For those who don’t know what is A/B split testing, it is a technique of using two versions of a web page, landing page used for PPC advertising or an email marketing campaign, and then comparing the performance of both the versions. Then you can select the version that performs better. You can keep on iterating until you have reached the best possible campaign or webpage. A/B is such an important part of serious online marketing that Google Analytics has made it an integral part of its analytical data.
According to this blog post, the basic purpose of A/B split testing is to gauge the reaction of your audience and then take decisions accordingly. Does your audience prefer red color or blue color? Does it prefer a fixed-with layout or a flexible layout? Does they react better to long copy or shorter copy? On which days of the week your email marketing campaign elicits the maximum response? The list can go on.
How to carry out an A/B split test
As is the case with any other regular campaign, first of all you need to know why you want to conduct an A/B split test? You will need to define your parameters against which you can evaluate the performance of both the options. Mind you it is not a one-time affair. It may take you months of different A/B split tests before you can find the perfect thing that works for you.
Let us suppose you observe that there is a particular page on your shopping cart funnel from where most of the people moving towards the checkout page (where they actually buy your item and pay for it) abandon your website or go to another page. You have an idea of why people abandon your shopping cart but you’re not sure. Suppose there is an image that you think causes the disruption, but you had put the image in the first place because you thought it facilitated the checkout procedure. You want to find out whether the presence or absence of this image affects the abandonments or not. So you create two versions of this page, one with the original image, and one with no image.
Suppose you discover that it doesn’t matter whether the page has an image or not and abandonments are still occurring. So you decide to stick with the image or perhaps then you shift onto a page with a different image as an alternative. On the other hand, if the rate of abandonments decreases then you know for sure that this image is the culprit.
Similarly you can test out various components of your website, for example the headline, the banner, the call to action buttons and the components of your navigation system. You can also have multiple versions of the copy you have on your selected web page.
After you have created alternative versions of the webpages you want to test, you will need some tools to analyse user behavior and the most easily available tool is Google Analytics. It allows you to create parameters that let you decide the performance of the pages you want via A/B split testing. It might be submitting the contact form, clicking a particular link or purchase of a particular item.
Later on you can use the option that gives you a better result.
The greatest benefit of using A/B split testing to increase your conversion rate is that you get to test various options and then choose what’s best for you. And you don’t have to stop at that. Once you have found the best performing headline for you you can move on to your images. Once you have found out that images cannot be improved further, you can start focusing on your copy. By the end of the day, every real estate business wants to improve its conversion rate and this can be accurately achieved through A/B split testing especially when you have clearly defined trackable data to analyse.