A/B Testing

What is it?

A/B testing (also referred to as split testing) is a method used to compare multiple (typically 2) different versions of the something to determine which one performs best.

You can compare almost anything on your site from different versions of a specific page, to different copy for a particular area or even the availability of a particular feature.

Let’s say you already have a banner on the top of your home page but are unsure whether that or another one you have would work best on your site and be more favourable to your users.

Instead of having to guess or make that decision uninformed A/B testing can be used to give you the information you require to make the right decision on which should be placed on your home page.

How does it work?

Going back to our home page banner example you would create what’s called a variation on your home page where you have a version with the alternate banner.

Then using A/B testing tools you could direct 50% of your website traffic towards your current page and 50% towards the variation of that page.

After a certain amount of time, let’s say a week, you could then view the statistics to see which one performed better based on whatever metric you choose to use e.g. purchases completed, specific category page visited, products reviewed etc.

You of course aren’t limited to a single variation or indeed the positioning of an element. Again in our banner example we could have set up 3 variations, and chose not just a different banner but different placements on the home page.

Why should you use A/B testing?

The main reason to use A/B testing is to allow you to make the right decisions for how your site looks, behaves and supports your users based on actual proof on what your users prefer and how they use your site.

To often we assume what our users will like, particularly when we’ve been in our industries for a long time and have relationships with those that use the site but unfortunately it’s incredibly hard to truly determine what works best without actually experimenting and tracking what does in fact work.

How do I implement A/B testing on my site?

A/B testing strategies can vary from incredibly simple (like our home page banner example used in this post) to very complex ones where there are several variations, metrics, specific audience types etc.

We have a suite of tools at Terracor that we can use to implement whatever strategy you want so just reach out and we’d love to work with you on implementing a test plan for your site.