Digital ad fraud costs businesses billions annually. Indeed, every day tens of millions of micro-crimes are committed, and those crimes are aided and abetted by legitimate companies, many of them publically traded, with directors and executives who, at least under Australian law, could be considered willfully blind.
Imagine, for instance, if as much as 20 per cent of the money you spend on technology, solving the organisation's problems, was diverted by fraud to criminals or criminal syndicates. Picture yourself explaining that to your CEO.
No reasonable CIO would tolerate it, yet, marketers routinely accept such losses in their campaign spending due to the impact of digital ad fraud.
The problem is especially pronounced in Australia, where so much digital advertising is purchased programmatically. In 2019, for instance, global fraud intelligence company Pixelate identified Australia as having the third-highest rate of digital ad fraud in the world, a figure it arrived at after analysing billions of impressions across millions of domains.
According to Lizzy Foo Kune, VP and analyst at Gartner, "Programmatic is a complex system with lots of different places not only for inventory to exchange but for dollars to exchange, and there's not a lot of visibility, in terms of what's happening, where. That's problematic."
The issue is further complicated because there are intermediaries involved, she says. "You've got your agencies involved, you've got the planners themselves involved, and then all the technology components to every area as one thing passes to the next it just creates an opportunity to introduce chaos, and one might think that fraud is an element of chaos."
The complexity of the programmatic adtech ecosystem means that the only way to tackle the problem of ad fraud is with a cross-functional approach, says Foo Kune.
While marketing and IT are working closer together, Foo Kune says in order to successfully prevent and attack ad fraud, the approach must come from all sides.
“The organisations that are effectively tackling fraud, are bringing together identity and access management, they're bringing their infrastructure and cyber security teams to the table, alongside legal and compliance, customer experience, marketing, customer support, finance, and increasingly fraud analysts and data scientists that specialise in some of the more advanced techniques to identify and potentially mitigate fraud,” says Foo Kune.
The ways that brands purchase their advertising or marketing promotions has a direct effect on the type of fraud they may experience.
According to Shailin Dhar, CEO of Method Media Intelligence (MMI), while the ad fraud network is diverse, the majority of it “is based on non-human traffic, where you design and program automated browsers to actually click on particular links or advertisements.”
“Most of the programmatic ad space is based on a chain of servers or server connections. So you're very rarely actually measuring the device that an ad is being served to. And it's in those gaps of measurement where most the ad fraud occurs,” says Dhar.
New York based ad measurement business MMI seeks to separate robotic traffic from human traffic, for the purposes of analytics, advertising and monetisation.
“If you have a polluted data set, no matter how smart the algorithm, it will make bad decisions over time.”
With approximately 90 per cent of all automated traffic on the web being driven by data centres, rather than malware, not all bot data is malicious, says Dhar. But it is still widely disruptive.
“50 per cent of the web at this point is not human activity. And so it's not just affecting advertisers in the term in the sense of ad fraud. It's affecting every web based business, it messes with analytics, it pollutes datasets,” says Dhar.
“It doesn't have to be based on a cybercrime, to manufacture and monetise bot traffic.”
Dhar says, that there is a misconception around ad fraud that it is the work of malicious and sophisticated cyber criminals. She cites Reddit, and the Subreddit “Beermoney”, as a forum where civilians learn how to participate in the industry.
“Very much what a lot of ad fraud is based on is people that have day jobs, but have found that they can make an extra $1000, $2000, maybe $5000 a month, just doing small little tidbits of web activity.”
