Pop Traffic: Which Landers To Cut (Just Checking: Did I get this right?) (2)
Pop Traffic: Which Landers To Cut (Just Checking: Did I get this right?)
Hey,
so right now I'm running my first campaign on Zeropark (Pop Traffic). I have been split testing one offer with 3 different landers for a little over three days.
I had the impression one of the landers was performing pretty bad (from an ROI perspective), but I wanted to be sure before doing anything wrong.
So here is what I did, would love some input on whether I'm doing this right (well actually raher on what I'm doing wrong):
Approach 1 - Using the Bayesian Calculator
So the first thing I did was use the Bayesian Math Calculator (Tutorial is here) but I wasn't very satisfied with the results:
In his Tutorial Caurmen said you should aim for 90% certainty that one is going to be the best before cutting the others, so this doesn't help (right now I don't even want to know which one is the BEST, I just want to find out whether it's safe to cut the WORST one).
I'm also not sure this is a viable approach to POP traffic, because it doesnt take into account the CTR of the Lander.
Approach 2 - Calculating using the "scary statistical Math calculator"
Next I went for the "scary maths calculator" (again Following Cuarmen's other Tutorial).
These are my Stats from
And here is how I put them in the stats Calculator:
Now Calculating these times 100 as specified in the tutorial I came up with the Minimum and Maximum Conversion Rates for my Landers:
Lander A: 4 Conversions @ 2817 Views: 0.13 - 1.21
Lander B: 1 Conversion @ 2782 Views: 0.00 - 0.69
Lander C: 2 Conversions @ 2690 Views: 0.03 - 1,02
So Far So good.
Applying this to my Numbers
Now according to the tutorial I need to see whether my minimum viable conversion rate is even achievable with those landers. This is the case, if the maximum achievable conversion rate is bigger than my minimum
Now comes the part where I feel unsure:
My bid on Zeropark is 0.006. So if I calculate this * 1000 I get to 6$ which at the payout of my offer (2.40$) would mean I'd need 2.5 Conversions per 1000 Pops in order to be profitable. I.e. my minimum useful Conversion Rate is 0.25% in order to break even.
So that means I should keep all the landers. BUT: That one lander is running at -90% ROI, whereas the others are running at -30%, so should I still leave it in (for now - and check again later?)
Honestly I'm not using these calculators, but what I can say about the screenshot:
You have a CPA offer that pays $2.40. You say you're bidding 0.006 (this is average bid on zeropark), so basically you've spent a total of $17 roughly per lander.
Now what I would do is go check how much you've spent effectively per placement. To get profitable on pops, you need a mass-appeal offer, no doubt about it. But the way your spends are from the screenshot, I'd check the following:
1) How much did you effectively spent per placement?
2) Are there placements that took a big percentage of the whole spendings
3) Are there any placements that look extremely weird in terms of CTR (think 0% / 50% CTR, basically any crazy stats that don't match the overall impression).
A total spent of $17 per lander with a payout of $2.4 is actually spending 7x the CPA in total per lander. Certainly not enough to decide if it's a zero or a hero overall.
LP CTR depends on the offer and payout. I personally had pop campaigns (although desktop for this example) running with a CTR of 1% average (high paying CPA) and was running over 40% ROI after scaling. Obviously with mobile and low payouts you probably need a way better CTR, but given your first LP is just -36% ROI, I'm sure it's possible to get it profitable.
Doublecheck how much you're spending per placement, because even in a very unlikely scenario in which there are just 10 placements running and sending traffic to your landers, you haven't even spend 1x the CPA per placement.
Home > The Newbie Zone > Questions and Answers