SaaS Churn
HealthyReal-time production health
Known immediately — no outcomes needed.
Predictions / hour
Latency
Inputs, context & predictions
Input drift vs reference (PSI)
Prediction distribution
Routing
- Backend
- Managed by Zero Grad
- Context
- pinned to v3
- Fallback on timeout
- Managed C · 0.02% of requests
Matured outcome quality
Performance metrics are computed only for predictions with observed outcomes. Cancellations and non-renewals take up to ~90 days to arrive, so recent predictions aren't scored yet.
Business impact
Jun 1 – Jun 30 · 98% coverageJune cohort. Retention attribution uses a 10% holdout group that received no outreach.
Model quality
Jun 1 – Jun 30AUC — matured outcomes
By prediction week. Recent cohorts are provisional until their outcomes arrive.
Outcomes
Ground truth flows back by prediction_id. It improves evaluation and routing — never model weights.
Send outcomes
Every prediction returns a prediction_id. Send the real outcome whenever you learn it.
# Later, when the real outcome is known
zg.capability("saas-churn").send_outcomes([
{"prediction_id": "prd_01J9X4P7QA", "outcome": "churns",
"observed_at": "2026-09-30T23:41:00Z"},
])Outcomes never update model weights — Zero Grad does not train a customer-specific model. They are used to measure real production quality, compare foundation models and context-selection strategies, improve routing, and decide whether a challenger beats the current production configuration.
What Zero Grad learned from your outcomes
No training doesn't mean no learning.
- Context: adding product-usage events to the examples improved matured AUC by +.012 (v4 challenger).
- The Enterprise slice under-performs (.842). Routing it to TabFM is being evaluated in shadow.
- Calibration drifted after the August pricing change; recalibrated on matured outcomes, ECE 0.031 → 0.018.
What adapts: which foundation model serves which slice, how in-context examples are selected, and calibration. What doesn't: model weights.