Usage Intelligence
Three usage-analytics lenses in the Analytics workspace — Usage Anomalies (Z-score spike/drop detection), Usage Trends (linear-regression projection), and Usage Cohorts (multi-dimension comparison).
Usage Intelligence
When you need to spot abnormal consumption before a customer does, project where a metric is heading, or see which customers drive the most usage, reach for the three usage-intelligence lenses in the Analytics workspace.
Open Intelligence → Analytics and pick a thumbnail from the landing grid: Usage Anomalies, Usage Trends, or Usage Cohorts. All three read your metered usage.
These three lenses were a standalone Usage Intelligence menu item until 2026-06-16; they now live as thumbnails inside Analytics. A bookmark to the old /advanced-analytics page redirects to Usage Anomalies.
Usage Anomalies
Z-score spike/drop detection over a 90-day lookback (default threshold 2.5σ, 7-day rolling window). Each anomaly reports:
Usage Trends
Historical trend plus a 30-day forward projection per metric, with the growth rate, daily growth rate, R² fit score, and a day-of-week seasonality index. The same view lists the top growing and top declining metrics so you can scan the whole catalog at once.
Usage Cohorts
Usage variance across a chosen dimension, with a spread ratio (top vs bottom cohort multiplier):
Each cohort bucket reports customer count, total usage, and average / median / p95 per customer.
These lenses analyze metered usage patterns. For revenue retention by cohort, use the separate MRR & Cohorts thumbnail in the same Analytics workspace. For the storefront visitor funnel, see Storefront Analytics; for cost/margin analytics, see COGS & Margins.