Definition:
A custom metric is a quantitative measure configured in an analytics tool to analyse data that is not covered by its predefined metrics. In Google Analytics 4 (GA4), it allows numerical values sent through event parameters to be used.
These metrics help analyse business-specific results and can incorporate information from other systems, such as a CRM, when an integration sends it to Analytics. Creating a metric does not establish that connection or automatically start data collection.
Creating a custom metric
In GA4, custom metrics are event-scoped. Before registering one, the transmission of a numerical parameter from the website, application or relevant integration must be configured, and its receipt by Analytics must be checked.
Creating a custom metric requires the Editor or Administrator role on the property. The process consists of these steps:
- Select the property: open the GA4 property receiving the data.
- Open custom definitions: in Admin, go to Data display and select "Custom definitions".
- Start creating the metric: open the "Custom metrics" tab and click "Create custom metric".
- Assign a name: use a clear, unique name that identifies the measure in reports. A description can be added to explain what it represents.
- Specify the event parameter: enter the name of the numerical parameter being sent. It must match the name used in the implementation.
- Choose the unit of measurement: select the unit corresponding to the data, such as standard, currency, distance or time. For monetary metrics, also specify whether they represent costs or revenue.
- Save the configuration: once the data has been sent and the metric created, it usually takes between 24 and 48 hours to become available in reports.
The unit must correspond to the value being sent. If a duration is recorded in milliseconds, it should not be configured as seconds without the appropriate conversion.
Standard properties allow up to 50 custom metrics, and Analytics 360 properties allow up to 125. Before registering a new one, it is useful to check whether a predefined metric already meets the need. The full procedure is available in the GA4 custom metrics documentation.
Examples of custom metrics
The following examples illustrate measures that can provide specific information for analysis:
- Points earned on completing a level: an application can send a numerical parameter with the points earned each time a level is completed. The metric allows accumulated scores to be analysed and compared across levels.
- Duration of a configuration process: a website can send the time spent completing a product configurator. The metric allows that duration to be analysed alongside characteristics of the process.
- Cost of handling a request: an integration with the CRM can send the cost associated with a handled request. The metric allows that amount to be analysed by request category, provided that the integration supplies the necessary data.
In these cases, what each value represents and when it is sent must be defined. Recording the same result several times can cause totals to count it more than once.
Use in reports
Custom metrics extend the possibilities of web analytics by allowing business-specific results to be measured. They can be included in compatible reports and explorations and analysed alongside other metrics.
A metric quantifies a value, whereas a custom dimension describes a characteristic. For example, points earned are a metric, while the difficulty of a level can be a dimension. Combining them allows results to be compared across categories.
They should not be confused with calculated metrics, which are obtained through formulas applied to other existing metrics.