A/B testing Tips for Digital Marketers

A true A/B test is defined as changing a single variable while keeping all others the same. This way, we can see if the variable we changed influences customer behavior. Sound familiar? Yep, should ring true to how your middle school science textbook defined an experiment. The two versions could be served randomly at a 50/50 split or if you have a large enough audience, you could also do a 20/20 split and the remaining 60% can receive the winning creative – ensuring you net the most possible conversions while still learning from the test.

What can you A/B test?

Below are the ten most tested elements in digital ads:

Text – We can test the font, color or messaging. (There are some exceptions as Facebook and Instagram captions are unable to be changed from the default size, color and font.We can change these items within the image part of these ads.)
Image – Changing or adding an image can make for an interesting test.
Call-to-Action Button – We can change the size, shape, text, etc. (This will not work on assets that don’t rely on clicks, such as most CTV inventory.)
Color Scheme – Perhaps the brand or company you’re representing wants to make use of that tertiary custom color that has been gathering dust in the brand guidelines PowerPoint …
Offer – This is a big one! Changing the offer will likely have a larger impact than most, if not all, of these items on the list.
QR Code – This makes the most sense for inventory that is not able to be clicked on like linear or connected TV. Billboards and digital out-of-home ads also make use of these. Most other ad formats are seen via phone. so a QR code doesn’t make much sense.
Video – This should be measured between two different creatives of the same length or similar creatives of different lengths.
Ad layout – Flipping the offer from the left to right side or from the top of the landing page to the bottom.
Headers – Having two landing pages with different headers and subheaders.
Price – Does it make sense to discount this product? Will the increased sales cancel out the decreased margin? The only way to find out is to test it!

A/B Test Pro-Tips:

Test subtle creative or offer changes. A/B testing does not work as well for major changes. If the offer goes from “Buy One, Get One Free” to “Buy One, Get Three Free,” that is not a good A/B test.

Start with a good sized audience. A common pitfall is a small audience size. When you have a small audience, it will take a longer time frame to get a number of conversions required to get a statistically significant result.

AI + A/B

AI tools are used by most advertising platforms to determine which creative version is performing better and to allocate more impressions from a given budget to the better-performing creative.

Other ways that AI can help with A/B testing are:

Generating potential A/B test ideas
AI-Generated Offerings – An example of this is in search engine optimization where a person provides keywords and AI puts together a description for a landing page based on those keywords.
Optimizing the places where ads are being served.
Real-time recommendations
Predictive behavior for future tests

A/B testing is a great way to optimize future creative and offers. AB testing allows us to gather meaningful results about the best way to engage a customer.While a square- or circle-shaped button may not seem like it would make a huge difference on the surface, it could drive more clicks that lead to additional sales.

Our Multichannel Direct and Scaled Video solutions are designed to handle A/B tests with ease. If you are interested in incorporating an A/B test on your next campaign, let’s talk.