Publishers, are you making the most of your traffic?


If you are a digital publisher and if most of your revenues come from selling ads, you might already be working with ad verification vendors such as DV, IAS, etc. But do you know, you might be losing a significant share of your revenues because of inaccurate IVT (Invalid Traffic) detection techniques of ad verification vendors?

Recently, at ShieldSquare, we studied web traffic of 300 publishers and made a few insightful discoveries. For example, most of the ad verification vendors consider all the traffic, coming from data centers as Non-human. We found out that many organizations use Secure Web Gateway (SWG) hosted on cloud data centers to access the Internet. Consequently, all the traffic coming from data centers is not Invalid.

47% Of Internet Traffic
Originating From
Data Centers Is Legitimate.

Surprisingly, we also discovered that a significant number of users coming through data centers were highly active on certain publishing sites. Hence, rather than rule-based detection techniques, Deep Behavioral Models that learn from closed-loop feedback systems are required to accurately detect Non-human traffic.

It is also widely known that ad verification vendors follow an opaque model for traffic verification. Their explanation for this opacity is to protect the detection methodologies from fraudsters, who can reverse-engineer the techniques.

At ShieldSquare, we propose that opacity is not the best approach to counter ad fraud. We advocate transparency. We recognize the increasing amount of non-human traffic (NHT), but we are against misclassifying the legitimate traffic as Invalid Traffic (IVT).

If you are interested to know more about, why should you ask for transparency from Ad verification vendors, read our report.