Fair Isaac Research Indicates Click Fraud Is Potentially Large Problem in Parts of Search Engine Advertising
Monday, 21 May 2007
Fair Isaac Corporation (NYSE:FIC), the leading provider of analytics and decision management technology, on Friday announced preliminary results of its research study into click fraud in pay-per-click (PPC) advertising. Early results indicate that fraudulent clicks can amount to 10-15 percent of advertisers’ billed click traffic. The study applied Fair Isaac’s market-proven artificial intelligence and fraud-fighting technologies to assess the scope of the problem and test effective ways to expose fraudulent clicks.
The announcement was made at Fair Isaac’s annual InterACT conference in San Francisco, where Fair Isaac presented a session on its research into click fraud.
The company launched the study in summer 2006 with support from the Search Engine Marketing Professional Organization (SEMPO). SEMPO members and non-member pay-per-click advertisers contributed anonymous click-stream data to the study in exchange for analysis of their search engine advertising and potential click fraud.
“These early results represent a significant step toward addressing the problem of click fraud in online advertising,” said Dr. Joseph Milana, chief scientist in Fair Isaac’s Research and Development group. “Based on these preliminary results, we believe Fair Isaac’s expertise and proven fraud detection technologies can be effectively brought to bear in helping to detect fraud in the pay-per-click marketing space.”
Dr. Milana also indicated that more research into the problem is necessary.
“These are early results based upon a limited view of the market,” said Milana. “We’re looking for more advertisers to contribute to the study to help us arrive at a solid picture of the problem’s size and scope across the broader marketplace and different vertical markets.”
Click fraud occurs when advertisers pay for ad clicks that come from fraudulent sources instead of legitimate web traffic. A primary form of click fraud takes place when unscrupulous search engine affiliates/publishers in the ad networks arrange for repeated clicks on the site ads in order to increase revenue. Online robots or “botnets” as well as “click farms” of low-cost workers in remote countries are two mechanisms that perpetrate this type of fraud for profit.
Another common form of fraud occurs when employees of companies click on a competitor’s ads to deplete advertising budgets and dilute results. According to Fair Isaac’s study, fraudulent traffic as a result of malicious clicks by a competitor appears to be relatively contained and does not seem to have a significant impact on advertisers’ bottom-line.
“Marketers have embraced pay-per-click advertising, but ways to defraud the system also have become more sophisticated,” said SEMPO Chair Gord Hotchkiss. “Based on these early results, we believe the opportunity exists for our members to participate in an expanded study that can help further determine the extent of the problem and point the way to industry-wide measures for dealing with them.”
In addition to using the company’s patented, multi-entity profiling technologies used to successfully fight payment card fraud for more than 15 years, the study uses an anomaly detection engine — initially deployed in Fair Isaac’s Network Assurance solution — to analyze advertisers’ click-stream data, identify predictive patterns and generate a “pathology” score indicative of click fraud.
Fair Isaac’s Falcon™ fraud detection solutions are known worldwide for their effectiveness in helping businesses in the financial services, telecommunications, healthcare and pharmaceutical industries reduce fraud losses and protect their customers from fraudulent transactions. The company’s Falcon™ Fraud Manager solution protects 65 percent of the world’s credit card transactions.