Tyler Loechner, Marketing Manager at Pixalate, sat down with RhythmOne’s Karim Rayes, Senior Vice President, Products and Business Development, to discuss the unique challenges of battling invalid traffic (IVT) within mobile apps.
Tyler Loechner, Pixalate: What does RhythmOne consider to be the biggest challenge when it comes to combating IVT within mobile apps?
Karim Rayes, RhythmOne: The greatest challenge is that today’s fraudsters are using highly sophisticated, ever-evolving tools to replicate in-app consumer behavior. This includes app installs, action-based like clicks, even attribution. There isn’t a fixed solution to unilaterally identify fraud particularly since new schemes are constantly being introduced, but rather a dynamic set of tools that can screen and filter potentially fraudulent traffic.
Loechner: What is RhythmOne doing at the organizational level to reduce IVT within mobile apps?
Rayes: We’ve built a bank of proprietary data filters to recognize current and emerging fraud types, as well as associated ad request variables, to better identify IVT. RhythmGuard, RhythmOne’s brand-safety technology that is built into our platform, RhythmMax, screens all inventory and eliminates suspicious or underperforming traffic pre-bid. It does this by leveraging supply-side block lists and traffic scoring algorithms built from first- and third-party verification data. But prevention is only one part of the solution. We also monitor traffic post-bid, providing creative and domain-level verification.
Loechner: With consumers spending so much time in-app, what do you say to advertisers eager to take advantage of the audiences but nervous about ad quality?
Rayes: I would encourage them to ensure that their programmatic partner is fully committed to brand safety and has been recognized for quality and integrity across channels, platforms, and devices. RhythmOne was thrilled to place first in Pixalate’s Q2 2018 Programmatic Quality Report for the Global Seller Trust Index, and the second spot in their Mobile Seller Trust Index.
Loechner: How do you define quality in mobile apps?
Rayes: Traffic quality is an empirical metric that doesn’t vary from device to device, nor does it vary from one ad format to another. It’s a real human who the advertiser intended to deliver an ad to, and quality is based on what this individual saw, clicked on, or otherwise took action, based on the serving of an ad. What does change, however, is how different variables and metrics can be used to identify a potentially fraudulent source, request, or session.
On mobile devices, we can assume one user per device, which you can’t do on a desktop. Essentially, mobile devices allow a form of “user fingerprinting” which generalizes behavioral trends. When that trend is not followed, the risk of IVT goes up. IVT for mobile apps acts a bit differently, because once it gets inside the app, it has a longer lifespan than mobile web. Mobile apps also tend to cluster by geographic location, so to identify IVT, we might flag activity that falls outside of these location clusters. Clicks to install and post-install behavior are also normalized, so activity that follows a different pattern, for example, short session lengths, might serve as a flag to identify potentially fraudulent behavior. We identify new fraud types constantly, based on behavioral aberrations, or differences.
Loechner: What are some of your best tips for maintaining a low level of IVT within mobile apps?
Rayes: The best tip I can offer is that there is no single answer or fix to maintaining low IVT in mobile apps. It is a multi-pronged strategy that begins with the initial app vetting phase, working with a partner, like Pixalate, who can apply scientific analysis to discover and avoid apps that have a higher preponderance of suspicious activity, so they never become part of your media buy in the first place. Second, it is ongoing behavioral monitoring in-app to identify non-human or suspicious activity. And lastly, it’s ensuring that you are conducting post-bid analysis and using accredited verification tools.
Loechner: What do you think the industry needs to know about mobile app ad fraud?
Rayes: Marketers have invested heavily in mobile advertising, specifically in-app. But AppsFlyer recently reported that a growing number of marketers believe that a large percentage of their mobile ad budgets are exposed to in-app fraud. We have to make reducing in-app fraud a top priority, across the industry, if marketers are going to continue to trust us with their ad dollars.
Loechner: Is there anything the industry at large isn’t focused on that you think needs to be a bigger priority when it comes to fighting ad fraud in mobile apps?
Rayes: The industry has done a good job in placing greater emphasis on reducing IVT across the board. Increasing the focus on preventative pre-bid efforts to stop fraud before it ever enters a campaign is the single most important area for us to focus on. Collectively, it’s important that the industry continues to embrace standardization, like the Open Measurement SDK (OMSDK), which will enable advertisers to view IVT and brand safety rankings from an objective standpoint. We need industry standards for IVT so that there is an apples-to-apples comparison that advertisers can make from campaign to campaign, particularly if they are working with multiple vendors.
Loechner: With millions of mobile apps available in the major app stores, how do you determine which apps are good partners? How do you ensure your list of partners stays ‘clean’?
Rayes: There is a certain level of quality control that must happen to vet inventory supply sources and new apps for consideration as part of an onboarding process. So it’s not just a matter of vetting an app’s description in the store, but reviewing the in-app content to ensure it adheres to our content guidelines, as well as verifying the owner or publisher of the app is a legitimate and verifiable source.
There is an obvious and direct correlation between quality apps — which I would define as apps that are well managed, maintained, regularly refreshed and updated by their publishers — and the performance of ads that are leveraging their inventory. There should be an incentive for well-run apps, and that incentive is increased demand for their inventory on the basis of prior performance. It may sound like utopia, but it is also the reality of our experience with the publishers we work with at RhythmOne.
This Q&A on ad fraud within mobile apps also appears on RhythmOne's blog.
Karim Rayes, Senior Vice President, Products and Business Development at RhythmOne, leads the company’s publisher development and product management organizations. Karim joined RhythmOne in 2007 and prior to his current post served as Vice President, Publisher Development where he was managing RhythmOne’s supply side platform, responsible for product direction and commercial relationships. With more than a dozen years of experience in digital media, Karim is an expert in publishing, programmatic advertising, media buying, and business development & affiliate marketing and management. He holds a B.A.Sc. in Computer Science from the University of Montreal.
Disclaimer: The content of this page reflects Pixalate’s opinions with respect to the factors that Pixalate believes can be useful to the digital media industry. Any proprietary data shared is grounded in Pixalate’s proprietary technology and analytics, which Pixalate is continuously evaluating and updating. Any references to outside sources should not be construed as endorsements. Pixalate’s opinions are just that - opinion, not facts or guarantees.
Per the MRC, “'Fraud' is not intended to represent fraud as defined in various laws, statutes and ordinances or as conventionally used in U.S. Court or other legal proceedings, but rather a custom definition strictly for advertising measurement purposes. Also per the MRC, “‘Invalid Traffic’ is defined generally as traffic that does not meet certain ad serving quality or completeness criteria, or otherwise does not represent legitimate ad traffic that should be included in measurement counts. Among the reasons why ad traffic may be deemed invalid is it is a result of non-human traffic (spiders, bots, etc.), or activity designed to produce fraudulent traffic.”