Help center Reporting and performance Investigate a KPI trend before deciding what to change Updated Wednesday, October 28, 2026 · Marcel Gravelle A KPI trend helps you ask whether a performance change is sustained, isolated or associated with a change in the work. Supportbench lets you request a trend from a metric in a case scorecard while retaining its selected target and reporting context. Use that history to guide investigation; a rising or falling line alone does not explain the cause or prove an individual’s performance. At a glanceFor: support supervisors, team leads and operations analysts with scorecard access. Applies to: Supportbench Core.Start here: Reporting → KPI Scorecards → choose a case-scorecard scope Open a metric’s trend · Understand what the trend represents · Worked example: investigate longer resolution times · Use the surrounding scorecard controls deliberately · A practical weekly review routine · Troubleshoot missing or unexpected trends Open a metric’s trend Open Reporting → KPI Scorecards. Choose an available Agent scorecards, Team scorecards, Division scorecards or Organizational scorecard view. Your role determines which options are available.Choose Period and its date controls, such as Day, Week/Year, Month/Year, Quarter/Year or Year. Confirm the displayed report period.Where relevant, choose Agent team and Select targets, then Use selected targets. Check the selected target chips so the report represents the intended agents, teams or divisions.Find the metric in the case scorecard and select its metric name/button, identified by View KPI trend.Read the KPI trend dialog, including the metric label, displayed date range, target series labels, time buckets and any explanatory status. A clickable metric can still have no saved history for the selected context.Close the dialog with Close and investigate the underlying operational context before deciding on an action.Article-scorecard metrics are not clickable trend controls in the current implementation. This procedure concerns case scorecards; do not assume every metric-looking number elsewhere in Supportbench offers the same trend behavior. Understand what the trend represents The trend uses persisted scorecard-history snapshots, grouped into time buckets for the selected scale. It is not an unrestricted live recalculation of every historical case. The request carries the scorecard’s scope, selected targets, period and team context.ContextWhy it mattersAgent / Team / Division / OrganizationDifferent populations answer different operational questions. Keep scope consistent across comparisons.Selected targetsA multi-target selection is different from one person’s result. Verify the included targets before discussing a change.Period and time scaleThe trend depends on saved history for that scale. Do not compare differently sized buckets as though they measured the same interval.Metric definition and unitsCase volume, response/efficiency, effort and quality measures describe different work. Confirm the definition in the scorecard guide.History availabilityA supported metric may have no saved snapshots for the selected time scale/date range, or the account’s history storage may not contain that metric. Missing data is not a zero value. Worked example: investigate longer resolution times A B2B support team sees its resolution-time measure increase during a period with several complex integration incidents. The useful question is not simply “Who is slower?” It is whether the work mix, unresolved dependencies, handoffs or service process changed.Open the same team or division scope used in the regular operational review and confirm the period.Request the metric’s trend. Look for whether the change appears in one bucket or persists across available history.Review the actual cases and account impact through an appropriate view. Separate routine requests from complex incidents before drawing a conclusion.Check changes in workload, staffing, customer dependencies and investigation complexity. A trend can suggest a question; it does not establish which factor caused it.Agree a specific action, such as improving a specialist handoff or a diagnostic article, and decide what evidence will show whether it helped.Revisit the same metric definition and comparable scope in a later review. Record what changed rather than attributing every movement to the intervention. Use the surrounding scorecard controls deliberately Compare period helps review the scorecard’s comparison context, while Show target and weight exposes configured target/weight information. These are different from the historical trend dialog. A score, a target and a raw metric are related but not interchangeable.If a metric improves while an overall score changes differently, inspect the metric direction, configured weights and the other contributing measures. Do not assume that a single trend should reproduce the overall score. Use the KPI scorecard guide for definitions and interpretation. A practical weekly review routine Start with a customer commitment or operational question, then choose the metric that helps investigate it.Keep the population, period and definition visible in your notes.Look for missing history and unusual sample/workload conditions before interpreting a pattern.Inspect representative underlying work and discuss the context with the people responsible.Record an accountable action and a follow-up check. Avoid treating the graph itself as the completed improvement.RecommendationUse trends for coaching and process improvement alongside case context. A team handling complex B2B escalations may have a different work mix from a team resolving routine requests; a direct ranking can hide that difference. This is an interpretation practice, not a configurable product default. Troubleshoot missing or unexpected trends ProblemChecks and correctionNo trend opens from an article metricArticle metrics are not trendable here. Use a case-scorecard metric.The dialog says history is unavailableCheck the selected period, scope and targets. The metric may lack saved snapshots or the applicable history capability. Do not interpret the gap as zero performance.A metric is clickable but has no plotted valuesClick availability does not guarantee historical data for every metric and period. Read the explanatory state.The graph does not match a freshly filtered case listThe trend uses saved scorecard history and its definitions; the list may use different filters, dates or current states. Compare like-for-like scope first.Two reviewers get different resultsCompare role-visible targets, Agent team, selected targets, period and metric.The change is treated as a proven causeUse underlying case evidence to test the hypothesis. Correlation in the trend does not identify the cause by itself. Continue learning Understand KPI scorecards and metric definitionsBuild a view to investigate the underlying casesExport a scoped list for a review Was this article helpful? Yes No Thank you for your feedback.