In wholesale, orders arrive as emails, Excel sheets, customer PDFs and EDI messages from retail chains, and the sales rep retypes them into SAP Business One by hand. The agent recognises customer codes, checks stock and contract prices and prepares the order in SAP Business One or NetSuite, which the rep approves with one click.
Built for ISO 27001NIS2 DORA
In a typical wholesale company, a sales rep receives dozens of orders a day in different formats: one customer writes free text to an Outlook mailbox such as orders@, another sends their own Excel sheet with their own product codes, a third a PDF from their procurement system, and retail chains send orders through EDI. Each one has to be read, the products found in your SAP Business One records, stock in the warehouse checked, the price agreed with that customer recalled and the order entered. Hours disappear here, and errors appear where the customer code "AB-100" looks like your "AB-1000" or where a contract discount is forgotten.
The agent connects to the shared orders mailbox in Microsoft 365 and to the EDI feed: document recognition extracts lines from Excel and PDF, and a Claude model maps customer codes and names to your product records, based on the history of previously approved orders. The agent then checks stock and that customer's contract prices in SAP Business One or NetSuite and creates an order document with the status "unconfirmed". Lines that could not be matched, or where stock is missing, are flagged for the rep. The rep reviews the document in the usual SAP Business One or NetSuite window, corrects it and confirms it. Only then does the order become valid, and the customer receives a confirmation with a delivery date. We connect to SAP Business One through its integration interface or import files, and to NetSuite or Business Central through an API, file export or email flow, depending on your version. The model runs on the Anthropic API or Amazon Bedrock in an EU region, your data is not used to train the model, and if your ERP runs on managed ERP hosting, the integration is deployed without additional servers.
Four steps from input to a record in your system, with human approval wherever there is risk.
The customer sends the order to the usual Outlook address, for example orders@, and a retail chain sends it through EDI. The agent recognises that it is an order and not a question or complaint, extracts the lines from the text, Excel sheet, PDF or EDI message, and identifies the sending company by email domain, GLN code and your customer list in SAP Business One.
Each line is mapped to your product record in SAP Business One or NetSuite by code, barcode, name and that customer's previous orders. The agent checks stock in the warehouse, applies the contract price or discount from the customer record and builds a draft order with notes on doubtful lines.
The agent creates the order document in SAP Business One or NetSuite with the status "unconfirmed" and attaches the original email. In the usual system window, the sales rep sees the original and the built order side by side, corrects what is needed and confirms. Only then does the order go to the warehouse, and the customer receives a confirmation with a number and a delivery date.
An unknown code, a new customer without a record in SAP Business One or missing stock does not enter the system automatically: such lines wait for the sales rep's decision. A confirmed mapping is remembered for next time, and every month we review the log and extend the rules.
The agent is connected to your Microsoft 365 Outlook mailbox and your EDI feed, reads emails, attachments and EDI orders, and does not touch any other mailboxes.
A dictionary of customer codes and the matching product records in your SAP Business One or NetSuite, which grows with every approved order and reduces the number of exceptions.
The "unconfirmed" order in your SAP Business One or NetSuite window with the original attached, stock levels and doubtful lines flagged, with no additional application.
Integration with SAP Business One through its interface or an import, with NetSuite or Business Central through an API or export, a confirmation to the customer and a full action log of who approved what and when.
Document recognition + Claude agent via API + SAP Business One integration through its interface or import files; with NetSuite and Business Central we connect through an API, file export or email flow, depending on the version. We choose the platform by the task and by where your data may be: read more about this service and its limits on the service page.
The line is marked as unclear and shown to the sales rep with the closest guesses from your SAP Business One records. The rep picks the right product and the choice is remembered, so next time the same code is mapped automatically.
We connect to SAP Business One through its integration interface or import files. With NetSuite and Microsoft Dynamics 365 Business Central we integrate through an API, file export or email flow, depending on your version and licences. We choose the integration method after a check in your environment, and in every case the order is created as an unconfirmed document.
A first working solution with one Outlook mailbox and one SAP Business One or NetSuite integration usually takes 2–4 weeks. From you we need access to the mailbox, an export of product records and a few dozen examples of earlier orders to build the mapping dictionary. We add the EDI feed in a second stage.
Every supplier sends a catalogue in its own format: different columns, units, language and codes.
At month end accounting matches statements to invoices by hand and writes reminders to debtors one by one.
Customers call and write about stock, their prices and order status, and sales reps answer the same questions dozens of times a day.
A new price list or quote for every customer group means a day of work in Excel and Word.
This service is delivered by specialists with valid international certifications. Their expertise is confirmed by independent organisations, not by us.
AI systems are deployed with protection against data leakage and model abuse.
The AI security programme and its controls are managed following the ISACA methodology.
AI risks are identified and managed like any other IT risk.
We reply within 1 business day
