Thursday, 25 October 2012 04:37

Case Study - Premier Data Systems A-B Testing

Premier Data Systems is a company which provides customer relationship management (CRM) software solutions for organizations. They have exceptional customer service and industry specific plugins to help companies streamline their operations.

Problem Posed:

Premier Data Systems came to FiG back in July 2011 in need of:

  • A more user friendly website

  • An e-commerce function

Service Solutions:

  • FiG revamped the existing website

  • Search Engine Optimization (SEO) of the new site

The site was launched September 2011 and after a year, FiG completed a conversion analysis and noticed the analytics were not where we wanted them to be. There was no clear answer or solution as to the best way to change this so to find out what was going on, we began A-B testing of the home page.

The point of an A-B test is to improve something on a site and reach a certain goal.  It is best to make improvement with reasoning behind it. This test looks at two pages where something is different, whether is be the way things are worded, a call to action is different, the layout is different, or it is an entirely different design.

For Premier Data Systems, we looked at the original home page design and then did a complete layout change, creating a second home page to help test. The goal was to find out if we could make a change and receive more clicks and longer visit times.


FiG designed the test with one goal of receiving increased click through and longer page visits and two presentations to provide goal completion data. We were looking for how many visitors of design A completed the goal, and how many visitors of design B completed the goal.

Through the use of code, Google dynamically served up one page or the other as individuals came to the Premier Data Systems website. Over 3 months, the results were recorded within the analytics.

The test resulted in the discovery that the original design had a better completion rate than the alternative design, so no change was needed. The test was successful because it told us that to change the design would do nothing to improve the site and my actually confuse or deter visitors from looking further.

Not happy with how your website is performing? Thinking of changing your website?  Visit FiG or contact us to find out how we can help!