How A/B Testing Improves Your Digital Marketing Efforts

If you are in regular communication with your digital agency, you've probably heard the term "A/B" testing thrown around. At least I hope so, since it's a pretty important part of optimizing your digital marketing campaigns.

Since the world of digital marketing is full of jargon and unfamiliar terminology, I'd like to dig into one particular aspect of it and explain how A/B testing works.

Simply put, A/B testing is the practice of testing multiple variations of a design, call to action (CTA), or phrase in order to see which one performs better.

Sounds simple, right? Let's see how it works.

Let's take a look at one of our clients - a global manufacturer named Hy-Pro Filtration. Hy-Pro manufactures hydraulic filters and other similar products. There is a call to action (we'll say CTA from now on) placed on the home page of the Hy-Pro website that invites the visitor to click and download a reference chart. We could have just placed a single CTA there and hope we designed it will enough to provide maximum effectiveness. But how do we know?

Well, in this case, we are actually running an A/B test on two different versions. Here they are:

As you can see, the first CTA ("A") has a dark blue button while the second CTA ("B") has an orange button. When someone visits the website, one of the CTAs is displayed at random. If someone clicks on it, it is registered as a "conversion" which means someone took action based on the CTA. Over time, we will start to see which CTA has the higher click rate and this one will emerge as the winner.

Once this happens, we can then remove the loser, and try a new version in the interest of seeing if we can beat the winner. For example, let's say we let enough time pass to allow each CTA at least 1,000 exposures. At this point, we see that CTA "A" had a click rate of 3.5% while CTA "B" had a click rate of 1.7%.  Obviously "A" is the winner so we will remove "B". Our goal now is to try to come up with a CTA that gets an even higher click rate than "A". We repeat the process and see what happens.

As you can see, over time we will continue to increase the click rate until we have a highly optimized CTA.

This practice can be used for website copy, email signup forms, banners, or anything else that encourages some sort of action by the user.

As you can see, A/B testing takes the guesswork out of designing good calls to action. With real data, we can easily see what works and what doesn't. Ask your digital agency about A/B testing and make sure you are getting maximum effectiveness from your marketing campaigns.