Performance Insights provides a focused analysis of agent communication quality and behavioral trends over time. It consolidates agent efficacy metrics — empathy, language quality, evaluation scores, and sentiment — into a single time-based visualization for easier interpretation and faster decision-making.
The Agent Performance Monitor includes a trend line for the Kore Evaluation Score, enabling comprehensive quality assessment alongside other performance metrics.
Enable the Conversational Intelligence toggle in Settings to unlock performance insights, agent insights, and conversation-level analytics.
Navigate to Quality AI > Analyze > CX Insights.
Filters
Date Range
Filter Performance Insights data by a specific date range (for example, Last 90 Days, Last 28 Days, Last 7 Days). The default is Today. Select Apply after selection.
Channel Filter
Filter to analyze performance across all channels or limit the view to a specific channel.
Agent Occupancy displays aggregate data across all channels.
Agent Efficacy
Agent Efficacy widgets summarize key communication quality metrics and highlight performance trends compared to the previous period.
Empathy Score
Measures how accurately agents respond to customers’ emotional needs.
How it works:
- Classifies customer utterances as empathy-seeking or non-empathy-seeking.
- Evaluates whether agent responses demonstrate empathy when required.
- Calculates a score based on the proportion of appropriate empathetic responses.
Widget shows:
- Average empathy score for the selected period.
- Percentage change vs. the previous period.
- Upward arrow = improvement; downward arrow = decline.
Crutch Word Score
Measures how frequently agents use filler words during conversations.
Common crutch words: um, uh, like, you know, so, basically.
Why it matters:
- Reduces perceived professionalism.
- Indicates uncertainty.
- Slows conversation flow.
Widget shows:
- Average crutch word usage score for the selected period.
- Trend percentage vs. the previous period.
- Upward arrow = increased filler word usage; downward arrow = reduced usage.
- Lower scores = clearer, more confident communication.
Visualizes relationships among key metrics to identify performance trends and correlations, enabling data-driven coaching decisions.
| Axis | Metrics |
|---|
| Y-axis (agent metrics) | Empathy Score, Crutch Word Score, Kore Evaluation Score. |
| X-axis (customer experience) | CSAT Score, Sentiment Score. |
How to Use
- Select metrics: Choose one or more Y-axis metrics.
- Analyze correlations:
- High Empathy + High CSAT → Empathy drives satisfaction.
- High Crutch Words + Low CSAT → Filler words impact satisfaction.
- Identify patterns:
- Top-right quadrant: High performers.
- Bottom-left quadrant: Agents needing support.
- Compare agents: Benchmark multiple agents.
- Track trends: Use the compare function to observe changes over time.
Visualization Details
| Element | Description |
|---|
| Type | Time-series graph showing performance trends over the selected period. |
| X-axis | Date range based on selected filter. |
| Y-axis | Score values showing performance variations. |
| Checkboxes | Enable or disable individual metrics to focus analysis. |
Actionable insights:
| Pattern | Action |
|---|
| Strong correlations | Replicate best practices from high performers. |
| Weak correlations | Investigate other influencing factors; consider training or resources. |
| Negative correlations | Identify and correct counterproductive behaviors. |
Playbook Adherence
Tracks how consistently agents follow predefined scripts during customer interactions. Supervisors can monitor adherence, identify gaps, and spot coaching opportunities.
Display format:
- Bar Graph: Each bar represents a playbook attribute and its compliance percentage.
- Trend Comparison: Shows percentage increase or decrease vs. the previous equivalent period.
Default Adherence Parameters
If no Agent AI playbooks exist, the system measures adherence against these default conversation etiquettes:
| Script | Example |
|---|
| Greeting | ”Hello, my name is John Doe, and I am your customer support executive. How may I help you today?” |
| Branding | ”Thank you for contacting XYZ.” |
| Privacy Policy | ”This call is being recorded for quality and training purposes.” |
| Hold Etiquette (Voice) | “May I place you on hold for a few minutes while I pull up some information?” |
| Customer Verification | ”Can I know your date of birth?” |
| Proper Sign-Off | ”Thank you for calling XYZ today. It was a pleasure to assist you. Have a great day!” |
Custom Playbook Selection
Use the Playbook dropdown to select a specific Agent AI playbook and view adherence for its configured steps. The system highlights the default playbook in the list.
Administrators can configure playbooks, stages, and attributes in Agent AI. Learn more.
- The Playbook Adherence section appears only when at least one playbook exists in Agent AI.
- If no playbooks or stages are configured, the system displays relevant empty states.
For insights on how agent actions affect customer satisfaction, see CX Insights.