[Infographic] A Newbie’s Guide to A/B Testing

Imagine that you’re redecorating your house. You head to the hardware store to buy some paint, where you and your friend flick through shades in the catalogue. However, while you like soft apple green, your friend likes blueberry white. So what do you do? You bring home two swatches and paint some test patches to see how they compare to one another before committing to a decision.

This is exactly how A/B testing – or split testing, as it’s sometimes known – works. It’s all about comparing two versions of your website to see which one performs better with your target audience, preferably with the upshot of improving key performance indicators (KPIs) like conversion rate and revenue.

You can use it to test and refine almost every facet of your website – anything from landing pages, promotions, calls-to-action, emails and search ads. Usually, A is the existing feature you’re currently working with (known as the ‘control element’) while B is the alternative option you want to test (‘the variant’). You then split your live traffic into two groups of subjects and direct them towards the two versions. When the experiment is done and dusted, you select the version that has produced the best results and move onwards and upwards to improving your bottom line.

Fortunately (or unfortunately depending on your outlook) only 52% of companies and agencies that use landing pages actually test them for ways to improve conversions. This means that businesses willing to experiment and adopt a nimble stance towards their business can stay one step ahead of the competition. With the right testing methods, companies can see up to 300% increase in their conversion rates.

But how to get started? In this new infographic, we’ll take you from being a beginner to an A/B testing power user – guiding you through the fundamentals of running a successful experiment.

For newbies, it might seem as though there’s a steep learning curve, but if you get the fundamentals right then there are substantial rewards to be gained – more than compensating for the time invested in running an experiment. In short, if you have a winning hypothesis, then you’re on the right path to learnings and significant improvements for your KPIs.