Tip
Actively shape the AI future of Xentral!
We don’t just want to fill Xentral with features — we want to develop exactly the intelligent solutions that save you real time in your daily work.
Do you have processes that frustrate you every day? Are there data volumes that are barely manageable manually? Share your vision with us! The more detail you provide about your use cases, the better we can understand how AI needs to work for you.
These AI features form the foundation for automated processes, data-driven decisions, and consistent customer communication in your Xentral.
-
AI chat (Co-Pilot) – Analyze your data, ask questions, and trigger tasks for your agents
-
Email agents for customer inquiries – Automate the processing of incoming inquiries and receive structured proposals
-
Custom agents – Create AI agents for your individual processes
-
Knowledge base – Build up your company knowledge and continuously train your AI agents
The AI chat (Xentral Co-Pilot) supports you in analyzing, planning, and carrying out your daily tasks in the ERP. You open it via the AI icon at the top right. It is available as a sidebar everywhere in Xentral.
You can ask the chat questions about your business, for example about revenue, customers, orders, or invoices. The Co-Pilot accesses your ERP data and provides you with understandable answers, analyses, or summaries right away – as a table or chart if you like. It also answers questions about using Xentral based on the Xentral handbook.
In addition, you can trigger actions via the chat, for example creating an order. The Co-Pilot creates a task for this, which the matching agent prepares and you approve in the Agent Hub. After your confirmation, the Co-Pilot can save rules and information you mention in the chat to your knowledge base.
This way, you can retrieve information faster, understand connections, and trigger tasks directly from the chat – without having to switch between different modules.
Tip
How to make your first requests and carry out actions: Getting started with the AI Co-Pilot (Chat)
You can find your AI colleagues in the Agent Hub on the Home page in the Your AI colleagues area. Each card displays what the agent can do (CAN) and which data it accesses (SEES). Each agent handles one type of request. Actions in Xentral, for example creating an order or sending a reply, are only carried out when you approve them.
You can find out what each agent checks in detail, what it proposes, and where its limits are in the article Your AI colleagues in detail.
It analyzes every incoming inquiry, breaks down multi-part cases, and distributes them to the appropriate specialist agents. It resolves simple cases itself. Finally, it combines everything into one task with exactly one reply email – even if a customer sends several concerns in one email.
It finds the sales order, reads the shipping provider and tracking number, and proposes a reply with the status and a tracking link. For DHL, the agent retrieves the shipment status in real time (if you have created the shipping method in Xentral and connected it to your DHL account). For UPS, DPD, Hermes, and GLS, it provides tracking links.
It checks the case against your return and complaint rules and recognizes existing returns or credit notes, which prevents duplicate credit notes. It then proposes a return, a credit note, a cancellation, or a replacement delivery, together with a reply to the customer.
It recognizes customer, products, and quantities from the email text or an order PDF, checks for duplicate orders, delivery blocks, and availability, and prepares new sales orders, offers, order changes, or cancellations – and a new customer if needed.
Alternatively, you can copy an order request into the AI chat (Co-Pilot) or paste it via Create New Task in the Tasks tab and ask the agent to create an order from it.
It answers questions about properties, variants, and availability directly from your product data and your knowledge base – even based on customer photos. It does not pass on internal information, for example purchase prices or data of other customers, to your customers.
It reads supplier invoices, matches them against supplier and purchase order, and proposes a liability with pre-accounting. It uses previous bookings of the supplier as a template, for example for the nominal account and cost center.
It matches every line item of a payment advice against your invoices, credit notes, and the open transactions on your bank accounts and reports discrepancies such as overpayments or underpayments or missing documents. After your approval, it books the listed documents against the payment.
Tip
How to forward customer inquiries and have them processed automatically: Getting started with customer inquiries (emails)
With custom agents, you create AI agents for your individual processes that run automatically in your email workflow. You can find them on the Home page in the Custom agents area.
You describe to an assistant in your own words what your agent should be able to do. The assistant creates the agent from this: its behavior description, the topics it is used for, and the tools it is allowed to use to access Xentral. Incoming inquiries on these topics are automatically forwarded to your agent. Just like the built-in agents, your custom agent only prepares actions – they are only carried out after your approval.
Distinction from Flows: Flows automate clearly defined processes based on fixed rules and throughput. Custom agents complement operational business by understanding context and responding dynamically to different situations.
You can learn how to create a custom agent in the article Creating custom agents.
In the knowledge base, you store all the information your AI agents need to work in line with your requirements. You distinguish between the types Company, Expert knowledge, and Templates.
Here you store general information about your company, for example: communication style (for example formal/informal address); brands, company structure, and roles; business model or internal guidelines. This knowledge helps the agents act consistently and in line with your brand.
Here you store everything related to your products and handling customer inquiries: product information and frequently asked questions (FAQ); specific instructions (for example references to the website or tools). In addition, you define your guidelines for customer communication here, i.e., how inquiries should be answered.
Here you store reusable reply templates for typical customer concerns (for example vouchers, returns, size questions). You can use a template, for example, when you edit a reply in a task yourself using the manual option with a template.
The knowledge base helps you continuously improve your AI agents.
While you process inquiries, you can give your agent corrections and rules directly in the task. The system recognizes such rules and suggests saving them as a knowledge entry. This works the same way in the AI chat (Co-Pilot).
In addition, the Agent Hub analyzes your previous requests and your knowledge base and gives you recommendations: It displays knowledge gaps, issues in your master data, and contradictory or duplicate entries in your knowledge base. This way, you can quickly recognize in which areas it is worth storing further knowledge and optimizing your agents in a targeted way.
This way, your knowledge base grows continuously – based on your real daily work.
Tip
How to build up your knowledge base and continuously improve your agents: Getting started with your knowledge base