Editorial draft available to read. Factual and language review is pending. Examples are illustrative.
Decision and method
Define which filters affect each indicator's numerator and denominator. Some indicators show activity during a period; others describe an entire cohort. Do not let an unchanged label hide different calculations. If a filter is meaningless for an indicator, disable it or clearly explain its limited scope.
Illustrative example
Illustrative example: five sales from twenty opportunities belonging to team A produce 25%. Filtering sales alone while retaining one hundred opportunities across all teams would produce a misleading 5%. A contribution view could legitimately use that denominator, but would need a different name and explanation.
Acceptance check
Keep a small table with manually calculated expected results. Test individual and combined filters, including an empty selection. The dashboard must reproduce those calculations and display no data when no valid denominator exists. Confirm that shared links and exported files retain the active filter settings.
Key takeaways
- Every filter must preserve the population defined by the indicator or explicitly explain the change.