Every Click Should Count: Microsoft Discusses Online Ad Fraud at NYC Symposium

Posted by Richard Boscovich
Senior Attorney, Microsoft Digital Crimes Unit

What if one out of every four cups of coffee you purchased turned out to be an empty cup? You’d quickly switch to another coffee shop and you’d have some very negative reviews of the coffee seller who shorted you. Now let’s say that just one out of every 10 cups of coffee from your new coffee seller is empty – better, but you’d still want a new barista because you’re entitled to the cup of coffee you paid for.

Online ad clicks, unfortunately, are a lot like those empty coffee cups. A distressingly high percentage of the clicks that advertisers pay for are not made by human beings with genuine interest in that ad. Instead, they’re the result of automated Web traffic and other criminal activity that endangers the online marketplace for everyone who uses it - merchants, Web content providers and ultimately, consumers, who benefit from free, ad-supported online content.

Microsoft takes this problem very seriously, which is why we are partnering with the U.S. Secret Service to host the Online Advertising Fraud Symposium today in New York. The symposium is an opportunity for online advertisers, content providers, ad exchange platforms and law enforcement to join in a thoughtful discussion about the online ad fraud issue and how to re-vamp the current ad model. Without a substantive shift to inject greater integrity into the online advertising ecosystem, the entire business model by which a substantial portion of the Internet is funded may be at risk.

Spending on Web-based ads in the U.S. topped $26 billion in 2010, a new record for annual online ad revenues. As consumers continue to embrace mobile technology, that figure is likely to grow. With such high stakes, advertisers should take seriously the value of their online ad spend, yet the level of fraudulent ad clicks hovers around 25 percent, representing hundreds of millions in wasted ad dollars for advertisers. Ad fraud is so common that some have come to accept it as a cost of doing business – at Microsoft, we disagree.

The online advertising business model – industry wide – is highly vulnerable to exploitation. While reputable ad exchanges can and do monitor ad traffic to prevent fraud, the current business model continues to reward fraudsters at the expense of advertisers. The business model places little, if any, incentive for such exchanges to protect advertisers. The truth is that as long as advertisers don’t demand more than the status quo from the ad exchanges that place their ads, those ad exchanges have little incentive to do so on their own.

Malvertising (in which malicious computer code embedded in online ads can infect computers), click fraud (in which advertisers pay for clicks that are not the result of legitimate, interested human users) and domain squatting (in which cyber pirates profit off of URLs with slight variations on well-known legitimate brands) are all forms of abuse in which the ad exchange stands to receive illegitimate revenue at the expense of the advertiser and the consumer, creating an inherent conflict of interest in the business model.

While it may surprise some to hear Microsoft say that people aren’t demanding enough of ad exchanges, we fundamentally believe that the future of the Internet is at stake. As an ad exchange provider, we know that we have a choice to make. We can stand by, downplay the problem with advertiser refunds and apologies, and simply look the other way as fraud occurs, or we can work to do something about it. At Microsoft, we do not believe that looking the other way is an option.

Advertisers can and should demand more from their ad exchanges. They have the power, but they need to be better informed about the vulnerabilities in the current business model and what they can do to ensure that their interests are protected. Advertisers should start by asking their ad exchange providers some basic questions about how their ads and ad dollars are being used, including:

· What mechanisms do you have in place to ensure that my ads are viewed by actual consumers?

· What is your definition of a valid click, and do you differentiate between invalid and low quality clicks?

· Do you discount for low quality ad traffic?

· Can I know and control where my ads appear to most effectively promote and protect my brand?

· What action do you take against publishers or individuals who are found to be generating invalid/fraudulent traffic? Can I expect repeat behavior?

Our Traffic Quality team in Microsoft Advertising and the Microsoft Digital Crimes Unit works hard to make sure every ad click counts – through stringent standards, technical tools, investigations, careful monitoring and by going after the criminals who try to exploit this model in court when necessary. While we will probably never completely wipe out online crime like ad fraud, we will continue to diligently invest in innovation and action to protect our platform, our advertisers and our customers and set a high bar for integrity in the online advertising platform.

However, no single company can solve this problem. It must be an industry-wide effort if we are to set a new standard, and it starts with advertisers demanding that they get what they pay for.