A/B Testing Explained Graphic

A/B Testing Explained

What Is A/B Testing, and How Heavy Do You Rely on Its Results?

A/B testing is one of the easiest and most straightforward ways to get meaningful insights that improve PPC performance. A/B testing is a split test that shows two different versions of something to two other people – so one person will get the “A” variation, and another person would call the “B” version. Landing pages and ad copy are the most common things you would want to A/B test to optimize your performance continually.

However, A/B testing is not testing two different variations of something. For example, you wouldn’t test your current landing page with a different layout, copy, and images or an ad with a completely different copy from the other.

Instead, you would swap out a headline or description on one ad and then compare that to the headline or description of the original ad; or swap out a different image or call to action on one landing page and compare it to the original image or call to action. This way, you can figure out the winning element and why performance has either improved or fallen off. By testing two completely different pages or ads, then it is pretty much impossible to figure out what exactly was the driving force behind the change in performance.

How Much Should You Rely on Results?

In terms of how heavily you can rely on the results, you’re getting depends on your mileage. If your landing page isn’t getting conversions to start with, and your A/B test isn’t moving the needle, then there might be a more significant issue at play with your page. Conversely, if you’re already having success converting people and image B leads to a 5% increase in conversion rate than the original. You can take those results and implement them.

A/B testing is all about taking things that are already good and making them better. And often, the results you get aren’t necessarily eye-popping numbers but instead more minor and more incremental. But even if your test doesn’t perform as well as expected, you’ve figured out what didn’t work, which can be equally as valuable as finding what works.

About the Author

Colin Brotherton is a Paid Search Specialist running campaigns on Google, Microsoft, and Facebook. When he’s not clicking away at work, he enjoys being outside as much as possible, playing music, and watching sports.

Scroll to Top