Business Week Cover Story: Click Fraud

A BusinessWeek investigation has revealed a thriving click-fraud underground populated by swarms of small-time players, making detection difficult. "Paid to read" rings with hundreds or thousands of members each, all of them pressing PC mice over and over in living rooms and dens around the world. In some cases, "clickbot" software generates page hits automatically and anonymously. Participants from Kentucky to China speak of making from $25 to several thousand dollars a month apiece, cash they wouldn't receive if Google and Yahoo were as successful at blocking fraud as they claim. Check out the FULL STORY.

Getting Paid to Commit Click Fraud

A recent article in The Washington Post profiles a "professional clicker". Jackie Parks, a disabled stay-at-home mom in Iowa, spends hours each day clicking Google AdWords ads for 1/2 a penny each. She makes about $300 a year doing this.

Jackie is one of thousands of people around the world who receive a daily email list of sites to visit and ads to click. Recruiters for these organized rings of clickers call them "pay to read" networks but they're really just "pay to click" networks. Call it what you want ... it's still blatant click fraud.

Advertisers Finally Waging War Against Click Fraud

Click fraud is undermining faith in pay-per-click ads and search engines are on the defensive. Advertisers have finally tired of not being adequately protected against a practice that's lining the pockets of scam artists and artificially inflating ad rates.

Source: Business Week

Search Engines Unite to Fight Click Fraud

The Internet’s leading search engines are teaming up with an advertising trade group to find a better way to identify and measure "click fraud," a scam that has raised doubts about the Web’s trustworthiness as a marketing vehicle.

The initiative, to be announced Wednesday morning by the Interactive Advertising Bureau, will draw upon the expertise of Google Inc., Yahoo Inc. and Microsoft Corp. - the owners of the top online search engines - to attack a problem threatening to erode their profits. Combined, the three companies control 86 percent of the lucrative U.S. search engine market, according to comScore Media Metrix.

Two smaller search engines, InterActiveCorp.’s Ask.com and LookSmart Ltd., also have joined the alliance along with the Media Rating Council, a nonprofit group formed 42 years ago at the urging of Congress to help track and validate the sizes of advertising audiences.

The collaboration of three fierce rivals like Google, Yahoo and Microsoft shows how serious the industry has become about curbing click fraud before the issue squelches an online advertising boom that’s enriching the search engines and their partners. Google, the pacesetting search engine, earned $1.3 billion on revenue of $4.7 billion through the first half of this year alone.

Click fraud has attracted an increasing amount of attention amid class-action lawsuits and industry studies asserting advertisers have been collectively overcharged by more than $1 billion for bogus sales leads during the past four years.

Google and Yahoo contend that those estimates are gross exaggerations generated by opportunistic lawyers and online advertising consultants hoping to cash in on the anxieties triggered by their calculations.

Part of the difficulty in fighting click fraud so far has stemmed from the inability to come up with a precise definition of the practice - a problem the alliance hopes to solve.

Click fraud has different twists, but the end result is usually the same: Merchants are billed for fruitless traffic generated by someone who repeatedly clicks on an advertiser’s Web link with no intention of ever buying anything.

The search engines and their partners are supposed to collect a commission for every click on short advertising links, typically displayed at the top or sides of Web pages, even if the activity doesn’t culminate in a sale. The commissions range from a few cents to more than $20 per click, depending on the words entered in a search box.

John Slade, senior director of Yahoo’s defense against click fraud, predicted the alliance’s guidelines "will be a game-changing step in measuring and fighting click fraud."

It may take more than a year before the guidelines are finalized, said Greg Stuart, chief executive of the Interactive Advertising Bureau.

The decision to develop the guidelines reflects the Internet industry’s "commitment to being the most accountable advertising medium and providing marketers with the highest level of transparency," Stuart said.

Source: BostonHerald.com

An Advertising Model That Does Not Click

Is the type of advertising that turned Google from just one more cool internet start-up without a business model into a corporate superstar too good to be true?

“Cost-per-click” advertising is one of those alluring ideas that make the internet so disruptive. By letting advertisers pay only when someone clicks on their message, and by targeting advertisements – either by linking them to the results of internet searches, or by fitting them to the subject of whatever someone happens to be reading online – it represents a huge leap forward in efficiency.

Powerful as these attractions are, though, there are inherent weaknesses. As cost-per-click advertising spreads round the world, more efficient approaches to online advertising are already being perfected. The next wave of successful internet companies will be the ones that master those techniques first – and there is no guarantee that it will be the ones who dominate in the cost-per-click world.

Click-based advertising is founded on one of the favourite myths of the first decade of online activity: that the click is the purest declaration of intent in the online age. As the basic action that drives internet activity, clicking on a mouse has come to be seen as the fundamental unit of measurement for online behaviour. To click is to cast your vote amid the tide of virtual endeavour. It is a statement of preference, an expression of desire, a prelude to action. If I clicked on your advert, I must be thinking of buying something.

This overly simplistic formulation is inherently flawed. Not all clicks were created equal. This is not just because of outright fraud, though that is certainly a sore point for advertisers. So-called “click fraud” is one of those problems where the perception may end up being worse than the practice.

In a court-ordered report, Alexander Tuzhilin, a professor of information systems at New York University, has just pronounced Google’s effort to stop fraud to be “reasonable”. Google seeks out those clicks that have been made for no other reason than to generate a payment from an advertiser – most commonly they come from unscrupulous online publishers who make up the wider Google network, carrying ads from the search engine on their sites and taking a share of the money that advertisers pay.

Yet, however good Google’s detection methods, it may never escape the perception that fraud is rife. By definition, it cannot explain to advertisers why it has decided that some clicks were fraudulent but charged them for others – that would give away its detection methods and expose the system to even greater fraud. Nor, given that its own short-term profits stand to benefit from illicit clicks, will it ever be able entirely to shake off a suspicion that it has a conflict of interest.

There are many other reasons to question the primacy of the click. To understand the reason behind every click, as Prof Tuzhilin points out, you would have to examine “deep psychological processes, subtle nuances of human behaviour and other considerations in the mind of the clicking person”. Say you entered the term “asbestosis” in a US search engine, then clicked on one of the adverts placed by a lawyer trawling for clients among the ranks of mesothelioma victims. That click would have cost the advertiser around $5. You may, though, have clicked merely out of curiosity, or because you are another lawyer trying to find out about a competitor, or because your click was a mistake.

This is not to say that the booming click-based advertising business is out of control. Google and other search engines have a clear interest in both preventing fraud and helping advertisers to analyse and understand the clicks they receive. A booming new advertising industry depends on it. Also, the pricing mechanism for clicks is to some extent self-correcting. If a certain proportion of clicks is redundant, then this will eventually be reflected in the amount that advertisers are willing to pay.

Eric Schmidt, Google’s chief executive officer, has recently been lambasted by bloggers for pointing this out. His choice of words was certainly unfortunate: “There is a perfect economic solution [to click fraud], which is to let it happen.” But he has a point. Yet Google’s mission is to bring greater efficiency to the advertising world. It hardly helps to have your CEO explain that a level of inefficiency is baked into the system – particularly if it is impossible for the average internet advertiser to assess exactly what that level is.

That should be a warning to advertisers who are only now trying out cost-per-click advertising for the first time. It should also serve as a reminder that, like many other things on the internet, advertising is still in its experimental stage. Potentially more efficient methods are already being tried. So-called “cost-per-action” approaches – only charging advertisers when a potential customer does something that points to a genuine interest, such as putting an item into an online shopping cart or filling out a form – are generating much of the buzz. As video spreads online, there is also likely to be a revival of traditional “cost-per-mille” advertising, where advertisers pay based on the size of an audience.

Google does not play much in either area, though it has started to experiment. As online advertisers look beyond the click, there should be plenty of scope  for new rivals to emerge.

Source: The Financial Times

AdWords Displays Click Fraud Stats

This week, Google AdWords advertisers will now be able to get invalid click data with their Campaign Performance and Account Performance reports. The new section displays both the number and the percentage of clicks that Google’s automated system has deemed invalid. Google doesn’t charge advertisers for false clicks. The new addition comes as Google continues to mop up after the Lane’s Gifts class action lawsuit, which Google settled and agreed to compensate AdWords click fraud victims up to $90 million for charges incurred by click fraud.

According to a study by Outsell, click fraud cost the ad industry more than $800 million last year. The AdWords click fraud stats are a way for Google to be more transparent about the way they handle the problem. Now advertisers can look directly to Google, instead to a third party researcher, to find out how many of their clicks are bad. The click fraud data displayed by AdWords only goes back to January 1, 2006.

Despite the transparency, Google is limiting the amount of click fraud information distributed to the public to keep click frauders from reverse-engineering their click fraud detection algorithms. Last week, independent researcher Dr. Alexander Tuzhilin of NYU released a report saying that “Google’s efforts to combat click fraud are reasonable.”

Source: Adotas

Google To Provide More Click Fraud Info

Google said Tuesday that it would begin offering AdWords advertisers reports on the number of invalid clicks that Google filters from their accounts.

Google is billing the move--which comes the same week that a federal judge in Arkansas is weighing whether to approve a $90 million settlement of a class-action lawsuit concerning click-fraud--as an attempt to give advertisers insight into click fraud.

Previously, Google told advertisers how many clicks they were being charged for; those advertisers then could figure out the number of invalid clicks by comparing their site logs to the number of clicks billed to their accounts. Now, however, the number of clicks Google deems invalid will be easily available to all advertisers, even those who do not employ analytics, said Shuman Ghosemajumder, Google's business product manager for trust and safety.

Ghosemajumder said the company was offering marketers the additional information to provide them with more insight and hard data about click fraud. "There's a lot of misinformation and confusing information that have been provided to advertisers," Ghosemajumder said. "We see a lot of greatly exaggerated reports," he said. The extent of click fraud remains unknown, but some researchers have pegged the problem has significant. Outsell recently reported that search marketers believe that around 15 percent of all clicks are invalid.

The reports on invalid clicks will be available on a yearly, quarterly, weekly, or daily basis, but will not be provided on a granular, per-click basis--information that some advertisers were hoping to receive.

A Google spokesman said that the decision to keep the clicks aggregated on a daily basis was made for security purposes. "This is obviously a concern and, while we wanted to be transparent, we also did not want to put our filtering systems and advertisers at risk, so we were very careful about it," the spokesman said. "We are confident that this info is useful to advertisers but not malicious actors."

Joshua Stylman, managing partner at search engine marketing firm Reprise Media, said that the move, while not necessarily a major one, was still heartening. "It may not be the ground-breaking click-fraud news, but it is a development, and I think it's a positive development for the industry," he said.

Stylman said that while sophisticated marketers could already glean the data that Google is now providing, this would give the data to any advertiser who was running a pay-per-click campaign through AdWords. "Believe it or not, there are advertisers who are not yet tracking," he said. "Anytime there's something that hits the market like this, it's not just for the big guys, it's for the common denominators."

Source: MediaPost.com

Google Comments Taken Out of Context?

A click-fraud activist believes that recent comments by Google CEO Eric Schmidt about solving click fraud by ignoring it, though apparently taken out of context, still belie an intrinsic problem.

The alleged misunderstanding began in March when Schmidt answered a question after his remarks at the SIEPR conference at Stanford University.

"Let's imagine for purposes of argument," Schmidt began, "that click fraud were not policed by Google and it were rampant.

"Eventually the price that the advertiser is willing to pay for the conversion will decline because the advertiser will realize that these are bad clicks.

"In other words, the value of the ad declines. So, over some amount of time, the system is, in fact, self-correcting. In fact, there is a perfect economic solution, which is to let it happen."

In an attempt to calm a turbulent blogosphere, which ran with Schmidt's comments, Google posted on its blog that Schmidt's comments were written out of context.

In the post, Google said that Schmidt was answering a hypothetical question with a hypothetical answer, adding that Schmidt was not describing Google's approach to the click fraud problem, only answering economics question at an economics conference.

But at least one click fraud activist, who is currently involved in litigation against Yahoo and Google and thus spoke only under the condition of anonymity, says that Schmidt's answer and the thinking behind it were flawed -- hypothetical or not.

The click fraud activist told internetnews.com that Schmidt's hypothesis presupposed that advertisers can tell the difference between good clicks and fraudulent ones.

That's not always true, the activist argued, suggesting that advertisers' sales are hard to correlate with online activity.

Take a car dealership, the activist said. Potential buyers might click on a sponsored link, look at a car and be very interested, but then not perform any transaction to signal the click was not fraudulent, thus making fraudulent clicks more difficult to detect.

Schmidt's theory presupposes that advertisers can always tell how many clicks are worth paying for, the activist said.

The activist also noted that advertisers who are already paying the minimum bid would have a hard time adjusting lower to compensate for click fraud.

The activist argued that Schmidt's comments belie Google's internal thinking that it doesn't need to worry too much about click fraud, that they do not have to go back to the drawing board.

But Google said in its post that it "strives to detect every invalid click that passes through its system, and to prevent those clicks from ever reaching an advertiser's account."

The company also said that "Anyone who has followed Google knows" they have "devoted significant resources to manage" click fraud.

Source: InternetNews.com

Bogus Clicks Still on the Rise

Swindlers have stepped up their effort to fleece millions of dollars from online advertisers who use lucrative marketing networks run by Google Inc. and Yahoo Inc., according to a quarterly report to be released Monday.

The sales referrals generated by clicks on the brief advertising links popularized by the two Internet powerhouses are a sham 14.1 percent of the time, based on information collected from 1,300 online marketers.

That's up from a click fraud rate of 13.7 percent three months ago, according to Click Forensics, a San Antonio-based consulting service that compiles the index.

The statistics jibe with other data asserting advertisers are paying a significant sum to Google, Yahoo and their partner Web sites for phantom shoppers even as more resources are devoted to thwarting scammers.

A recently released survey of 407 online advertisers by market research firm Outsell Inc. estimated click fraud cost advertisers $800 million last year.

Click fraud is a highly sensitive subject for Mountain View, Calif.-based Google and Sunnyvale, Calif.-based Yahoo because it raises doubts about the trustworthiness of the advertising model that drives their profits and stock prices.

Google, Yahoo and partner Web sites get paid each time someone clicks on advertising links usually displayed at the top and on the side of Web pages. Advertisers pay the commission even when the click doesn't produce a sale — a system that inspired bilking schemes.

The motives for click fraud vary. Most often, Web site owners repeatedly click the ads on their own sites to generate money for themselves. In other cases, advertisers target the ads of their rivals to drain their marketing budgets.

As click fraud becomes more prevalent and attracts more media attention, advertisers are becoming more aggressive about demanding refunds and better protection, said Tom Cuthbert, Click Forensics' president.

"Advertisers aren't satisfied with the status quo," he said. "They don't want to keep losing sleep at night wondering how much money they are losing to click fraud."

Reflecting those concerns, about 900 advertisers have joined Click Forensics' anti-fraud network during the past three months.

Google and Yahoo are better at weeding out click fraud than smaller Web sites, but Click Forensics still concluded both companies are being hard hit. About 12.8 percent of the clicks on ads served up by Google and Yahoo are deceptive, up from 12.1 percent three months ago.

Cuthbert said Google and Yahoo may be identifying some of those fraudulent clicks and removing fees from advertisers' bills. Both companies are tightlipped about how they monitor for click fraud, another factor that has frustrated some advertisers that want more transparency.

Google Chief Executive Eric Schmidt acknowledged click fraud remains an ongoing headache, but disputed the notion that the problem is becoming more prevalent.

"Smart people are trying to break the law, but we have even smarter people trying to prevent it," Schmidt said during an interview at a conference that concluded Sunday in Idaho.

Yahoo CEO Terry Semel declined to discuss the latest data on click fraud, saying he intended to address the issue Tuesday when the company is scheduled to release its second-quarter earnings. "We will be very proactive about it," Semel said during the same Idaho conference.

Both Google and Yahoo have agreed to settle class-action lawsuits to limit their potential liability for past click fraud. If approved, the two settlements would address any click fraud that occurred amid more than $22 billion of ad spending.

A two-day court hearing on Google's offer to pay up to $90 million in refunds and attorney fees is scheduled to begin July 24 in an Arkansas court. Yahoo's proposed settlement, which doesn't limit how much the company might pay, isn't scheduled to be reviewed in a Los Angeles federal court until late this year.

Source: Yahoo News

Click Fraud: Inherently 'Self-Correcting'?

Last week, market researcher Outsell released the results of a study on click fraud in the pay-per-click advertising market. Based on interviews with 407 advertisers representing approximately $1 billion in ad spending, the study found that confidence is dropping in pay-per-click advertising as advertisers estimated that more than 14 percent of the clicks that they're billed for are fraudulent. This would represent more than $800 million in wasted spending last year. The results of the study were reported broadly in the online ad trades, since $800 million is a big number, but to many the story didn't seem much different than the dozens of other click fraud stories that we have become accustomed to reading over the past year or two. But that changed.

Earlier this week, Donna Bogatin in her ZDNet blog uncovered comments from Google's CEO Eric Schmidt made in March of last year discussing Google's point of view on click fraud. He described how the pay-per-click advertising model was inherently "self-correcting" and that click fraud should be viewed by Google advertisers as a "cost of doing business." Google's logic, it seems, is that to the extent that click fraud becomes rampant, advertisers will understand its extent and will factor the cost of those "bad clicks" into what they are willing to bid and pay for ads. Thus, even if it becomes a serious problem, it will "self-correct." Advertisers will not overpay for the value that they actually receive. If the click fraud becomes too great, they will stop buying Google ads or paying the bid prices.

Essentially, Schmidt is saying that Google knows that some of its distributors of ads (Web sites) are falsifying clicks so that they can get paid more than they deserve and that this problem isn't that much of a problem since advertisers seem to be happy paying the inflated numbers. Thus, the logic goes, the clicks must have been undervalued to start with.

This certainly raises some questions in my mind, such as:

What about the advertisers? How do they feel about this? I suspect that they do not feel the same way. I suspect that most of the hundreds of thousands of Google advertisers, particularly the small "mom and pops," have no idea about the level of click fraud that exists and that if they all knew, they would not be happy paying for the fraudulent clicks. Many are turning to third parties like ClickFacts to help them recover dollars wasted by fraud.

Is Google right? Is click fraud self-correcting? Of course not. That's not how the business works. The only way that Google's argument holds water is if their advertisers have perfect information. This would only work if the advertisers could perfectly track each and every click to each and every site visit to each and every paid sale--and only if they could see all of Google's data across all of the sites and advertisers, and if they had extraordinary analytics tools, to be able to learn for themselves the validity of each and every click. They don't and can't, by the hundreds of thousands.

Is there any harm done? I think so, and not only to the advertisers that were the direct recipients of click fraud. What about all of the advertisers that bid against them? What about all of the advertisers that lost their top ranking to "click-frauded" competitors? It seems to me that this click fraud problem has a network effect just as powerful as the network effect in the auction component of Google's business model. If fraudulent clicks inflate the prices, it artificially inflates the auction and it impacts everyone. It makes more money for Google. It makes more money for the Web sites. But, unfortunately, it hurts the advertisers.

Is this much different than the newspaper circulation fraud that was so much in the news over the last year or two? In those cases, it was discovered that several newspaper distributors (bulk wholesalers) had fraudulently inflated their circulation numbers so that they would be paid more. And, of course, since these numbers were used in computing the respective newspapers' rate cards, advertisers ended up paying for the fraudulent circulations--fraudulent clicks, as it were. Following Google's logic, shouldn't this have also been a "self-correcting" situation? If advertisers paid the higher rates and it included the fraudulent circulation, they must have thought that the media value they received was worth the price that they paid. If not, they would have stopped buying it. Of course, newspaper ads are not nearly as measurable on an ROI basis as pay-per-click ads, but in the grand scheme of things, it's just a matter of degree.

Is this self-correcting? Whether or not Google is right, it doesn't matter. We cannot afford to wait. Michael Caruso at ClickFacts asserts that 14 percent is a very low estimate for click fraud, and he has clients where 30 or 40 percent is more the norm. The world, particularly the advertisers and financial markets, watch Google and Yahoo and other pay-per-click advertising providers with bated breath. They are our barometers. They are the basis upon which almost all of us are judged. We benefit from their successes. We are hurt with their failures. In this case, the trust of our industry is at stake. We can not wait to see if it self-corrects. We (Google and Yahoo) need to get out in front of this issue and solve it now. We (they) owe it to the advertisers in whose trust and pocketbooks our futures lie.

Source: MediaPost.com

Fear of Click Fraud Cuts Paid Search Spending

Fear of click fraud has caused more than one in four search marketers to cut paid search spending in the last year (according to a May survey of 407 advertisers conducted by Outsell). The report estimates that the cutbacks have cost search engines about 500 million in revenue. Furthermore, an additional 10% of advertisers replied that they plan to decrease their paid search spending in the future. Ouch!

Click fraud a huge problem / Study finds practice widespread; many cut back online ads

How much do you think internet advertisers paid for bogus clicks last year? According to a study by Outsell Inc., the number is $800 million ... representing 14.6% of all clicks! There is also mounting evidence that advertisers are finally beginning to change their spending habits in response to concerns over widespread fraud. Eventually this may force the Google, Yahoo and others to become more transparent regarding click fraud. If not, more advertiser money will continue to move away from PPC and towards other performance-based types of advertising such as pay-per-action.