cfx211 - 4:02 pm on May 6, 2011 (gmt 0)
@cfx211: If you don't even target by age group, scroll up to my post that says
repeat: facebook sucks, facebook is stupid, facebook is for children
Geotargeting, for some products is a given. Like if you are selling english language books from a us warehouse, probably you are only going to target usa and canada. But that is really the bare bones. Everybody so far I talked to about facebook not working has no idea what demographic actually buys their product. Until you know that, save your money or give it to Google
I should have clarified, I am age targeting everyone 18 and over.
Let's try a different track here, because maybe I did not do a good job of explaining my FB marketing problem and right now I am trying to make FB work for me not bash it.
Our customers are people from country X who now live in country Y. It's a niche product in that respect, but amongst that group of people our product is used across an entire gamut of age/sex/education etc...We have had success marketing our product via display ads simply by getting in front of people from country X who now live in country Y.
I took this approach to FB by putting together a group of interests that are fairly specific to people from country X. I then geotargeted the ad to people who live in country Y. I then age targeted to people 18+ because kids don't buy our product, but college students do. Let's say this gave me 500k people and I am assuming that they are my known universe.
I ran a 2 week campaign in January and got a 0.7% conversion rate (which is comparable to semi-targeted display campaigns I run), but returned a $160 CPA which is 5x over a comparable display campaign. The problem? CPC was in the $1.20 range which is also 5x what I pay for display.
Now I hope you understand my original gripe about FB. I am paying a 5x premium on clicks for a comparable level of performance to display.
The campaign was turned off not only because of the high CPA, but because FB only brought me 20 new accounts. The sample size was large enough to tell me this test was a dog and to take my money elsewhere. Had FB delivered 500 accounts at that CPA, well I would have sliced and diced to optimize the campaign down to a more reasonable CPA but it didn't deliver enough volume to justify that.
So in summing up my problem:
A. I need to identify my target market by collecting a bunch of precise interests because I cannot think of a way to identify someone as originating from country X otherwise.
B. Combining Step A with Geo and basic Age targeting gives me a fairly large universe of 500k users, but FB wants $1-$1.50 a click for them.
C. FB delivers the same conversion rate for me as display but at a 5x CPC.
D. FB was not returning enough new accounts to warrant optimizing my campaigns.
To make my latest pass at FB work, one or more of the following needs to happen:
A. My universe of potential customers needs to expand. It could probably max out at about 1 million based on what I know about my customer base and FB usage in the US.
B. CPC needs to be brought down.
C. Conversion rate needs to increase upwards of 2-3x.
D. Realize that even though FB might be 1/3rds of ads served, realistically it will probably max out at 5% of paid accounts for me, and to get those effectively I will need to cut up my known universe into a lot of demographic slices to find the few that will accomplish B&C.
D. Change my view of FB from direct conversions to thinking of it as a seeder for my remarketing lists. This can only happen if B occurs.
So enlighten me Facebook sophisticates...
What practical advice do you have for this hayseed neanderthal marketer as he tries to effectively hunt the swift gazelle of social marketing after years of feasting upon the lumbering mammoth of adwords/adsense?