Read about how A/B testing can drastically improve the sites that you design.
A/B testing is an easy yet effective way of assessing what might work best for the site that you’re designing. When it comes down to it, the only real way to know which elements of design are better for your site is by trying it out.
What is A/B testing?
A/B testing consists of serving two versions of a web page and then comparing the results to see which one leads more visitors to complete a call to action. The call of action could be making a purchase, signing up for a subscription or free trial, or clicking on certain pages on the site.
UX designers and web designers use A/B testing as informative research. A/B testing helps designers decide what to change on their site to make it perform better. For example, a designer may be looking to see if a blue theme or red theme will lead more customers to make a purchase, which we call an increase in conversions. The most efficient and accurate way to see which theme will increase conversions (visitors taking action) is by running an A/B test.
A/B test results for transactional pages are measured by the conversion rate. This rate is found by dividing the total number of visitors by the amount of visitors that complete the call to action.
Other pages use measurements such as ad click throughs, engagement, read time, etc.
What to test
Once you know what you’re looking to improve on your site, determine which design element you’re going to test.
Here are some commonly tested elements:
- Color (could be an overall theme, or even just the color of the call to action button)
- Position on the page of a call to action
- Different fonts, size, placement, etc
- Amount of text and wording
A change in something as simple as color can lead your users to action. Be willing to experiment with different design choices to see exactly what your visitors prefer.
How it works
Generally you start with your original model and call this test A (referred to as the control). Then, you make an alternate version with either significant or subtle changes, depending on what you want to test. This is test B.
To run the actual test, you must have software that will complete your A/B test. Visual Website Optimizer is a great tool that provides you with the necessary resources to complete an A/B test (along with much, much more). Visual Website Optimizer, or VWO, offers a free trial that allows you to see everything VWO provides. If you’re a designer, it’s worth investing a bit of money in a potent tool to take full advantage of A/B testing.
Your A/B testing tool will direct a specified number of visitors to site A and site B. This is where your conversion goal, or call to action, comes into play. Your tool records the results and informs you which model – A or B – led more people to the conversion goal. Once you have your results, it will be obvious which design you apply to your site.
Note: There are free tools available such as Google Experiments. This can work well because it integrates results into Google Analytics. You may not get as many additional features with a free version, but it’s worth checking out.
Designing is all about exploring and improving ideas. Running tests is an effective way to enhance your design and really hone in on what your visitors want.
If you’re not convinced that A/B testing can work for you, check out this post about three examples of highly successful A/B tests.
Let us know how your A/B testing works out by commenting below or sending us a tweet @resellbiz.