Fraud is a plague spreading through the online advertising industry that is about as difficult to fight as online piracy – you can shut down one site and another will pop up in its place within seconds. Fraud is a pervasive problem because there is money to be made in serving ads to eyeballs, real or fake. Advertisers are willing to spend a pretty penny on eyeballs, and publishers are motivated to deliver these views in order to turn a profit.
Scammers take advantage of this cycle by delivering fake traffic through increasingly complex methods. These bad actors have done an excellent job at masking their traffic as genuine and often function within the legitimate ecosystem undetected.
It’s easy to blame publishers for buying false impressions, but there is no sense in pointing fingers. Most of the time publishers don’t even know that they are serving up ads to fraudulent traffic. They often work with several partners to help optimize traffic and can be unaware of a leak that is letting in bot traffic.
It’s time for us to stop playing the blame game and instead, come together as an industry to develop solutions to the fraud problem. This of course is easier said than done, because there is no simple technology fix for the issue; rather, it will require completely rethinking how we price inventory and measure success. Here are some ways we can get started on the path to a fraud-free digital marketing ecosystem.
Only pay for real conversions – If we readjust our inventory pricing scheme so that advertisers only pay when a consumer has actually bought something or has downloaded media, we can take control back from the fraudsters. It may sound radical, and indeed it would require a complete overhaul of media buying and selling practices across the industry, but on the other hand, it would not require any major technological adjustments. Media transaction platforms could still function in much the same way, but post-conversion payment would ensure that there is a real person at the other end of an ad, not a bot, and that advertisers could be certain that the results they see represent real consumer engagement.
Use data analysis to police fraud – Data is not only our most powerful resource for targeting and engagement; it can also be helpful in identifying fraud by distinguishing non-human patterns on both the front and back ends of a media transaction. Advertisers should employ sophisticated data analysis to search for red flags; for example, 100,000 impressions that cost $5 indicate that traffic is likely to be not only low quality, but probably bot-driven as well.
Rethink how we measure ad effectiveness – Measurement and benchmarking, as important as they are to the digital advertising ecosystem, are really the main culprits behind the fraud epidemic. Rather, fraud has become something that we resign ourselves to because we rely on an antiquated system of measurement that does not distinguish real impressions from false ones.
It’s not only foolish to measure online ads in the same way that we measured television, especially considering all of the data that we can access through digital channels, it also traps us in a system that rewards fraudsters, rewards publishers that sell fraudulent inventory (knowingly or not,) and even rewards marketers who include those false impressions in their results. That’s why the fraud issue is so insidious – in the end, everyone’s balance sheets show good results so it looks like there are no problems when, in fact, many of those impressions never translate to revenue so that money is still being wasted.
Adopting an entirely new approach to measuring ad effectiveness is probably the most important step we can take as an industry in the fight against fraud. In the age of Big Data, there has got to be a better guiding metric for success than the click, one that we can also use to prove that there is a human really viewing the ad? Why are we using metrics designed for television in online media?
The fraud issue won’t be solved overnight, and these are just a few of the solutions we should examine. There are lots of ways to get around the fraud problem, but it requires new kinds of thinking and new approaches to doing business. We are a creative, forward-thinking industry, so there is no reason we should not be able to reinvent our business models, even if it means a little disruption to the status quo.