> For the complete documentation index, see [llms.txt](https://academy.gooey.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://academy.gooey.ai/ai-for-impact/module-11/how-to-get-ai-agent-analysis.md).

# How to get AI Agent Analysis

{% embed url="<https://youtu.be/b4X4B1N_Gss>" %}

#### 1. Accessing Your Integration Analytics

1. Go to your Gooey.AI Agent dashboard.
2. Find your AI Agent integration (for example: on Web, WhatsApp, or Voice).
3. At the bottom of the integration details, you will see:
   * A **View Analytics** button
   * An **Analysis Workflows** section

#### 2. Viewing Analytics

1. Click on **View Analytics**.
2. The analytics dashboard will display:
   * When the AI Agent was created and last updated
   * Number of users and trends
   * Breakdown of topics users asked about (e.g., Pricing, Product)
   * Tables listing all conversations and messages
   * Filters for:
     * All messages sent
     * User answers
     * Feedback (positive/negative)
     * If the question was answered successfully or unsuccessfully

***

#### 3. Testing Analytics with New Messages

1. Ask the AI Agent Integration a few questions
   1. It cannot answer (example: “What is the weather?”).
   2. Ask a relevant question (example: “How do I use the AI animation tool?”)
   3. Give feedback using the thumbs-up/thumbs-down options.
2. Refresh the analytics dashboard and check updates

***

#### 4. Setting Up Richer Analysis using LLM Scripts

**What is an Analysis Workflow?**

Analysis workflows use an LLM script to categorize each question and answer. The script creates structured JSON data for better analysis and charting.

**How to Set Up an Analysis Workflow**

1. Go to your integration and find the **Analysis Workflow** section.
2. Click to add a new Analysis Script.
3. Configure the LLM script. The script should:
   * Categorize each Q\&A pair (e.g., “answer missing” or “answer found”)
   * Identify the subject (e.g., product workflow, pricing, delete account, unrelated)
   * Optionally, tag the language of the conversation
4. Provide example Q\&A pairs in your script to help the LLM categorize accurately.
5. For each new message, the script will analyze and output structured data (JSON) with fields like `subject`, `workflow`, `language`, and if the answer was found.

#### 5. Using the Analysis Dashboard

1. After you have some messages, go to the **Analysis Results** section.
2. Add fields from your analysis JSON (e.g., `subject`, `language`) to the dashboard.
3. Choose how to display each field (for example: Pie Chart).
4. Example: Chart showing question topics—Product, Pricing, Unrelated.
5. Example: Chart showing user languages—English, Spanish.

***

#### 6. Multi-language Insights

1. Ask the AI Agent questions in different languages (e.g., Spanish).
2. Refresh the dashboard and see the language breakdown update.

***

#### 7. Health Bot Example

1. The same analysis setup can be used for other AI Agents, like a health bot on WhatsApp.
2. Example fields:
   * If the patient’s health was OK
   * Type of visit (monthly, special, house call)
   * Common health concerns (e.g., flu, fever)
3. View counts and breakdowns for each category.

***

#### 8. Improving Your AI Agent

Regularly check your analytics dashboard:

* See which topics are most asked
* See what is not being answered
* See user feedback
* Analyze conversation language and user intent

Use this information to refine your AI Agent and improve user experience.


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://academy.gooey.ai/ai-for-impact/module-11/how-to-get-ai-agent-analysis.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.