Often, the goal that people set for their websites is to have a lot of traffic. Another goal that they should have is to increase their conversion rates from this traffic. This is a goal that many webmasters don’t have because they are too hung up on the numbers they get from their traffic reports. But the question is, of what use is all that traffic, when you aren’t converting from it and making sales or getting people to sign up?
A/B testing is useful for a variety of purposes. It is a method that presents site visitors with two versions of a website where only one element has been modified. The reason for this is so that webmasters can quickly pinpoint the reason for the varied results that each version will get during the course of the test.
However, it is also common practice to vary a number of elements at once in an A/B test. This has the advantage of reducing the number of tests to conduct and the testing time required in order to come up with an optimized design for the site.
Where is A/B Testing Done?
A/B testing is often done on a variety of websites, such as online retailers, news sites, blogs, and corporate sites. Some big-name websites that make use of this method include Amazon, eBay, Google, Microsoft, and the BBC. Elements that can be tested on websites include site content, font size and formatting, image placement, color themes, and page layout. Specific elements within these can also be modified and tested, such as the types of images used and whether or not these images will have any accompanying text.
Aside from websites, A/B testing can also be used by marketers in their email marketing or advertising campaigns. Elements that can be tested in email copies include the greeting, content, image usage, font size and formatting, and the call to action.
Do’s of A/B Testing
- Do conduct a series of initial tests to determine the optimum test duration. Running the tests with a duration that’s too short will waste your time, since you won’t be able to use the data from it. Running tests for durations that are too long will also be a waste of time in a sense that you can spend it instead conducting more tests to further optimize your site.
- Do conduct tests again and again.
Conduct A/B tests regularly so you can continuously improve your website to get better results from it each and every time.
- Do be consistent with your A/B tests.
Make sure you apply your variations consistently all across your site, wherever the element might occur in the page. This will prevent site visitors from getting confused and will ensure that the results that you will get are reliable.
Don’ts of A/B Testing
- Don’t test the variation before testing the control.
This is perhaps the basic thing to remember when conducting any test, and not just A/B tests. Testing variations without testing the control first will render the results you get useless, since you won’t have baseline data to compare it to.
- Don’t surprise your site visitors.
Before conducting the test, make sure you inform your site visitors via email or by posting a notice on your site beforehand.
- Don’t ignore the test results.
Some webmasters and designers tend to be more intuitive than practical, meaning they will rely more on their gut feeling than on the test results obtained. Intuition counts, but don’t ignore what the test results are saying. There’s no point in doing the A/B tests if the results will merely be ignored for intuition.
Ruben Corbo is a freelance writer and writes for a number of online marketing websites including those that help online businesses improve A/B Testing techniques to increase sales conversions. When Ruben is not writing he’s producing or composing music for short films or other visual arts.