What is A/B testing?
A/B testing, in the digital marketing world, is the science of comparing 2 versions of a digital marketing asset to measure which has a better performance. Generally, A/B testing is conducted on ads, landing pages, audiences or EDMs. Common measures of success are the conversion rate or the click-through rate but there are a lot of other possibilities specific to each type of test.
The ‘science’ aspect refers to the statistically significant difference between the two versions. There can be a difference in click-through rates between 2 ads due to random variance rather than a performance difference. Statistical significance requires a combination of the performance metric and the number of overall ‘tests’. We won’t get into the maths here as there are plenty of easy to use A/B Test Calculators available online. https://neilpatel.com/ab-testing-calculator/
One of the key points with A/B testing is to keep all other variables the same to remove their bias in the test.
Why run A/B tests?
Simple, to continually improve the performance of your digital marketing campaigns.
What are some A/B tests I can try?
The list of possible digital marketing A/B tests is very long so let’s just stick to some high impact ones.
Audience selection is one of the most important aspects when building a high-performance campaign. Spending big money on a perfect product with a perfect ad targeting the wrong person is absolutely pointless. You need to show your ads to the right audience.
For example, testing your lookalike audiences against an interest-based audience to determine which should get scaled up, budget cut or just completely paused.
Facebook’s automated bidding strategies have made this a bit easier for advertisers so we can be a bit broader with our initial audience selection. Facebook is essentially running some A/B testing (as a small part of their much bigger and more complicated algorithm) to help target ads to people that are likely to complete the bid strategy goal.
Testing 2 creative variations on Facebook is a great way to see which has a higher click-through rate. For example, a people-centric image against a product-centric image or a testimonial video against a testimonial image. There are endless opportunities with creative options on Facebook.
Google Ads Headlines
Generally, the first thing you see you on a Google Search Ad is headline 1. It needs to grab the attention of the user while also being informative about your product/service. Driving clicks from your top-performing keywords are the key to a good return on investment for search campaigns. A good headline test includes adding numbers to your headlines to test click-through performance.
Similarly to Facebook’s bid strategies, Google’s responsive ads use algorithms (which probably include some form of A/B testing) to deliver ads to people that are more likely to convert.
A/B testing is a great way to maximize account performance over time. It does require time and effort but the results are absolutely worth it. There are many ways to implement A/B testing in your digital marketing campaigns. Remember to keep other variables fixed to avoid any outside bias on your results.
Author: Tyron Zeelie – Campaign Manager at Sprocket Digital