Push Notification Traffic Quality?

11 replies
Hi all. New here, but not new to the ad tech space. Have a question, I probably cannot post it yet in the appropriate forum, but maybe I can still get some help here please.

So, I have a bunch of my own push notification users which are not particularly high quality (collected through run of network traffic); however, this universe of push users still performs significantly better than the traffic I buy from networks.

I've tried a bunch, Ad Maven, Propeller Ads, Hilltop, Adcash, richpush, and while Ad Maven and Propeller are not completely horrible, they still perform way worse than my users (again, my users are not targeted users of any kind).

Has anyone else experienced these quality issues? Are there better networks I should be testing? It makes me suspicious most of these networks are diluting their push traffic with non-push or something else since how can you perform much worse than completely untargeted "randomly" acquired users, but maybe I'm missing something?

Thanks
#notification #push #quality #traffic
Avatar of Unregistered
  • Profile picture of the author professorrosado
    And we're missing a lot too!

    What are you casting to your non-targeted group?
    Are you saying that your targeted push purchases are performing worse with the same ad or content than your non-targeted group?

    Are you sure there is absolutely no other difference? You say that your group are "your users"? Doesn't that mean they have more of a reason to click through than a completely external group?
    {{ DiscussionBoard.errors[11558958].message }}
  • Profile picture of the author manonpoint
    Same offer and creative is being sent to both test sets.

    Unfortunately, I cannot get more specific/share on how I acquire the push users, but I know the collection process, and the users are, at best, a random collection of users who surf the net. I am not sending push related to Mens products to users who accept push notifications on a Mens Style site or anything even remotely that targeted.

    So, there is no targeted/ un targeted groups ... my users are also completely untargeted; however, they perform much better than every network I have tested. Since I know my users are random users and no targeting is done whatsoever, the main conclusion I arrive at is that the networks are blending in bots/or other bad traffic, non-US traffic, mobile but calling it desktop, etc.. But I can't imagine EVERY push network is doing that...
    Signature
    {{ DiscussionBoard.errors[11558961].message }}
    • Profile picture of the author savidge4
      Originally Posted by manonpoint View Post

      So, there is no targeted/ un targeted groups ...
      You need to simply start targeting... I would push a bit of a questionnaire maybe? At the end you could engineer the answers segregating your list, that would be oh so awesome.

      Think of it this way.. I give you a list, that happens to be the white pages in a phone book. Granted this is actually kinda a targeted list based on Geography.. but aside from that just a bunch of names and numbers. Some how some way I would need to segregate the list to make it of use.

      I would start with calling each number and noting who answers ( M / F ) I would also pay attention to answering machines.. you can usually tell M and F as well as single, couple or Family. Once the first round was complete I then have a segregated list that I can maybe direct an offer towards.

      So maybe for you the answer maybe ask a simple question.. Are you Female or Male? press 1 for female 2 for male. Next round through segregate by age Under 21, Over 21. from there you then have some amount of demographics and the ability to send an offer of some kind.

      You could also take the simple 1 or 2 questions further if you have a specific offer you are wanting to send, to identify of the base group who would be most qualified to send the offer to.
      Signature
      Success is an ACT not an idea
      {{ DiscussionBoard.errors[11559217].message }}
  • Profile picture of the author manonpoint
    What you are saying makes a lot of sense, but the performance issues are not CTR, they are conversions post CTR. Since my list is already un-targeted, I'm at a loss as to how the performance from most of the networks can be so much worse than even my untargeted users once they click. The main things I can think of are non-US clicks (my traffic will only convert w/US), fraudulent/junk clicks, or mobile clicks despite filtering for only desktop on the platforms (mobile conversion is way lower for my offers).

    Either way, has anyone else noticed a significant quality difference between the different networks, even if you haven't experienced the same "all the networks suck" that I have so far.
    Signature
    {{ DiscussionBoard.errors[11559236].message }}
    • Profile picture of the author savidge4
      Originally Posted by manonpoint View Post

      What you are saying makes a lot of sense, but the performance issues are not CTR, they are conversions post CTR. Since my list is already un-targeted, I'm at a loss as to how the performance from most of the networks can be so much worse than even my untargeted users once they click. The main things I can think of are non-US clicks (my traffic will only convert w/US), fraudulent/junk clicks, or mobile clicks despite filtering for only desktop on the platforms (mobile conversion is way lower for my offers).

      Either way, has anyone else noticed a significant quality difference between the different networks, even if you haven't experienced the same "all the networks suck" that I have so far.
      Yes, there are inconsistencies, and they all suck LOL - that is the nature of the beast.

      I think at this point you need to be looking at not so much WHERE you are getting the "List" from but HOW you can segment the list once you have it. You technically should be able to see Location, Device, and Browser type information.

      something like this: https://vwo.com/blog/audience-segmen...ion-marketing/

      You then need to figure that part out, and the rest becomes easier.
      Signature
      Success is an ACT not an idea
      {{ DiscussionBoard.errors[11559239].message }}
  • Profile picture of the author manonpoint
    yea that i can do easily... but I don't have enough of my own users (tens of thousands right now I want more like 150+k) so I wanted to supplement with ad networks, Also, my list even without any segmenting and targeting, just knowing that they are actually US desktop users and no fraud (since I control the push and tracking) already performs fine.. but, network performance is another story. Ad Maven and Propeller seem to perform slightly above terrible, but still way worse than my blind pool, and was just trying to see if others experienced the same or know which of the networks should be avoided for selling mostly bot/junk traffic disguised as real push users. Because, I can segment bot traffic all day long, it isn't going to help.
    Signature
    {{ DiscussionBoard.errors[11559252].message }}
    • Profile picture of the author savidge4
      Originally Posted by manonpoint View Post

      yea that i can do easily... but I don't have enough of my own users (tens of thousands right now I want more like 150+k) so I wanted to supplement with ad networks, Also, my list even without any segmenting and targeting, just knowing that they are actually US desktop users and no fraud (since I control the push and tracking) already performs fine.. but, network performance is another story. Ad Maven and Propeller seem to perform slightly above terrible, but still way worse than my blind pool, and was just trying to see if others experienced the same or know which of the networks should be avoided for selling mostly bot/junk traffic disguised as real push users. Because, I can segment bot traffic all day long, it isn't going to help.
      You have 10k+ and would rather have 150k+..and you are now experiencing "You Get what you pay for" Im not about to suggest a "Better" push pool because I have never found one... BUT I can suggest segmenting bot traffic into a DEAD file and filter the keepers. Yeah sure it takes an amount of work, but in the end.. you have what you want and that's 150k

      Its kinda the price you pay... you can bring your efforts to scale - or you can pay for crap from others and sort.
      Signature
      Success is an ACT not an idea
      {{ DiscussionBoard.errors[11559272].message }}
  • Profile picture of the author manonpoint
    I see that makes sense. Will take a look and see which networks let me get the most granular what what I'm buying / give me the best reporting.
    Signature
    {{ DiscussionBoard.errors[11559274].message }}
  • Profile picture of the author volcan
    I tried a couple of networks too , Adsterra or Peopellerads had similar reports and CPM. I started, however, to use Push notifications with AdMaven and rate increased with time. , I think now I have a lot of mobile traffic, mobile users have better performance from desktop, By mobile i mean Android users, perhaps you should focus more on that, An admaven representative contacted me from their support and gave me some few good tips I can use, so maybe you should ping them also, can help for long term no matter who you choose to work with.
    {{ DiscussionBoard.errors[11559396].message }}
  • Profile picture of the author manonpoint
    My current demand source will not work on mobile, but I will try for other demand. I have contacts at AdMaven.
    Signature
    {{ DiscussionBoard.errors[11559493].message }}
  • Profile picture of the author mdgolammostafa
    Hello, Push traffic enables associates and publicists to advance various offers, blueprints, presentation pages, edges and can be modest and effectively versatile.

    Anyway, consistently watch out for quality and before sending large volumes, contact your partner director to request authorization.

    You should consider message pop-up in your advancement system while it is as yet hot in offshoot advertising industry.

    Also, you, how would you see push traffic in the long haul? Amped up for testing this arrangement for your battles? Kindly don't stop for a second to share your contemplations beneath
    {{ DiscussionBoard.errors[11560239].message }}
Avatar of Unregistered

Trending Topics