ByteFlow AI Labs
All projects

Client project · Nova Digital Store

Receipt verification on WhatsApp — Nova Digital Store

A WhatsApp workflow that reads payment-receipt screenshots, checks them against the order ledger, catches repeat submissions and confirms or rejects the customer, with staff alerted for edge cases.

The problem

Customers sent payment-receipt screenshots on WhatsApp to complete their purchases. Staff checked each image by hand, matched the reference number against the order records and approved or rejected the order, which was slow on busy days.

What we built

A 69-node n8n workflow on the client's server:

  • Customers send payment-receipt screenshots on WhatsApp
  • OCR reads the amount and transaction reference
  • The details are checked against a Google Sheets order ledger
  • Repeat submissions of the same transaction are caught
  • The customer gets a confirmation or a rejection; staff are alerted for edge cases

Documentation and handover were delivered in the client's working language (Spanish). This client's payments were not bKash or Nagad, but the receipt-checking method is the same. No time-saved or accuracy figures are published.

Workflow

How the workflow is connected

Receipt check

  1. Receipt on WhatsApp
  2. OCR extraction
  3. Ledger check
  4. Duplicate detection
  5. Reply or staff review
Simplified view based on the delivered workflow; client system details not shown.

Errors and human handoff

  • Receipts that fall outside the agreed rules go to staff for manual review instead of being approved.
  • A repeat submission of the same transaction is flagged.
  • Staff are alerted only for the cases that need a person.

Stack

  • n8n (self-hosted)
  • WhatsApp Business (Evolution API)
  • OCR
  • Google Sheets
  • VPS

Want something like this for your business?

A free 30-minute consultation. Tell us where enquiries or orders get stuck, and we will tell you honestly what is worth automating.

Book a free 30-minute consultation