ServiceNow Performance Analytics: Building Dashboards That Actually Help

// Example: Incident volume broken down by assignment group
Indicator: Active Incident Count
Table: incident
Condition: active=true
Aggregate: COUNT

Breakdown Source: Assignment Group
Breakdown Field: assignment_group

PA Dashboards — design for decision-making

The most effective PA dashboards are built around specific questions that specific stakeholders need to answer. Before building a dashboard, identify: who is the audience (IT leadership, team leads, operations staff)? What decisions will they make using this data? What metrics directly inform those decisions? A dashboard built for an IT director answering "Is our ITSM performance improving?" needs different metrics than one built for a team lead answering "How is my team performing this week?" Clarity of purpose produces dashboards that get used; dashboards built without a specific audience and question in mind get ignored after the initial demo.

Scheduled data collection best practices

Run PA data collection during off-peak hours — the collection job queries large tables and adds load to the instance. A daily collection job scheduled at 2am produces the same data quality as one scheduled at 2pm with significantly less impact on business-hours performance. For weekly or monthly indicators that do not need daily granularity, schedule collection accordingly rather than collecting unused data daily. Monitor collection job execution time in the Scheduled Job History — a job that takes 30 minutes to collect suggests indicator filter conditions that are not selective enough or breakdown sources that generate too many dimension values.

Related: PA basics guide · GlideAggregate · SLA management · Scheduled Jobs

Indicators: The Core Building Block

A Performance Analytics indicator is a defined measurement that tracks a specific platform metric over time. Each indicator specifies a table to query, a condition that filters the relevant records, and an aggregation method (typically COUNT, but also SUM for numeric fields). The indicator stores snapshots of its value at each collection interval — hourly, daily, or weekly depending on the indicator configuration. What makes PA powerful compared to standard reports is this time-series storage: a report shows current state, while an indicator shows how that metric has changed over days, weeks, and months. For ITSM teams, the minimum useful indicator set is open incident count, incident backlog by priority, mean time to resolution, and SLA breach rate — these four metrics tell most of what you need to know about service desk health.

Breakdowns for Dimensional Analysis

Breakdowns add a dimensional axis to indicator data. Without a breakdown, an indicator tracks a single number over time: "total open incidents." With an assignment group breakdown, it tracks "open incidents split by assignment group" — you get a separate time series for each group value. Breakdowns can be stacked: assignment group and priority together let you see high-priority incidents by team. The tradeoff is storage and collection cost — each additional breakdown multiplies the number of data points collected per interval. For high-cardinality fields (like individual users), breakdowns can generate millions of rows of PA data quickly. Limit breakdowns to dimensions that drive actual decisions and avoid using free-text fields as breakdown sources. The GlideAggregate guide explains the underlying query mechanism PA uses for breakdown collection.

Automated Analytics and Threshold Alerts

PA can generate automated alerts when indicator values cross defined thresholds. A threshold for "SLA breach rate > 15%" configured on the incident SLA breach indicator will trigger a notification when the rate exceeds that value, enabling proactive response before the situation appears in a weekly review meeting. Threshold alerts integrate with the ServiceNow Notification Engine and can target specific groups or individuals via email or SMS. For escalation workflows, combine threshold alerts with Flow Designer triggers that watch for PA score changes — this enables automated workflow responses to metric deterioration rather than just passive notification.

Scores and Targets

PA Scores aggregate multiple indicator values into a composite health metric. A "Service Desk Health" score might combine incident volume, resolution rate, customer satisfaction, and SLA compliance into a single 0-100 number that represents overall performance. Scores use a formula that weights each component indicator and maps the result to the score range. This abstraction is particularly useful for executive reporting — rather than presenting multiple charts, you present one number with a trend line and drill-down capability for anyone who wants the underlying detail. Pairing scores with PA targets (defined expected values for each indicator) turns the score into a deviation measure: how far is current performance from the defined standard?

Data Collection Troubleshooting

When PA indicators show unexpected flat lines or missing data, the diagnosis starts with the PA Collection Job — the Scheduled Job that triggers indicator data collection at configured intervals. Check whether the collection job is running on schedule and completing without errors. If the job completes but indicator data is missing for specific indicators, check whether the indicator's data collection is enabled and whether the source table query is returning results. A condition that accidentally excludes all records (because a filter references a deleted value or a field that no longer exists) will produce zero data silently. The GlideAggregate COUNT guide shows how to manually run the equivalent of a PA collection query to verify that the expected data exists before debugging the PA configuration itself.

Calculated Indicators for Ratio Metrics

Some of the most useful PA metrics are ratios rather than raw counts — SLA compliance rate (breached / total as a percentage), first-call resolution rate, or reopened incident percentage. PA Calculated Indicators compute these ratios from existing source indicators, automatically maintaining the ratio as both components change. A calculated indicator for "SLA compliance percentage" divides the "SLAs met" indicator by the "SLAs total" indicator and multiplies by 100, producing a clean percentage score that is easier for stakeholders to interpret than two separate trend lines. Pair calculated indicators with PA Targets to show compliance against a defined goal — a target of 95% SLA compliance with the current trend line makes performance gaps immediately visible in dashboards without requiring any manual analysis.

PA Widget Configuration in Service Portal

Performance Analytics data is most accessible when surfaced in Service Portal widgets that operational teams can access without navigating to the PA module. ServiceNow ships PA widgets for the Service Portal that display indicator trend lines, score gauges, and breakdown charts. Configuring these widgets requires specifying the indicator sys_id, the time range, and any breakdown dimensions to display. For executive dashboards, PA score gauge widgets provide the single-number-with-trend-line view that is most effective for stakeholder communication. For team dashboards, breakdown charts showing the same metric split by assignment group or priority give teams actionable visibility into their specific workload. The widget configuration is done through the Service Portal designer, and the PA data populates automatically as long as the indicator collection jobs are running on schedule.