Most dashboards show what changed.
This shows what diverged.
Anomalies Engine detects cross-system divergence, scores it against five deterministic factors, and only surfaces what crosses a materiality threshold. Everything else is tracked silently.
Alerts fire on thresholds. Anomalies Engine detects when systems begin telling different stories — sales says one thing, finance says another, and the gap is widening.
Every detected anomaly is scored across five weighted factors. Only anomalies that cross the materiality threshold reach the executive surface. Everything else is tracked silently.
The scoring model is visible. Weights are declared. Thresholds are explicit. No hidden ranking, no opaque AI confidence — the same input always produces the same output.
Each scored above 74 on a five-factor weighted composite. Signals below threshold are tracked silently.
Every detected divergence is scored deterministically. Same input, same output. No hidden weights.
Anomalies scoring below this are tracked but suppressed. Only signals that cross this boundary reach the executive surface.
- • Divergences that lack persistence or recurrence
- • Signals with low data confidence behind them
- • Narrow-scope anomalies confined to a single system
- • Changes within normal operating variance