Want to know how to AB test your banner ads? We’ve got the answers.

Online advertising has become much more demanding, meaning any small beneficial changes you can make to your ad banners can, potentially, have a huge impact on your overall campaign performance. But how do you know what changes you should make? This is where AB testing comes in.

Rather than relying on your gut feeling, your opinion or your friend’s opinion (as valuable as all these things may be) really, the most important thing is what actually works.

While, yes, there are “universal truths” and guidelines for creating good ad banners you should always be led by what the data tells you. So, what should you AB test then?

How to A/B test - What is A/B testing

AB Test, Test, Test

  • Design

    • Try playing around with the use of colors; light backgrounds, dark backgrounds, strong contrast between visual elements. Also, think about arranging elements differently in your creative. Does using photos result in more engagement than illustrations?
  • Ad Copy

    • You can experiment with different wording in your ad copy; would your ad perform better by emphasising the benefits of your product or service? Or would it work better by emphasising its features? Also, don’t forget you can play around with the formatting of the text: different fonts, font sizes and weights and, of course, capitalisation. Does having everything in caps make it more noticeable, or do people find it off putting?
  • Call To Action

    • As covered in our last post about creating awesome ad banners, your creative should have a clear CTA. But you can test different wording, like “Join Now” or “Sign Up”. Does red colour really work the best for CTAs?
  • Pricing

    • Using different pricing levels or promotional offers is a good thing to test in your ad creatives. For example, using $0.99 rather than $1; which one will drive more engagement? Would offering free shipping or a discount on the final basket total work better?

Things To Keep In Mind…

Take One Thing At A Time

One thing at a time

Don’t modify and test too many different elements at once. If, for example, you want to test ad copy, only test that – all your design elements, pricing and CTA should be the same. The reason being is that, if you change too many things at once, it’s going to be much harder for you to see later what change led to your improved results.

ab testing, how to ab test, great things take time

It takes time, but it’s worth it

AB testing can sometimes seem like a daunting task because it’s seen as adding another step to the process of creating your ad campaigns, taking up more of your valuable time. Many of us work, flat-out from project to project, so the thought of having another thing to cover can be off putting.

But, think of it this way: rather than breaking your campaign down into “Concept – Creation – Targeting – Optimization” phases, think of the “Optimization” phase as an overall guiding principle that dominates all.

You should also avoid jumping to conclusions too early. Depending on the volume of your traffic, you should let your AB test run for around 4 weeks, continuously, then look at the data to understand which one performed better.

Image Source: giphy

Image Source: giphy

What are you waiting for?!

It may seem obvious to test and optimize every element of your ad campaigns, but a lot of companies fail to do so, meaning they’re missing out on one of the huge advantages that digital marketing has over traditional marketing. With digital marketing, you can gather information and insights, allowing you to react and improve constantly.

Are you already running AB testing for your ad campaigns? If so, what trends or insights have you managed to gather? Let us know in the comments below!

Pin It on Pinterest

Shares
Share This