Home > Paid Traffic Sources > Mobile

How do you measure cross network cannibalization (8)

01-29-2014 08:39 AM #1 grandtheftpixel (Senior Member)
How do you measure cross network cannibalization

Considering that exchanges are powering a lot of mobile ad networks, how are you measuring the effect that cannibalization is having on your campaigns.

cannibalization = when you're buying traffic from the same site/app across different traffic sources.

Given that a lot of the time all you get is a site ID, how are you identifying when a new source is fucking with your profitable one?

01-29-2014 06:18 PM #2 bbrock32 (Administrator)

Hmm that's a good question but I don't think there is an answer.

Yep, there are a few networks like InCrowd that provide you the exact placement but that's about it.

02-15-2014 01:31 PM #3 auditor (Member)

Quote Originally Posted by bbrock32 View Post
Yep, there are a few networks like InCrowd that provide you the exact placement but that's about it.
How about that HumanDemand.com network? I just found an old thread here on STM that introduced them. Their USP was that they have transparency of which app is working for you.
I didn't know about InCrowd, still need to run a test with hem.
Is anyone still using HumanDemand? Perhaps it's limited to mainly US?

02-15-2014 07:37 PM #4 stackman (Administrator)

I've never really thought about this, but i bet there's quite a bit of cannibalization if you're advertising on 5+ major networks

02-16-2014 12:01 PM #5 The Angry Russian (Moderator)

It depends if the network is pure RTB or a more traditional network.

A lot of times publishers allocate their impressions to different networks based on contracts or whatever stupid mediation layer they are using. If you're buying the same placement from 5 different networks you're minimizing your risk of the traffic just dying when that publisher decides to send more inventory to InMobi instead of Millenial as one example.

02-16-2014 01:26 PM #6 Mr Green (Administrator)

Quote Originally Posted by auditor View Post
How about that HumanDemand.com network? I just found an old thread here on STM that introduced them. Their USP was that they have transparency of which app is working for you.
I didn't know about InCrowd, still need to run a test with hem.
Is anyone still using HumanDemand? Perhaps it's limited to mainly US?
Last time I checked humandemand it was horrible. It was also built like some bizopp, and was asking $1500 deposits.

Incrowd has full transparency.

Quote Originally Posted by The Angry Russian View Post
If you're buying the same placement from 5 different networks you're minimizing your risk of the traffic just dying when that publisher decides to send more inventory to InMobi instead of Millenial as one example.

I don't know if that benefit outweighs the cost of you bidding against yourself....With that being said, I know incrowd say they built the platform serve 100% of traffic to the highest bid. However when I tried created 3 identical campaigns but with 3 different bids, the lower bids still got a portion of traffic. So that shows even if you have one campaign bidding the highest possible, you still might not be getting all the traffic that's available.


---

In terms of how to measure cross network cannibalisation, to find that out you really need to know the in workings and deals of the networks you are running with. Some networks won't be happy sharing that information.

02-16-2014 07:51 PM #7 auditor (Member)

@MrGreen that's valuable advice. Appreciated.

02-16-2014 10:00 PM #8 nefig (Member)

Dont touch Human Demand. Horrible..

Home > Paid Traffic Sources > Mobile