Health Scores

Saturn assigns every monitor a health score (0-100) and letter grade (A-F) based on reliability, performance, and incident history.

Scoring System

Score Ranges

Score	Grade	Status	Interpretation
95-100	A+	Excellent	Near-perfect reliability
90-94	A	Very Good	Occasional minor issues
85-89	B+	Good	Acceptable performance
80-84	B	Above Average	Some concerns
75-79	C+	Average	Needs attention
70-74	C	Below Average	Action required
60-69	D	Poor	Serious issues
0-59	F	Critical	Failing

How Scores are Calculated

Health scores combine multiple factors:

Health Score = (
  Uptime Weight × Uptime Score +
  Incident Weight × Incident Score +
  Anomaly Weight × Anomaly Score +
  Consistency Weight × Consistency Score
) / Total Weight

Factor 1: Uptime (40% weight)

Uptime Score = (Successful Pings / Total Expected Pings) × 100

Example:

Expected: 30 pings (daily for 30 days)
Successful: 28 pings
Uptime: 28/30 = 93.3%
Uptime Score: 93.3

Factor 2: Incidents (30% weight)

Incident Score = 100 - (Incident Penalty × Number of Incidents)

Penalty by Type:

• MISSED: -5 points per incident
• LATE: -2 points per incident
• FAIL: -5 points per incident
• ANOMALY: -1 point per incident

Example:

Last 30 days:
- 2 MISSED incidents: -10 points
- 5 LATE incidents: -10 points
- 1 FAIL incident: -5 points
- 3 ANOMALY incidents: -3 points

Incident Score = 100 - 28 = 72

Factor 3: Anomaly Frequency (15% weight)

Anomaly Score = 100 - (Anomaly Rate × 100 × 5)

Example:

Anomalies: 4
Total runs: 100
Anomaly Rate: 4%

Anomaly Score = 100 - (0.04 × 100 × 5) = 80

Factor 4: Consistency (15% weight)

Consistency Score = max(0, 100 - (CV × 200))

Where CV = Coefficient of Variation = StdDev / Mean

Example:

Mean duration: 12 minutes
Std Dev: 3 minutes
CV = 3/12 = 0.25

Consistency Score = 100 - (0.25 × 200) = 50

Combined Example

Monitor: Daily Backup

Uptime Score: 93.3
Incident Score: 72
Anomaly Score: 80
Consistency Score: 50

Health Score = (
  0.40 × 93.3 +
  0.30 × 72 +
  0.15 × 80 +
  0.15 × 50
) = 37.32 + 21.6 + 12 + 7.5 = 78.42

Grade: C+

Time Windows

Health scores are calculated for multiple time windows:

Window	Use Case
7 days	Current health, recent trends
30 days	Monthly SLA reports
90 days	Quarterly reviews
All time	Historical baseline

View all windows in the dashboard.

Org-Level Health

Organization health = weighted average of all monitors:

Org Health = Σ(Monitor Health × Monitor Weight) / Σ(Monitor Weight)

Monitor Weights:

• Critical monitors: 3x weight
• Normal monitors: 1x weight
• Low-priority monitors: 0.5x weight

Set weight in monitor settings:

{
  "name": "Production API",
  "priority": "critical"  // 3x weight in org health
}

Health Dashboard

Org Dashboard Widgets

Health Distribution:

A: ████████████████ 45%
B: ████████ 25%
C: ████ 15%
D: ██ 10%
F: █ 5%

Top/Bottom Monitors:

Best:
1. Daily Backup (A+, 98)
2. API Health Check (A, 94)
3. Log Rotation (A, 92)

Worst:
1. Legacy ETL (F, 45) ⚠️
2. Weekend Deploy (D, 63)
3. Cache Rebuild (C, 74)

Trend:

7-day trend: ↗ +5 points
30-day trend: → stable
90-day trend: ↗ +12 points

Improving Health Scores

Scenario 1: Low Uptime Score

Current: 75 (C+)
Uptime: 85%

Actions:

1. Fix root cause of MISSED incidents
2. Adjust grace periods to reduce LATE incidents
3. Add redundancy/retries to jobs
4. Impact: +10-15 points

Scenario 2: High Incident Count

Current: 70 (C)
Incidents: 15 FAIL in 30 days

Actions:

1. Review and fix failing jobs
2. Add input validation
3. Improve error handling
4. Monitor dependencies
5. Impact: +15-20 points

Scenario 3: Frequent Anomalies

Current: 82 (B)
Anomalies: 12% of runs

Actions:

1. Investigate performance degradation
2. Optimize slow queries/operations
3. Tune anomaly thresholds (if false positives)
4. Impact: +5-10 points

Scenario 4: High Variance

Current: 78 (C+)
CV: 0.4 (very inconsistent)

Actions:

1. Identify and fix variable performance
2. Consistent resource allocation
3. Remove dependencies on shared resources
4. Impact: +5-10 points

SLA Reporting

Export health scores for SLA compliance:

Via Dashboard

1. Go to Analytics → Health
2. Select time range
3. Click Export Report
4. Choose format (PDF/CSV)

Report Contents

Saturn Health Report
Organization: Acme Corp
Period: Oct 1-31, 2025

Summary:
- Overall Health: 87 (B+)
- Total Monitors: 45
- Uptime: 97.2%
- Incidents: 23

Grade Distribution:
- A: 18 monitors (40%)
- B: 15 monitors (33%)
- C: 8 monitors (18%)
- D: 3 monitors (7%)
- F: 1 monitor (2%)

Critical Monitors:
- Production API: 95 (A)
- Payment Processing: 93 (A)
- User Auth: 91 (A-)

[Detailed per-monitor breakdown...]

Health Alerts

Get notified when health drops:

{
  "name": "Production Services",
  "healthAlerts": {
    "enabled": true,
    "thresholds": [
      {
        "score": 80,
        "channels": ["email"]
      },
      {
        "score": 70,
        "channels": ["slack:oncall"]
      },
      {
        "score": 60,
        "channels": ["pagerduty"]
      }
    ],
    "frequency": "daily"  // or "immediate"
  }
}

Benchmarks

Industry Benchmarks

Industry	Avg Health Score
SaaS / Tech	88
E-commerce	85
Financial Services	92
Healthcare	90
Media / Publishing	82

By Monitor Type

Type	Avg Health Score
Health Checks	95
Backups	89
ETL / Data Pipelines	83
Report Generation	86
Cleanup Jobs	91

API Access

# Get monitor health
curl -X GET https://api.saturn.example.com/api/monitors/YOUR_MONITOR_ID/health \
  -H "Authorization: Bearer YOUR_TOKEN"

Response:

{
  "monitorId": "mon_abc123",
  "health": {
    "score": 87,
    "grade": "B+",
    "trend": "improving",
    "breakdown": {
      "uptime": {"score": 95, "weight": 0.4},
      "incidents": {"score": 82, "weight": 0.3},
      "anomalies": {"score": 85, "weight": 0.15},
      "consistency": {"score": 78, "weight": 0.15}
    }
  },
  "period": "30d",
  "calculatedAt": "2025-10-14T10:00:00Z"
}

Next Steps

• Uptime & SLA — Track uptime percentages
• MTBF/MTTR — Reliability metrics
• Percentiles — Performance distribution