INVOICE PROCESSING

Map Invoice Lines to GL Budget Categories in BigQuery

On each new vendor invoice, matches every line item to a GL budget category using your BigQuery mapping table.

CategoryInvoice Processing
Enginesim
Difficultyintermediate
Triggerwebhook
Steps6
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerNew invoice received via webhookHTTP webhook
  • ActionQuery BigQuery for vendor GL mapping rulesGoogle BigQueryBigQuery
  • LogicScore each line against GL categories by confidence
  • LogicBranch: confident match vs. uncategorized
  • ActionInsert matched lines into categorized-spend tableGoogle BigQueryBigQuery
  • OutputAppend unmatched lines to controller review queueGoogle BigQueryBigQuery

What it does

When a vendor invoice lands, this workflow reads its line items and matches each one to a general-ledger budget category using a mapping table in BigQuery (vendor + description keywords to GL code). Confidently matched lines are written to a categorized-spend table; lines it can't map are flagged as uncategorized for controller review.

When to use it

Run this when AP receives invoices faster than your team can hand-code them to the budget, and you want a single source of truth in BigQuery for categorized vs. uncategorized spend before month-end close.

How it works

  1. 1A new invoice arrives via webhook from your AP intake system.
  2. 2The flow queries BigQuery for the vendor's historical GL mappings and active keyword rules.
  3. 3For each line, it scores the best GL category match against a confidence threshold.
  4. 4A branch splits confident matches from low-confidence or unmatched lines.
  5. 5Matched lines are inserted into the BigQuery categorized-spend table.
  6. 6Unmatched lines are appended to a controller review queue table with the invoice reference and reason.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect HTTP webhookTrigger any URL on agent actions.
  2. 2
    Connect BigQueryDatasets, queries, schemas.
  3. 3
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  4. 4
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  5. 5
    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.