Generic Outlier Detector using VC
-You don’t need to change the visual or drag KPIs into it.
Just:
-Add your KPIs to the KPI_Selection table
-Point [Selected KPI Value] to the right KPI
-Reuse the same outlier logic on any visual and data
Outlier detection (per selected KPI over the visual’s rows)
Is Outlier → returns Yes/No depending on which method is selected in the slicer.
Outlier Severity → measures how far a value is beyond the chosen fence, scaled to 0–1 (0 = not outside, 1 = extremely far).
The setup is modular:
-You can add more rules, change thresholds or switch to one-sided detection without changing the overall structure.
-Just manage your KPIs in KPI_Selection, map them in [Selected KPI Value], adjust the rules to your domain, and reuse the same framework everywhere.
By Hadi Kheireddine
