> ## Documentation Index
> Fetch the complete documentation index at: https://knowledge.bitbybit.studio/llms.txt
> Use this file to discover all available pages before exploring further.

# How to configure AI Tagging

> Automatically categorize conversations using smart labels, sentiment analysis, and custom rules.

<Info>
  **What is AI Tagging?** AI Tagging allows your AI Agent to "listen" to conversations and automatically apply labels based on the context. This helps you filter chats later for analytics or follow-up without manual sorting.
</Info>

## Default System Tags

By default, the AI Agent comes equipped with three essential tagging capabilities that require no configuration:

1. **Language:** Automatically detects and tags the language the customer is speaking (e.g., `English`, `Spanish`, `Indonesian`).
2. **Priority:** Analyzes the sentiment and urgency to tag conversations as `High`, `Medium`, or `Low`.
3. **Intention:** Summarizes the user's goal (e.g., `Support`, `Sales`, `Inquiry`).

## Creating Custom Tags

You can create specific tags to track product interest, campaign success, or specific customer types.

<img src="https://mintcdn.com/asmaraku/WlT_6AuzhITpdgTf/images/Screenshot2025-12-23101209.png?fit=max&auto=format&n=WlT_6AuzhITpdgTf&q=85&s=b0841379a4aaff17185f9b52ca730fa9" alt="Screenshot2025 12 23101209" width="848" height="689" data-path="images/Screenshot2025-12-23101209.png" />

<Steps>
  <Step title="Add a New Tag">
    1. Navigate to \*\*AI Studio Sidebar \*\*> **AI Tagging**.
    2. Click the **Add new tag** button in the top right corner.
  </Step>

  <Step title="Configure the Tag Logic">
    Define how the AI recognizes and labels the specific topic.

    <AccordionGroup>
      <Accordion title="1. Label Name">
        Give your tag a unique internal name.

        * *Example:* `Smartphone_Interest`
      </Accordion>

      <Accordion title="2. Label Description (The Prompt)">
        This is the most critical step. You must tell the AI **what to look for**. Think of this as a mini-instruction.

        * *Example:* "Identify the specific smartphone model the customer is asking about or interested in buying."
      </Accordion>

      <Accordion title="3. Values (The Output)">
        List the specific tag values the AI is allowed to apply. If the customer mentions a model, the AI will pick the matching value from this list.

        * *Example:* `iPhone 15, Samsung Galaxy S24, Google Pixel 8, iPhone 14`
      </Accordion>
    </AccordionGroup>
  </Step>

  <Step title="Save">
    Click **Save** to activate the tag. The AI will now scan future conversations for these keywords and contexts.
  </Step>
</Steps>

## Example Use Cases

| Goal               | Label Name        | Description                                | Values                                       |
| :----------------- | :---------------- | :----------------------------------------- | :------------------------------------------- |
| **Track Products** | `Laptop_Model`    | Identify which laptop the user asks about. | `MacBook Air`, `Dell XPS`, `Lenovo ThinkPad` |
| **Lead Source**    | `Referral_Source` | Identify how the user found us.            | `Google`, `Facebook`, `Friend`, `Newsletter` |
| **Complaint Type** | `Issue_Category`  | Categorize the type of problem reported.   | `Shipping`, `Defect`, `Payment`, `Login`     |

<Tip>
  **Best Practice:** When entering **Values**, separate them with commas. Ensure the "Label Description" is specific enough so the AI knows exactly which value to pick from the list.
</Tip>