Agent analytics and monitoring

Edit online

Important: This experience is supported only on AWS SaaS Commercial and IBM Cloud instances, including Trial, Essential, and Standard (non-isolated) plans. It is not supported on AWS GovCloud (US), Premium (data isolated), and on-premises deployments.

The Analytics page provides insights into how your agents are performing in production. It helps you monitor conversation activity, track user engagement, identify trends, and gather feedback to improve agent effectiveness.

Agent analytics helps you understand usage patterns, measure performance metrics, and make data-driven decisions about optimization and resource allocation. It provides visibility into real-world performance, such as response speed, user satisfaction, and potential quality issues.

Note: The legacy Analytics experience is deprecated and will eventually be removed. It is recommended that you start using the new analytics experience.

Before you begin

The Analytics dashboard requires administrator or builder access.

Only users with administrator or builder permissions can access analytics data and monitor agent performance.

Accessing Analytics

To access Analytics:

From the main navigation menu, select Analyze.
The Analytics page displays performance data for all agents by default.

Key metrics

The Analytics page displays three primary metrics at the top of the page:


Metric	Description
Total conversations	The total number of conversations across all selected agents during the specified time period
Unique users	The number of distinct users who interacted with agents during the specified time period
Avg conversation duration	The average length of time users spend in conversations with agents

Filtering and viewing data

You can filter analytics data to focus on specific agents or time periods:


Filter	Description
Agent filter	Select specific agents from the dropdown menu to view their individual performance. By default, all agents are selected. The filter displays the total number of agents (for example, "43 All agents").
Time period filter	Choose from predefined time ranges: Past 7 days (default) Past 15 days Past 30 days

Agent trend

The Agent trend section displays a horizontal bar chart showing conversation activity for your agents.

Use this chart to:

Identify which agents are most frequently used
Compare conversation volumes across agents
Spot usage patterns and trends

You can customize the chart view:


Option	Description
Top N filter	Display the top 5, 10, or 15 agents by conversation volume
Metric selector	Choose which metric to display: Conversations - Total number of conversations per agent Unique users - Number of distinct users per agent Avg conversation duration - Average conversation length per agent
View options	Use the toolbar icons to: Switch between list and chart views Expand to full screen Access additional options

User Feedback

The User Feedback section displays user satisfaction ratings for each agent.


Column	Description
Agents	List of agent names
Thumbs down	Number of negative feedback responses
Thumbs up	Number of positive feedback responses

Use the search field to quickly find specific agents in the feedback table.

The table supports pagination, allowing you to:

Adjust items per page (10, 20, 50, etc.)
Navigate between pages of results
View total item count and current page position

Interpreting analytics data

Use analytics data to:

Identify popular agents: Agents with high conversation volumes indicate strong user adoption and utility
Detect underutilized agents: Low conversation counts may indicate agents that need better promotion or improvement
Monitor user satisfaction: High thumbs-down counts suggest agents that may need refinement or additional training
Assess engagement: Conversation duration and unique user counts help measure how effectively agents engage users
Track trends over time: Compare metrics across different time periods to identify growth patterns or declining usage