Introduction to goals in Sitecore Personalize
The purpose of running an experiment is to test changes to what a customer sees in order to improve the outcome of a business goal. The business goal can be anything such as increasing purchases, reducing clicks, or improving website metrics. To test changes and track performance, you must assign a primary goal to the experiment. This involves adding the specific business goal as the first goal in the experiment.
Sitecore Personalize uses the primary goal to:
-
Calculate the guest or session count that must be reached to declare a variant winner or declare the test inconclusive.
-
Measure variant performance when determining a leading or winning variant.
-
Determine the next steps for the experiment after statistical significance is reached, if post-test actions are configured.
If there's additional experiment data you want to analyze, you can add a secondary goal. For example, you can add a revenue goal as the primary goal and the average order value as a secondary revenue goal. After the experiment concludes, you see a 5% increase in the conversion rate for the primary goal, but also see that the secondary goal shows a decrease in the average order value. You might use the results of the secondary goal to run another experiment with average order value as the primary goal.
Any goals you add to the experiment after the primary goal are automatically set as secondary goals. Sitecore Personalize does not use any secondary goals as factors when declaring a variant winner.
After the experiment runs for 24 hours, you can see how your variants are performing against each other on the Performance screen. You can view quantitative data for each variant, eliminating any speculation about which design or webpage element results in more conversions, or other goals.
We recommend you familiarize yourself with experiment metrics and calculations to understand how Sitecore Personalize declares a variant winner, as shown here:
