Facebook marketers have been asking for one thing and one thing only when it comes to campaign creation – and that is the ability to A/B test their creatives.
Dynamic Creative is Facebook’s way of addressing that fundamental marketing need, making split testing easier and less time-consuming.
As any marketer knows, you’re only as good as your theory – and only through testing our images, headlines, and other factors can we find what truly resonates with our audience (and saves $$$ on advertising in the process).
Before Dynamic Creative’s inception, adding in variations could take hours of copying and editing, but this new addition to the Facebook arsenal makes it easier than ever.
In this guide, we’ll go over the features of this new tool, how to get the most out of it, and how you can analyze your results with ease.
With the invention of the internet, customers now have the whole world at their fingertips. They’re able to search for what they want, when they want it, and customize their experience (or product) as fits their needs.
But, what if a major factor in what they choose to buy isn’t the product at all, but how it’s displayed to them?
What if the one thing preventing you from selling your product isn’t the product – but the image you choose to advertise it? Wouldn’t you want to know what the problem was, how bad it is, and how to solve it?
This, my friends, is where the beauty of split testing lies.
The importance of split testing
Split testing (also referred to as A/B testing) allows us to choose multiple elements of our ads, like the image or headline, to test them to see which one gives us the better result we’re looking for. An easy way to think of this process is like an experiment that you might see in a lab using the scientific method.
Just like a scientist, you may find that your predictions about performance are often wrong, which is completely normal.
The only true way to know if your assumptions are true is to test them and view the results yourself – which is where split testing comes in.
Take this example campaign, for instance. We created a campaign whose ultimate goal is to generate signups for a trial of AdEspresso. To see if our images made a difference in this case, we chose to split test 2 of them – and the results were as predicted.
After running the campaign, we were able to see that there was a big difference in the cost per conversion of the two images we ran. We were then able to pause the image that was not performing as well (and costing is us more in the long run) in order to favor the image that performed better.
If you’re looking for an in-depth guide of things you can test, you can check out our Split Testing 101 guide, which goes over all the aspects of split testing, like initial testing ideas, budget, structure, and more!
Now that we have a basic understanding of split tests and their importance we can move on to the next piece of this puzzle – Facebook’s Dynamic Creative!
Where does Dynamic Creative Come In?
As we said before, Dynamic Creative is Facebook’s way of addressing the need to split test within their own system. Before Dynamic creative, you’d have to:
- Create an ad
- Copy it
- Edit your needed variables (headlines, etc)
- Save it
- Clone another ad
- Repeat ad infinitum
Now, this might not seem too labor intensive for testing one or two images, but what if you wanted to create several variations of the headline, image, and text?
Even if you only have three variations for each, that means you’ll end up with a total of 27 variations. If that’s not time-consuming enough, don’t forget that you’ll probably want to run multiple variables in different campaigns.
Thankfully with Dynamic Creative, you can add your multiple creative elements all from one screen, so you never have to copy things endlessly again. Talk about a game changer.
Now that we’ve learned what the Dynamic Creative feature is and why it’s important, we can now move on to the part you all have been waiting for – setting up a campaign with this feature!
We’ll also show you how you can do it with AdEspresso, and how to analyze your results on both platforms.
How to setup Dynamic Creative on Facebook
First, let’s look at how to get started on Facebook.
To start, you want to head over to Ads Manager. From here, select Create Campaign on the left-hand side of the page:
For this next step, we ideally just want to set everything up as a draft.
To use this feature you want to make sure you are using the Quick Creation method (and not the guided).
Then, give a name to your campaign, adset, and ad.
From here, you now need to edit your adset.
Just select the adset name on the right-hand side and you will be brought to the adset page where you can toggle the Dynamic Creative feature on or off.
The Dynamic Creative feature is not compatible with all objectives or placements, so any creative elements not fitting the format will be changed or removed, which is something you cannot undo later.
The objectives that are currently supported via Facebook are:
- App installs
Now that we’ve got that out of the way, we’re onto the real fun part – the testing itself! From here we can start to add all of our different creative assets.
To get started, you need to edit your ad.
From this page, you can now add multiple headlines, images, text, and CTAs.
You can view all of your variations and their placement-unique previews by clicking the View More Variations button near the top right-hand side of the Ad Variations box.
The maximum number of assets you can have for Dynamic Creative is 30, and there are also specific maximums regarding each of the creative elements:
- Title: maximum of 5 titles
- Image/Video: maximum of 10 images or 10 videos
- Text: maximum of 5 bodies of text
- Description: maximum of 5 descriptions
- CTA: maximum of 5 CTAs
When creating your ads in this method, a good thing to keep in mind is that your goal is to ultimately create one single ad – therefore you cannot choose which variations you would like to exclude (if any). Before you publish your campaign, make sure that all of your variations ‘make sense’ and there are no confusing matches in the bunch.
Once you’ve done that, just confirm your campaign settings and give yourself a pat on the back. You’ve just successfully created your first campaign using Dynamic Creative!
How to Analyze Results (The Facebook Way)
In addition to being able to see your statistics the ‘normal’ way, Dynamic Creative will allow you a new option to sort your results via the Dynamic Creative Asset breakdown.
To use this feature, navigate to the Ads or Adsets tab and then select Breakdown.
From there you can sort by any of the creative elements in your campaign (although you can only do it with one element at time).
Now, I’d love to say that this is the part where I can start to wow you with Facebook’s reporting showstoppers – but unfortunately, as far as Dynamic Creative goes, the buck stops here.
There are no additional reporting benefits, flexibility, or unique treats from using this feature. You can simply see the metrics compared to one creative element without being able to pause a unique combination that’s not performing well.
Another problem that comes into play is the social proof. Facebook creates a unique ID for each ad, so the social proof (aka likes, comments, and shares) are not combined into one place and are instead spread across multiple ads.
While the feature aims to save you time in creating these variations and allows you to view how different elements are performing, the inability to adjust during the campaign’s lifetime or view them compared to another variable really puts a damper on what could be a powerful tool in the Facebook arsenal.
Luckily, that’s where AdEspresso comes in.
How to Split Test Creatives using AdEspresso
Simply put, split testing is AdEspresso’s raison d’être – so the setup will be a bit quicker than through Facebook already.
To get started with AdEspresso, just select your campaign type on step 1.
In AdEspresso, we support all objectives, so any of these options will allow you to test your creative elements.
Step 2 is where we start adding in all of our variations.
On this step, you can choose the ‘Create All Permutations’ option to start your testing. From here, simply add in your choice of headlines, CTAs, images or even video elements.
Once you’ve added in your creative elements, you can view all of your variations across different placements using the Previews & Placements box.
We have a maximum of 50 variations on this step, giving us about 20 more variations than available through Facebook itself.
Creating these variations in AdEspresso also means that if you want to change or pause a combination after you create your campaign you can do so at any time!
Bonus Points: Audience Testing
While split testing the creative elements itself is a snap, another powerful feature of AdEspresso is the ability to easily test your audience!
Want to know whether the U.S or the U.K is performing better? What about males or females? Age ranges? In this day and age, the only way you can be certain is to test, test, and retest.
Doing so on Facebook would mean creating individual tests one at a time, but in AdEspresso we speed up this process tenfold.
Once you select “proceed” after creating the variations, you’ll be brought to the audience page where you can add in your targeting parameters.
This includes things like location, gender, interests, and even language options.
After entering in your details, you can then proceed to the final split test selection popup.
At this time you can select the additional tests you’d like to run, like placement, gender, location, and more.
The best part of this entire process is that we do all the heavy lifting for you – no more duplicating ads or adsets and tweaking things for hours – AdEspresso will set up the campaign according to your chosen options and wham – you can create a campaign with 24 ads (as we did in our example) in under 2 minutes.
How to Analyze Results (The AdEspresso Way)
Ads Reporting – Using the Ads Reporting tab, we’re able to view an in-depth analysis of the different experiments against any metric we choose.
In this example, we decided to test some different images and countries for a signup campaign to see if certain countries would have a higher cost for us than others. Using this feature, I can select both the Image and Location experiments and see how they stacked up against the cost per conversion.
From this data, we’re able to see that in this case, the Netherlands (the darker green bar) had a high CPA for image two and three, but a cheaper cost for image one. We can also see the various other countries and how they related to the cost at a glance.
If we scroll down, we can also see these results in a more defined table and even pause the image and country combinations that are costing us too much and reducing profitability.
Top Factors – Even if you’re not Facebook Blueprint Certified, AdEspresso makes it easy to understand what’s working and what’s not – take our Top Factors for example. As soon as your campaign starts to deliver, AdEspresso will display a side-by-side table with the results of your experiments directly on your dashboard. We’ll also give you suggestions based on your KPI’s (things like CPA, CPC, etc) while letting you know how much we believe it’s impacting your campaign performance.
In the example above, the system is telling us that the cost per lead in the United States is higher than our cost in Malaysia, so we’re prompted to pause this experiment to save us some more precious budget.
Daily Actionable Tips – Yet another no-brainer feature is our Daily Actionable Tips. Our proprietary Optimization Algorithm analyzes your campaign’s metrics in real-time and recommends improvements for your campaign’s performance. Every time you check your dashboard, you’ll find updated suggestions on potential next steps. Thanks, Walter!
So, what’s next?
Now that you’ve learned why split testing is important and how to set one up (either on Facebook or AdEspresso), you’re ready to get to work and start your own experiments.
If you’re looking for some creative inspiration, be sure to check out our eBook which has 500+ examples of some of our favorite tactics or a 101 on some of our best practices here.
What are YOU planning to test? Let us know about your experiments and their results below!