But the problem here is pinpointing what is fraud and what isn't fraud, and that's the real problem.
Obviously 100 clicks from a single IP address in an hour or a day or even over a month are suspicious. However, how does Google or anyone else detect a network of seemingly unrelated IPs or proxies making a bunch of malicious clicks that are too subtle to track? Even spammers have taken to sending a little spam from a bazillion addresses (instead of one mega-spam blast from one IP like back in the old days) just to make it next to impossible to lock them out.
You have to prove pattern and intent to prove it's fraud, otherwise it's just a high traffic ad, maybe overhyped in it's advertising, taking the wrong people to the wrong page with poor ROI.
Who draws the distinction of whether:
- the ad targets the wrong people, gets good CTR but bad ROI dragging in the wrong crowd
- the ad gets good CTR but the product just stinks so no ROI as nobody will purchase
- the ad is getting fraud clicks
It's a very slippery slope where any whining MLMer can supposedly make a claim of fraud just because none of the 10,000 clicks he paid for wanted to be the next tier in his pyramid - WAH!