What stays with people
Orders and uncertain matches flagged by the agent, invoice discrepancies and the customer relationship.
From manually entered orders to orders processed from multiple incoming channels, with review where needed.
A wholesaler with 300 retail stores as customers and its own delivery trucks. Orders arrive by email, as PDFs, as photos of handwritten notes or by phone to the sales rep.
It reads orders from connected channels, identifies products and quantities, and matches them to the company's product codes. When the match is reliable, it creates or prepares the order in the ERP according to the company's approval rules. Unknown product codes, unusual quantities or uncertain matches are flagged for human review.
When confirmed loading or route-planning information is available, it informs the store of the expected delivery day and available delivery window.
It compares each supplier invoice with the purchase order and goods received, and prioritises for review any discrepancies in product codes, quantities, prices or totals, as well as items that cannot be matched reliably.
Orders and uncertain matches flagged by the agent, invoice discrepancies and the customer relationship.
Clean orders can move into the system much faster, while exceptions are sent for review. Accounting can focus on documents that show a discrepancy or need confirmation.
At a wholesaler like this one, the first step is the study. We talk to the sales reps, the people who enter orders and the accounting team, and map which channels orders come from, where errors happen most often and how invoices are checked today. That shows which agent to start with.
The company defines the approval rules for orders, what counts as an unusual quantity and who reviews uncertain matches. It also defines which system provides the confirmed loading and route information.
Before they take on any work, the agents are tested with real orders and invoices from the company. When something does not fit their rules, they stop and flag it for review instead of guessing. After handover, we monitor how they work and adjust them when something changes, such as new product codes, customers or suppliers.