SE Advertising

[VMG]

How do search engines protect advertisers from click fraud?

Thanks

[TallTroll]

<grin> The first word of that question is a superfluous, surely? </grin>

Seriously though, there are a few obvious things that can be done.

Limit any given IP to generating 1 "paid for" click on any one advertisers account in a given time period (12 hours, 24 hours)
Monitor IPs generating multiple paid clicks for signs of fraud (subsecond click intervals, high money term clicks only, temporal click clustering)
Monitor certain IP blocks (AOL proxies etc) to establish "baseline" activity for compare and contrast when investigating click fraud
Maintain a "blacklist" of known fraud IPs
Employ a core of experienced PPC types who can manually review click data, and identify fraudulent traffic
Carefully evaluate advertiser feedback / complaints

[Jeff Martin]

Even with all that TallTroll mentioned, you should be monitoring your PPC campaigns for fraud. The more familiar you are with your campaign activities the better you can spot irregularities. There are both free and pay services available for click fraud monitoring.

In the end, and despite all of the technology the search engine's use, you may still be the best detector of suspicious activity.

[VMG]

Thanks Jeff! What should I look for?

[Jeff Martin]

I would first recommend a tool/service to use unless:
1. You have little PPC activity or
2. You have enough time to plow through all of your data and historical data

You can do a search for 'click fraud' and find alot of paid click fraud services, I operate VeriClix which, as far as I know, is the only free click fraud monitoring service.

You have to look beyond the IP address and look at the characteristics of each visitor geography, IP block, user agent, user agent plug-ins, keywords used, ad clicked on, date/time of each click, etc.