> ## Knowledge Base Index
> Fetch the complete knowledge base index at: https://help.chrisp.chat/sitemap.xml
> Use this file to discover available pages before exploring further.
> Pure-Markdown content can be obtained by appending a '.md' suffix to the content URLs listed in the sitemap (without the trailing slash).

# How to Classify Customer Support Topics with Hugo AI

__This article explains how Hugo Analytics helps you monitor AI Agent activity, outcomes, and improvement opportunities.__

Analytics make it easier to understand what Hugo handles, where conversations are escalated, and which topics generate the most volume. Use these insights to improve training resources, instructions, routing, and the overall automation strategy.

---

# Where to monitor the AI Analytics?

__Hugo Analytics allow you can monitor two types of data in the AI Analytics section:__

1. Conversation Funnels
2. Topics volume

To find this data go to: ****AI Agent → Evaluate → Analytics****

#### Conversation Funnels

This metric will show you the dispatch of conversation from the first entry to which workflow triggered or if the conversation has been escalated.

![Example of data](https://storage.crisp.chat/users/helpdesk/website/-/3/1/f/2/31f2b8b7eb841c00/0658dfa8-eaa4-4ff1-a698-af73cd_tvgfq1.png)

You can also click on any branch to access more details about the conversations.

From there, you can download the conversations associated with this step to analyze them more deeply.

![Download conversations](https://storage.crisp.chat/users/helpdesk/website/-/e/0/8/b/e08bb3ca53f62000/01130e7d-d4f1-472c-a8a8-e58df7_4vdfs7.png)

This is especially useful to:

* Review how Hugo handled specific cases
* Identify patterns in user requests
* Improve your training data and workflows based on real conversations

#### Workflow heatmap

The workflow heatmap gives you a visual overview of how your workflows perform in real conditions.

It highlights the most triggered paths and steps in your workflows based on real conversations.

![Workflow heatmap](https://storage.crisp.chat/users/helpdesk/website/-/4/1/b/2/41b2444acab33400/59dcfa66-2b83-4a75-b98e-40c590_1czkdd.png)

This allows you to quickly identify:

* Which parts of your workflows are used the most
* Where users drop or get stuck
* Which paths might need optimization

By analyzing this, you can simplify your workflows and improve the overall experience.

#### Topics volume

This value indicates what topic have been triggered the most and the data associated. If you want to find out how to set up a Topic, you can refer to [this article](https://help.crisp.chat/en/article/what-are-hugo-topics-cv1bd0/).

![Topics volume](https://storage.crisp.chat/users/helpdesk/website/-/e/1/4/6/e146d57aa689e800/b8e7a545-4b46-4e12-aed2-4d28e4_1f6vndd.png)

Here as well, uou can hover each of the topics to have an insight on the numbers and percentage for each value.

#### In a nutshell 🌰

__Which data can you monitor thanks to the AI Analytics?__

* View the ****topic**** recurrence of conversations received
* Compare ****escalation**** and deflection rates
* Review the steps and ****routings**** taken by Hugo when handling those conversation
* Download and review real ****conversations**** handled by Hugo
* Visualize workflow usage with the ****heatmap****
* ****Compare**** the data on the topics
* Analyse which ****workflows**** are the most triggered by your users