Sephrata
Active Member
I thought I might post this here to show my recent experience with Propeller Ads and bot traffic.
Now bot traffic is part of media buying. If we identify it, we have to try and block it or the source of where it comes from.
Networks should also take action to block and ban people found abusing their sites with bot traffic.
I left a campaign (on SmartCPA) running with Propeller only to find one 'zone' dominated my visits. So I analyse the data in Voluum and see that that entire zone is getting traffic from one ASN, an internet data centre. When I look at the user agents, it's quite clearly bot/fraudulent traffic that I've been receiving:
Now most networks are fairly good dealing with fraudulent traffic when it is identified. As advertisers, we usually get refunded or compensation paid for the traffic and the publisher should get banned.
I had an interesting experience with Propeller (on the most part, their network is quite good and has relatively low bot traffic). The first answer I get is that, if I am unhappy with the quality of the traffic then I should block that zone. Ok, yes I should but there are two problems here. Firstly, you cannot block a zone after starting the testing phase of SmartCPA. Secondly, you should immediately ask the advertiser for the records of the traffic so it can be investigated.
So, I escalate the issue to their support with full records of the fraudulent traffic. I was expecting a completely different answer, but I get told this time that "You campaign has such a high frequency (5 ads per 1 hour) , which leads such results" and "if you do not like some zone id , feel free to exclude them". Not exactly impressive answers!
For a data centre in a tiny village (population less than 500 people) in the Czech Republic to be sending over 20,000 visits from mobile devices in a 48 hour period, with not a single device using Czech or Slovak as the system language, is strange to say the least.
When I look at the zone in question here, this is first one I block from every single campaign I run. It usually dominates the visits in most geos and you never get a single conversion from it. Unfortunately (and this is an error by me), I did not add this zone to the block list before running the SmartCPA campaign. However, it does raise very serious question marks as to how Propeller deal with reports of fraudulent traffic and why this zone has not been flagged up before by them.
Now bot traffic is part of media buying. If we identify it, we have to try and block it or the source of where it comes from.
Networks should also take action to block and ban people found abusing their sites with bot traffic.
I left a campaign (on SmartCPA) running with Propeller only to find one 'zone' dominated my visits. So I analyse the data in Voluum and see that that entire zone is getting traffic from one ASN, an internet data centre. When I look at the user agents, it's quite clearly bot/fraudulent traffic that I've been receiving:
Now most networks are fairly good dealing with fraudulent traffic when it is identified. As advertisers, we usually get refunded or compensation paid for the traffic and the publisher should get banned.
I had an interesting experience with Propeller (on the most part, their network is quite good and has relatively low bot traffic). The first answer I get is that, if I am unhappy with the quality of the traffic then I should block that zone. Ok, yes I should but there are two problems here. Firstly, you cannot block a zone after starting the testing phase of SmartCPA. Secondly, you should immediately ask the advertiser for the records of the traffic so it can be investigated.
So, I escalate the issue to their support with full records of the fraudulent traffic. I was expecting a completely different answer, but I get told this time that "You campaign has such a high frequency (5 ads per 1 hour) , which leads such results" and "if you do not like some zone id , feel free to exclude them". Not exactly impressive answers!
For a data centre in a tiny village (population less than 500 people) in the Czech Republic to be sending over 20,000 visits from mobile devices in a 48 hour period, with not a single device using Czech or Slovak as the system language, is strange to say the least.
When I look at the zone in question here, this is first one I block from every single campaign I run. It usually dominates the visits in most geos and you never get a single conversion from it. Unfortunately (and this is an error by me), I did not add this zone to the block list before running the SmartCPA campaign. However, it does raise very serious question marks as to how Propeller deal with reports of fraudulent traffic and why this zone has not been flagged up before by them.
.
