Use A/B Testing to Improve Your Website

Posted by in Content Strategy, Data and Analytics, Design, Digital Marketing tagged with

Graphic representing A/B Testing

Want to know what’s really working on your homepage and what’s not? Try an A/B test.

A simple method for tracking the performance of any element on a web page, A/B tests can help you make data-informed decisions on how best to improve content, design, layouts and interactions while providing a better user experience in the process. In this post—with the help of Optimizely co-founder Dan Siroker—we explore how you can use A/B testing to improve your website and get better results.

Your conversion rate is defined as the number of people who take a desired action on your website. Inspiring people to take online action requires a fair bit of trial and error to figure out what works and what doesn’t. A typical website has many elements that can either distract users from tasks at hand or entice them to make choices that are beneficial to both their needs and business goals. Text placement, headlines, copy choices, calls-to-action (CTAs), and button colors are just a few. A/B Testing these elements can help you decipher which ones work for users and which don’t. This information can help you make valuable decisions to optimize site elements and thus improve conversion rates.

A/B Testing, sometimes also called split testing, can help online marketers test two aspects (the A and B) of an online campaign with the goal of improving conversion rates. When running these tests, users are fed one of two options—such as a red button vs. a green button, for example—and click-through rates for each option are tracked. The better performing option is typically what ends up being used on a website.

Dan Siroker, Co-Founder and CEO of Optimizely

While working on Return on Engagement we talked to many industry professionals about how they measure website success. During his tenure as Director of Analytics for the 2008 Obama campaign, Dan Siroker came up with the idea for Optimizely, an online tool that can help individuals with or without technical expertise conduct A/B and multivariate tests on their websites. Some of his insights from our interview are below. The full interview can be found in the ‘Creating Usable Designs’ chapter of Return on Engagement: Content Strategy and Web Design Techniques for Digital Marketing from Focal Press.

What is an A/B Test?

An A/B Test is “a simple way to test changes to a website page against the current design in order to determine which design produces positive results,” says Dan. “It’s a way to validate a change ahead of time, so that you know that it will improve your conversion rate before you alter your site code.” Put another way, A/B tests help you determine the best use of time for your design team.

The Reason A/B Tests are Good For Your Website

In our interview, Dan noted that “improved user experience leads to higher customer conversion, higher customer retention, and improved customer satisfaction. A/B testing is an important tool for improving user experience for websites, and is critical for customer-facing websites.” In other words, you may have the best looking website, but is it achieving the results that you want? Are customers signing up for your newsletter, downloading software, or buying your products? If the answer is “no”, you can always run experiments to find out why.

What Website Elements Should You A/B Test?

You can A/B test any element of your website. Some of the most commonly tested items are:

  • Calls-to-action (i.e. “Buy Now” vs. â€œPurchase Here”)
  • Button placement, colors, designs, etc.
  • Headlines
  • Product pricing
  • Product photos and placement

A/B Tests don’t have to be run on elements being added to the page. They can help you figure out which elements to remove as well. Dan offered an example of Electronic Arts, which used A/B testing methodologies on the website for their popular SimCity game series. By removing a large call-to-action—a banner that read ‘Pre-order and get $20 off on your next purchase’—they actually improved conversion rates by 43%. “What so amazing about this,” Dan says, “is that it was just by removing something from the page, not adding something. They removed a number of things that were on the page that might distract the user or a choice that would prevent them from actually finishing the entire purchase process.”

An A/B Testing Example

Using Optimizely, we tested our own website to determine which iteration of the call-to-action on our homepage lead to higher click-throughs. Our variables were the layout and copy. We chose the click-through rate as our key performance indicator.

Side by side screenshots of the mightybytes homepage with different layouts
Version A (Left): Our site used the grid format and had no clear call to action. When testing this version of the site using Optimizely, we received a dismal 6% click through rate.  Version B (Right): We used a clear call-to-action button “Our Work”. In this version we saw a 21% click through rate. Clearly, this was the winner.

 So Get Testing!

Dan noted that while A/B Testing shows promise, “we’ve got a long way to go – 90% of businesses aren’t even testing. A majority of companies (53%) spend less than five percent of their total marketing budgets on optimization activities.” Yet websites that employ these techniques achieve measurable performance improvements, making the case clear that more companies should get on board the testing train.

For more information on A/B Testing and performance optimization, check out the Optimizely blog for many great stories and tutorials on how to optimize your website. More information on this topic is also included in our book Return on Engagement: Content Strategy and Web Design Techniques for Digital Marketing by Tim Frick and Kate Eyler Werve. Or contact Mightybytes through this website. We’re happy to connect!

Mightybytes is a Chicago-based digital agency and Certified B Corporation. Connect with us on Twitter or get in touch via our contact form.