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

# Attributes

> Custom fields for granular tagging

Attributes are custom fields with predefined options for more granular conversation tagging.

{/* Screenshot: Attributes page showing list of attributes with their options */}

## What are Attributes?

Unlike categories (one per conversation), attributes are additional fields that can capture multiple dimensions:

| Attribute     | Options                           |
| ------------- | --------------------------------- |
| Priority      | High, Medium, Low                 |
| Product Area  | Dashboard, API, Mobile App        |
| Customer Tier | Enterprise, Pro, Free             |
| Sentiment     | Positive, Neutral, Negative       |
| Issue Type    | Bug, Question, Request, Complaint |

A conversation can have values for multiple attributes.

## Creating Attributes

<Steps>
  <Step title="Navigate to Attributes">
    Go to **Tag → Attributes** in your dashboard.
  </Step>

  <Step title="Click Create Attribute">
    Click **Create Attribute**.

    {/* Screenshot: Create attribute dialog */}
  </Step>

  <Step title="Name the Attribute">
    Choose a clear, descriptive name:

    * Good: "Priority", "Product Area", "Customer Tier"
    * Avoid: "Field 1", "Attribute"
  </Step>

  <Step title="Add Options">
    Define the possible values:

    {/* Screenshot: Adding options to an attribute */}

    For Priority: High, Medium, Low
  </Step>

  <Step title="Choose Selection Type">
    * **Single-select:** Only one option can be chosen
    * **Multi-select:** Multiple options can be chosen
  </Step>

  <Step title="Save">
    Click **Save** to create the attribute.
  </Step>
</Steps>

## Single vs Multi-Select

| Type              | Use When                       | Example                       |
| ----------------- | ------------------------------ | ----------------------------- |
| **Single-select** | Options are mutually exclusive | Priority: High/Medium/Low     |
| **Multi-select**  | Multiple can apply             | Product Areas: Dashboard, API |

**Single-select examples:**

* Priority (can only be one level)
* Sentiment (positive, neutral, or negative)
* Customer Tier (one subscription level)

**Multi-select examples:**

* Product Areas (issue might span multiple)
* Tags (multiple descriptors)
* Features Mentioned (could discuss several)

## How Auto-Extraction Works

The AI extracts attribute values based on:

1. **Conversation content** — What was discussed
2. **Customer information** — Available metadata
3. **Context clues** — Tone, urgency, specifics mentioned

For example, if a customer says "This is urgent, I need this fixed today!", the AI would likely set Priority to "High".

## Common Attributes

### Priority

```
High — Urgent, blocking, immediate attention needed
Medium — Important but not urgent
Low — Minor issue, no time pressure
```

### Sentiment

```
Positive — Happy, thankful, complimentary
Neutral — Factual, matter-of-fact
Negative — Frustrated, angry, disappointed
```

### Issue Type

```
Bug — Something is broken
Question — How-to or informational
Request — Feature or change request
Complaint — Expressing dissatisfaction
```

### Product Area

```
Dashboard — Main application interface
API — Developer integration
Mobile App — iOS/Android app
Billing — Payment and subscription
```

### Customer Tier

```
Enterprise — Large business customers
Pro — Paid individual/small business
Free — Free tier users
Trial — Users in trial period
```

## Using Attributes in Analytics

### Breakdown by Attribute

{/* Screenshot: Bar chart showing distribution by attribute values */}

See distribution across attribute values:

* 45% Low Priority, 35% Medium, 20% High
* Dashboard: 60%, API: 30%, Mobile: 10%

### Filter by Attribute

{/* Screenshot: Run history with attribute filter applied */}

View only High Priority conversations, or only API-related issues.

### Cross-Analysis

Combine category and attribute analysis:

* "High Priority Technical Issues"
* "Enterprise Billing Questions"

## Assigning Attributes to Agents

<Steps>
  <Step title="Open Agent Configuration">
    Go to **Build → Agents** and click on an agent.
  </Step>

  <Step title="Go to Classification Tab">
    Select the **Classification** tab.
  </Step>

  <Step title="Select Attributes">
    Check which attributes this agent should extract.

    {/* Screenshot: Agent classification tab showing attribute checkboxes */}
  </Step>

  <Step title="Save">
    Save the agent configuration.
  </Step>
</Steps>

## Best Practices

* **Start with essentials** — Priority and Product Area are common starting points
* **Keep options clear** — Each option should be distinct
* **Use descriptions** — Help the AI extract accurately
* **Review extraction** — Check that attributes are being applied correctly
* **Don't over-attribute** — 3-5 attributes is usually sufficient

## Next Steps

<CardGroup cols={2}>
  <Card title="Categories" icon="folder" href="/tagging/categories">
    Set up high-level categories
  </Card>

  <Card title="Resolution Rules" icon="circle-check" href="/tagging/resolution-rules">
    Define resolution criteria
  </Card>
</CardGroup>