January 10, 2019

Types of Ad Fraud (and What You Can Do About Them)



Amanda Bleich

Product Marketing Manager

Ad fraud is a big, costly problem in our industry. To fight it, you have to understand the different forms it can take. The brand safety experts at Peer39 typically divide ad fraud into three categories: non-human traffic (i.e., bots); ads with zero chance of being seen (i.e., zero-percent viewability); and intentional misrepresentation. The imposters who are responsible for these kinds of fraud are savvy, and they are continually finding new and more sophisticated ways to make money by defrauding advertisers.


Here’s a closer look at some of the most common types of fraud:

Bot basics

General invalid traffic (GIVT) are scripts that run from a server such as Amazon Web Services or some other hosting provider. As their name implies, these bots are usually easy to identify because they have a static IP, user agent, and cookie ID. This makes fingerprinting them pretty easy using DSP auction logs or even web server logs to spot abnormally high clickthrough rate (CTR) or unexpected spikes in traffic that are the signatures of simple bots.

Sophisticated invalid traffic (SIVT) is not as easy to identify. These bots rotate user agents, using random proxies to rotate IP addresses, and they mimic normal “human” CTRs, so they are more challenging to detect. They are also now capable of completing complicated tasks like filling out forms or completing videos. Sophisticated bots can even put items in shopping carts and visit multiple sites to generate histories and cookies—making them look attractive to advertisers and publishers.

The unviewable

Ad stacking is a common way that fraudulent publishers get credit for running an ad that is actually hidden behind other ads and not viewable. The publisher can thereby generate multiple impressions for a single page view, even when only the top ad in the “stack” is ever seen.

Site scams

Domain spoofing is a scheme employed by deceitful publishers, ad exchanges, or networks to obscure the nature of their traffic to resemble legitimate websites. For example, an advertiser might sign off on a contract to run a campaign on a legitimate entertainment website with very high monthly traffic, but instead its ads end up on an unknown site. This practice is most prevalent in the programmatic space where publishers are sometimes allowed to declare their own domains and label their own site IDs. Spoofed domains are not just fake website addresses, they are also banner farms that contain bad content.

Ghost sites are among the most difficult fraud methods for advertisers to spot. Fraudsters create content farms and use bots to mimic human traffic. The sites may then be introduced to a legitimate ad exchange, where ad impressions are made available for advertisers to buy programmatically. Exchanges usually spot these schemes quickly, but even a short lifespan can be profitable to the ghost site creators.

Zero-ad sites are those where advertising is forbidden, such as government or educational sites. But fraudsters still find ways to inject ads into them when a user downloads and installs a browser extension or app (such as a free PDF converter or browser toolbar) bundled with software that quietly injects unwanted ads into the user’s browser.

Fraud is lucrative

The scale of online ad fraud has a significant impact on advertising ROI and advertiser confidence because all those falsified impressions and clicks cost money without yielding conversions or revenue. It’s estimated that fraud consumes $1 of every $3 spent on digital advertising. In 2018, and advertisers lost an estimated $51 million every day to fraud, a figure that is expected to more than double by 2022. Time and time again, advertisers unwittingly reinvest in fraudulent inventory because it appears on reports to be driving results. Worst of all, ad fraud is not technically illegal, so there is minimal risk for bad actors.

Protection is possible

Because fraud schemes continually evolve, effective fraud prevention requires staying one step ahead of their game. Peer39 does this by tackling the problem from every angle, both before and after the buy. Peer39 pre-bid antifraud targeting helps marketers exclude fraud from the buy up front, eliminating zero-ad sites and other fake inventory. Peer39 post-buy solutions offer multichannel, AI-driven fraud detection and filtration that enables you to monitor viewability and detect bots and other invalid traffic threats—even difficult-to-detect schemes and domain spoofing.


Fraud isn’t going anywhere, but with vigilance, you can significantly reduce your exposure and protect your investment.

Contact a your account rep or a Peer39 account manager to keep your next campaign fraud-free.