Revity Learn•Billing observability 101

Learn the foundations of billing observability

This guide explains the core concepts behind billing observability: how metrics become analytics, how alerts are triggered, and how teams use these signals to keep revenue healthy.

Core concepts

The building blocks of Revity

These concepts show how raw Stripe data becomes the analytics and alerts that drive action. Each term is used precisely throughout the product.

Observability

Observability is the ability to ask new questions about billing health without guessing. It combines metrics, events, and context so you can explain what changed and why. In billing, this means correlating invoices, subscriptions, payments, refunds, and disputes to understand both revenue and customer impact.

Metrics

Metrics are raw measurements like counts, totals, rates, and ratios. They tell you what happened, but not always why. Good metrics are consistent, defined precisely, and tied to a specific time window.

Analytics

Analytics are interpreted views built from metrics. They turn raw numbers into trends, comparisons, and insights you can act on. Analytics help answer "why did this change?" and "what should we investigate next?"

Alerts

Alerts are signals that something is off or needs attention. A good alert includes context and thresholds so teams can triage quickly and avoid alert fatigue.

Guardrails

Guardrails are pre-flight checks and invariants that prevent billing mistakes before they reach customers. Think of them as automated QA for pricing, discounts, metadata, and configuration.

Anomalies

Anomalies are unexpected spikes or drops (failed payments, refunds, past due). They often indicate revenue risk, operational issues, or changes to billing logic.

Metrics

Raw measurements

Counts: open invoices, failed payments, cancellations.
Totals: MRR, refunds, outstanding balances.
Rates: collection rate, churn rate, past-due rate.
Time-based: average days to pay, retry cadence.
Precision: define exactly how each metric is calculated and what time window it uses.

Metrics answer "what happened?" and are the foundation for deeper insight. When you track a metric, write down the formula and the data source so it stays consistent.

Analytics

Interpreted insight

Trends: MRR trend vs last 30 days.
Comparisons: payment failures this week vs last week.
Breakdowns: churn by plan or customer segment.
Insights: top movers and emerging risk signals.
Context: add thresholds, baselines, and "normal ranges" to interpret change.

Analytics answers "why it happened" and "what to do next." The same metric can produce multiple analytics depending on the question you're asking.

From data to action

How billing observability works

1. Ingest

Read Stripe Billing data like invoices, subscriptions, charges, refunds, and disputes. The goal is comprehensive coverage, not just a single object type.

2. Normalize

Normalize raw objects so metrics and analytics stay consistent across accounts, currencies, and billing models.

3. Analyze

Calculate metrics and analytics per window (7/30/90 days) to highlight changes and compare to a baseline.

4. Detect

Guardrails and anomaly detection surface billing drift and revenue risks early.

5. Act

Alerts provide context, likely causes, and suggested next steps to resolve issues quickly.

Analytics library

Five core analytics groups

These are the analytics groups used in the Stripe extension. Each group answers a specific business question and rolls up multiple metrics into a single view.

Revenue health trends

MRR trend, net revenue retention, churn rate. Answers: Are we growing cleanly? What changed vs last period?