Google Releases Click Fraud Details

March 2, 2007: As the industry continues to raise concerns over the amount of fraudulent clicks occurring in search engine supplied, pay-per-click advertising services, Google has revealed how much they actually believe is going on and what they intend to do about it.

This week Google revealed on its AdWords blog, that ‘invalid’ clicking reaches up to ten percent of the total amount of pay-per-clicks that actually occur in its services. Google estimates the problem costs the industry around $US100 million a year, can be likened to ‘spam,’ yet is not as problematic as some auditors and advertisers suggest.

Google says they have a three layered system for invalid click detection which includes real-time filters, offline analysis and reactive investigations. Through this system, the search giant says it can identify and track levels of clicking activity and can safely make the assumption that ‘invalid’ clicking activity never exceeds more that ten percent of its total pay-per-click service.

Unfortunately, exactly what ‘invalid’ means is not explicitly defined by Google. Realistically, such a term could merely cover simple clicking errors or accidentally repeated clicks, rather than organised fraudulent activity.

“Click fraud protection is something we take very seriously, and it requires a great deal of research and development to do effectively,” says Shuman Ghosemajumder, Google’s business product manager for trust and safety on the Google Adware blog. “We believe we lead the industry in terms of our level of investment as well as the effectiveness of the protection we provide.”

Google backs up its claim that invalid clicking activity consistently remains in the single digits by revealing that the percentage of invalid clicks actually caught by advertisers but not detected by Google, consistently hovers around 0.02 percent.

But according to click fraud auditors and a number of disgruntled advertisers, the Google figures simply don’t add up. An auditing firm called ‘Click Forensics,’ estimates the figure on fraudulent clicks to be more like 14.2 percent of the industry’s total pay-per-click activity.

As Google released its hard facts of click fraud, it held no punches in lashing out at click auditors like Click Forensics. The auditor, which claims to have held a number of discussions with Google, has responded, claiming they are not compensated on the amount of fraud they find occurring but rather “each time a fraudulent click occurs and is missed by a search engine, they get paid for it!”

“Our goal is the continue to build a great company that helps advertisers ensure they get what they pay for in the tradition of Nielsen and Arbitron,” says Tom Cuthbert, president and CEO of Click Forensics, in a written statement. “Google would do well to take a less confrontational and more cooperative approach embraced by their competitors.”

Overall, advertisers should welcome the fact that Google acknowledges its service has been abused. Google is bringing a number of tools into the scenario to help combat the problem and intends to soon allow advertisers to block certain IP addresses that are suspected of click fraud.

Advertisers might feel confident with the knowledge that Google will offer them a credit if they find that clicking activity has been fraudulent. But the real question still remains about the number of fraudulent clicks not detected by Google, advertisers or even the auditors.

Comment on this story.