Using Split Tests to Understand Users

Knowledgeable product managers often deploy split tests, such as A/B tests or multivariate tests, to help them understand the impact of the changes they've made to their products, websites, and apps. This course explores some of the best tools for running split tests with a focus on how to set up and run experiments in Google Optimize. In addition, the course describes the details of reviewing split test reports, including how to review the results of a split test in Google Analytics. After this course, participants can begin running their own split tests to find new ways to gain insight into how their users interact. While not a requirement, learners will gain the most from this course by first viewing the O'Reilly Media video course “Using Analytics to Inform Product Design.”

• Understand what split tests are and when split tests are the most effective
• Explore alternative methods to use when split tests are not a viable option
• Learn how to use events, goals, and other data points to determine what to test
• Pick up the distinctions between A/B tests and multivariate tests
• See how split tests are set up using Google Optimize and other available tools
• Learn how to review the results of split tests in Google Optimize and Google Analytics
• Understand how long split tests should run and when they've hit a statistically significant conclusion

Matthew Edgar is a partner at Colorado based Elementive, a consulting firm specializing in web analytics, CRO, UX, and technical SEO. Author of the O'Reilly video series "Using Analytics to Inform Product Design" and the book Elements of a Successful Website, Matt is Google Analytics certified and has worked in the web performance analytics field since 2001. He holds a Master's in Applied Science, Information, Communication and Technology from the University of Denver.