One of your principal aim is to give to your users a good experience on your website and to turn them into customers. This conversion rate is very low if we don’t count the pickup sector. If a website has between 1 and 3% of conversion by visits, it’s average. There are numerous ways to improve the conversion rate but one of them speak speaks itself these last years: The A/B testing method.
What is A/B testing?
The method allows websites to compare two different versions of the same page in order to see which one is more effective. The interest (the aim) is to present to the same number of visitors one of the pages and to see how they’re dealing with it. Are they leaving the website? Are they buying the item in question? Are they subscribing to the newsletter?
In a simple way: Are they doing what we want them to do? Are they increasing our conversion rate?
The modification between the two versions of the webpage can be very specific. Little changes like the color of a button or the place of a text can be a real game changer for the actions of your visitors.
At the end, you will analyze which one of the modifications was the most efficient regarding to your goal. The successful one will be the one to put online.
Why should you do an A/B testing?
Like I said at the beginning, the conversion rate is, on average, very low for the number of visits. So, any improvement is welcome to help reaching your goal.
Imagine that your website receives a lot of visits but those visits rarely end by a sell. This issue can be due, for instance, to some bad visibility of the information . Some studies show that it’s useful to apply this method when a website has over 100.000 visits by month.
If, for example, you don’t understand the behavior of your customers during navigation, it will also give you some guidance to see why they react in some way or another.
It’s also a way to collect data about the navigation on your website and to see how people move around it. It can give you great indication of what to put in some landing pages and way more.
Instead of simply assume without being sure of its success, A/B Testing is a way to know for a fact which solution is the better one.
Some guidance to set up a good A/B testing:
At first, you have to define your goals. What do you want to improve?
Then, take time to collect some data from your customers, see various rates about your website (rebound, number of visits per pages, …)
Those two t steps will allow you to have a first idea of the changes to make.
You can also try it on your newsletters, Facebook Pages etc. But don’t try all the changes, that you want to make, in the same time. You will risk to lose yourself into the data you have collected.
Don’t neglect small modifications. That can also make big improvement on your conversion rate.
And keep in mind that you want to develop a long-term strategy, so do not hesitate to run regular analysis and then an A/B test to find new ways to improve the user experience.
Camille SCHALLER
Etudiante en Master 2 Commerce Electronique à l’Université de Strasbourg, je suis passionnée par les multitudes de leviers qu’offrent l’e-marketing pour les marques, l’e-branding et l’évolution des communautés sur le Web.
