Sitecore Experience Platform

Experience Optimization and content testing glossary

Abstract

Important terms and concepts for Experience Optimization and content testing.

One type of A/B test is the basic content test, which compares the effectiveness of two versions of a page. Another type of A/B test is a component test in which you compare different variants of a single component. If there are more than two variants, it is an A/B/N test. Page tests are also either A/B or A/B/N tests, depending on how many pages you test.

The rate of change between the highest tested trailing value per visit and the original trailing value per visit.

A test of the different variations of a single component.

The probability in percent that the difference between the results of the tested experiences after a test has reached the specified number of forecasted contacts is statistically significant.

A test that compares the performance of a new version of the content associated with a page item to the old version of that content.

The percentage that represents how much higher the trailing value is for a tested component compared to the original component. If the original component performs best, the effect is 0.

The combination of the testable content that a contact sees on a single page.

A multivariate test is a test which contains variants of more than one component on a page. The variants can differ in content or personalization. In a multivariate test, there can be many different combinations. For example, if there are 3 components with 3 variants each, they can be combined to make 27 different pages.

A test that compares one complete page against another page, for example, different versions of the same page or different pages.

A test of a combination of personalization rules that measures whether the rules have a positive effect on engagement value or goal conversion rates.

The number of contacts who have viewed a specific personalized variation of the content.

The minimum and maximum duration of a test.

The effect of the test multiplied by the number of visits made to the page.

The measurement that you use to compare test results, for example, the trailing value of visits or the conversion rates of specific goals.

The total engagement value for the variant of the content that is being content tested and that is based only on the page views that occur after the contacts have been exposed to the test, divided by the number of visits to the site with the variant.