Audience measurement is an increasingly important topic for brands that advertise online. In fact, most of the dollars spent on digital ad buys today are directly tied to the size of audiences and the impressions seen by digital users. But accurately measuring these audiences and impressions has only gotten more difficult due, in large part, to the amount of invalid, non-human traffic that drives up the numbers of impressions served in digital advertising.
Non-human traffic refers to online page views and clicks generated by “robotic,” (rather than human) website visitors that creates inaccurate, inflated measurements. Extreme Reach is a key partner to advertisers in this fight against non-human ad fraud. How do we detect this non-human traffic? And how can we help ensure advertisers get what they pay for?
Detecting non-human traffic is a multi-stage process, requiring a combination of internal tools and external standards defined by outside partners. As a starting point for detecting non-human traffic, the industry looks to guidelines established by The Interactive Advertising Bureau (IAB) and the Media Rating Council (MRC). You can read more about these guidelines here.
In addition to using the industry guidelines, Extreme Reach relies on its own proprietary measurements to monitor for non-human page visitors. Revealing too many details of a fight against invalid traffic only helps educate the “bad actors,” but there are a couple of things we can share about our measurements. We pay a lot of attention to the timing and frequency of ads that are viewed. The number of times a website visitor returns to a site can reveal whether a human is actually viewing the page. Similarly, the “patternistic” behaviors of site visits also provide important clues. Return visits to a site at the exact same time interval over a specified amount of time are unlikely the habits of a human visitor.
Staying on top of the constantly evolving tactics used by advertising “bots” generating non-human traffic is a real challenge. But considering the potential money at risk for advertisers who end up paying for such fraudulent ad impressions, the stakes are high. One ad fraud detection firm that spoke to the Wall Street Journal estimated that close to $6 billion dollars was lost to non-human ad fraud in 2014. With this kind of money on the line, it’s critical that the industry continues improving its intelligence for the ongoing fight against such non-human traffic.