I’m about to unveil a Facebook campaign we recently rolled out at Digital Marketer… and it’s not your average campaign.
I’m not only going to show you the ad copy and images we tested…
I’m also going to show you the exact Facebook split testing structure we used:
That allowed us to split test…
- 2 “Marketing Messages”
- 2 Audiences (for both marketing messages)
- 3 Ad Copy Variations
- 4 Ad Images
…for a total of 36 different ad variations within 4 different Facebook campaigns (144 ad variations total)!
This is how you find PERFECT Facebook Ad(s).
But, before I dive into the campaigns you need to understand a major change in the way Facebook organizes your ads…
How to Organize Facebook Ad Campaigns
On March 4th of this year, Facebook rolled out a new “campaign” structure. The campaign structure is the way that Facebook organizes all of your ads within your account.
There are three “levels” to the structure:
- Ad Set
Think about it like Russian nesting dolls…
- The campaign is the “house” in which everything lives (the doll on the far left).
- The ad sets live within the campaign.
- The ads live within the ad sets.
Here’s a screenshot from our Power Editor to help you get a better idea.
You have multiple ad sets within a campaign, and multiple ads within those ad sets…
In the example above, you can see that we have 6 campaigns selected, within those 6 campaigns there are 18 ad sets, and within those 18 sets there are 72 ads total…
You get the idea!
Facebook created this hierarchy to help us stay organized.
They also (and more importantly) created this hierarchy to help us TEST!
So, why does this matter to you? You could have Googled this “Campaign Structure” stuff, right?
Well, one of the biggest questions we get inside of Digital Marketer Lab is, “How do I actually split test with Facebook ads?”…. and more importantly, “How do I test multiple variables at once?”
The way Facebook describes it is that you should test the “objective” (website conversions, website clicks, event RSVPs, etc.) at the campaign level…
That you should use the ad sets to test the audience, or targeting…
…and that you should use the ads to test your “creatives” (ad copy, image, etc. — all in ONE test).
But, what if you want to test ad copy and images at the same time? How can you run a proper split test if you’re throwing them all in the same ad set?
What if you want to test ad copy, creative, and targeting all in one campaign?
As digital marketers, we know that a simple tweak in ad copy, image, or targeting can have a MAJOR affect on your campaign.
Believe it or not… there’s a major performance difference between this ad…
and this one….
What’s the difference? The ad copy!! The first ad is generating leads for almost a DOLLAR less than the second ad.
So, if we’re not able to properly test all variables… how do we optimize the campaign to achieve the PERFECT ad?
Now let me actually show you how we overcame our testing frustrations with our “Double Your Sales” ad campaign…
Facebook Ad Split Testing Case Study: The Double Your Sales Campaign
Double Your Sales is a free, 6-week course presented by Infusionsoft and Digital Marketer. The course is aimed for business owners, and teaches them how to DOUBLE their business with one simple formula.
The ultimate goal of the course is to eventually sell prospects Infusionsoft (a CRM) and our email marketing product, The Machine.
When we first began planning the campaign – we knew we wanted to shoot for two different “marketing messages”.
One was more direct and spoke to the end results of Double Your Sales… the premise was “6 Ways to Double Your Business in 2015”. The images were to be more professional.
The second spoke to a pain point… “Is your business stuck?” We used images of cute animals that looked “stuck” and they were to be more light hearted and eye catching.
The marketing messages defined the ad copy and images for their respective campaigns.
So there we sat again, wondering the best way to structure this campaign to test as many variables as possible… and here is what we devised…
As you can see, the first marketing message is at the top, labelled “Double Your Business”. The second is below, labelled “Are You Stuck?”
We selected two audiences that we would target, and we wanted to run BOTH marketing messages to BOTH audiences to see which would perform better.
But, instead of using the campaign level to test “objective” (we already knew we wanted to optimize for conversions because we were running traffic to an opt-in form) we decided to use the campaign level to test different audiences.
So, as you can see in the green below – we have FOUR different ad campaigns set up, two for “Double Your Business” and two for “Are You Stuck?”. This allows us to test both marketing messages to two audiences at the same time.
Now, let’s talk about the “Ad Set” level. Facebook says that this should be used to test the targeting. Well, we used the ad sets to test COPY…
We created 3 different copy variations for both marketing messages. One was short, one was longer, and one leveraged the Digital Marketer and Infusionsoft brands (we’ll show the ads in just a bit).
This allowed us to run a DIRECT split test between three different copy variations. The copy variations are the same for both “Double Your Business” campaigns, and the copy variations are the same for both “Are You Stuck?” campaigns.
Now, let’s talk ads.
At the ad level, Facebook is actually going to optimize for the best performing ad and show that ad to more people. Although it’s not a “true” split test…. It’s the perfect place to test ad IMAGES.
The IMAGE is usually the first variable in an ad to fatigue, and having them all in one ad set allows you to decide a winner out of the gate, and be able to pause and give the other breathing room if the initial winner starts to fatigue.
For Double Your Sales, we created 4 ad images for each marketing message.
So, as you can see in the orange… we tested each ad image with each copy variation…. bringing us to a total of 12 ad images per campaign.
Do you see why this is so exciting!?
With only FOUR Facebook campaigns we’ve been able to test marketing message, targeting, copy, AND image.
I know this may seem a bit confusing. So, let me show you the actual ads we ran for this campaign, to give you a bit more context…
We’re going to look at each ad set for each marketing message to give you a peek inside.
First, let’s talk about the “Double Your Business” campaign…
Remember, everything for both of these campaigns (labelled Audience 1 and Audience 2) is EXACTLY the same, we are just testing two different audiences at the “campaign level”.
At the ad set level, we are testing 3 different copy variations.
The “short copy” ad set is punchier and to the point. This is the kind of copy we used when we started running Facebook ads a few years ago, let’s take a look…
The “short copy” still works, especially to people who already know your brand.
But, we are finding that as people see more and more ads online, they need MORE information, they need to be further persuaded to make the click, this lead us to our second ad set…
This ad set features very LONG copy (you can use Power Editor to extend the character limit). We’ve found that when running traffic to an offer that is going to require TIME from the prospect (a webinar, a mini class, a course) this long copy ALWAYS converts better.
Why? It’s a harder sell. This isn’t a traditional Lead Magnet. In order for the prospect to consume the information, they’re going to have to show up and they’re going to have to invest more than their email address.
As you can see above, we simply used the “short copy” and then added specific details about each lesson.
This takes us to the next ad set, the “DM & IS” copy. This is something new we wanted to test, especially since Infusionsoft was involved in the branding.
You’ll notice that this ad set was the “Short Copy” with “Brought to you by Digital Marketer & Infusionsoft” simply added below. We wanted to test to see if adding credibility from the brands would make any sort of difference.
Now that you’ve seen inside the “Double Your Business” campaigns, let take a look at the “Are You Stuck?” ads…
The structure behind these campaigns is identical to the “Double Your Business” campaigns, we’ve just tweaked the copy and ads to reflect the “Are You Stuck?” marketing message.
Here’s the first ad set, “Short Copy”…
Again, we were very punchy and to the point here. We added “Is your business “stuck”?” in order to immediately hit a pain point with the prospect.
The second ad set we tested was “Long Copy”…
Again, very similar to the long copy ad set from the previous campaigns, we’re hitting a pain point with being “stuck” and then really explaining what they’re going to learn in the course.
The third and final ad set was “DM & IS”…
Again, we’re maintaining the marketing message of “Are you stuck?” but testing how the use of brand names affects the ads.
Structuring our campaigns in the way I’ve revealed above allowed us to test 144 different variations in only FOUR ad campaigns. This allows for simplicity when scaling and optimizing.
When testing campaigns on Facebook, you should test variables that make sense for YOUR business.
What do I mean? Not every business should test using their name in the ad to see if it adds credibility. Some businesses don’t necessarily have that credibility yet (but you will)!
Think of other variables that you can test… should you try calling out your audience in the first few words? How about black and white images versus color?
The biggest takeaway here is to have a plan.
Planning and structuring your ad campaign is one of the biggest factors to success.
When you have an offer that you want to put in front of an audience, think about the most creative and interesting way to catch their attention…. but also don’t be scared to do the “boring” stuff, that’s usually what wins.
Have a question?
Ask the DM team and 9,036 other members in the DM Engage Facebook Group!
Not a DM Lab Member? Learn more here.