How to Do Website A/B Testing

A/B Testing or Split Testing is a very powerful tool that can help you improve your website. The idea behind it is very simple and it consists of running a live experiment on your site to test and measure the results of changing a single element on a web page. In the following paragraphs we will explain how you can use A/B Testing to increase conversion (in this example, conversion is the ratio of visitors who fill out a form and give you their contact information out of the total number of visits) and use data to drive changes to your site.

Imagine you want to increase conversion on a landing page, which was developed with the objective of capturing lead contact information. You currently drive traffic to this page and offer a free download of a whitepaper in exchange for users filling out a short form like the one shown below.

Form SampleAt the bottom of the page you see the orange “Submit” button (also known as Call to action), where users click to provide their information and proceed to download their free whitepaper.

After many months using this landing page, you have measured that only one visitor out of every 100 visits completes the form and submits their information. You realize that this 1% conversion rate seems very low and want to do something about it. This is where A/B Testing comes into play. The first experiment you might want to run can be around changing the text in the button, from “Submit” to something more descriptive and direct like: “Get your free whitepaper”. You set up a second landing page with everything exactly the same as the original (control) but you change the text in the orange button:

Lead capture forms

Once this all set up, you direct half the traffic to the original version (A) and the remaining half to version (B). It’s very important to run the experiment on both pages simultaneously, to make sure you keep every other external element constant and get valid results. Now you wait and measure the conversion for both pages over a period of time (you must wait until you have enough data to make the results representative and this period will depend on the number of visits the page gets). Once you feel you have enough data you can sit down and analyze your results. You go ahead and compare the conversion of version A, to that of version B. In this case, let’s imagine that after two weeks version B shows a conversion rate of 3.2%, which is much higher than that of Version A.

You decide to keep the new text and make the change on your landing page.  However, you now begin to wonder whether changing the color of the button from orange to red will increase conversion even more? Or maybe making the font bigger and bold will help? You now have a tool to help you determine which changes improve conversion and which ones don’t.  We love to continuously experiment using A/B testing and we recommend that you do the same thing to improve your conversion and get more leads!