Why split testing audiences for Facebook Ads is a waste of time
Nowadays I rarely find myself split testing audiences. I’m talking about interests, demographics or behaviors. For example, setting one adset targeting interest “credit card” and another adset targeting interest “online shopping” to see which will bring the best result.
And here are the reasons why.
I don’t have the resources to perform proper multi variable testing
One person has lots of interests and those interests are determine by Facebook based on our activities on Facebook. You can check out the interests you yourself are under by going to your profile settings > Ads > Your interests.
Now you see that there are lots of interests that advertisers can choose to target to get their ads in front of us. To get to the right audiences, you will need to use the “narrow audience” option and add another level of interests.
But by doing this, you will make things more complicated because you’ve just added another variable in your experiment.
If you ignore the additional variables and continue to run your test, the results you get will not be strong enough to make a concrete decision.
I let Facebook optimize ads delivery for me
It’s important for advertisers to understand what each of Facebook’s campaign objectives mean and how it will affect the delivery of your ads.
If you select website traffic campaign objective, Facebook will optimize the delivery of your ads for website traffic. This means that Facebook will show your ads to people who are most likely to click on the link of your ads.
Say the targeting options you set up gives you a potential reach of 500,000 people. But your daily budget will only get your ads shown to 500 people a day.
So how does Facebook decide which 500 people to show your ads to out of the 500,000 people?
That’s where delivery optimization comes in. Facebook will optimize based on the objectives that you select for your campaign.
This is the reason why I do a lot of conversion campaigns. I can tell Facebook to show my ads to people who are most likely to register, to add to cart, or to purchase.
Because of this feature, I just do broad targeting like “Facebook page admins in Malaysia”, which has like over 3 million potential reach and I still get good results (ROI) once it’s optimized.
I put more effort on the strategy and not techniques
When I work with businesses, I’m the one planning their strategies as well as executing it. They don’t deal with me and then I pass it to an account manager to execute.
If I focus too much on the small platform specific techniques, I will take away focus from the strategy part of it. Facebook Ads is just a tool. Without a great strategy, you will not get great results that matter to your business.
Put more effort on understanding consumer behavior, how people use Facebook and Instagram, creating a funnel, and the overall business strategy.
What I do test instead
Even though I don’t split test audience by interests, I do split test adsets. Testing is still a huge part of building a scalable growth engine. There’s no doubt about that.
What I test instead are offers, ad creatives, types of ads, campaign objectives, placements, delivery optimizations.
Pretty much whatever questions I have, I will come up with a hypothesis and test it. For example, I would like to know whether having a call-to-action button in the ad makes a difference.
I will run 2 ads, one without a CTA button and another with a ‘learn more’ button. It’s a simple test that doesn’t take much effort.
Ready to get serious about growing your business with Facebook Ads?
I will come up with a custom Facebook Ads Strategy specific to your business & industry.
Click here to get your customized Facebook Ads Strategy.