Testing website data, design and content: A guide to your first A/B test

A/B testing is an easy way to test data, design, and content on your website. The process allows you to learn more about your site visitors and tailor the experience to their specific needs.

What is A/B testing?

A/B testing compares two versions of a webpage against one another, then declares a winner based on the most successful conversion rate. When a visitor lands on a webpage, they are served up either the A test, or the variation, called the B test. It is very common to test call-to-action copy, website copy, specific interactions, or design elements.

A/B Testing is a crucial means of measuring that allows you to see into your users behavior. It gives you the opportunity to learn more about your audiences preferences and test various design solutions without an entire revamp of your website. Additionally, it gives your team the freedom and power to constantly innovate, allowing your website to become an organic and ever-evolving product. By learning more about your visitors, you are able to optimize your website, creating a voice between the website and your audience.

What can you test?

You can test a variety of content options, from copy to design. Here is a list of A/B testing possibilities to get you thinking:
  • Headlines
  • Sub headlines
  • Paragraph text
  • Testimonials
  • Call to action text
  • Call to action Button
  • Links
  • Images
  • Content near the fold
  • Forms

How to get started

  1. Get to know your data
    The first step to beginning any A/B testing experiment is to begin with metrics. Analyzing your website data, bounce rates, conversion rates, and page views will help you establish a baseline.
  2. Identify testing goals
    After combing through the data, the next step is to specify what to test. Whether it’s a specific call-to-action, button, or other type of content, you need to identify what to test and set a goal.
  3. Execute the test
    Once you decide what you are testing, create your opposing test. For example, if you are testing content, you should create a second version of the content to test against. Then you're ready to run the test.
  4. Compare results and take action
    We typically recommend allowing two weeks for a test to run before making any assumptions based on the results. Results can significantly change depending on the amount of visitors to your site. Visual Website Optimizer has a great tool that calculates this for you. Once a winner is declared, you can immediately implement the successful test and see improved website performance.
That's it! By following these steps you have the opportunity to fine-tune your website according to your user's needs. Keep in mind, you may not always get a declared winner and additional testing may be necessary.

More resources

Smashing Magazine's ultimate guide to A/B testing
Usability Geek's intro to A/B testing
Mashable's tools and resources for AB testing

Popular tools for A/B testing

Optimizely
Visual Website Optimizer
Unbounce

Need help planning or executing your next A/B test? TBH Creative can help. Contact us today to get started.
TBH Creative

About TBH Creative

TBH Creative is an Indianapolis-based internet strategy agency. Our services include web strategy & planning, website design, mobile solutions, application development, and analytic reporting.

Find out more

Receive articles in your inbox