Ads that don’t get tested don’t get optimised – and if you’re not optimising, you risk wasting money on ads that don’t work. By testing everything from audiences and placements to images and headings, you can increase your conversions and lower your costs.
In this guide we go through how you can A/B test your ads on platforms like Google Ads and Meta Ads – ensuring that you get the most out of your ad spend.
What is A/B Testing and Why is it Important?
A/B testing is the process of comparing two or more ad formats to see which one performs best.
By A/B testing your ads you can:
- start making data-driven decisions
- invest your budget more efficiently
- optimise ads towards your goals
For example, if you want to generate more purchases, lower the cost per lead, or reach a certain return on ad spend (ROAS).
Here's What You Should be A/B Testing
There are a number of factors that affect how an ad performs. Here are the ones we recommend you test continuously:
Content
Test different types of ad formats, like videos, images, graphics, and carousel ads. It can even be worth testing different variations of each format – different types of videos, images, and so on. The end goal is to compare everything from the overarching message of your ads to details in design and colour choice.
Audience
Methods like audience segmentation are particularly useful when A/B testing. You can build audiences using your own data – for example lookalike audiences to find more users that resemble your current client base, and remarketing audiences where you target your ads to audiences who have shown interest in your offer but haven’t converted yet.
Messaging
Experiment with your campaign messaging, word choice, and tone. Try different lengths, angles, and ways of framing your offer.
CTA (call-to-action)
Test different CTAs to see which one drives the most conversions.
Placement
There are several different ad placements available that are worth testing. While you might be familiar with feed placements in Meta Ads, it’s worth testing Facebook Marketplace or even Messenger ads. On Google Ads, you can experiment with Youtube placements as well as different sites within Google’s Display network.
Landing Pages
Finally, it’s important to remember to optimise your landing page. Here, it’s worth experimenting with everything from layout and images to headers and forms, to improve your conversion rate.
How to A/B Test
1. Ensure that your tracking is set up correctly
Before you start A/B testing it’s important to ensure that you are collecting the right data – and as much of it as possible. Incorrect or incomplete tracking can produce misleading results, which in turn lead to making decisions based on faulty data.
We recommend all of our clients to implement server-side tracking as a complement to traditional pixel tracking. This helps you collect more data, providing a more complete foundation for testing and optimising.
2. Test one thing at a time
In order to get clear results, you need to isolate the variable you’re testing. If you change picture, text, and audience at the same time you won’t be able to tell what is affecting what – making it difficult to draw concrete conclusions.
3. Choose the number of variables
Although it’s called A/B testing, you aren’t limited to two variables. We recommend testing at least three variables (A, B, and C). It is however important to adjust this based on your budget. If you spread your budget too thin across too many variables it will take longer to collect data and identify a winner.
4. Conduct your tests in a structured way
In order to maintain control and flexibility over your tests we recommend that you conduct them manually using standard campaigns, instead of using the platform’s built-in A/B testing tool.
When you test a specific variable, for example an image or a heading, you can create different ads within the same ad group.
If you are testing larger variables, such as audiences and budget strategies, you should instead create separate ad groups or campaigns to compare with each other.
5. Analyse and act on the results
It can be tempting to end a test as soon as you start seeing signs that one variable is outperforming another. But it’s important to be patient and give your test time to collect sufficient data to draw a reliable conclusion.
As a rule of thumb, every variable should generate at least 100 conversions before you start reviewing the results – though the exact number depends on your budget, your goals, and how large the difference between variables is. The smaller the difference, the more data you need to confidently identify a winner.
Once you see a clear pattern – and the data shows that the difference is statistically significant – you’re ready to start acting on the results. Pause the underperforming ads and scale up your budget on the variable that works best.
After that it’s time to test the next variable and optimise step-by-step.
Do you want help getting started with A/B testing?
Collecting, analysing, and acting on data is incredibly important for creating cost-efficient and growth-driven advertising. And that’s exactly what we specialise in at Social Zense!
Get in touch with us, and we’ll take a look at how we can increase the effectiveness of your advertisements.
