This is not an easy campaign to start out with -- I tried to hit it and have since left it on the backburner as my budget simply isn't big enough right now to adequately optimize it.
Most I spent in a day on this campaign was around ~$250 and even that wasn't enough for me to find reliable profit pockets. It also seemed to perform differently from day to day for me. One day it would have a phenomenal ROI and convert like gold, and then go on a cold streak the next day.
Like I said, my budget simply wasn't big enough to let this campaign run all day on full blast -- especially when you don't see any conversions come in for hours at a time. In my case at least.
If there's any advice I'd give, it's to focus on one area at a time to optimize and figure out if it works or not.
One carrier, one banner size, one device type. You can even go as far as targeting only select placement options and get a couple of days data. That may not be the best or correct way to test an offer, but if you must take on these higher payout offers with a limited budget I don't see an other way of doing it.
You can't go into an $18 payout offer and only spend $50 a day with a hundred different variables. I thought I could tackle it and learned the hard way. I did have some lucky days when I made a good chunk of the money back, but it wasn't something solid that I knew I had achieved because of calculated optimizations. Like Brent told me over Skype, I was gambling.
Hope that helps.