Create metrics

  • Updated

This functionality is now available to all users. Contact your Customer Success Manager or Account Executive for more information.

Metrics in Warehouse-Native Experimentation Analytics are computed properties that can be used in analyses. For example, a Revenue metric can aggregate the Sale Price column from a Transactions dataset to calculate total sales. Metrics are not tied to a specific dataset.

  1. Click New > Metric on the home page. Select an existing template (Simple Aggregation, Filter, And/Or, or Formula) or click Custom to create a custom metric. This choice helps you start quickly; you can change it later without losing any work.

    metrics-1.png

  2. Use the blocks to define the computation for your metric. Click on the title to change the name and add a description.

    Screenshot 2025-03-28 at 7.05.07 PM.png

  3. Validate the metric and click Run Metric to sample the computed values. The results display the metric's values, along with any dimensions. Click Save.

    run-metrics.png

You can now use the metric as a catalog entity anywhere the Catalog is accessible, like inside exploration templates or the block editor.

The Stats Engine supports conversion and numeric metrics. Configuring a custom metric that deviates significantly from these types may not be compatible with the Stats Engine, which can lead to inaccurate results. Ensure your metric aligns with supported formats for reliable analysis.

See Metrics for information.