Optimization and Split Testing: The Split-Test Affiliates Forget to do 99% of the Time
I’ve always stressed that the most important aspect of any campaign is the offer. The best landing pages, ads, and angles in the world don’t matter if the offer doesn’t convert.
Most affiliates will do two things in order to optimize the offer.
1) The first is they’ll ask for a pay bump. If the offer pays $3 street, then getting it bumped to $5 will instantly make your campaign more profitable.
2) Split-test offers. We’ll use USA mainstream dating as an example to make it simple. If you’re running USA dating then perhaps you would split-test match vs eharmony vs localusadating (made-up offer). You’d run a 33/33/33 split test in CPVLabs and just see which one makes the most money.
In this scenario people would get the paybump for localusadating, and they see that localusadating is the winner. They are “done” with optimizing the offer and move on to other variables of the equation.
There is a missing 3rd step and that would be split-testing the offer on different affiliate networks.
My friend John was running localusadating and getting a $5 payout and didn’t understand why he would split-test on different networks. He asked around and no one could match the $5 payout, but someone else offered $4.75.
He did a 50/50 split-test and the network that paid less per lead ended up generating more revenue.
His filthy newbie mind couldn’t understand why.
Here are some possible reasons:
1) The affiliate networks are using different servers and different tracking platforms. Click losses and pixel misfires happen in every campaign.
2) Most affiliate networks scrub. What is scrubbing? It means they purposely don’t credit you leads or sales that you earned.
Why would they do this? Money. It’s simple as that. Affiliates are stupid and only look at the payouts. The fact is if everyone’s offering $5, you HAVE to offer $5 or else people won’t run with you. Why not give them the highest payout possible and make our profit on the backend by scrubbing leads?
Another idea is perhaps they have to hit certain goals. End of the quarter is coming and they’re not near their revenue goals? Scrubbbbb
I’m friends with several advertisers and it’s funny when they’ll test their own offers in different networks and see the extreme variance.
This is why I like affiliate networks that are lean and have low overhead. See a network with a fancy office? You’re paying for it. I don’t like the networks owned by flashy people.
3) The advertisers are scrubbing the entire affiliate network. Maybe affiliate network A sent some horrible quality leads in February and the advertiser lost money. The advertiser decides to scrub network A for this month in order to make more money, and you’re running with network A. Network B on the other hand could have an amazing relationship with the advertiser and you’re not affected.
This is not a rant against networks at all. There are good ones and there are bad ones. They have their place in the industry.
What I’m telling you to do is be smart and do your own research.
Men Lie, Women Lie, The Numbers don’t – Jay Z
– Ngo Knows