What is a Comparison metric?
Comparison metrics validate consistency between two datasets — computing a metric value from each source and evaluating the difference, ratio, or percentage deviation between them. They are the primary tool for cross-environment validation, migration integrity testing, pipeline checkpoint verification, and ongoing reconciliation between source systems and downstream targets. Type: User-defined — Comparison metrics are created manually. Navigate to Metric → Add and select Subcategory: Comparison. Comparison metrics answer the question: Do these two datasets agree — and by how much do they differ? Prizm computes the configured metric from both sources on each run, records the deviation, and trends it over time. Alerts fire when the deviation exceeds the configured tolerance threshold or when a baseline drift is detected.Metrics
Setup
Step-by-step guide to creating and configuring a comparison metric.
Use Cases
Real-world examples — migration validation, staging vs. production parity, pipeline checkpoints.