A/B Testing: Tips, Tricks, and Uses


If you run your own site, then you’re probably always on the lookout for ways and means to drive more people to your site and convince them to subscribe or make a purchase. Some webmasters keep trying to launch campaigns to bring more traffic to their sites, but what they can do instead is work to increase the conversions they get from their current traffic.

A/B Testing: How It Can Increase Conversions

One thing they can look into is A/B testing. This is a marketing test method where a baseline sample is pitted against alternative ones in order to determine which one is better. In this case, our control is a website design that will be compared against the same website which has had one element that has been varied. A/B testing is pretty rigorous in the sense that only one element is to be modified for every test run. This is unlike multivariate testing, which allows the variation of any number of elements in the test. However, it does have its advantages. For one, the reason for the varied results obtained from the test can be easily pointed out, since only one element has been modified in each test run.

How to Run A/B Tests

Conducting an A/B test is relatively easy and simple. Before anything else, an assessment of the website or email campaign should be done. It is in this phase where elements that seem out of place or outdated are noted and listed down. It will be these elements that will be subjected to the tests later on. Only one element from this list is selected for the actual test, and two variations of it should then be made. The process of running the A/B tests can be done with the aid of software and apps, which more or less automates the entire process. An example of such tool is Google’s Website Optimizer, which can be accessed for free online. Users only need to sign up for a Google account and register their website so they can start running A/B and multivariate tests on their pages.


Tips and Tricks of A/B testing

Even though A/B testing is one of the easiest testing methods out there, it’s important not to overdo or oversimplify things. Below you’ll find a list of things you should keep in mind when you’re ready to run your own A/B tests:

• Do test both the control sample and test simultaneously. There’s no way to gauge the effectiveness or impact of your campaigns if you don’t run both versions at the same time.

• Do choose an appropriate duration for your test. The length of time when you’ll need to run your test will depend on several factors, one of these being your site traffic. If you have a large following, then you might be able to run your test for a shorter period of time. If you have a smaller site traffic, then take this into consideration when you decide the test duration so you will be able to gather enough data once the test is underway.

• Do make your A/B test consistent throughout the site. When you run your test, make sure you’ve implemented the variation all throughout your website. This way, you won’t confuse any users and you can be more assured of the reliability of the data that you’ll be able to gather from the test.

• Do conduct many tests. When you’re done testing out a certain element, move to the next one on your list and run more tests. A/B tests become easier to run, especially if you’ve already got a few completed runs under your belt.

This is a guest post by Ruben Corbo. He is a freelancer writer specialized in different topics including technology, online marketing with AB Test and Product Recommendation Engines, music, art, and motivation. When Ruben is not writing he is producing and composing music for short films and other forms of visual arts.

This entry was posted in news. Bookmark the permalink.