Home >
General >
Affiliate Marketing Forum
ProplerAds PUSH - how to start, optmizie and scale campaign? (9)
07-23-2019 07:04 AM
#1
peweb2005 (Member)
ProplerAds PUSH - how to start, optmizie and scale campaign?
So I have a good offer. AM confirmed it works great for other affiliates and I want to promote it as well.
I want to test 3 angles, one lander (this is the lander everybody uses) and 3 ads (one everybody uses and two mine). When I run CPC push in Propler I can choose between high, medium and low user acivity.
When I run 3 campaigns - one for each user activity with suggested bid and I see there is a big overlap between zones so the ad from campaign 2 (medium) could appear 3 times on the same zone - and each time with different bid because of this overlap. This happens and views and clicks it gets are different - the higher bid is, the more views and clicks it gets.
Example:
Zone id 1524741 is part of the high, medium and low user activity. Campaign targeting the high user acitity got 47852 views 174 clicks at 3.8 price. Campaign targeting the medium user activity got for this zone 112 views 2 clicks at 0.072 price and this same zone id as part of the low user activity got 2385 views 49 clicks at a price of 1.788
What is the best way to start, optimize and scale a campaign? Should I first run one campaign targeting all three user activities, run 3 campaigns - one for each activity or run one campaign for one activity - which one to start with?
This is a sample data of each user activity

You see, low gets most of views but CTR is low, CR is 0. The high campaign made 2 conversions but how to scale? The bid is already set at max so I can see how to get more traffic - it would be great if I can get 10x more traffic and make 20 instead of 2 conversions at this CR but no idea how to do it.
07-23-2019 01:04 PM
#2
matuloo (Legendary Moderator)
I had the best results with the "high activity" subs at propeller. These users were converting the best, period.
That doesnt mean I wasnt using the other activity settings though. In order to get volume, I targeted all 3 groups, just always in separate campaigns. And in most cases, I was starting with the "high" subs, then scaling to medium and low. This worked the best for me.
07-23-2019 03:17 PM
#3
peweb2005 (Member)

Originally Posted by
matuloo
I had the best results with the "high activity" subs at propeller. These users were converting the best, period.
That doesnt mean I wasnt using the other activity settings though. In order to get volume, I targeted all 3 groups, just always in separate campaigns. And in most cases, I was starting with the "high" subs, then scaling to medium and low. This worked the best for me.
Thank you!
This was my hypothesis now backed up with expert advice!
When you find good / bad zones in high do you Black / white list them in middle and low campaigns?
07-23-2019 08:54 PM
#4
matuloo (Legendary Moderator)

Originally Posted by
peweb2005
Thank you!
This was my hypothesis now backed up with expert advice!
When you find good / bad zones in high do you Black / white list them in middle and low campaigns?
Depends on how bad the zones are. In some cases it's the problem of the subscribers "age" so what doesnt work with low, can still back out with high. But many zones are simply bad, no matter how fresh subs they deliver. Based on my own data, usually when a zone is totally useless, it's working bad across all the activity levels and it's visible pretty early.
07-24-2019 08:26 AM
#5
peweb2005 (Member)

Originally Posted by
matuloo
usually when a zone is totally useless, it's working bad across all the activity levels and it's visible pretty early.
How would inedtify a zone as totally useless? I run a bot test in two GEOs and looks like Propler Push has no bots, or my test can not find them so the only thing I can think of is spent 2-3x with no conversion.
Propler has auto optimization algo that can auto block list zones if you set up target CPA. But I have found out they divide a zone id into slices and I get "some slices were blocked" even when spent per zone is 2-4x lower than target CPA - still no idea how this algo works so I turned it off.
07-24-2019 09:43 PM
#6
matuloo (Legendary Moderator)

Originally Posted by
peweb2005
How would inedtify a zone as totally useless? I run a bot test in two GEOs and looks like Propler Push has no bots, or my test can not find them so the only thing I can think of is spent 2-3x with no conversion.
Propler has auto optimization algo that can auto block list zones if you set up target CPA. But I have found out they divide a zone id into slices and I get "some slices were blocked" even when spent per zone is 2-4x lower than target CPA - still no idea how this algo works so I turned it off.
There are bots there too, less than in case of POPs, but there is still a lot of them.
How would inedtify a zone as totally useless?
If I see a zone performing poor in "high" and it also does poor in "medium" and "low", I check other campaigns to see if its showing the same behavior there and if it does, I block it everywhere.
Let me add that I do such "deep analysis" only with zones that send a significant amount of traffic, I don't feel like bothering with all the small ones. I know I'm leaving money on the table, but I simply don't have time for this.
I don't use the auto algo by propeller, so can't really comment here, unfortunately
08-06-2019 01:43 PM
#7
expertmobi (Member)
here is one more thing. if you are working with big sources you have to be ready to see bot traffic. you can block some zones but when their publishers see that they stop earning with them, they can easily change their IDs and your black-lists won't be relevant anymore. so some zones can become profitable after some time and vice versa. it isn't the universal truth but the info to take into consideration.
victoria
08-06-2019 04:17 PM
#8
matuloo (Legendary Moderator)

Originally Posted by
expertmobi
here is one more thing. if you are working with big sources like PropellerAds you have to be ready to see bot traffic. you can block some zones but when their publishers see that they stop earning with them, they can easily change their IDs and your black-lists won't be relevant anymore. so some zones can become profitable after some time and vice versa. it isn't the universal truth but the info to take into consideration.
victoria
I'd actually say this statement is somewhat "universal", thanks for pointing that out. I've noticed this across several sources and niches/verticals... those who are gaming the system try to refresh their bot operations from time to time, in order to be able to sell as much of their crap as possible. So me networks are quite good at catching this, with some it happens more frequently. This is one of the reasons why it's important to monitor campaigns often and blacklist whatever crap appears
08-07-2019 02:46 AM
#9
erikgyepes (Moderator)
What is the best way to start, optimize and scale a campaign? Should I first run one campaign targeting all three user activities, run 3 campaigns - one for each activity or run one campaign for one activity - which one to start with?
Better to separate as the bids are different and also the user base is different.
High user activity are basically fresh users on the list, that's why they tend to convert best as Matej pointed out (I also got best result with this targeting, even it's the most expensive one).
Medium can also sometimes work, and low in my experience is the most garbage, you have volume and cheap bids there, but those users already been bombarded with many ads, so they not response as well.
Also if you wan to get more impressions you can try their CPM bidding model. It's more dangerous as you will pay for each impression, but if you combine it with HIGH user activity and you already weed out bad zones, you can have good results with it.
Home >
General >
Affiliate Marketing Forum