Warehouse native metrics

Overview

This topic explains how to configure a warehouse native metric for use in warehouse native experiments. Warehouse native metrics are metrics that measure events stored in an external warehouse that you specify. Warehouse native metrics are available for use in only warehouse native experiments.

You cannot create a warehouse native experiment until you have created a warehouse native metric for the external warehouse you want to experiment on.

Prerequisites

Before you can create a warehouse native metric, you must configure a warehouse native Experimentation integration, such as Snowflake.

You should also create a new metric data source for the integration, which you can use to create a warehouse native metric. To learn more, read Metric data sources. You can optionally create a new data source when you create a new metric.

Warehouse native metric requirements

Warehouse native metrics must:

• be either custom conversion binary, custom conversion count, or custom numeric
• use the “average” analysis method

Warehouse native experiments cannot use:

• clicked or tapped metrics
• page viewed metrics
• metric groups
• LaunchDarkly hosted metrics that measure events from LaunchDarkly SDKs
• a percentile analysis method

Create a warehouse native metric

To create a warehouse native metric:

1. Navigate to the Metrics list.
2. Click Create metric. The “Create metric” dialog appears.
3. If you use warehouse native experimentation, you have the option to select either LaunchDarkly hosted or Warehouse native as the source for metric events. Select Warehouse native to measure events from an external warehouse, such as Snowflake.
4. Select an existing data source from Metric data source, or click + Create to create a new data source.

5. Finish creating the metric by continuing the procedure as described in custom conversion binary, custom conversion count, or custom numeric.

You can now create a warehouse native experiment by selecting this metric.