As Facebook advertisers we all strive to find the perfect audience and maximize profit from it. But most of us also know, that once we find this “goldmine”, so to say, it’s a matter of time before we should find another. Often because a good audience is a narrow audience, which means limited volume and thereby limited profit.
Through my time as an advertiser on Facebook, I’ve tried out a lot of different strategies to avoid lacking traffic on my good performing campaigns.
What I’ve often done, is probably the same as you. I find and extract either all users or only the best, and upload these as an audience on Facebook. Right away I create my different Lookalike audiences and look forward to see the great results. But often a 1% or 3 % Lookalike audience is rather narrow. Especially if you, like me, primarily drive traffic to geo-specific offers. (I’ve mostly worked with offers in the Nordics.)
Currently I work with financial offers in Norway and Finland, and my new strategy for creating audiences has improved the quality and volume I deliver radically.
The vast majority of citizens in the nordics already has debt, and are interested in finding a new loan, with a lower interest rate. In general, just like you and me, everyone is interested in money. So, instead of focusing on including the absolute top notch users in a market, I focus on finding the absolute scum of a market, only to exclude them. So how do I do this?
Userdata
With the campaigns I’m running I have the possibility to extract userdata from the system. I extant the wort users I’ve previously delivered, all duplicates and low quality. And make sure to exclude these on Facebook. I create lookalike audiences (5% diversion) on these users, and exclude the lookalike audience as well.
Interests
I also see if I can find any interest-based audiences on Facebook, that most likely won’t be the best users. So with the financial offers I’m running, I can only assume that users whom Facebook categorize as “interested in loans” probably has been seeking loans for a long time. Chances are that by including these I will see the duplicate and low-quality rate will raise. So instead of including them, I make sure to exclude them.
Demographics
Peoples educationlevel and/or worktitel often tells you a lot about them. When looking at financial offers, I’m not interested in the top nor the bottom of society, but everyone in the middel. In the States you can actually target people depending on their estimated income. Given that this isn’t an option in Europe yet, I test out excluding people with a high education-level, top positions, very low/no educationlevel and low paying jobs.
———
Depending on the offers you’re running, I hope you can implement some of my ideas, or twist them to suit to your business case. Let me know if this is helpfull. What do you do to enhance you quality and traffic?
Great insights. Sometimes you are astonished what type of people are into certain offers. Your logic makes sense, but be careful with assuming too much. Always test!
Great posts coming around facebook lately! Creating our John Doe is always useful, it saves money for sure. Testing the audience is best so test always! A good read though.
LeadCloak
Yes, defiantly. The post however is a result of testing. This doesn't mean that we won't keep on testing. :-)