Hello everyone,
When using a cloaker like blackjack or traffic armor, you are able to see all incoming traffic and what type it is. Bots (data centers) are auto blocked and rerouted to the safe page. Unfortunately even though many traffic sources say that you are credited back for bot clicks, the numbers never ever match up and bot clicks get paid for. I'm thinking that there must be thousands of domains/ widgets as most major native traffic sources have enormous networks so there is lots of room for blacklisting... but surely even premium domains must have SOME bot traffic? Or are those sites able to block it themselves? My question to you today is: how many registered bots do you pay for before you ban a domain?
There are always bots more or less. I only looking at the margin I want to achieve and if this is possible with the bots, I keep it.
When dealing with bots, there are 2 situations you must consider.
1. starting green, without a proven offer or LPs or creatives.
2. scaling an already proven campaign with a solid funnel.
In case of the 1st situation - you need to be more conservative and pause placements with high bot %, because you are testing a campaign and need to pick the best ad/offer/lander ... and obviously, testing on BOT traffic won't help you in any way.
Once you have a proven funnel, you can restart the paused placements, and look just at the margin as seonos suggested, as that's the only metric that matter in the end.
Caurmen wrote an article about this, couple months ago, check it out : http://stmforum.com/forum/showthread...ut-Bot-Traffic
On some traffic sources you are able to blacklist by IPs.
So before ban domain you can get a list of bot ISPs and IP ranges and blacklist it.
This way you remove a lot of bots traffic from all domains.
And only after that you should start to blacklist domains.