Search
  • Gail Malcolm

Why you should A/B test social media ads


A/B testing is a digital advertising strategy that allows you to continuously fine-tune your ads until you find the best ‘formula’ that will help you to understand your audience on a higher level while finding the best formats, copy and creative that will resonate with your audience.

There are so many different variants you can choose to A/B test, including:

  • Copy

  • Creative

  • Format

  • Call to action

  • Landing page

  • Headline

  • Audience

And there’s more where they came from! With so many different variants to A/B test, it’s important to come up with an A/B testing strategy first so that you are able to easily track what you’re testing, what you’ll change as a result and what your next steps will be.

For example, if you’re testing copy, make you sure you know what it is about the copy you are testing. Is it a question vs a statement? A list vs a paragraph? This will be crucial when you come to do your final analysis.

Let’s say you’re testing questions against statements, using the same creative. You might find that in this instance, the questions are the top performers for your audience. Your next steps should be either:

  1. Test the questions only to learn which kind of questions are most effective

  2. Change the creative variant and test this to see if this had an impact on the audience

But why should you continue to test your ads and why is each variant just as important as the other?

Why you should A/B test copy

There are several different ways to use copy in your ads. If you have never changed your copy style before, how do you know if your current style is really working? Here are some copy variants you could try to test:

  • Long form copy

  • Short form copy

  • Style

  • Tone of voice

  • Use of emoji

  • Use of numbers and lists

  • Hashtags

Why A/B test Call to Action

Your call to action could be the key factor in getting your audience to make a final decision. Some call to action A/B tests you could try are:

  • CTA buttons

  • CTA in copy

  • Questions

  • Commands

Why A/B test imagery

Like copy, imagery can be extremely flexible, with many different variations of imagery resonating with different segments of your audience. Imagery to consider A/B testing includes:

  • Lifestyle

  • Products

  • Animation

  • Photography

  • Close ups

  • Light colouring

  • Dark colouring

  • With logo

  • Without logo

  • Price of product/service

Why A/B test video

Video testing is very similar to imagery testing. You could test all the above points in imagery or you could branch out and try testing:

  • Long form

  • Short form

  • Voice over

  • Subtitles

  • No sound

  • Interview

Why A/B test format

Formats are so important when you consider the attention span of your audience or how interactive they are. If you are selling a product that requires a quick decision, you might find that images work best. If you are selling a product where the audience will need more information, video might be the best performing variant – There’s only one way to find out! Some variants to consider are:

  • Image

  • Video

  • Image carousel

  • Video carousel

  • Catalogue

  • Canvas

  • Story

Why A/B test landing page

Once your audience clicks from your ad to your website, your work doesn’t just stop there – ultimately, your overall end goal will be to convert the audience into a customer. Once a user has landed on your website, there may be a few more things to consider to help them on their journey:

  • Page layout

  • Information on the page

  • Lead forms

  • Pop ups

Why A/B test headlines

Headlines can be useful in providing your audience with that extra piece of information that they are looking for. For example, if you are selling a product you may want to include the price or a feature or benefit.

Why A/B test audience

You won’t have one overall audience for your product or service – you’ll have a range of different primary and secondary audiences that you should be tapping into. If you have something that you can sell globally, you might find that different demographics will respond differently to your ads. Consider A/B testing:

  • City

  • Country/State

  • Age

  • Gender

  • Job title

  • Industry

  • Education

  • Interests

  • Behaviours

Other variables you may want to consider testing

  • Placement: Testing a story ad against a newsfeed ad might be insightful! Or seeing how your ads perform on Instagram as opposed to Facebook might tell you that your audience are more responsive to your ads on one of these platforms

  • Time of day: Try testing running your ads early in the morning of later at night to see how they fare; or during travel times

  • Device: Your audience might be using mobile more than desktop