability change, so confirm details with the vendor. Tool names are trademarks of their respective owners.
Related workflows
- Accounts Payable
Auto-Code PO and Non-PO Invoices with Stampli's Billy the Bot
Stampli's blog walks through how its AI assistant, Billy the Bot, extracts invoice data and then codes it. For PO invoices Billy associates the purchase order, pulls GL codes and matches invoice details to the PO, flagging discrepancies; for non-PO invoices it identifies cost center and expense type and flags details it needs clarified, then learns to suggest the right approver over time.
Stampli, Billy the Bot
Stampli Blog (vendor article) - Reconciliations
Reconcile a bank statement export against your chart of accounts with Claude Cowork
Claude Academy walks through giving Cowork a folder with a checking account statement export and the chart of accounts, then asking it to match transactions to accounts, flag anything that does not line up, and draft journal entries for discrepancies. Claude may ask clarifying questions and build a plan for review, then returns the reconciliation logic, specific discrepancies and journal entries ready for review. Suggested follow-ups include exporting entries as CSV and creating a reconciliation summary for the records.
Claude, Claude Cowork +1
Claude Academy (Anthropic tutorial) - Reconciliations
Build a Formatted Reconciliation Workbook from CSVs with Claude
Anthropic's help article says Claude can execute code to create and work with files directly in a conversation, producing Excel spreadsheets, PowerPoint presentations, Word documents and PDFs that you can download or save to Google Drive. It says code execution and file creation are available on Free, Pro, Max, Team and Enterprise plans, on the web, Claude Desktop and Claude Mobile. Its examples include building financial models in Excel with working formulas and performing advanced analyses on uploaded data, which is what a reconciliation workbook built from bank and ledger exports needs.
Claude, Microsoft Excel
Claude Help Center - Reporting and Analysis
Build a Rolling 13-Week Cash Flow Forecast with AI and Excel
In this Mostly Metrics tutorial, the author records building a 13-week rolling cash forecast for a fictional software company from eight messy data tabs, including an invoice export with inconsistent terms, collections history, AP, a payroll calendar and one-off items. Claude first works out each customer's actual days-to-pay from collections history rather than nominal terms, then builds a forecast with beginning cash, collections, disbursements, net change and ending cash chained week to week. The model includes a covenant line with an automatic OK/BREACH flag.
Works in ChatGPT, Claude, Grok, Gemini
Microsoft Excel
Mostly Metrics (newsletter)
For more ready-to-use prompts for ledger checks like this one, the Finance Prompt Builder (Close and Reconciliation) is on our sister store.