Broadleaf's new Content Testing module allows the Broadleaf Content Management System (CMS) to be used with Google Analytics’ A/B testing tool, known as Google Experiments. A/B testing is the process of showing users one of two variations of a singular web page (i.e. “Variation A” and “Variation B”) in an effort to maximize an outcome of interest. Examples of this are: page views, time on site, bounces, and conversions. By linking your Google Analytics account to your Broadleaf website, data from Content Tests will be sent to Google Experiments to determine which variation maximizes the selected metric of interest.

What is Google Experiments?

Google Experiments allows site managers to determine whether a user finds a certain font style, layout, content, image, etc., more appealing than another. With traditional Google Experiments, when an experiment is created, a team member selects at least two web pages to be tested. “Variation A” is generally the control, or unedited, page while “Variation B” contains the new element to be tested against. Users then set the length of time for which the experiment will occur.

For the duration of the experiment, page visitors are presented with either Variation A or Variation B, selected at random from a set of weights tied to each variation. While the visitor is on the site, page traffic data tied to that variation is sent to Google. Google then takes this traffic data and applies the Multi-Armed Bandit Algorithm to determine which variation is generating the best traffic based on the selected metric of interest, slowly adjusting the weights of the variations over time. After the pre-set minimum amount of time for the experiment has passed, Google will choose whether Variation A or Variation B is the “winner” based on the available traffic data - ending the experiment, and providing the final results to the user. With Broadleaf, we’re able to take A/B testing a step further.

Why Content Tests?

While Google Experiments allows different web pages to be shown to users, a unique URL must be provided for each variation. This typically requires users to build an entirely new page with a different URL than that of the controlled, or original, page. Utilizing Broadleaf’s CMS, Content Tests provides the ability to change a specific element of a page for each variation without building new pages. Each variation has an "action", which denotes the change that will be made to the page.

The following actions are currently supported by Content Tests:

- Show a different page

- Forward to a different URL

- Use an alternate template

- Change template path (for templates on the file system, and not in the admin)

- Show an alternate Content Item in a Content Zone

- Change a Structured Content Data priority within a Content Item

- Change a system variable

In addition, the option for multiple actions is available, allowing a list of actions to take place for each variation.

Site managers can also identify specific visitors or customer segments to participate in the Content Test based on defined rules. These customizable rules range from determining the percentage of visitors to using Customer Segments to identify test groups. Google Analytics syncs general data from the test every 12 hours, allowing users to view basic Content Test information from the admin. Specific traffic data is available on the Google Analytics’ Experiments page.

Accompanying the release of Broadleaf version 5.2, the Content Tests module is available for Broadleaf enterprise users. For more information about the Content Tests module, or to learn more about Broadleaf, please contact: