Click Forensics has announced a new version of its traffic quality management platform, which will now allow ad networks, publishers and advertisers to identify traffic based on its likelihood to convert.
The pay-per-click traffic quality management and click fraud prevention service provider said the newly-architected version is able to do this thanks to the company’s ability to monitor traffic across more than 300 ad networks as well as billions of clicks from search engines, publishers and advertiser Web sites.
“We know what clicks are converting for advertisers,” said Tom Cuthbert, president of Click Forensics. “To date, our system has been exceptionally good at identifying bad traffic that ad networks and publishers should not be selling and now the product has gotten to the point where we have the ability to identify the good traffic.”
The challenges, he added, included the fact that the company’s predictive modeling had to have the ability to adapt to different types of traffic, such as traffic from a certain country or that of different ad networks.
The new version also uses a machine-learning approach, he said.
“Each click learns from other clicks, finds click patterns and shares that information with the entire community,” Cuthbert explained, adding that anomaly detection, in which unusual traffic patterns are monitored, also was essential.
“The product highlights the value of individual clicks,” he added. “It allows publishers to commute the quality of that click based on its propensity to convert.”
