INVOICE PROCESSING

Three-Way Match Vendor Invoices Against PO and Goods Receipt

When a vendor invoice lands in a shared inbox, matches it line-by-line against the purchase order and goods-receipt record in Airtable.

CategoryInvoice Processing
Enginesim
Difficultyintermediate
Triggerevent
Steps5
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerNew invoice email in AP inboxGmailGmail
  • ActionExtract line items, PO number, totals from PDFOpenAI
  • ActionLook up matching PO and goods receiptAirtableAirtable
  • LogicCompare qty, price, received amount vs tolerance
  • OutputMark Approved or Exception in Airtable queueAirtableAirtable

What it does

Receives an incoming vendor invoice, extracts its line items and totals, and reconciles them against the matching purchase order and goods-receipt entry held in Airtable. Invoices where quantity, unit price, and received amount all agree within tolerance are marked approved automatically; anything outside tolerance is held as an exception so AP staff never pay an unverified bill.

When to use it

Run this when your AP team manually opens every invoice, looks up the PO in a spreadsheet, and eyeballs whether the goods actually arrived. It removes the repetitive matching step and only surfaces the invoices that genuinely need a human.

How it works

  1. 1A new invoice email arrives in the AP Gmail inbox and triggers the flow.
  2. 2The invoice PDF is parsed into structured line items, vendor, PO number, and total.
  3. 3The matching PO and goods-receipt records are pulled from Airtable by PO number.
  4. 4A match check compares quantities, unit prices, and received-vs-billed amounts against the configured tolerance.
  5. 5Clean matches update the Airtable invoice record to Approved; mismatches set it to Exception with the failing fields noted.
  6. 6The result and any variance summary are written back to Airtable for the AP queue view.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect GmailRead, draft, send, label.
  2. 2
    Connect AirtableBases, tables, views, automations.
  3. 3
    Connect OpenAIModels, embeddings, files.
  4. 4
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  5. 5
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  6. 6
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.