We are announcing the ability to analyze user sentiment from agent conversations in Microsoft Copilot Studio. This feature will reach general availability on March 31, 2026.
How does this affect me?
Copilot Studio uses AI models to analyze the tone, language, and user interactions throughout the conversation to calculate an overall sentiment score. The new sentiment analysis is an additional metric to understand customer satisfaction. Sentiment analysis works alongside existing CSAT scores and will help you quickly identify sessions with negative sentiment, enabling you to investigate issues and continuously improve your agent’s performance with the following features:
- Analyze a sample of all sessions.
- Automatic availability for customers with at least 10 sessions with questions that have triggered generative answers in the past 7 days.
- Sentiment is calculated for each user message in the sampled sessions and then all messages are aggregated for a single sentiment score for the session.
- Each session is classified as positive, negative, or neutral, based on all user messages within that session.
- The sentiment score is calculated and refreshed once a day.
This message is for awareness, and no action is required.
Source: Microsoft
Latest Posts
- AWS End User Messaging and Amazon SES now offer AI agent skills for the AWS MCP Server
- Amazon EC2 M8i and M8i-flex instances are now available in additional regions
- Amazon EC2 R8i and R8i-flex instances are now available in additional regions
- Amazon EC2 C8i and C8i-flex instances are now available in additional regions
