> For the complete documentation index, see [llms.txt](https://help.sipgate.de/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.sipgate.de/ai-agents/en/behavior/kundenfragen.md).

# Customer questions

In the customer questions section, store typical inquiries about opening hours, services, prices, or processes, as well as the corresponding answers. The clearer the questions are formulated, the more reliably your agent will recognize the request and respond precisely.

You can find the customer questions in sipgate sona by opening the agent overview, clicking your agent, and **Knowledge** select.

If you created your agent via your website, some initial customer questions are already stored there. Check the questions and answers, adapt them, and add any missing topics.

{% hint style="info" %}
If you created your agent by entering your URL, some initial customer questions are already stored there. Check the questions and answers, adapt them, and add any missing topics.
{% endhint %}

### How to formulate customer questions correctly

{% stepper %}
{% step %}
**Clearly separate different topics from one another**

Different topics should be clearly distinct from one another. Questions that are too similar on different topics can confuse the AI agent and reduce recognition quality.

**Not recommended:**

* How do I reach the hotline?
* How do I reach customer service?

These questions are phrased almost identically and refer to the same information. The AI agent can hardly distinguish between them and does not know which answer is meant in which context.

**Recommended:**

* How do I reach the hotline?
* At what times is the hotline staffed?

These questions address different aspects of the topic (contact options vs. availability times). The AI agent can clearly recognize which information is being requested and provide the appropriate answer.
{% endstep %}

{% step %}
**Store multiple phrasings for the same topic**

You can enter multiple phrasings for the same request per customer question. This is useful because customers often ask the same question in different ways. The AI agent learns to recognize different ways of expressing the same request and assign them correctly.

**Example - all three questions lead to the same answer:**

* How long does delivery take?
* When will my order arrive?
* What is the delivery time?
  {% endstep %}

{% step %}
**Formulate specific rather than general questions**

The more specific the question, the better the AI agent can assign it. Instead of general questions like "What do you offer?", it is better to formulate specific questions like "What plans are available for business customers?"
{% endstep %}
{% endstepper %}

### What does not belong in customer questions

**Appointment-related topics when calendar integration is active.** Your agent automatically recognizes appointment requests and books directly in Cal.com. Additional customer questions about appointments, bookings, or availability can cause it to classify the request incorrectly and not start a booking. More about this under Calendar.

**Content that is already in the knowledge base.** Duplicate or conflicting information reduces answer quality. Customer questions are suitable for individual, clearly defined questions, while the knowledge base is for more extensive content.

### Formulate answers correctly

Answers should contain all important information in two to three sentences, without filler words or embellishments.

Not recommended:

* Our opening hours are designed to be very customer-friendly. We are open Monday to Friday, and we are also available on Saturdays. Specifically, that means: Monday to Friday 9 a.m. to 6 p.m., Saturday 10 a.m. to 2 p.m.

Recommended:

* We are available Monday to Friday from 9 a.m. to 6 p.m., and on Saturdays from 10 a.m. to 2 p.m. We are closed on Sundays.

### Derive customer questions from support tickets

The best customer questions can be found in your existing support requests. They show which topics actually matter to your customers.

Here's how to proceed:

* Analyze past support tickets and email inquiries
* Note the most common topics and recurring questions
* Turn them into clear question-and-answer pairs
* Prioritize questions that are asked particularly often

### After setup

Once you have stored the customer questions, test your agent thoroughly. Call it and ask the questions in different phrasings. Adjust the questions if recognition is uncertain or incorrect answers are given.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://help.sipgate.de/ai-agents/en/behavior/kundenfragen.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
