I’m on a flight from NYC right after ASE16 on my way back to Houston my 2nd home. I know it’s probably a #FirstWorldProblem but there’s no internet, nor hot meal on this flight even tho I’m flying BusinessFirst on United. At least I have a big room for my laptop and to kill time I have 2 options: watch TV to get brainwashed OR write a killer article to educate my fellow affiliates.
So here I am thinking about a very interesting topic…
What’s your tactic to choose the right offer? What are the main metrics you look at when you’re selecting an offer for testing? How exactly do you measure a GOOD offer?
Let me take a wild guess! You look at the payout first… You’re probably signed up for a multiple CPA Networks and what you probably base your decision on is the payout.
Ever wondered why some networks offer a much higher payout for the very same offer? Let’s say “CPA Network A” offers $12 payout per lead while “CPA Network B” offers only $9 as a payout. What makes such a huge difference?
Read carefully, here comes the trick: Networks that offer a higher payout generally scrub a LOT. What is scrubbing? It may also be called shaving… Basically you get fewer conversions reported in the network’s system, which you can’t even check unless you have access to the advertiser’s CRM platform. Let’s say you send a 100 conversions but you only see 80 in the reports and the 20 is taken by the affiliate network. You will never know because it’s an automatic feature in their system.
Be mindful of CPA Networks that use Cake or HasOffers as these systems have built-in features that allow for lead manipulation. CAKE in particular has a % scrub option, essentially. The network can for example set a 20% scrub rate, which means you won’t see 20% of your leads, and only get credited for 80% of it.
Oh and don’t get me wrong, the issue isn’t with these systems… It’s more of a moral question on the CPA Network’s end.
So what’s the solution?
Split-testing between CPA Networks. Same offer, different payouts from multiple networks. Split test and let the data decide which one has the lowest scrub rate. I’d be extra cautious with networks who use CAKE because the may be too tempted to scrub just because it’s to hard not to do.
It would be far more legit to just reduce payouts. But the competitive marketplace requires networks to offer higher payouts to get affiliates running with them. Affiliates are lazy with split-testing networks so are more likely to just pick higher payouts initially. Networks could technically run at a loss e.g. they bump payout to $5 on an offer they make $4.90 on, but then add 20% scrubbing – they make far more that way.
Don’t select offers based on Payouts. Make sure you request EPC and CVR reports before you decide what offers to test.
Always split test between networks and use your own tracking system.
If you’re running volume request CRM access to check what leads you weren’t credited for.