The Basics of A/B Testing
Firstly, what is A/B testing? I'm sure we've all heard the term before but do we really know what it means and what it's used for? A/B testing, also known as Split Testing, is an approach to web design and UX design that is used to find out which changes to a website maximise or minimise interest from users. There are always two versions presented to users (an A and a B) which have one difference that might impact the users behaviour. The reason for A/B testing is to define a metric that offers the best success for your website and your business.
Each A/B test is obviously different as the outcome of the test is dependant on the clients goals. However there are some common elements that are usually tested:
- Calls to action - their size, colours, content, where it's positioned on the site (i.e. above the fold, below etc).
- Main heading wording - is your message getting across with the first line?
- Overall layout and style of the website
- Ads and promotional items on the page
- Amount of text on a page - is it pre-empting the user to want to scroll to read it or is it overbearing?
Once a client knows what they would like to test for then they can commence the process. There are a few things that they should keep in mind through the experiment though. The main being, do not jump to any conclusion before testing is finalised. It is very easy to let your own assumptions and wants overrule the test results. The outcomes of the tests can be surprising and unintuitive, which doesn't mean they are incorrect. In saying that, a client has to know how long they wish to run the test for before coming to a conclusion.
A/B testing isn't as common as other marketing subjects/tools (such as SEO), however it is still very powerful and could be what separates your website from your competitions.
