Understanding Customer Sentiment & Agent Empathy Distributions

What These Dashboards Show

This multi-chart view provides a comprehensive analysis of how customer sentiment and agent empathy are distributed across your organization over time and by team.

Four Main Visualizations:

1. Time Series (Top Left): Trends of customer sentiment, agent empathy, and case volume over time
2. Team Scatter Plot (Top Right): Customer sentiment vs. agent empathy by team
3. Distribution Charts (Middle): Percentage breakdown of sentiment/empathy ranges over time
4. Team Breakdown Bars (Bottom): Sentiment and empathy percentages by individual team

How to Read the Time Series Chart

Three overlapping lines:

• Purple line: Agent Empathy %
• Blue line: Customer Sentiment %
• Light blue area: Total Cases Solved (volume)

What to look for:

• Lines moving together: Good correlation between agent empathy and customer satisfaction
• Lines diverging: Disconnect between how agents feel they're doing vs. customer perception
• Drops to 0%: Service outages or no data (like weekends)
• Volume spikes: Check if quality metrics drop during high-volume periods

Example Insight from Screenshot:

• Oct 1: Customer Sentiment 87%, Agent Empathy 77%, 478 cases
• Empathy is slightly lower than sentiment, which is healthy (agents being realistically critical)

How to Read the Scatter Plot (Team View)

Each bubble = one team positioned by:

• X-axis: Agent Empathy (how empathetic agents are in interactions)
• Y-axis: Customer Sentiment (how satisfied customers are)

Ideal quadrant: Top right (high empathy, high sentiment)

Example from Screenshot - Team NL:

• Agent Empathy: 50%
• Customer Sentiment: 23%

Interpretation: Even though agents are showing moderate empathy, customers are very dissatisfied. This suggests:

• Empathy alone isn't solving the problem
• Possible product/policy issues
• Agents may lack knowledge to actually resolve issues
• Communication style may not match customer expectations

The Four Quadrants: Team Scatter Plot

Top Right (High Empathy, High Sentiment) ⭐

• Teams that both care and satisfy customers
• Action: Study and replicate their approach
• Characteristics: Balanced soft skills and problem-solving

Top Left (Low Empathy, High Sentiment)

• Customers are satisfied despite lower emotional connection
• Interpretation:
  • Efficient problem-solvers who get results
  • May handle technical issues where speed matters more than warmth
  • Could be even better with empathy training
• Action: Add soft skills training to complement their efficiency

Bottom Right (High Empathy, Low Sentiment) ⚠️

• Agents are empathetic but customers are still unhappy
• This is a critical insight - empathy isn't enough
• Possible causes:
  • Agents lack the authority to resolve issues
  • Poor product/policies create unwinnable situations
  • Long resolution times despite good communication
  • Agents over-apologize without solving
• Actions:
  • Empower agents with more resolution authority
  • Review policies that prevent resolution
  • Focus training on solution-finding, not just communication
  • Escalate to the product team about the root causes

Bottom Left (Low Empathy, Low Sentiment) 🚨

• Neither emotional connection nor satisfaction
• Urgent intervention needed
• Actions:
  • Comprehensive training overhaul
  • Check for team burnout or morale issues
  • Review team leadership effectiveness
  • Consider team restructuring

How to Read Distribution Charts (Stacked Bars)

Customer Sentiment Distribution (Middle Left): Shows the percentage of cases falling into each sentiment bracket over time.

Healthy distribution:

• Majority (>60%) in purple (>80% sentiment)
• Minimal red (<19% sentiment)
• Small pink section (20-39%)

Warning signs:

• Growing red section = declining satisfaction
• Shrinking purple = fewer highly satisfied customers
• Large middle sections = mediocre, inconsistent experiences

Agent Empathy Distribution (Middle Right): Shows the percentage of cases falling into each empathy bracket.

Healthy distribution:

• Majority (>70%) in purple (>80% empathy)
• Minimal red (<19% empathy)

Warning signs:

• Large red section = agents are disengaged or burned out
• Growing lower tiers = declining morale or training issues
• Check if volume increases correlate with empathy decreases

How to Read Team Breakdown Bars (Bottom)

Shows each team's sentiment/empathy distribution as horizontal bars.

Example from Screenshot:

• Boston Support: 6% red sentiment (bad), 10% high sentiment (good)
  • Analysis: Very polarized - either customers love or hate their service
  • Action: Investigate why experiences are so inconsistent
• San Francisco Support: 20-39% sentiment bracket dominant
  • Analysis: Mediocre performance, customers lukewarm
  • Action: General improvement needed across the board
• Miami Crew: 10% low sentiment, 10% high sentiment
  • Analysis: Very small team with limited data, or highly variable performance

What to look for:

• Long red bars: Teams in crisis
• Long purple bars: Excellent teams
• Long middle bars: Inconsistent or mediocre teams
• Similar patterns across teams: Systemic issue, not team-specific
• One team with a different pattern: Isolate what makes them different

👩‍🏫 Operational Decisions You Can Make

1. Correlation Analysis Use the scatter plot to identify:

• Do empathy and sentiment correlate in your org? If not, something deeper is wrong (policies, product, processes)
• Which teams buck the trend? Learn from positive deviations, fix negative ones

2. Trend Identification Use time series to spot:

• Seasonal patterns: Does sentiment drop during holidays/sales?
• Volume impact: Do spikes in cases lower quality metrics?
• Training effectiveness: Did metrics improve after training rollout?
• Long-term trends: Are things generally improving or declining?

3. Team-Specific Interventions Use team bars to:

• Prioritize coaching: Focus on teams with the largest red bars
• Share best practices: Teams with the largest purple bars
• Balance workload: If empathy drops with volume, redistribute

4. Resource Planning

• Hire for high-empathy teams with low sentiment (bottom-right) - they need problem-solvers, not just communicators
• Hire for high-efficiency teams with low empathy (top-left) - they need people skills
• Clone teams in the top-right quadrant

5. Policy Review: If you see:

• High empathy, low sentiment across all teams = policy/product issue, not agent issue
• Escalate to leadership for systemic changes

🏆 Best Practices

1. Daily time series check for sudden drops
2. Weekly distribution review for trend shifts
3. Monthly team comparison to identify outliers
4. Quarterly scatter plot analysis for strategic planning
5. Combine with case volume data to understand context

👩‍🔬 Advanced Analysis Techniques

Correlation Testing:

• If empathy increases but sentiment doesn't → Empathy training alone won't fix issues
• If sentiment increases without empathy increase → Efficiency improvements working

Segment Analysis:

• Filter by date range to see specific campaign or event impacts
• Filter by team to compare against the company average
• Filter by contact reason to see if patterns differ by issue type

Anomaly Detection:

• Sudden drops in time series = investigate immediately
• Teams moving quadrants = understand why before the shift becomes permanent
• Distribution shifts = early warning of broader issues

📝 Example Action Plan: Bottom-Right Team (High Empathy, Low Sentiment)

Week 1 - Diagnose:

• Interview team members: "What prevents you from resolving customer issues?"
• Review sample cases: Are agents doing everything right but still getting poor feedback?
• Check policies: Are there rules preventing resolution?

Week 2-3 - Experiment:

• Give the team more resolution authority
• Reduce approval requirements
• Test faster refund/replacement policies

Week 4-8 - Monitor:

• Watch scatter plot: Team should move up (higher sentiment)
• Check empathy: Should stay high or improve (less frustration)
• Measure time-to-resolution: Should decrease

Success Metric: Team moves from bottom-right to top-right quadrant