Chapter 7: Facebook Ads Testing
As any experienced online advertiser (or marketer) knows, testing your campaigns is essential.
The primary reason for this is that I can't tell you what will work for your business with Facebook Ads.
I can tell you what's worked for me, what's worked for people like you, and what should work based on all of the current conversion optimization psychology out there.
And all of it will be simply recommendations. They're not a strategy for your business.
And that's not even considering that this article is only up-to-date until Facebook changes the platform, which they do about once every half hour…
What works for me today might not work for me tomorrow, let alone working for you. But that's okay, so long as you're not just taking anybody's advice at face value and throwing your hard-earned ad budget willy-nilly into this fantastic ad platform without testing first.
Note: When testing, avoid being distracted by cost per impression, as it's almost completely tangential. Focus on CTR (which tells you whether or not your ad is optimized in the first place) and cost per desired action (which tells you if your target audience is a good one and your ad is convincing that audience to ultimately convert).
Here are the five Facebook Ad elements I recommend you test, particularly in the early stages of running your campaigns:
#1. Campaign Objective
When I was talking above about how Facebook Ads are structured, I mentioned that Campaigns are above Ad Sets and Ad Sets above individual Ads. They're structured this way because Facebook (and I) want you to create different campaigns for each objective.
Of course that doesn't mean you have to use every potential campaign objective (for instance, your brick and mortar coffee-shop may not have an app to promote…), but you should nonetheless be testing different objectives against each other with the same ad layouts, budgets and designs.
Does Website Clicks drive more success than Website Conversions? At what cost? Go find out!
#2. Your Audience
Your ad might be as optimized as it's possible to get: beautiful, contrasting, appealing copy and delivering awesome value. But unless it's being seen by the right people, you're just throwing money down the drain.
In my experience, changing target audiences is perhaps the most influential element you can change. And that makes sense, right? Otherwise it's like promoting false teeth exclusively in a frat house.
Not gonna' get far…
Newsfeed is definitely the most expensive route, but is it worth it when compared with the low click-through-rates of the sidebar or the finicky-nature of mobile users? Then again, while sidebar does have low CTR, your CPA might actually be better.
Or is Instagram's ad platform a winning one for your ecommerce site?
My recommendation would be to create multiple ad sets for each placement and divide your budget evenly among them. You should be able to (within reason) run a placement test only once every few months.
Bidding structures are one of the most changing elements of Facebook Ads, so it's always a gamble for me to recommend anything.
Will website clicks deliver a better ROI than impressions? Or will paying by impression yield more clickthroughs than paying by action. And what about that whole "manual vs automatic" bidding thing?
Nothing for it but to test it.
#5. Design and Copy
I only put design and copy at the end because I want to emphasize that, while color, text, CTA button etc may be the most obvious things to test, they're not the only ones.
That said, of course it's essential that your Facebook Ad stands out from the rest of the newsfeed and grabs the attention of your target audience.
A few things to test:
- Your image: Engage with an image, convert with a value proposition.
- Your title: After your image, your title is the most important factor in catching the eye of your audience.
- The details: A border around your image, a slight shadow, the word ‘now’ within your body copy, the subtlest color change… all can surprisingly boost your CTR.
- Your body copy: Words like "Free," "now," "today," and "limited-time" have all proven to boost conversion rates before. Will it work fo you though?
- Your CTA: Facebook has a few options here. See if it matters which you choose.
How to set up your Facebook Ad test:
Of course it depends on what variable you're testing. If objective, you'll test from the campaign level down. However, that can get a bit confusing very quickly. Here's a more manageable split test:
Top Tip:: Facebook often determines which of your Facebook Ads is the "winner" before the results are actually in. This can be super frustrating.
As a workaround, keep a close eye on your "being tested" ads. If one starts seeing more impressions than another, this is a sign that Facebook has decided that one's the winner. Turn it off. If another ad you're testing then starts seeing all the reach, turn it off as well. Only turn your (winning) ads back on once each ad has at least 1,000 impressions and you can better determine which is the actual winner.