What exactly is split testing? Why would you use it?
Put simply, split testing is a way to test exactly what functionality or graphic design works on a website and what doesn’t. It is the method of using two possible experiences for a visitor. Included on a website are experimental variations of the same row or module. The measured outcomes provide data showing which variation has given the best results or conversion rate.
Split testing is sometimes referred to as A/B testing or multivariate testing. Traffic to a website is distributed between an A (control) and a B (variable). Without the visitor knowing they are part of a randomised testing event. This creates a completely controlled experimental environment. The site visitors behaviour, or lack of behaviour is noted and able to be analysed. These behaviours could include interactions such as clicks, form completions and product purchases. The purpose is to determine which of the variants, A or B, creates the best outcome or desired result.
Although the best or optimum outcome differs for all sites, increased sales or lead generation is what we are looking to achieve most of the time. At Woodswork Design we like to say that wherever a goal can be measured, it can be improved. This marketing method can be used to test alternative strategies for any website feature and are often used on calls to action, signup forms and registration pages. Once the split test is in place, the tester waits for any significant data changes to emerge.
What would you change?
The reality is if you are getting visits but not sales, calls or some other desired outcome of your website, it could be time to test a change. For example, testing changes to a checkout logic and functionality could help to determine if any factors may increase conversions from one page to the next and would lead to increased orders rather than page exits.
What may seem to be a subjective choice about web design can be made objective using split testing. Button colours for example are a simple visual distinction. Would a red button work better than a grey button? With split testing it’s easy to see since the data collected from experiments will either support or undermine a hypothesis on which design will work best.
Demonstrating ROI (return on investment) for a testing platform can be measured easily because tests are created with a quantifiable goal in mind.