ManufacturingComplex automationSeveral jobs working together: multi-stage pipelines with state, branching on rules or AI decisions, approvals or escalations, 4+ systems, AI agents that act, or reconciliation across sources. Usually rolled out in stages.

A driver's WhatsApp photo of the signed note issues the invoice

For: Packaging manufacturer with its own delivery vans

Typical impact

Invoices out the same day as delivery

After each drop, the driver photographs the signed delivery note and sends it on WhatsApp. AI reads the note number, signature and handwritten corrections, matches the shipment in the ERP and issues the invoice with the signed note attached as proof of delivery.

Typical results

Same day

Invoice issued after delivery

≈4 days

Earlier invoicing on average

≈6 hrs/wk

Billing clerk time back from matching paper notes

All invoices

Carry a signed proof of delivery

Figures reflect typical results for this kind of automation.

The challenge

Drivers bring signed delivery notes back in a folder at the end of the day or the week, and the billing clerk matches each one to a shipment before invoicing. Invoices go out four or five days after delivery, handwritten shortages get missed, and customers dispute invoices when the signed note cannot be found.

Before and after

Before With the automation

Hours from delivery to invoice

−96%
100 hrs
4 hrs

Billing clerk hours per week on delivery notes

−75%
8 hrs
2 hrs

Invoice disputes over missing proof per month

−83%
6 disputes
1 disputes

The automation

After each drop, the driver photographs the signed delivery note and sends it to a WhatsApp number. AI reads the delivery note number, checks for a signature and picks up any handwritten quantity corrections. The note is matched to the shipment in the ERP, and the invoice is issued in the accounting system with the signed note attached as proof of delivery. Notes with recorded shortages or no signature go to the billing clerk before anything is invoiced.

How it works

  1. 01

    Photo after each drop

    The driver photographs the signed delivery note and sends it to the company's WhatsApp number before leaving the customer.

  2. 02

    AI reads the note

    AI reads the delivery note number, checks for a signature and picks up any handwritten quantity corrections.

  3. 03

    Matched to the shipment

    The note is matched to the shipment in the ERP, and delivered quantities are compared with what was loaded.

  4. 04

    Invoice with proof attached

    Clean notes trigger the invoice in the accounting system with the signed photo attached as proof of delivery.

  5. 05

    Exceptions to billing

    Notes with recorded shortages or no signature go to the billing clerk, who confirms quantities before anything is invoiced.

Starts from

  • Signed delivery note photo
  • Driver WhatsApp message

Delivers

  • Invoice with proof attached
  • Shipment matched in ERP
  • Shortage alert to billing

The impact

Invoices go out the same day as delivery instead of four or five days later, which brings payments forward by about the same margin. The billing clerk spends around 2 hours a week on exceptions instead of 8, and disputes over missing proof of delivery become rare.

Questions about this automation

What if the photo is blurry or the note number cannot be read?

The driver gets a WhatsApp reply within seconds asking for a clearer photo while still at the customer. If the number still cannot be read, the note goes to the billing clerk together with that van's deliveries for the day, so it can be matched by hand in a minute.

Does it need both an ERP and an accounting system?

Not always. It uses a business WhatsApp number for the drivers, the ERP for shipments and quantities, and the accounting system that issues invoices. If the ERP issues invoices itself, that becomes a single connection. This is a multi-step build, usually rolled out in stages, starting with one van and its routes.

How are handwritten corrections handled?

When a customer crosses out a quantity and writes a lower one, the AI reads both numbers and the invoice is held. The billing clerk sees the photo next to the shipment, confirms the corrected quantity, and the invoice goes out with the adjusted lines.

Manufacturing

Related automation examples

View all 20 examples

Want this in your business?

We map it to the tools you already use and show you where it saves the most time, in a free 30-minute call.