Churnly AI
History & Analytics
Track every prediction over time, audit model performance against logged outcomes, and drill into individual cases.
Total predictions
8
Accuracy
86%
7 reviewed
Precision
75%
Recall
100%
Risk distribution
High3
Medium2
Low3
Predictions over time
Confusion matrix (reviewed predictions)
True Positive
3
Predicted churn · churned
False Positive
1
Predicted churn · stayed
False Negative
0
Predicted stay · churned
True Negative
3
Predicted stay · stayed
Probability distribution
All predictions (8)
| Customer | Probability | Risk | Source | Actual | Date | Actions |
|---|---|---|---|---|---|---|
| Aiden Reyes | 87% | HIGH | batch | CHURNED | 6/15/2026, 11:56:05 PM | Analysis Review |
| Maria Chen | 22% | LOW | manual | STAYED | 6/15/2026, 5:56:05 AM | Analysis Review |
| Jamal Patel | 64% | MEDIUM | manual | CHURNED | 6/14/2026, 11:56:05 AM | Analysis Review |
| Sofia Romano | 12% | LOW | batch | STAYED | 6/13/2026, 5:56:05 PM | Analysis Review |
| Lukas Becker | 73% | HIGH | manual | STAYED | 6/12/2026, 11:56:05 PM | Analysis Review |
| Priya Nair | 41% | MEDIUM | manual | Pending | 6/12/2026, 5:56:05 AM | Analysis Review |
| Noah Williams | 91% | HIGH | batch | CHURNED | 6/11/2026, 11:56:05 AM | Analysis Review |
| Emma Larsen | 8% | LOW | manual | STAYED | 6/10/2026, 5:56:05 PM | Analysis Review |