Since the dawn of online marketing, A/B testing various Ad Creatives has been a common best practice. Depending on marketing channel, it can be difficult executing tests with statistical significance, accuracy and getting reliable metrics. Same goes with Facebook, where Ad Split Testing has been something the user had to perform – which often results in hastily made decisions, due to the human factor and lack of knowledge and insights in the results.
For the first time ever, highly effective Ad Creative split testing is possible. It’s called Dynamic Creative Optimization and it is exactly what it sounds like – a way to dynamically optimize and split test Ad Creatives.
Briefly – what is DCO?
Let’s say you have 3 images, 2 titles, 2 body texts, 1 description and 2 call to actions you want to include in an Ad Set with Dynamic Creative Optimization. This will give you 24 variations. In a “regular” Ad Set, these 24 Ads would be included in the delivery of your Ad Set and depending on their initial performance, 1-3 of these Ads will be the ones shown the most – for the whole audience in your targeting.
What DCO does, is that it divides your audience into sub-audiences, since different parts of your audiences and responding differently to the variations of the Ad. Instead of showing the generally best performing Ad to everyone – each user in this audience will be shown a specific variation depending on the sub-audience this user is in.
A DCO “Ad” will in Ads Manager be visible as 1 Ad, instead of 24 Ads. This is because data isn’t relevant on Ad variation level anymore, but on Creative variable level (e.g. image, title, body, call to action).
How to set up Dynamic Creative Optimization
In the Zalster platform, you’ll now find Dynamic Creative Optimization in our section for Automated Campaigns. In order for us to help you set up this, we’ll need to know:
Where do you want to perform DCO?
- Facebook Page & Instagram Account
What accounts should be the advertised ones?
- Daily Budget
- Targeting / Audience
What audience do you want to run DCO towards?
And lastly, we need to know what variables you want to dynamically test with:
- Body texts
- Website URL
- Call to Action buttons
You’ll also have the opportunity to choose if you want us to split test DCO vs. non-DCO, to see whether it help you increase ROI/decrease CPA or not. If you select this, a second Ad Set will be created with all Ad variations – to see whether static Ads is performing better than Dynamic Optimized Creatives for your specific audience and purpose.
Have in mind that a DCO test should be run for at least 5 days and that you should have a significant daily budget, so that any statistical conclusions can be made.