Home > Paid Traffic Sources > POP / PPV / Redirect

PopAds ISPs - Don't Optimize for What You Can't Target! (11)

03-31-2016 12:14 AM #1 xxf8xx (Member)

Thanks Amy this post is gold! I'm targeting wifi myself right now but this is great to know for the future!

04-01-2016 02:51 AM #2 sushiparlour (Member)

Thanks vortex, I only recently realized this problem as well and have just added in the tokens

04-01-2016 05:34 PM #3 vortex (Senior Moderator)

Regarding the original post above, there's something I forgot to point out / clarify. When I said "don't optimize for a carrier you can't target JUST carrier traffic for", what I meant was don't just test offers that convert well on carrier traffic for that specific carrier. I wasn't trying to suggest to stay away from those carriers!

The point I was trying to make, is that we need to be prepared to monetize the "tag-along" wifi traffic, such that we'd still be getting a positive ROI for the ISP/Carrier OVERALL.

What I like to do for situations like these - where you can't JUST get carrier traffic without the wifi - is to set up tracker rules to test offers for wifi and carrier separately.

Example: Going to use Viettel again here. Remember that even when we target Viettel with "connection types" set to "cellular" and "connection speeds" set to "cellular/carrier", we'll STILL get 80% wifi traffic. So what we can do, is set up a rule in Voluum that says "If mobile carrier is Viettel", and list offers that accept Viettel carrier traffic. Then in the default path (which the wifi traffic will go through by default), list offers that accept wifi traffic. That way you'll find the best offer for EACH of Viettel carrier traffic and wifi traffic.

Another alternative - and this is what most people are doing - is to simply redirect the wifi traffic to a monetizer service like monetizer.co or YTZ. I personally prefer knowing exactly which offer it is that I'm promoting, so I tend to test individual offers instead of rerouting to monetizer services. But you can for sure add these to the list of offers for wifi while split-testing.

I'm in the process of writing up another post on granular offer testing that will illustrate this process, so if you'd be interested please stay tuned!



Amy

04-02-2016 12:54 AM #4 memo87 (Member)

Great post, thank you for sharing. I had no idea about the popads "token details" tool! I wish Voluum had a way to cross-reference IP's and ISP's natively. Is there any tracker on the market that can do this?

04-02-2016 03:18 AM #5 cloudalien (Member)

Using Funnelflux (http://funnelflux.info) I can create a conditional node that will separate the wifi from the carrier traffic using a wildcard. Example: If Connection ISP is Viettel and Connection IP is 125.235.13.*(wildcard) then route traffic to main offer else route to other offer.

Viettel uses these public IP's 125.235.13.* (https://db-ip.com/all/125.235.13) When a Viettel user is on cellular they are assigned one of these public IP's but once they switch to wifi they will be assigned a local IP from their router. Hope this helps.
-Michael

04-03-2016 05:16 AM #6 sushiparlour (Member)

@vortex - what are your thoughts on even more granular testing whereby breaking down campaigns for each carrier? Since that way you can pay different bids for different carriers as this in a sense is an extension of spliting wifi vs carrier. Just wondering if its worth the try or is the amount of traffic too low to bother from your experience?

04-04-2016 04:01 PM #7 vortex (Senior Moderator)

I wish Voluum had a way to cross-reference IP's and ISP's natively. Is there any tracker on the market that can do this?
Hm! I don't know if you could do that within Voluum and I haven't used that many other trackers. I only know of sites where you could look that up - for example: http://ipinfo.io/countries/


Using Funnelflux (http://funnelflux.info) I can create a conditional node that will separate the wifi from the carrier traffic using a wildcard. Example: If Connection ISP is Viettel and Connection IP is 125.235.13.*(wildcard) then route traffic to main offer else route to other offer.

Viettel uses these public IP's 125.235.13.* (https://db-ip.com/all/125.235.13) When a Viettel user is on cellular they are assigned one of these public IP's but once they switch to wifi they will be assigned a local IP from their router. Hope this helps.
-Michael
This is MASSIVE help - thanks very much Michael! I've been trying to figure out how to achieve the same thing on funnelflux to be honest - now I know thanks to you!


@vortex - what are your thoughts on even more granular testing whereby breaking down campaigns for each carrier? Since that way you can pay different bids for different carriers as this in a sense is an extension of spliting wifi vs carrier. Just wondering if its worth the try or is the amount of traffic too low to bother from your experience?
VERY solid thinking! I actually JUST finished writing up that exact method - part 1 can be found here. The amount of traffic can be too low for us to bother, but all it takes to find out is like $5-10's worth of testing to see where the carrier traffic is. I went into way too much detail in that post.



Amy

03-16-2017 08:57 PM #8 davidep (Member)

Hey Amy, great thread as usual. I've made some test on Popads but it seems to me that it doesn't have much carrier/3g traffic. It looks like something like 80% of the traffic is wifi and wired. So I suppose that we should move to other TS if offers we run accept only carrier traffic.

In your experience, is this true?

03-16-2017 09:41 PM #9 matuloo (Legendary Moderator)

Quote Originally Posted by davidep View Post
Hey Amy, great thread as usual. I've made some test on Popads but it seems to me that it doesn't have much carrier/3g traffic. It looks like something like 80% of the traffic is wifi and wired. So I suppose that we should move to other TS if offers we run accept only carrier traffic.

In your experience, is this true?
I'm not exactly a popads expert, but the breakdown you mentioned 80/20 will be similar in any other network, there is simply less people browsing on 3g compared to wifi. The percentages will fluctuate, but wifi always wins in terms of volume. So it is more about the overall volume of a network, if it's low, 3G volumes will be low too.

03-18-2017 02:21 PM #10 davidep (Member)

Thanks Matuloo! Now I realize it was a question with a quite obvious answer....

03-18-2017 05:58 PM #11 vortex (Senior Moderator)

Quote Originally Posted by davidep View Post
Hey Amy, great thread as usual. I've made some test on Popads but it seems to me that it doesn't have much carrier/3g traffic. It looks like something like 80% of the traffic is wifi and wired. So I suppose that we should move to other TS if offers we run accept only carrier traffic.

In your experience, is this true?
What matuloo said.

I just want to add: It would be a very good exercise to go to a few traffic networks that have traffic estimators, and spend some solid time to look up the volumes of each carrier and log them into a spreadsheet for future reference. Then you know where the volume is and just target those carriers.

The same applies to any other targeting, not just carrier: Look for the traffic first, then spy to see what's running well there, then start testing.



Amy

Home > Paid Traffic Sources > POP / PPV / Redirect