WHAT IS AD FRAUD AND ITS IMPACT?
Ad fraud happens when online ad impressions are fraudulently represented in order to generate revenue.
The risk of digital ad fraud is estimated to be $22.4 billion globally. In the United States, that estimate is $2.5 billion. The biggest marketer of ad fraud is China at $18.7 billion. In the OTT space, analysts estimated in March 2019 that ad fraud comprised 18% of open exchange inventory. However, by December 2019 that estimate had increased to 20%. Based on this percentage, the estimated waste for OTT ads in 2019 is $1.4 billion; this is an issue that is not going away.
The digital space has been dealing with ad fraud for years. OTT is still a relatively young medium in the history of modern media formats, but its growth has catapulted over the past five years, and with that comes increased chances for fraud. Advertisers and ad agencies need to take note and make plans.
THE GROWTH OF OTT (AND CTV)
During OTT’s infancy, the space was considered expensive compared to linear TV. This was due to scarcity in the amount of inventory. In the last two years, however, the industry and inventory has exploded, especially with the rise of CTV*. From the days of just Netflix and Hulu, we now have even more major players such as Amazon Prime, Peacock, Disney +, Apple TV+, YouTube TV, Tubi, Crackle, Xumo, Pluto, and more. Many of these are ad-free, but with the growing list of ad-supported services (Peacock is the latest entry), agencies and buyers need to be vigilant. With increased inventory, comes more opportunity for bad actors to take advantage of unwitting advertisers via fraud.
Here are the three most common types of OTT ad fraud to watch for:
- Misrepresentation: This type of fraud can occur across a variety of devices and environments. In some scenarios, sellers are selling ad impressions they claim are shown to U.S.-based IP addresses, but ads are instead shown on devices outside of the United States. Fraudsters do this to inflate impressions. In other cases, Supply Side Platforms (SSPs) can manipulate ad buys even if buyers think they are buying premium publishers. SSPs and ad networks will buy an entire pod, cut it into small ad slots and resell (and give themselves a free ad slot). Similar tactics are also used to misrepresent app inventory as premium when it actually isn’t.
- Device-based fraud: In these cases, a single device might show a very high number of ads during a given period (50 ads per minute to one device is not possible). This type of fraud is commonly referred to as looping. But there is a caveat here: the nature of how OTT livestreaming works (multiple ads are fetched for multiple users simultaneously) can render false positives for looping. While livestreaming doesn’t represent all OTT viewing, it does make up a considerable amount. To correct, fraud detectors need to filter this activity out by comparing the ads that were fetched with the ads that were ultimately viewed.
- App-based fraud: As the name implies, this type of fraud occurs solely within app environments. One example is when an OTT-supported app shows an unrealistically high rate of ad activity at all hours, without the natural ebbs and flows you would expect to see with a human viewer (versus a bot).
While the range of potential ad fraud issues may seem daunting, agencies and advertisers can easily overcome these threats by creating rules and frameworks for how to select, buy, and verify OTT inventory.
RULES, RULES, RULES…
It took years to put together the tools and frameworks needed to combat digital ad fraud, but even now, with the ever-changing technical world, there is room for improvement. Advertisers who don’t continually address fraud head-on can face risks, including financial (spending money on fraudulent inventory) and reputational (being associated with questionable sites or platforms). These concerns factor into overall ad fraud, and have multiplied across OTT as delivery and inventory options continue to expand. Now, advertisers are facing a much larger number of factors to consider when combating fraud. The good news? With the right approaches, technology, and verification practices, these risk are fairly easy to overcome, even if the threats are ever-changing.
While advertiser and agency approaches sometimes differ, KSM’s basic rules of the road to combat OTT ad fraud include:
- Implement formal anti-fraud practices: KSM is currently enrolled in the Verified by TAG (Trustworthy Accountability Group) program and also achieved the TAG CAF (Certified Against Fraud) seal. In our effort to uphold TAG’s mission to eliminate fraudulent traffic, combat malware, prevent Internet piracy, and promote greater transparency in digital advertising, we require all partners to either confirm that they are currently Verified by TAG and TAG CAF, or if they aren’t, detail how they ensure 100% of all monetizable actions are filtered for GIVT (General Invalid Traffic), Domain, Data Center IP and App threats.
- Utilize in-house verification tools to self-audit: Upon campaign activation, KSM prioritizes quality over key performance indicators, using MOAT to measure and protect against fraudulent activity. And beyond ensuring ads are actually being delivered to people instead of bots, MOAT also helps confirm whether or not an ad is viewable. After all, if an impression is not seen, what value does it have? KSM’s viewability rates are above industry standard, because we make ongoing optimizations to campaigns by balancing AI technology with manual optimizations to ensure that we are achieving the best performance possible.
- Consider the content delivery model: OTT is content that can be delivered across a range of devices, including mobile, desktop, and CTV. While pre-bid features alone on desktop and mobile will prevent fraud for any OTT running on those devices, CTV requires some additional considerations. For CTV, advertisers must leverage specific pre-bid technology to block fraud (at this time, Double Verify is the only provider that has the technology built out on their end to get ahead of fraud/IVT within the CTV space), run only on PMPs (private exchange) never open exchange, and have deals with publishers directly and not SSPs.
- When dealing with CTV, buy directly with trusted partners: When KSM integrates CTV into a campaign, we always buy directly from trusted partners, and for programmatic we only buy on private exchanges. This ensures we can individually audit partners to be selective about the ones we work with, and maintain the most control over the types of inventory that clients are running on.
KSM employs these anti-fraud practices (and more) via our High Caliber Programmatic Network, designed to connect brands with the most trusted content providers and high-quality audiences in the programmatic space. And when not buying programmatically, we implement these same standards in our vetting, investment, and auditing process. To that end, these efforts drive a greater demand for quality and premium content overall. If this type of quality threshold is emphasized across the digital universe (including OTT), it will go a long way toward combating the overall ad fraud threat.
ON THE HORIZON
A mix of transparency, selective partnerships, and verification technology are really the keys toward largely extinguishing current ad fraud threats across the OTT universe. And when it comes to transparency and verification, Blockchain is showing signs of promise as a potential resource down the line.
By using blockchain and cryptography, demand- and supply-side players can achieve proof-of-concept projects to fight against the common types of OTT ad fraud. Several blockchain tests have indicated that it can help with ad fraud on websites and mobile apps, but more testing still needs to be done in the OTT space. With common issues that overlap between OTT and the digital space, there needs to be a common solution. Blockchain testing appears to be one good option, and non-profit industry groups like AdLedger are continuing to explore the possibilities.
But even with some promising tools being added to the arsenal of anti-fraud weapons, the ebbs and flows of threats will continue to keep marketers and the industry at large on its toes. For instance, post-COVID data is showing just how much of a potential fraud threat may be lurking on the horizon for OTT. As stay-at-home orders went into place, OTT (inclusive of CTV, which are devices on which OTT streaming video is served) experienced extreme increases. Early March pre-COVID numbers had CTV households spending 2.7 billion hours watching TV. CTV households rose by 48% to nearly 4 billion hours by early April. These CTV increases have boosted OTT consumption by nearly 40%. With increased consumption, however, comes an increased risk of ad fraud. In Q1’20, DoubleVerify claims a 161% increase in CTV fraud versus Q1’19.
As OTT and CTV usage increase, fraudulent activity will follow. But the rules are still the same: marketers need to remain vigilant, and transparency is key. Like the pre-COVID world, marketers must embrace transparency. Analysis, scenario management, and technical research will be the perfect combination of AI and human expertise. By investing in technology, applying human expertise, and committing to transparency, marketers can continue to fight fraud in this COVID world.
FINAL THOUGHTS
What is the definition of fraud? According to the Oxford English Dictionary, it is “wrongful or criminal deception intended to result in financial or personal gain.” Words like “wrongful” and “deception” aligned with “financial gain” will not help build strong partnerships. As marketers, our goal is to drive business growth through the power of media.
With technology ever evolving, marketers need to evolve just as fast. Publishers, buyers, and advertisers have to continue to push as we find the correct path to combat ad fraud. The consistent messages to keep in mind are quality and transparency. By choosing the best partners, and exchanging the right information, ad fraud in the OTT space can be overcome.