# Analytics

## Outputs

The **Outputs** tab in your agent workspace shows every conversation your agent has handled. Each conversation includes:

| Column              | What it shows                                                    |
| ------------------- | ---------------------------------------------------------------- |
| **Resolution**      | Whether the conversation was resolved without human intervention |
| **CSAT**            | Customer satisfaction rating (Great, Good, Fair, Poor)           |
| **Gap in training** | Whether the agent lacked knowledge to answer the question        |
| **Channel**         | Which integration or channel the conversation came from          |

### CSV export

You can export your conversation history and metrics as a CSV file. The export is generated and sent to your email.

### Using Outputs to improve your agent

**Look for gaps in training.** Filter by conversations flagged as having a gap — these show you exactly where you need to add more knowledge.

**Review low CSAT scores.** Conversations rated Fair or Poor highlight areas where your agent's responses need improvement.

**Check resolution rates.** A high resolution rate means your agent is handling conversations well. If it's low, review the escalated conversations to understand why.

## Reports

> Reports is coming soon. [Contact us](https://docs.eesel.ai/support/contact) and we'll let you know as soon as it's ready.