ZeroGrad
Capabilities · Classification

SaaS Churn

Healthy
In-context predictionEndpoint saas-churnProduction v3Challenger v4

Real-time production health

Known immediately — no outcomes needed.

Last 24 hours
Predictions
18.2K
Availability
99.998%
P50 latency
12ms
P95 latency
31ms
Error rate
0.00%

Predictions / hour

4657541K00:0004:0008:0012:0016:0020:00

Latency

0ms19ms38ms00:0004:0008:0012:0016:0020:00
P50P95

Inputs, context & predictions

Schema
Healthy
Feature drift
Moderate
seats_active distribution shifted after pricing change
Missing values
Normal
OOD events
0.8%

Input drift vs reference (PSI)

0.1 · 0.25
seats_active_30d
0.14
plan_tier
0.06
support_tickets_open
0.04
last_login_days
0.03
invoices_overdue
0.02

Prediction distribution

today vs reference · no shift
0Predicted churn probability1
Live Reference examples

Routing

In-context prediction
v3 · Managed A100%
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.

Evaluation window
Jun 1 – Jun 30
predictions made in this period
Outcome coverage
98%
have an observed outcome
Median outcome delay
90 days
cancellations and non-renewals
Last outcome received
Sep 30, 18:02
via outcomes API

Business impact

Jun 1 – Jun 30 · 98% coverage
Accounts flagged
1,284
At-risk ARR
$2.31M
Retained after outreach
$412K

June cohort. Retention attribution uses a 10% holdout group that received no outreach.

Model quality

Jun 1 – Jun 30
AUC
.890
Precision
62%
Recall
58%
Calibration
Healthy

AUC — matured outcomes

By prediction week. Recent cohorts are provisional until their outcomes arrive.

.876.889.902Apr 6May 4Jun 1Jun 29Jul 27Aug 24Sep 21
AUC · maturedProvisional (< 60% coverage)
Outcome coverage by cohorttoday →

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

Simulated

No training doesn't mean no learning.

  1. Context: adding product-usage events to the examples improved matured AUC by +.012 (v4 challenger).
  2. The Enterprise slice under-performs (.842). Routing it to TabFM is being evaluated in shadow.
  3. 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.