> 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/conversation-intelligence/conversation-intelligence-en/getting-started.md).
# Getting started
Overview of Conversation Intelligence: features, requirements, and first setup steps for admins.
#### What is Conversation Intelligence?
Conversation Intelligence is a standalone product fully integrated into the sipgate cloud phone system. With it, you can analyze the content of your inbound and outbound calls: phone calls made via the sipgate telephony infrastructure are automatically transcribed and evaluated with AI support.
The goal is to gain business-relevant insights (business insights), e.g. to optimize processes, products, and service quality.
The features:
* Structured storage of all call metadata
* Optionally transcribe all calls, with granular settings for which calls should be transcribed and which should not
* Definition of content ([Custom Fields](http://custom-fields-verwalten.md/)) that is extracted from the conversations and stored in structured form
* Specification of the expected output formats of the custom fields, for example yes/no, number, date, free text, or selection from a predefined list
* Contact integration across all common CRM systems
* Central, filterable, and searchable overview of all calls, all collected custom fields and metadata in the [Data Hub](http://data-hub.md/)
* A database that scales with your call volume
* [Webhooks](http://webhooks-einrichten.md/), to further process the captured custom fields and metadata in downstream systems
* Access via REST API and via the [MCP server](http://mcp-server-verbinden.md/)
* Compliant with Article 50 of the EU AI Act for employees
* [GDPR-compliant deletion](http://anrufdaten-loeschen.md/) of records
* Optional anonymization of personal data records (PII)
#### What requirements and plans are there?
To set it up, you need admin rights in your sipgate account. Which plan you need depends on whether you want to use Conversation Intelligence or AI Assist:
| What you need | Conversation Intelligence | AI Assist |
| ------------------------------------------------------------------- | ------------------------------------------------------------- | --------- |
| **Business plan** (Business S, L, XL or Enterprise) | Yes | Yes |
| **Business or Premium app** (additional app for your users) | No | Yes |
| **Conversation Intelligence package** (Small, Medium or Enterprise) | One of the three packages, Contact Center or AI Agents / sona | No |
For Conversation Intelligence, it is sufficient **one** of the following options. You can also combine several:
* a Conversation Intelligence package (Small, Medium or Enterprise)
* Contact Center
* AI Agents / sona
#### What is the difference to AI Assist?
AI Assist and Conversation Intelligence both evaluate calls with AI, but they are aimed at different target groups:
| | AI Assist | Conversation Intelligence |
| ---------------------------------- | ----------------------------------------- | -------------------------------------------------------------------------- |
| **Who activates it?** | Each user for themselves | Admins, for the channels in your account |
| **Whose calls are evaluated?** | The user's own calls | The calls in the activated channels |
| **Where are the results visible?** | Directly on the call event in the app | In the Data Hub or in external systems |
| **Plan** | Business plan and Business or Premium app | Business plan and one of the options above, but no Business or Premium app |
You can find out more about AI Assist in the article [What are the sipgate AI features?](https://help.sipgate.de/cloud-telefonanlage/sipgate-nutzen/die-sipgate-app/was-sind-die-sipgate-ai-features).
#### How do I activate Conversation Intelligence?
You can find Conversation Intelligence in the main navigation. There you have access to the Data Hub, Custom Fields, Webhooks and settings.
{% stepper %}
{% step %}
## As **Admin** log in
Log in as **Admin** an.
{% endstep %}
{% step %}
## Open settings
Open **Conversation Intelligence → settings**.
{% endstep %}
{% step %}
## Activate channels
Click in the **Activation** section on the gear icon for the respective product (e.g. phone system, user or contact center) and specify for which channels the analysis should be activated. The overview shows you in how many channels Conversation Intelligence is already active.
{% endstep %}
{% endstepper %}
As soon as Conversation Intelligence is activated for a channel, your employees are automatically shown a notice while making calls in the sipgate app that conversations in this channel are analyzed with AI support. You cannot deactivate Conversation Intelligence for yourself.
{% hint style="warning" %}
Inform your employees and callers before you activate Conversation Intelligence for a channel. What to consider is explained under [privacy](http://datenschutz.md/).
{% endhint %}
#### How do I connect my CRM system (contact integration)?
With contact integration, you connect your CRM system to Conversation Intelligence. We resolve the contact name and company name via the phone number and write both as fields into the [Data Hub](http://data-hub.md/). This allows you to find all calls from a company there, for example using the filter.
You can connect all systems listed under [sipgate.de/integrationen](https://www.sipgate.de/integrationen) .
{% stepper %}
{% step %}
## Open settings
Open **Conversation Intelligence → settings**.
{% endstep %}
{% step %}
## Add CRM system
Click in the **Contact integration** on **Add** and select your CRM system.
{% endstep %}
{% step %}
## Check connection status
The status in the list shows you whether the connection exists, e.g. **Connected**.
{% endstep %}
{% endstepper %}
You can remove an integration via the trash icon in the integration row.
#### How do I get insights?
This is how you quickly gain new insights with Conversation Intelligence:
{% stepper %}
{% step %}
## Ask a business question
Ask a business question, e.g. “How can I optimize support costs?”
{% endstep %}
{% step %}
## Define custom fields
Define which content should be collected from the conversations in order to answer this question, e.g. “Which topics concern the calls in the support channels?” or “Which topics escalate?” How this works, read under [Manage custom fields](http://custom-fields-verwalten.md/).
{% endstep %}
{% step %}
## Filter results and create dataset
Narrow down the call data in the [Data Hub](http://data-hub.md/) specifically, e.g. to a call duration of over 5 minutes.
{% endstep %}
{% step %}
## Evaluate signals in the dashboard
See at a glance how a signal develops, and analyze patterns directly.
{% endstep %}
{% step %}
## Export or further process
Further process the dataset, e.g. in Excel, via [Webhooks](http://webhooks-einrichten.md/), via the [MCP server](http://mcp-server-verbinden.md/) or via our REST API.
{% endstep %}
{% endstepper %}
Also, you can [notifications](http://benachrichtigungen-einrichten.md/) be automatically notified as soon as something in your conversations requires your attention.
---
# 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/conversation-intelligence/conversation-intelligence-en/getting-started.md?ask=&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.
