Click Fraud 101 - Part 1: Definitions and Culprits
|
Written By Reprise Media | May 22, 2006 | Share This
|
|
Many of our readers are curious about click fraud; it’s often in the news, and the term itself carries an aura that suggests there’s a darker, more mysterious side to our industry. Dark, maybe - mysterious, nah. While click fraud is definitely a problem, it’s not something that needs (or deserves) to be shrouded in inscrutability. We hope this primer will help take the edge off, and give you a basic idea how to keep track of and minimize the problems associated with click fraud.
Click Fraud - What is it?
Basically, click fraud is any instance when somebody - or something - clicks on an online ad with any intention other than to follow the ad to the advertised site. Since most every form of search advertising is of the pay-per-click variety (meaning the advertiser is charged every time one of their ads registers a click), it’s not hard to see how this can be a problem.
It’s not an issue that’s too popular with the engines, understandably - Google prefers the softer term “invalid clicks” when addressing it. We estimate that somewhere between 3 and 4 percent of all ad clicks on the Big 3 engines (Google, Yahoo! and MSN) are fraudulent.
While the engines do track click fraud and regularly refund a percentage of advertising fees, not everyone is satisfied with the recompense and resort to litigation. Google recently settled a class action click fraud lawsuit brought by Lane’s Gifts (now in the notification phase) which would divvy up 60 million dollars among all of the engine’s advertisers since 2002.
Who does it? The perps
Some click fraud is associated with an advertisers’ competitors or detractors; enough fraudulent clicks can eat up an ad budget and cause a company headaches. But a far greater threat, (due to the direct incentive involved) comes from publishers who self-click on their own ads to drive up revenue.
Of course, easily-detected fraudsters could see themselves getting kicked off a publisher ad network pretty quick. This has given rise to another, more complicated kind of click fraud perpetrated by a given entity’s adversaries in which the agents of the fraud hijack IP addresses to make it appear as though a publisher is clicking on its own ads. The innocent publisher is removed from the network (if temporarily) and their revenue stream shrinks or vanishes.
So how is it done?
The kind of click fraud that’s monetarily problematic isn’t just one person on one computer hammering on an ad and a ‘back’ button in succession. You may have heard stories about many people on many computers doing something like this - so called click farms, where people are paid to sit in a room and click on ads all day, usually in a place like China or India. But we feel that such representations are overplayed and even a little cartoonish.
Click fraud is a high tech crime, and as such, most of the heavy lifting gets done by tech. Usually the culprit is a bot, which is an internet program that’s automated to perform certain tasks; in this case, the task is to impersonate a user and click on given ads. Such bots are getting ever more sophisticated in order to escape detection - Search Engine Watch reports on a recently discovered bot called Clickbot.A that infected 34,000 computers; clicks from those bots were coordinated in such a way that no one machine registered an alarming number of clicks. Taken in combination, however, all fraudulent clicks originating from the bot network exacted a heavy toll.
Next: Identifying and dealing with click fraud.
Topics: Click Fraud |
Did any of you hear about something like the story I mention in my blog http://fightpc.blogspot.com/2006/05/what-if.html
???
Thnx