FINANCE

Expense Pre-Audit: OCR Receipts and Catch Amount Mismatches

When a receipt image is added to expense storage, this extracts the printed total via OCR, compares it to the claimed line-item amount in Airtable.

CategoryFinance
EngineSim + Paperclip
Difficultyadvanced
Triggerevent
Steps5
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerReceipt file uploaded to Drive expense folderGoogle DriveGoogle Drive
  • ActionOCR receipt for total and merchantOpenAI
  • ActionLook up matching expense line itemAirtableAirtable
  • LogicReconcile OCR total vs. claimed amount within tolerance
  • OutputFlag mismatches to finance channel with receipt linkSlack

What it does

Verifies that the amount an employee claimed actually matches what the receipt says. When a receipt file lands in cloud storage, the flow OCRs the document, pulls the printed total, and reconciles it against the claimed amount on the matching expense line item. Differences beyond a small tolerance (rounding, tax) are flagged as potential over-claims.

When to use it

Use this when employees attach receipt images and you want an automated tie-out between the receipt total and the claimed amount, catching inflated or transposed figures the human eye misses.

How it works

  1. 1A new receipt file uploaded to the Google Drive expense folder triggers the run.
  2. 2An OpenAI vision step OCRs the receipt and extracts the printed grand total and merchant.
  3. 3The flow looks up the matching expense line item in Airtable by report ID or filename.
  4. 4A reconciliation check compares the OCR total to the claimed amount against the allowed tolerance.
  5. 5A branch passes matches silently and routes mismatches to the finance channel with both figures and the receipt link for review.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect Google DriveDocs, sheets, slides, files.
  2. 2
    Connect OpenAIModels, embeddings, files.
  3. 3
    Connect AirtableBases, tables, views, automations.
  4. 4
    Connect SlackChannels, DMs, threads, mentions.
  5. 5
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  6. 6
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  7. 7
    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.