Skip to main content

Overview

This guide helps CSMs and analysts understand what health scores mean, how to interpret trends and category breakdowns, and most importantly, what actions to take based on different score patterns.
For CSMs: This is your practical guide to using health scores in daily customer management. Focus on the “What to Do” sections.

Reading the Health Score

The Overall Score (0-100)

What it represents: A composite measure of customer health based on Revenue, Usage, Engagement, Support, and Market Signals. Score tiers:
ScoreTierWhat It MeansTypical Action
80-100🟢 HealthyCustomer is thrivingMaintain cadence, explore expansion
60-79🟡 MediumCustomer is stableStandard touchpoints, monitor trends
40-59🟠 High RiskWarning signs presentIncrease engagement, address specific issues
0-39🔴 CriticalSevere risk of churnUrgent intervention required
The score is a starting point, not the full story. Always review the category breakdown and trend to understand WHY the score is what it is.

Understanding Category Breakdown

Click on a customer’s health score to see category contributions: Example:
Overall Health: 65 (Medium) ↓

Revenue:        85 (28% weight) → Contribution: 23.8
Usage:          45 (30% weight) → Contribution: 13.5  ⚠️ LOW
Engagement:     70 (20% weight) → Contribution: 14.0
Support:        55 (15% weight) → Contribution:  8.25
Market Signals: 80 ( 7% weight) → Contribution:  5.6
                                  Total:        65.15
Interpretation:
  • Problem: Usage (45) is dragging overall score down
  • Strengths: Revenue (85) and Market Signals (80) are good
  • Focus area: Investigate why usage is low
Identify the weak category - that’s where to focus your intervention.

Questions to Ask When Interpreting Scores

Look at category breakdown:
  • Which category is strongest?
  • Which category is weakest?
  • Are scores aligned with your qualitative assessment?
Example:
  • Score is 65, you thought customer was doing well
  • Breakdown shows Usage is 40 (low)
  • Investigate: Is usage data accurate? Is low usage a real problem or data issue?
Small fluctuations vs. real trends:
  • Score moving 2-3 points week to week → Normal variance
  • Score dropping 15+ points over a month → Real trend, needs attention
Check historical chart:
  • Is this a long-term trend or short-term blip?
  • Has score recovered from declines before, or is this new?
If score doesn’t match your intuition:Scenario 1: Score too low, but customer seems fine
  • Possible data quality issue (missing usage data, tickets not syncing)
  • Customer in onboarding phase (scores improve as they ramp)
  • Recent one-time event (e.g., single critical ticket dragging score down)
Scenario 2: Score too high, but customer has issues
  • Possible: Revenue masking engagement/usage problems
  • Possible: Recent improvement hiding longer-term concern
  • Action: Flag to admin for configuration review
Trust your instincts but use scores to catch what you might miss.
Three questions:
  1. Is immediate action needed? (Score critical or rapidly declining → Yes)
  2. What category needs attention? (Lowest category score → Focus here)
  3. What intervention makes sense? (Call, training, escalation, expansion talk?)
Rule of thumb:
  • Critical (<40): Action within 48 hours
  • High Risk (40-59): Action within 1 week
  • Medium (60-79): Standard cadence, monitor trends
  • Healthy (80+): Maintain relationship, explore expansion
Red flags for inaccurate scores:
  • Score dramatically different from your assessment
  • Customer in special situation (seasonal, pilot, unique contract)
  • Known data sync issues for this customer
  • Recent company acquisition/merger (confusing data)
What to do if inaccurate:
  • Document the issue
  • Flag to admin for investigation
  • Add manual note to customer profile explaining context
  • Don’t ignore the customer - use your judgment alongside scores

Next Steps


Key Takeaways

Health scores (0-100) provide a composite view, but always review category breakdowns to understand WHY.
Use scores to prioritize your day and week, focusing on critical customers first.
Trust but verify - if scores don’t match your experience, investigate data quality or customer context.