Sweepstake on Pop Follow Along (4)
Hi everyone,
As the ever-so creative thread title suggests, this is a follow-along of my efforts to run a profitable sweepstake campaign on pop traffic.
The campaign was launched on Friday 29th July, and has been spending the initial test-budget up until now. The offer is an iPhone giveaway in an eastern european country, with a $0.60 payout per lead. To begin with, I used spy tools to identify some of the most frequently seen landers for this type of offer in the geo, and tested the top 3 of these. I also created 3 versions of the campaign, with a low, average and high bid (thanks vortex!) to gauge the best bid for maximising ROI. I funded each campaign with $30 and allowed to run without alteration until all budget had been spent.
Below are the initial lander results, followed by the performance of each campaign:
Cost: $86.90
Conversions: 67
Revenue: $40.20
Profit: $-46.70
ROI: -53.47%
Initial Observations
For the first run, this is pretty in-line with what I expected. The low bid campaign delivered the best ROI (-38.4%), so this will be the bidding strategy for the next set of tests. The CTR on 2 of the landers seems quite poor- though ultimately these convert better than the most clicked lander. Reason why? I assume it's because the LP in question uses pops in a way designed to funnel the user through to the offer as quickly as possible- which doesn't seem to be translating into conversions quite as effectively.
The good news is that the offer seems to convert, despite the landers appearing to be less than optimal.
I also took a look at data on placements, OS, connection type etc, however there was no statistically/financially significant incentive to make any changes here at this point. If this campaign is to be profitable, I'll need to find an LP that converts more leads- once I've found this, I can then look at making backend changes to improve the campaign's margins.
Next Steps
For the next round, I'm culling the worst performing lander, and replacing it with 3 new ones. 2 of these are variations on the remaining 2 landers from the initial test, that will hopefully improve CTR and CR. The other is a completely new lander with a slightly more aggressive angle.
I've funded PopAds with $50- $10/lander, and will report back with the next set of data.
If you have any questions/critique/pointers, please do let me know- I'm an eager student and will take all the help I can get! 
Cheers to a great start! 
Considering you're only testing a single offer and 3 landers, the camp looks very promising!
Your first 2 landers are performing around the same - I would suggest to randomly cut one to leave the other one as the test control. Then start a new round of lander-testing like you were planning to do. Once you have a winning lander from that round, mass-test offers.
Eager to see the next round of stats!
Amy
Chapter 2: In Which The Protagonist Rushes Unwittingly Into The Unknown
Thanks for the encouragement Amy!
After running the campaign for a couple of days since the last update, the figures are as follows:
Cost: $20.95
Conversions: 18
Revenue: $10.80
Profit: $-10.15
ROI: -48.46%
Observations
I already know what the more eagle-eyed among you are going to say: 'But rhm, you said you'd allocated $10 to each lander and would let them spend the full amount again before making a decision!', and you'd be right to say so.
However, continued testing showed little difference in the performance of the remaining landers from the original set, with the new angle falling completely flat and one of the variations showing zero difference in either click through or conversions. I decided to bite the bullet and kill these landers off, as they were most unlikely to become winners without my intervention.
The spot of light from this round, however, was one of the tweaked landers, which performed at a VERY small profit. There is very little to get excited about here, other than the fact that the performance of this lander was boosted significantly by the simple adjustment of increasing the size of the headline text.
Another observation is that I'm finding it very slow going actually spending my budget at the current bid level, with an average spend of roughly $15/day. I'll have to see how the new test performs, and consider upping the bid if the current EPV holds out. Obviously this will hurt ROI, but if it increases total profit I won't be too fussed right now. Still someway off scaling this to other sources, but this could be an option for greater volume once I see further improvement.
Next Steps
I'll be running the control against a further variation to see if I can eke out more performance from the winning lander. Whilst this runs, I've used Adplexity to identify other networks that run similar offers- on Amy's advice I tried to find more offers to mass test from my current networks to test, however there were none! I'll be applying to these newly identified networks to test offers in a future stage.
I'll also get my thinking cap on for headline text, to see if running several variants of this will have any real impact on CVR.
any updates about this ?
Home > POP / PPV / Redirect > Follow-along Campaigns