Define the decision, calculation, and data boundary for each metric before designing its chart. A precise-looking number is not useful when teams disagree about what it includes.
Prepared with AI assistance. These are practical scoping recommendations; examples are illustrative, not client results.
Start with the action
Ask what someone will do differently when the metric changes. A count of unresolved jobs may drive reassignment; a total record count may not. Limit the first dashboard to measures with a clear operational use and identify the person responsible for acting on them.
Write the calculation
Specify the eligible records, relevant dates, exclusions, and treatment of missing data. For completion time, distinguish elapsed calendar time from active handling time. Record whether reopened work changes the original result. These choices can materially affect interpretation even when the underlying data is correct.
Show context
Display the reporting period and any freshness limitation. Consider whether users need a comparison or a way to inspect the underlying records. A summary without a drill-down can leave staff unable to explain a surprising result. Make it possible to connect the number to the work it represents.
Reconcile a sample
Calculate a small set manually with the business owner and compare it with the dashboard. Resolve disagreements in definitions before changing the visualization. Keep the agreed examples for later checks so a future workflow change does not silently alter the meaning of a metric that leaders already rely on.