Choose metrics

Metrics are a quantitative measure of the success of your experiment. They tell you whether the variations in an experiment are winning, losing, or inconclusive based on changes in visitor behavior in response to your experiment.

Let us use a simple example. Suppose your site has an Add-to-Cart button. You would use an event to track every time the button is clicked. You would use a metric to measure the percentage of users who added to the cart at least once or the average number of items added per user.

Choosing primary and secondary metrics

The primary metric is the one Optimizely Feature Experimentation uses to determine a statistically significant winning or losing variation. It’s the most important goal of the experiment and decides whether your hypothesis is proven or disproven. In Optimizely Feature Experimentation, the primary metric will always achieve statistical significance at full speed, regardless of any other goals or events added.

All other metrics are secondary or monitoring metrics; select metrics that will give you insights into long-term success. See Primary, secondary, and monitoring metrics.

Identifying the right metric is a huge factor in determining whether your experiment will have statistically significant results. See these support articles:

And, of course, if you run into trouble, check out our article on troubleshooting metrics.

How metrics are calculated

Optimizely Feature Experimentation creates metrics by aggregating events over time. Directly track actions like clicks, pageviews, form submissions, purchases, and scroll depth. After you create an event and add it to an experiment, you will decide how it is displayed as a metric.

Adding metrics to experiments

Every Optimizely Feature Experimentation experiment needs at least one metric. You can add or modify metrics at any time. For details, see Run A/B tests. View the metrics attached to your experiment on the Results page.