Available AI features

It automatically recognizes relevant information from the message, for example order number, customer data, shipping provider, and tracking number. The agent then matches the data with Xentral and — if available — with the shipping provider (available for the DHL shipping method — when you have set up the shipping method in Xentral and connected it to your DHL account).

Based on this, it creates a brief summary, evaluates the current delivery status, and suggests the next steps, for example to inform the customer or contact the shipping provider.

This allows you to check and respond to delivery inquiries faster, without having to manually gather all the information.

It automatically recognizes relevant information from the customer message, for example customer data, order number, affected items, reason for complaint, and uploaded images. The agent then checks applicable rules or processes, such as withdrawal, exchange, or replacement delivery.

Based on this, it creates a summary, provides a decision recommendation, and suggests the next steps, for example to initiate a return or create a new order for the exchange.

The agent checks your stored returns and complaints rules, for example return periods of 14 days or complaint periods of 2 years. If no rule applies, you receive a notification with a corresponding recommended action.

This allows you to process returns and complaints faster, more structured, and more consistently.

It automatically recognizes relevant information from the customer message, for example customer data, desired items, quantities, and other order details. The agent then matches the information with your data in Xentral and prepares the order.

This way, for existing customers, the agent can create a new order with the specified items when a customer orders products from your range by email.

Alternatively, you can also copy an order request directly into the chat and ask the agent to create an order from it.

This allows you to capture new orders faster, more structured, and with less manual data entry.

It automatically recognizes relevant information from the customer message, for example requested items, product properties, or specific questions about usage, availability, or compatibility. The agent then accesses your product data from the ERP, stored FAQ pages, and product knowledge from your knowledge base.

This way, the agent can formulate suitable answers based on your current product information and your defined customer communication.

Based on this, the agent creates a summary of the inquiry and suggests an appropriate answer, for example about product properties, sizes, materials, accessories, or frequently asked questions.

This allows you to answer product questions faster, more consistently, and with less manual research effort.

Here you store general information about your company, for example: communication style (for example formal or informal address); brands, company structure, and roles; business model or internal policies. 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); response templates for typical customer concerns (for example vouchers, returns, size questions); specific instructions (for example references to your website or tools). You also define your guidelines for customer communication here, i.e. how inquiries should be answered.

The Learning Loop helps you continuously improve your AI agents.

While you process inquiries — for example in tasks or emails — you can create new knowledge building blocks for specific cases directly via chat. The agent supports you in formulating this content meaningfully and saves it directly in your knowledge base.

In addition, you receive insights in the Learning Loop about which inquiries occur particularly frequently. This allows you to quickly identify which areas are worth adding further knowledge to and optimize your agents in a targeted way.

This way, your knowledge base grows continuously — based on your real everyday work.

The incoming invoice agent supports you in automatically capturing incoming invoices and preparing them as a liability in Xentral.

It recognizes relevant information from incoming emails and attached invoices, for example supplier, invoice number, line items, amounts, taxes, due date, and payment information. The agent then matches the data with existing liabilities, supplier data, and stored knowledge from the knowledge base.

Based on this, the agent creates a suggestion for the new liability and displays the planned steps and pre-filled fields. After your review, you can complete the action and create the liability directly in Xentral.

This allows you to process incoming invoices faster, more structured, and with less manual data entry.

The payment advice agent supports you in reviewing extensive payment advices.

It automatically extracts relevant information from the advice, for example total amount, individual items, invoices, credit notes, and associated amounts. The agent then matches the items with existing ERP documents such as invoices and credit notes.

This way, the agent recognizes whether the information in the advice matches the data in Xentral or whether there are discrepancies. In matching cases, you receive a clear response, for example that the advice has been checked and exactly matches the ERP data.

This allows you to review even extensive payment advices faster and significantly reduce manual effort in accounting.

With Custom Agents, we aim to enable AI agents to be individually adapted to your daily operational tasks.

The goal is to map your own use cases — for example specific review processes, individual decision logic, or company-specific workflows. Since requirements vary greatly depending on business model, company size, and setup, we are currently still in an early concept phase.

Focus: intelligent decision logic & flexible, individual use cases

Distinction from Flows: Flows automate clearly defined processes based on fixed rules and throughput. Custom Agents are intended to complement the operational business by understanding context and responding dynamically to different situations.