At ROCKETROI, our basic pillar for developing a good Facebook Ads strategy is testing.
When we prepare the Facebook Ads strategy for our clients, we can intuit, according to the insights that the client gives us about their business and based on our experience, which variables can generate better results. However, what makes the point of difference is measurement and A/B testing, which allows us to know with certainty what is optimal for each advertiser.
That is, these tests allow us to test different versions of our ads so that we can see what works best to further develop and improve our future campaigns.
Before developing our testing strategy it is very important to have clear objectives, a defined buyer persona and target, good creative and promote the best possible user experience in our eCommerce; based on this we can generate a testing plan that will be essential for the key to success and growth of the client.
How does it work?
A/B testing can be done with most of the targets we have available for the creation of campaigns, and the variables we can measure are:
At ROCKETROI, the testing that we tend to emphasize and constantly promote is that of creative. As you know, our storytelling and creativity are key to promoting a good user experience and having the highest user value possible.
On the other hand, it is also important that we have a budget that allows us to achieve the necessary results to determine with certainty that a strategy is the winner.
How is the audience split?
In an A/B test, the potential audience is randomly split - which ensures that the test is not altered by other factors that may affect the results and also ensures that each group of ads will have the same chance of winning - and shown the same ads, except for the variable we are testing.
When to use it?
Our recommendation is that it should be an "always-on". For example, you can test the same ad with two different audiences (Broad vs. LAL), or you can test two types of messages to see which of our potential users are more reactive.
So we can use the A/B testing tool to:
Know what content (ad format: video, image, carousel, collection) generates more probabilities of acquiring our target. In many cases, for prospecting, a mix of formats tends to work, while for retargeting, static formats tend to convert more. However, there are always exceptions and for this reason it is always advisable to test and not lose opportunities along the way.
Know which messages or CTAs generate more probabilities of acquiring our target. For example, if we have an offer it is interesting to test in the message if it works better to include a % with the DTO or Price Point.
Know what type of audience is more likely to acquire our objective. Think about all those potential audiences that make sense, making the most of what we have in our DB, the pixel, and our Buyer persona.
Knowing what type of optimization strategy is the most successful to acquire our target.
Knowing which locations (right column, feed, stories, marketplace, etc...) or platforms (FB, IG, Messenger, Audience Network) generate more probabilities of acquiring our target.
How to view and measure results?
In ROCKETROI, when we have an active A/B test we track daily from the ad manager in tests and results - it's quick and easy and allows you to clearly see the evolution of the test. At the end of the test, Facebook measures the performance of each set of ads based on the campaign objective and you can extract the results. The one with the best results wins, i.e. the one with the lowest cost will be the winner. Another factor also to consider is the confidence level, which represents the probability that we would get the same results if we were to run the test again. In this case, Facebook recommends that, for any test, a confidence level higher than 75% indicates that we should stay with the winning group.
