Sentiment page

Australia Conversational Interfaces

Release

australia

ft:locale

en-US

ft:publication_title

Australia Conversational Interfaces

ft:clusterId

convint

bundleId

convint

workflow

Platform

Sentiment page in Assistant analytics

Release version: Australia

Updated October 27, 2025

4 minutes to read

Summarize

Summarized using AI

Summary of Sentiment page in Assistant analytics

The Sentiment page in Assistant analytics provides ServiceNow customers with a comprehensive dashboard to analyze user sentiment derived from conversations with virtual assistants. Using inferred Customer Satisfaction (CSAT) scores and emotional indicators such as empathy, frustration, and confusion, the dashboard helps monitor and improve the quality of assistant interactions. This insight enables targeted enhancements to assistant behavior and overall user experience.

Show full answer Show less

Key Features

Overall Sentiment: Displays the average inferred CSAT score on a 0 to 5 scale, allowing tracking of user satisfaction trends over time and evaluation of assistant updates.
Conversations Analyzed: Shows the total number of conversations analyzed for sentiment, indicating the volume of data supporting the metrics.
High Empathy Rate: Measures the percentage of conversations where the assistant demonstrated high empathy, reflecting its ability to respond sensitively to user queries.
Conversations with Negative Emotions: Tracks the percentage of conversations exhibiting frustration or confusion, highlighting areas where user experience may be suffering.
Average Inferred CSAT Over Time: Provides daily trends of CSAT scores to identify periods of improvement or decline in user sentiment.
Transfers and Escalations Over Time: Monitors how often conversations are transferred or escalated to live agents, indicating when human intervention is needed.
Average Inferred CSAT by Interaction Type: Separately displays CSAT scores for Virtual Agent, Live Agent, and combined sessions, enabling benchmarking of assistant and agent performance.
Assistant Recommended Next Steps: Evaluates how clearly the assistant communicates subsequent actions to users, categorized as low, medium, or high clarity.
Conversation Insight - Inferred Resolution State: Indicates whether the assistant successfully resolved user issues, providing counts of resolved, unresolved, or unknown states.
Empathy Levels Distribution Over Time: Shows the distribution of empathy levels (high, medium, low) in assistant responses, supporting assessment of emotional intelligence.
Negative Emotion Feedback Over Time: Tracks trends in frustration and confusion feedback to help identify and address negative user experiences.

Key Outcomes

Using the Sentiment dashboard, ServiceNow customers can:

Monitor user satisfaction and emotional feedback to pinpoint strengths and pain points in assistant interactions.
Track empathy and negative emotions to improve the assistant’s responsiveness and reduce user frustration.
Analyze transfer and escalation trends to optimize when and how the assistant defers to live agents.
Benchmark assistant and live agent performance via inferred CSAT scores, guiding prioritization of improvements.
Gain actionable insights into conversation resolution and clarity of assistant guidance, enhancing the overall user experience and success rates.

Analyze user sentiment through customer satisfaction (inferred CSAT) score and CSAT factors such as empathy, frustration and confusion, transfers and escalations from conversations with assistants to improve the quality of user interactions.

The Sentiment dashboard page aggregates metrics related to user satisfaction, emotional feedback, empathy levels, and conversation outcomes. These metrics enable you to monitor inferred CSAT, track transfers and escalations, analyze empathy distribution, and review negative emotion trends. The insights from these metrics support targeted improvements to assistant behavior, response quality, and overall user experience.

Sentiment dashboard page in Assistant analytics. — Figure 1. Sentiment dashboard page in Assistant analytics

The visualizations on the Sentiment page help you with the following.

Monitor user satisfaction and sentiment trends to identify strengths and areas for improvement in assistant interactions.
Track emotional feedback and empathy levels, enabling you to address frustration, confusion, and other negative emotions.
Analyze conversation outcomes and recommended next steps to guide assistant optimization and enhance resolution rates.

Overall Sentiment: This area of the dashboard shows the overall average inferred CSAT score for analyzed conversations in the selected date range. The CSAT score is measured on a scale from 0 to 5, where 0 represents the lowest satisfaction and 5 represents the highest. Use this metric to track changes in sentiment over time and evaluate the impact of assistant updates.
Figure 2. Overall Sentiment
Conversations Analyzed: This area of the dashboard shows the total number of conversations analyzed for sentiment in the selected date range. This number indicates the breadth of data supporting CSAT scores.
Figure 3. Conversations Analyzed
High Empathy Rate: This area of the dashboard shows the percentage of conversations where high empathy was detected in assistant responses. It's calculated as ((Number of conversations with high empathy)/(Total number of conversations analyzed)) x 100. High empathy rate is an indication of the assistant's ability to respond with empathy to users queries.
Figure 4. High Empathy Rate
Conversations with Negative Emotions: This area of the dashboard shows the percentage of conversations where negative emotional feedback in terms of confusion or frustration was detected. It's calculated as ((Number of conversations with Frustration or Confusion)/(Total number of conversations analyzed)) x 100. This metric highlights the prevalence of negative experiences in assistant interactions. Use the metric to monitor and reduce negative emotion rates through targeted assistant improvements.
Figure 5. Conversations with Negative Emotions
Average Inferred CSAT Over Time: This area of the dashboard shows daily average of inferred CSAT scores in the selected data range. The CSAT scores are measured on a scale from 0 to 5, where 0 represents the lowest satisfaction and 5 represents the highest. This chart highlights periods of improvement or decline in user sentiment.
Figure 6. Average Inferred CSAT Over Time
Transfers and Escalations Over Time: This area of the dashboard tracks the number of conversations transferred or escalated to live agent. Hover over the trend line to view the number of conversations transferred or escalated to live agent on a given day. This chart helps you with how often assistants require human intervention.
Figure 7. Transfers and Escalations Over Time
Average Inferred CSAT (Virtual Agent): This area of the dashboard shows the average Inferred CSAT score for Virtual Agent interactions in the selected period. For conversations involving both Virtual Agent and live agent, this score reflects only the Virtual Agent CSAT. Scored on a 5-point scale, 0 = least satisfied and 5 = most satisfied. Use this metric to benchmark assistant performance and prioritize improvements where satisfaction is lowest.
Figure 8. Average Inferred CSAT (Virtual Agent)
Average Inferred CSAT (Live Agent): This area of the dashboard shows the average Inferred CSAT score for live agent interactions in the selected period. For conversations involving both Virtual Agent and live agent, this score reflects only the live agent CSAT. Scored on a 5-point scale, 0 = least satisfied and 5 = most satisfied. Use this metric to benchmark assistant performance and prioritize improvements where satisfaction is lowest.
Figure 9. Average Inferred CSAT (Live Agent)
Average Inferred CSAT (Session): This area of the dashboard shows the average Inferred CSAT score for all interactions handled by Virtual Agent or a combination of Virtual Agent and live agent in the selected period. Scored on a 5-point scale, 0 = least satisfied and 5 = most satisfied.
Figure 10. Average Inferred CSAT (Session)
Assistant Recommended Next Steps: This area of the dashboard shows how clearly the assistant explained what happens next or what the user should do. Low: Conversations where no clear guidance was provided. Medium: Conversations where some guidance was provided. High: Conversations where clear and complete guidance was provided.
Figure 11. Assistant Recommended Next Steps
Conversation Insight Inferred Resolution State: This area of the dashboard shows the conversations where the user's issue was resolved. Yes: conversations where the assistant met the user's needs. No: conversations where the assistant didn't meet the user's needs. Unknown: conversations where the resolution state couldn’t be determined.
Figure 12. Conversation Insight Inferred Resolution State
Empathy Levels Distribution Over Time: This area of the dashboard shows the distribution of empathy levels (High, Medium, Low) in assistant responses in the selected date range. Use this chart to assess the emotional intelligence of assistant interactions and target improvements.
Figure 13. Empathy Levels Distribution Over Time
Negative Emotion Feedback Over Time: This area of the dashboard tracks the feedback related to negative emotions: frustration and confusion in conversations with assistants. This chart helps you identify trends in negative user experiences and take necessary steps to reduce the negative feedback.
Figure 14. Negative Emotion Feedback Over Time