Late last month, MediaPost reported on an ad-fraud scheme, called out in a Pixalate blog, in which thieves exploited mobile app advertising security loopholes by using consumers’ devices as proxies.
That scheme — which became inactive as of mid January, but has served to spotlight the growing fraud problem — used code to disguise standard banner ads purchased on Grindr’s Android app to look like video ad slots on Roku CTV devices.
The spoof ads were then sold on programmatic exchanges. The objective: Make money by passing off a low-CPM mobile display unit as a much higher-CPM unit.
Grinder said it is working to address its problem.
Roku stressed that the company recommends that OTT ad buyers buy directly from Roku or publishers on its platform, and if buying from other sources (open exchanges in particular), use technology to verify the source of ad requests.
"When buying directly from Roku, we can assure buyers that their media plan is brand-safe and 100% viewable," stated ad industry PR manager Sarah Saul. "We certify streaming channels on our platform. Furthermore, we developed our operating system and own the ad tech stack and the first-party data that ensures that we offer a best-in-class ad experience and meet industry best practices."
Like Pixalate, DoubleVerify, another leading ad-fraud detection/prevention company, says it identified and began protecting customers from this particular scam starting back in first-half 2019.
“It’s important to recognize that this scheme, while sensationalized because it was centered on the Grindr app, was agnostic to the mobile app it was abusing,” stresses Roy Rosenfeld, head of DoubleVerify’s Fraud Lab.
“It was simply targeting cheap mobile app inventory that it could represent as more lucrative CTV inventory — a relatively common tactic in the CTV ecosystem. DoubleVerify’s technology can, and does, accurately identify such traffic as mobile app traffic on Android devices, and not CTV traffic.”
Rosenfeld agreed to share some more context and advice on CTV ad fraud, via a Q&A:
How widespread is OTT/CTV ad fraud now?
In the past 12 months, DoubleVerify tracked a 120% increase in fraudulent CTV and mobile apps. Further, we detect over 500,000 fraudulent CTV devices per day.
With eMarketer estimating that almost 60% of CTV inventory will be bought programmatically by 2021, the risk of fraud will continue to rise.
The more common varieties of fraud we’re seeing include fraudulent apps and SSAI bots. On CTV, fraudulent apps will manipulate the environment by creating automated, completely fabricated ad calls coming from non-existent devices: playing ads back-to-back and spoofing the “app name” parameter to appear as if they are CTV ads.
In turn, SSAI bots exploit the fact that much of the data coming from SSAI servers is self-declared, making it easier to spoof the user agent, client IP and/or bundleID.
What's behind the increase in ad fraud?
We expect CTV ad fraud to increase in volume and sophistication into 2020, fueled by premium CPMs, more advertisers and more inventory.
The appeal of CTV to advertisers is no surprise when one considers the fast growth of CTV audiences, the measurability of digital video, and the appeal of sound and motion on the living room’s biggest screen.
In fact, according to Tru Optik, U.S. CTV ad spend is projected to reach $20.1 billion this year — a year-over-year growth rate of over 50%.
Unfortunately, fraud follows ad spend, especially within emerging channels, where standards have yet to be established and demand outstrips supply.
Fraud in CTV is also no more difficult, and in some cases easier, than in other environments. The lack of standards mixed with an inaccurate perception that it’s a fraud-free environment equates to an easy target for fraudsters. They don’t even need a CTV device or app to pull this off.
For years we’ve detected and blocked the purchasing of mobile display ads from respectable publishers that are then resold as premium CTV video inventory. The media buyers believe they’re purchasing premium CTV video inventory. Publishers believe they’ve sold their standard display ads. Users see a random display ad, while all these CTV video calls go off in the background. None of these parties is aware of what’s going on, and it’s usually not easy to detect.
Third-party verification that’s dedicated to finding and blocking this fraud is really the only way to tackle these issues. As long as advertisers accept the myth (sometimes pitch) that CTV/OTT is fraud-free, and don’t use verification partners, fraud in this area will only continue to increase.
What can media buyers, marketers and CTV/OTT platforms do to protect themselves?
Advertisers need to be able to detect and prevent fraudulent activity, since it obviously can have an adverse impact on the success and ROI of their campaigns.
We encourage CTV buyers to take several steps to protect themselves:
Start measuring CTV inventory. Quality measurement starts with transparency. Advertisers must demand that transparency by insisting that supply partners become certified for measurement, and that their verification providers are able to scale across this medium.
Push industry groups to advance standards. CTV app identification standards and development of a shared technical solution for viewability measurement are needed to move CTV inventory into a more standardized framework.
The IAB’s OM SDK (Open Measurement Software Development Kit) working group has been tasked with establishing these standards to ensure consistent measurement. It’s critical for supply partners to adopt the latest version of the OM SDK, and for the SDK to support CTV environments.
Enlist brands in advocacy efforts. Brands must be educated about the need for independent measurement. They should insist on CTV targeting from their DSPs and buying platforms in order to enhance the delivery of quality CTV impressions in their programmatic buys.
Advocate with major CTV suppliers. CTV suppliers must also be educated about the need for robust third-party measurement to create a healthy and trusted ecosystem. We, as an industry, need to make sure a “walled garden” scenario, where individual providers are the only ones with access to quality data, does not arise within CTV.