performance-analytics

Build ServiceNow Performance Analytics — pa_indicators (count/sum/avg/percentage), pa_breakdowns, pa_thresholds with severity colors, pa_widgets, and pa_dashboards for KPI tracking.

SKILL.md

GitHub

Performance Analytics for ServiceNow

Performance Analytics (PA) provides advanced reporting and KPI tracking capabilities for measuring and improving business processes.

PA Architecture

Component Hierarchy

PA Dashboard
├── Widgets
│   ├── Scorecard Widget
│   ├── Time Series Chart
│   ├── Breakdown Pie Chart
│   └── Single Score
├── Indicators
│   ├── Number of Open Incidents
│   ├── Average Resolution Time
│   └── SLA Compliance Rate
├── Breakdowns
│   ├── By Priority
│   ├── By Assignment Group
│   └── By Category
└── Thresholds
    ├── Critical: > 100
    └── Warning: > 50

Key Tables

Table Purpose
pa_indicators KPI definitions
pa_indicator_breakdowns Indicator-breakdown links
pa_breakdowns Breakdown definitions
pa_thresholds Threshold rules
pa_widgets Dashboard widgets
pa_dashboards Dashboard containers

Indicators

Indicator Types

Type Aggregation Example
Count COUNT(*) Number of incidents
Sum SUM(field) Total cost
Average AVG(field) Avg resolution time
Percentage (A/B)*100 SLA compliance %
Duration Time calculation Avg time to resolve

Creating Count Indicator (ES5)

// Create "Open Incidents" indicator
var indicator = new GlideRecord("pa_indicators")
indicator.initialize()
indicator.setValue("name", "Open Incidents")
indicator.setValue("description", "Number of currently open incidents")

// Source configuration
indicator.setValue("cube", "pa_cubes_incident") // or table-based
indicator.setValue("facts_table", "incident")
indicator.setValue("conditions", "active=true")

// Aggregation
indicator.setValue("aggregate", "COUNT") // COUNT, SUM, AVG

// Display settings
indicator.setValue("direction", 2) // 1=Up is good, 2=Down is good
indicator.setValue("unit", "Incidents")
indicator.setValue("precision", 0) // Decimal places

// Scoring
indicator.setValue("frequency", "daily")

indicator.insert()

Creating Average Indicator (ES5)

// Create "Average Resolution Time" indicator
var indicator = new GlideRecord("pa_indicators")
indicator.initialize()
indicator.setValue("name", "Avg Resolution Time")
indicator.setValue("description", "Average time to resolve incidents")

indicator.setValue("facts_table", "incident")
indicator.setValue("conditions", "state=6") // Resolved

// Aggregation on duration
indicator.setValue("aggregate", "AVG")
indicator.setValue("field", "calendar_duration") // Duration field

// Time unit
indicator.setValue("unit", "Hours")
indicator.setValue("unit_conversion", 3600) // Seconds to hours

indicator.setValue("direction", 2) // Lower is better
indicator.setValue("frequency", "daily")

indicator.insert()

Creating Percentage Indicator (ES5)

// Create "SLA Compliance" percentage indicator
var indicator = new GlideRecord("pa_indicators")
indicator.initialize()
indicator.setValue("name", "SLA Compliance Rate")
indicator.setValue("description", "Percentage of incidents meeting SLA")

indicator.setValue("facts_table", "task_sla")
indicator.setValue("conditions", "task.sys_class_name=incident")

// Formula-based percentage
indicator.setValue("aggregate", "FORMULA")
indicator.setValue("formula", "(COUNT(has_breached=false) / COUNT(*)) * 100")

indicator.setValue("unit", "%")
indicator.setValue("direction", 1) // Higher is better
indicator.setValue("precision", 1)

indicator.insert()

Breakdowns

Common Breakdowns

Breakdown Field Use Case
Priority priority Incidents by P1/P2/P3
Category category Incidents by type
Group assignment_group Team performance
Location location Geographic analysis
Time opened_at Trend analysis

Creating Breakdown (ES5)

// Create "Priority" breakdown
var breakdown = new GlideRecord("pa_breakdowns")
breakdown.initialize()
breakdown.setValue("name", "Priority")
breakdown.setValue("description", "Breakdown by incident priority")

breakdown.setValue("facts_table", "incident")
breakdown.setValue("dimension_field", "priority")

// Sorting
breakdown.setValue("sort_field", "priority")
breakdown.setValue("sort_order", "ASC")

// Element mapping (optional)
breakdown.setValue("use_element_mapping", true)

breakdown.insert()

Linking Breakdown to Indicator (ES5)

// Link breakdown to indicator
var link = new GlideRecord("pa_indicator_breakdowns")
link.initialize()
link.setValue("indicator", indicatorSysId)
link.setValue("breakdown", breakdownSysId)
link.setValue("active", true)
link.insert()

Thresholds

Creating Thresholds (ES5)

// Create threshold for Open Incidents
var threshold = new GlideRecord("pa_thresholds")
threshold.initialize()
threshold.setValue("indicator", indicatorSysId)
threshold.setValue("name", "Critical Level")

// Threshold conditions
threshold.setValue("operator", ">=") // >=, <=, =, >, <
threshold.setValue("value", 100)

// Visual styling
threshold.setValue("color", "red")
threshold.setValue("icon", "exclamation-circle")

// Notification
threshold.setValue("notification_user", adminSysId)
threshold.setValue("notification_script", thresholdScript)

threshold.insert()

// Add warning threshold
var warning = new GlideRecord("pa_thresholds")
warning.initialize()
warning.setValue("indicator", indicatorSysId)
warning.setValue("name", "Warning Level")
warning.setValue("operator", ">=")
warning.setValue("value", 50)
warning.setValue("color", "orange")
warning.insert()

Widgets

Widget Types

Type Use Case Shows
Single Score Current value "127 Open Incidents"
Scorecard Value + trend Current + sparkline
Time Series Trend over time Line/bar chart
Breakdown By dimension Pie/bar chart
Comparison Multiple indicators Side-by-side

Creating Widget (ES5)

// Create scorecard widget
var widget = new GlideRecord("pa_widgets")
widget.initialize()
widget.setValue("name", "Open Incidents Scorecard")
widget.setValue("type", "scorecard")

// Indicator
widget.setValue("indicator", indicatorSysId)

// Time range
widget.setValue("time_range", "last_30_days")
widget.setValue("show_trend", true)
widget.setValue("compare_to", "previous_period")

// Display
widget.setValue("show_breakdown", true)
widget.setValue("breakdown", priorityBreakdownSysId)
widget.setValue("chart_type", "bar")

widget.insert()

Dashboards

Creating Dashboard (ES5)

// Create PA Dashboard
var dashboard = new GlideRecord("pa_dashboards")
dashboard.initialize()
dashboard.setValue("name", "Incident Management Dashboard")
dashboard.setValue("description", "Key metrics for incident management")

// Layout
dashboard.setValue("layout", "2-column")

// Access control
dashboard.setValue("public", true)
dashboard.setValue("owner", gs.getUserID())

var dashboardSysId = dashboard.insert()

// Add widgets to dashboard
function addWidgetToDashboard(dashboardId, widgetId, row, column) {
  var placement = new GlideRecord("pa_dashboard_widgets")
  placement.initialize()
  placement.setValue("dashboard", dashboardId)
  placement.setValue("widget", widgetId)
  placement.setValue("row", row)
  placement.setValue("column", column)
  placement.insert()
}

addWidgetToDashboard(dashboardSysId, openIncWidget, 0, 0)
addWidgetToDashboard(dashboardSysId, avgTimeWidget, 0, 1)
addWidgetToDashboard(dashboardSysId, slaWidget, 1, 0)

Data Collection

Manual Score Collection (ES5)

// Collect scores for an indicator
var job = new PAScoreCollector()
job.collectIndicatorScores(indicatorSysId)

Scheduled Collection

// PA uses scheduled jobs for data collection
// Default: Daily at midnight
// Configure via: Performance Analytics > Data Collection > Jobs

MCP Tool Integration

Available PA Tools

Tool Purpose
snow_pa_create (action='indicator') Create indicator
snow_pa_create (action='breakdown') Create breakdown
snow_pa_create (action='threshold') Create threshold
snow_pa_create (action='widget') Create widget
snow_pa_operate (action='get_scores') Retrieve scores
snow_pa_operate (action='collect_data') Trigger collection
snow_pa_discover (action='indicators') Find indicators

Example Workflow

// 1. Create indicator
var indicatorResult = await snow_pa_create({
  action: "indicator",
  name: "Open P1 Incidents",
  table: "incident",
  conditions: "active=true^priority=1",
  aggregate: "COUNT",
})
var indicatorId = indicatorResult.sys_id

// 2. Create breakdown
var breakdownResult = await snow_pa_create({
  action: "breakdown",
  name: "By Assignment Group",
  table: "incident",
  field: "assignment_group",
})
var breakdownId = breakdownResult.sys_id

// 3. Attach the breakdown to the indicator
//    (add_breakdown creates the breakdown + the indicator link in one step)
await snow_pa_indicator_manage({
  action: "add_breakdown",
  sys_id: indicatorId,
  breakdown_name: "By Assignment Group",
  breakdown_table: "incident",
  breakdown_field: "assignment_group",
})

// 4. Create threshold
await snow_pa_create({
  action: "threshold",
  indicator: indicatorId,
  type: "critical",
  operator: ">=",
  value: 10,
})

// 5. Create widget
await snow_pa_create({
  action: "widget",
  name: "P1 Incidents Scorecard",
  type: "scorecard",
  indicator: indicatorId,
  breakdown: breakdownId,
})

// 6. Get current scores
var scores = await snow_pa_operate({
  action: "get_scores",
  indicator_sys_id: indicatorId,
  time_range: "last_30_days",
})

Best Practices

  1. Direction Matters - Set correctly (up/down is good)
  2. Meaningful Thresholds - Based on business requirements
  3. Consistent Frequency - Match data volatility
  4. Use Breakdowns - Enable drill-down analysis
  5. Dashboard Purpose - One focus per dashboard
  6. Trend Analysis - Always show comparison
  7. Performance - Limit active indicators
  8. Documentation - Clear indicator descriptions