Wednesday, November 26, 2014

So How Exactly Does Adsense Track Fake Clicks

Advertisers depend heavily on Google's click-based Web advertising programs to grow their businesses, with many spending thousands of dollars annually to attract Web surfers to click on advertisements in Google. As a result, fake clicks, also known in the industry as "click fraud," are a major problem for advertisers who stand to lose significant amounts of money if the invalid clicks aren't detected. Google employs multiple techniques to identify click fraud through its AdSense program.


Background


Google's AdSense program is a pay-per-click advertising models through which advertisers pay a certain amount of money each time a person clicks on an advertisement. The amount paid per click is based on the popularity of the keyword and how frequently the ad is clicked compared to competitors. With Google AdSense, the ads appear on specific Web pages relevant to the ad content. Businesses who advertise through Google depend on the fact that a certain percentage of those people who click on their advertisement will make a purchase.


Click Fraud


Click fraud occurs when someone clicks on an advertisement -- often a large number of times -- without any intention of making a purchase. The intention of this is to make an ad campaign appear more effective than it actually is. Websites that host ads through Google's AdSense program earn a commission each time an ad is clicked, which means that publishers of such websites can repeatedly click on an ad appearing on their website to keep increasing its commission. According to PC Magazine, some websites have paid workers in third-world countries a few cents per hour to sit at a computer and click ads on their websites.


Monitoring


Google monitors clicks on Google AdSense ads to detect unusual activity. The company uses a proprietary technology to detect whether there is any unusual activity intended to drive up an advertiser's clicks. Numerous data points are analyzed on each click, including the IP address of the person who clicked on the ad, the time of the click, duplicate clicks, and other click patterns. If there are "excessively repetitive" clicks, those are filtered out by Google's system. If there are known sources of invalid activity, such as an IP address, those are automatically discarded, according to Google's advertising guide. Advertisers are not charged for clicks that Google determines to be invalid.


Legal Issues


Click fraud is a multi-million dollar annual problem for advertisers. In March 2006, Google agreed to pay a $90 million settlement to a group of its advertisers, including Lane's Gifts & Collectibles, which advertised through Google and headed a class-action lawsuit claiming it was being improperly charged by Google for fraudulent clicks. The lawsuit claimed that Google's technology to detect invalid clicks was not effective. Google has repeatedly said that it's not possible for its automatic filters to catch 100 percent of the invalid clicks, but that a very small number of clicks are fraudulent.