Understanding Unexpected Health Scores
Last Updated: January 7, 2026 Estimated Reading Time: 7 minutes
Overview
Health scores sometimes appear unexpected or incorrect. This guide helps you understand why a CI has a particular health score and how to investigate discrepancies. Most "incorrect" scores are actually revealing real data quality issues.
Common Health Score Questions
"Why did my health score drop suddenly?"
Possible Causes:
1. Data Became Stale
Explanation: CIs not updated in 90+ days receive lower freshness scores
How to Check:
- Click the CI to view details
- Check Health Breakdown → Data Freshness component
- Look at Last Updated timestamp
- If > 90 days ago, this is the cause
Solution:
- Update CI data in ServiceNow
- Set up automated discovery to keep data fresh
- Schedule quarterly data reviews
2. Relationships Were Removed
Explanation: If CI had relationships that were deleted, relationship health drops
How to Check:
- View CI details
- Check Health Breakdown → Relationship Health
- Look at relationship count
- Compare to previous state (use history/audit logs)
Solution:
- If relationships were incorrectly removed, restore them
- If intentional (CI decommissioned), update operational status
3. CSDM Violations Detected
Explanation: Sync may have detected new CSDM compliance issues
How to Check:
- View CI details
- Check Health Breakdown → CSDM Compliance
- Look for specific violations listed
- Use CSDM Workbench to visualize
Solution:
- Fix CSDM violations using CSDM Workbench
- See CSDM Compliance Guide
4. Required Fields Cleared
Explanation: If required fields were emptied, completeness score drops
How to Check:
- View CI details
- Check Health Breakdown → Data Completeness
- See which fields are missing
- Review ServiceNow change history for CI
Solution:
- Fill in missing required fields
- Investigate why fields were cleared
5. Criticality Changed
Explanation: If CI was marked more critical, health standards become stricter
How to Check:
- View CI details
- Check Criticality field
- If recently changed from Low/Medium to High/Critical, this affects scoring
Solution:
- Ensure critical CIs have complete data (100% fields filled)
- Accept that critical CIs have higher standards
Health score drops are often early warnings of data quality issues. Investigate rather than dismiss them.
"Why is this CI showing low health when it looks complete?"
Investigation Steps:
1. Check All Health Components
Not just data completeness. Review all 6 components:
- Click CI to view details
- Expand Health Breakdown
- Review each component score:
- Data Completeness (25%)
- Relationship Health (25%)
- Data Freshness (15%)
- CSDM Compliance (15%)
- Operational Status (10%)
- Business Criticality (10%)
2. Common Hidden Issues:
Missing Non-Obvious Fields:
- Fields may appear filled in UI but empty in database
- Check field values in ServiceNow directly
- Look for whitespace-only values
Invalid Relationships:
- Relationship exists but points to deleted/invalid CI
- Circular dependency (A → B → C → A)
- Wrong relationship type for CSDM
Stale Despite Appearing Current:
- CI was "updated" but only minor fields changed
- ServiceNow discovery not running
- Last meaningful update was months ago
CSDM Violations:
- Relationships don't follow CSDM rules
- Service hierarchy incorrect
- Missing required service layers
"Why does a decommissioned CI have low health?"
Answer: This is expected and correct
Explanation:
- Decommissioned CIs should have relationships removed
- Do not owner assignment needed
- Data freshness penalized (not actively maintained)
- Low health score indicates "don't rely on this data"
Solution:
- If CI is truly decommissioned, accept low health
- Archive or delete CI if no longer needed
- If CI is still operational, update status to "Operational"
Don't aim for 100% health across all CIs. Decommissioned, dev/test, and low-criticality CIs can have lower health scores.
"Why do identical CIs have different health scores?"
Common Reasons:
1. Different Last Updated Timestamps
Even identical data, if updated at different times, affects freshness:
- CI A last updated 30 days ago: 100% freshness
- CI B last updated 120 days ago: 50% freshness
2. Different Relationship Quality
Same number of relationships but different quality:
- CI A has valid, CSDM-compliant relationships
- CI B has circular or invalid relationships
3. Different Criticality
Health standards adjust based on criticality:
- Low criticality CI with 80% completeness = 90 health
- Critical CI with 80% completeness = 70 health
4. Environment Differences
Production vs. dev/test:
- Production CIs held to higher standards
- Dev/test CIs can have looser requirements
"My CI has 100% data completeness but low overall health. Why?"
Answer: Data completeness is only 25% of health score
Calculation Example:
Data Completeness: 100% × 25% = 25 points
Relationship Health: 0% × 25% = 0 points (orphaned CI)
Data Freshness: 50% × 15% = 7.5 points (stale)
CSDM Compliance: 100% × 15% = 15 points
Operational Status: 100% × 10% = 10 points
Business Criticality: 100% × 10% = 10 points
Total Health Score: 67.5 / 100 = 68% (Fair)
Solution: Address weak components (relationships and freshness in this example)
Investigating Specific Health Issues
Step-by-Step Investigation Process
1. View CI Details:
Navigate to CI Health Workspace
→ Search for CI
→ Click CI card
→ View Health tab
2. Check Health Breakdown:
Expand Health Breakdown panel
→ Review each component score
→ Identify lowest-scoring components
→ Click component for details
3. Review Specific Issues:
Scroll to "Issues Detected" section
→ Read specific problems identified
→ Note recommended actions
→ Prioritize by impact
4. Compare to Similar CIs:
Use CI Health Workspace filters
→ Filter by same CI class
→ Sort by health score
→ Compare to high-scoring CIs
→ Identify differences
5. Review History:
Navigate to CI History tab
→ View health score trend chart
→ Identify when score dropped
→ Correlate with ServiceNow changes
Using AI Assistant for Investigation
Ask AI:
- Navigate to AI Assistant
- Ask specific questions:
Example Prompts:
- "Why does CI PROD-WEB-01 have low health?"
- "Compare health of PROD-WEB-01 and PROD-WEB-02"
- "What's causing the relationship health issue for [service name]?"
- "Show me all servers with health score drop in last 7 days"
AI Assistant analyzes health components and provides specific remediation steps.
Health Score Edge Cases
High Criticality + Missing Data = Very Low Score
Why: Critical CIs must have near-perfect data
Calculation:
- Missing 1 field on Low criticality CI: Minor penalty
- Missing 1 field on Critical CI: Major penalty
Solution:
- Either fill all fields for critical CIs
- Or reduce criticality if CI isn't truly critical
New CI with N/A Health Score
Why: Health not yet calculated
When Happens:
- CI just created
- Sync in progress
- CI missing critical data needed for scoring
Solution:
- Wait for next sync (recalculates health)
- Manually trigger health recalculation: CI Health Workspace → Recalculate
CI with 0% Health Score
Why: Severe data quality issues
Common Causes:
- Completely orphaned (no relationships)
- All required fields missing
- Operational status unknown
- Not updated in 180+ days
- Multiple CSDM violations
Solution:
- Likely abandoned CI
- Either fix comprehensively or decommission
Production CI with Lower Health Than Dev CI
Why: Production CIs held to higher standards
Explanation:
- Same data, different environment = different expectations
- Production completeness target: 95%+
- Dev completeness target: 70%+
Solution:
- This is intentional design
- Focus on improving production CI data quality
Validating Health Score Accuracy
Manual Verification Process
1. Check Data Completeness:
In ServiceNow:
- Open CI record
- Count empty required fields
- Calculate: (Filled Fields / Total Required) × 100%
In cmdbx:
- View CI details → Health Breakdown → Data Completeness
- Compare percentages
- Should match exactly
2. Check Relationship Count:
In ServiceNow:
- Open CI record
- Navigate to Related Items
- Count relationships
In cmdbx:
- View CI details → Relationships tab
- Count relationships
- Should match (±1-2 due to sync timing)
3. Check Last Updated:
In ServiceNow:
- Open CI record
- Check "Updated" field
In cmdbx:
- View CI details
- Check "Last Updated"
- Should match (±24 hours due to sync delay)
4. CSDM Compliance:
Use CSDM Workbench:
- Load service in CSDM Workbench
- Click "Validate"
- Review violations
- Should match health breakdown CSDM component
If Health Score Seems Incorrect
1. Force Recalculation:
CI Health Workspace
→ Select CI or "All CIs"
→ Click "Recalculate Health Scores"
→ Wait 2-5 minutes
→ Verify score updates
2. Check Calculation Logs:
Settings > Sync Status → Health Calculation
→ Search for CI sys_id or name
→ Review calculation details
→ Look for errors or warnings
3. Compare to ServiceNow:
If cmdbx shows low health but ServiceNow data looks good:
- Verify data synced correctly (run manual sync)
- Check if cmdbx has stale cached data
- Review sync logs for errors
4. Contact Support:
If health score still seems incorrect after investigation:
- Document your findings
- Include: CI details, expected vs. actual scores, screenshots
- Email [email protected]
Improving Low Health Scores
See Improving Health Guide for comprehensive improvement strategies.
Quick Fixes:
For Data Completeness Issues:
- Fill missing required fields in ServiceNow
- Resync to update cmdbx
- Target: 90%+ completeness for all production CIs
For Relationship Issues:
- Add missing relationships (owner, hosting, dependencies)
- Remove invalid relationships
- Fix circular dependencies
For Freshness Issues:
- Update CI data during maintenance windows
- Enable ServiceNow Discovery
- Set up quarterly data reviews
For CSDM Issues:
- Use CSDM Workbench to model services correctly
- Follow CSDM relationship rules
- Insert missing service layers
Related Articles
- Health Scoring - How health scores are calculated
- Improving Health - Step-by-step improvement guide
- CSDM Compliance - Fixing CSDM violations
- Common Errors - Error reference
Need Help?
Use the AI Assistant to ask specific questions about your health scores, or contact [email protected].