So the last couple of days I've been learning the TrafficJunky software thanks to their amazing offer to match any first deposit. From what I've heard in the limited amount of time I've been in this industry, it seems like TJ's traffic is very premium. In the limited experience I have with them, it has been very solid and straight forward. The only other traffic source for adult I've used in ExoClick. So, I went back and looks at the statistics within ExoClick to compare the two and found something pretty, uh, odd...
On 9/25 of the campaigns I ran, the detailed stats show a strange "Site" with an abnormal number of clicks and CTR. I have a strong feeling the clicks are bots, but I wanted to run it by everyone to see if this is a legitimate traffic before I ask for a refund. If it turns out it is in fact bots, maybe some of you can recoup some stolen money and take another look at your borderline campaigns to see if they were actually profitable.
Here are some screenshots of the statistics for a handful of campaigns, and an Excel doc with the totals. If you notice, each one of the "Sites" has two decimals, as if they are different in some way. One of the campaigns has more traffic from "Other" than xHamster, which I find nearly impossible.
Would have posted an image here, but there was some major shrinkage :P
https://www.dropbox.com/s/juqoo4kow4xcvtz/all.jpg
I've had the exact same results on exoclick and I'm spending $2-300 a day. If anyone can shed light on this that't be great.
I sent an email to them yesterday and they said they are going to get back to me tomorrow. t seems they are at least pretending to be a little concerned, but I will let you know what they say either way.
Some publishers have smaller sites where they send your traffic from, this could be listed under 'others'. Beside that, sometimes when they update stats, stats are kind of 'stored' for a minute or two under other category and then moved back to the sites. Not sure if this is the case here and I'm also eager on what they respond.
Hmm - bot detection is tricky, but it's doable.
Probably your best bet would be visitor behaviour analysis. An obvious "tell" would be a visitor who clicks straight through your LP after 200ms or so. It'd also be worth looking at the browser types that are visiting your page - bots might be dumb enough to be showing a non-standard browser.
More advanced tricks would include looking at the mouse movements they make, or using some interactive elements onscreen and watching how fast and consistent their response times are.
You can expand the table, choose Top 500 at the options to view more sites you got impressions from.
I don't think there is something unusual as all sites listed on Others can have more traffic than xhamster combined.