The KPI Dashboard gives you an aggregated view of your AI agent's performance across all conversations in a project. Where the Conversation List focuses on individual conversations, the KPI Dashboard helps you spot trends and measure overall performance over time.

Available KPIs

The dashboard shows the following metrics, aggregated across all processed conversations in the selected time period:

Conversation volume & outcomes

• Total conversations — total number of imported and processed conversations
• Handover rate — percentage of conversations escalated to a human agent
• Automation rate — percentage of conversations fully resolved by the AI agent

Quality

• Average quality score — mean AI-generated quality score across all conversations
• Conversation length distribution — breakdown of conversations classified as short, medium, or long

Agent performance

• Average handle time — mean conversation duration in minutes
• Average bot response time — mean time the agent took to respond to a user message
• Bot verbosity — classification of how concise or verbose the agent's responses are on average (low / medium / high)

Time period

Use the date range selector at the top of the dashboard to focus on a specific time window. All KPIs update to reflect only the conversations within the selected period.

Using KPIs to find issues

A few patterns to watch out for:

• A rising handover rate may indicate the agent is struggling with new or shifted topics
• A dropping average quality score can signal a decline in response quality that warrants investigation