About
Conversational agent analytics help you understand how users interact with your agents and how agents respond.
You can use this data to monitor usage, identify trends, review user feedback, and see which knowledge sources agents reference when generating responses.
This article explains the metrics and data available on the Conversational Agents dashboard. For information about accessing, filtering, and working with the dashboard, see Work with the Conversational Agents dashboard.
Open the Conversational Agents dashboard
You can access the Conversational Agents dashboard in two ways:
From the Agents page
- Locate the agent you want to review.
- Click the three dots on the agent card and select Analytics from the menu.
From an agent
- Open the agent you want to review.
- Click the Analytics tab.
The Conversational Agents Dashboard opens for the selected agent.
Agent Insights
The Agent Insights panel shows the agent's most recent conversation activity. It displays up to five recent conversation threads, ordered from newest to oldest.
Use this panel to quickly see the topics users have recently discussed with the agent.
If the agent doesn't have any conversations yet, the panel is empty.
Click View Insights at the bottom of the Agent Insights panel to open the full thread history view for the agent.
For details, see Work with the Conversational Agents Dashboard.
Highlights
The Highlights section provides an overview of conversation activity for the selected time period.
- Threads created - Number of unique threads in the selected time period.
- Queries sent - Number of all queries across all threads.
- Avg. queries sent per thread - Average number of queries per thread, calculated as total queries divided by total threads.
- Unique users - Number of unique users that had at least one interaction in the timeframe.
Avatar conversations
The Avatar conversations section provides usage data for conversations with avatars.
- Number of avatar conversations - Number of conversations initiated with an avatar.
- Total minutes of avatar conversations - Total duration, in minutes, of all avatar conversations during the selected time period.
- Average avatar conversation time (in minutes) - Average duration of an avatar conversation, calculated as total minutes divided by total conversations.
Queries over time
The Queries over time chart shows query volume over time, helping you identify usage patterns and periods of higher activity.
You can view the data using the following time intervals:
- Daily - Groups queries by day. This is the default.
- Monthly - Groups queries by month
Hover over a data point to view the exact date and query count.
Click View Details below the chart to open the detailed view.
Response type
The Response type section shows the types of answers provided by the agent.
The chart includes the count and percentage for each response type, helping you understand which response formats the agent uses most often.
User feedback
The User feedback section shows feedback users provided about agent responses.
The chart displays:
- Green - Positive feedback.
- Red - Negative feedback.
The total feedback count and the percentages of positive and negative feedback are displayed below the chart.
Knowledge sources
The Knowledge sources section shows the sources the agent referenced when generating answers.
Sources are ranked by the number of times they were referenced. For each source, you can view:
- Source name
- Source type
- Number of times the source was referenced
- Number of unique users
Enter text in the search field to find a specific source.
Click a source link to open the source's entry dashboard.
Kaltura does not use customer data to train its AI models. To learn more, see Kaltura's Artificial Intelligence Principles.
