5 Steps to Start A/B Testing in Education Marketing

A/B testing is a great way to identify how design changes affect the success of your school’s web pages. By conducting randomized experiments between two design variations of a university’s site, we can slowly optimize pages to dramatically improve success over time. To start A/B testing as part of your education marketing plan, follow these steps:

1. Form a hypothesis.

Begin by forming a hypothesis. Examine each element and how it might affect your users’ actions. Keep a list of potential test ideas and theories, but don’t spend too much time outlining an entire test schedule or series of concrete tests upfront. Often the results of one test may trigger a new theory or eliminate others and may ultimately determine what you test next.

2. Create a test strategy.

Because each aspect of the page contributes to the holistic success of the page, try testing only one variable at a time for accuracy. This way you’ll know exactly what caused the change. Start with common elements such as headline copy, buttons or other calls to action. You can test multiple versions of the variation, but keep in mind that this will likely add to the amount of time it takes you to find a true winner.

3. Define your goal.

Decide on a success metric or goal. Try to align it as much as possible with the variation being tested. Again, it’s best to look only at one goal at a time for accuracy. If your traffic is segmented through a PPC or email campaign, your traffic source could help determine your goal.

4. Give it time.

Be sure you have enough traffic volume and conversions to make the test statistically significant. Be sure there is a clear winner and that your results are not simply caused by chance or noise. Declaring a winner prematurely could be worse than not testing at all if you incorrectly implement a losing variation. Your confidence level should be at least 95% or above to minimize the margin of error from test to test. Since there can also be variances by day of the week, let your tests run for at least a week, even if you have sufficient traffic and conversions.

Once you’ve determined a true winner of your A/B test, analyze your findings to determine what to test next. If you start to see consistent trends, you might decide to focus on one testing direction over another.

5. Document your results.

While many people overlook the importance of documenting their results, it’s important to build an archive of your findings. Knowing exactly what has worked and what hasn’t will help you start to form best practices that can serve as starting points for future redesigns. Documentation is not only helpful for future planning but also for onboarding processes and new recruitment plans.

***
Get deep analysis and take advantage of the latest marketing trends in higher education:

Edu Standard, 2nd Edition

Dusty Steinbrink

About Dusty Steinbrink

Dusty is a Designer and former Front End Developer on the Fathom Creative team. He is a graduate of Kent State University, with a Bachelor of Fine Arts in Visual Communication Design and a concentration in two-dimensional design. He also holds a Bachelor of Arts in Psychology from Lake Erie College and has studied clinical counseling and consumer psychology at a graduate level. His expertise lies in creating dynamic designs that are strategy-driven, conversion optimization, user experience, and thought process analysis.

Leave a Reply