OTT Is Growing Fast But So Is The Fraud
by Hunter Gebron, Director of Strategic Initiatives
Over-the-Top or simply OTT is one of the fastest-growing sectors of digital advertising and expected to reach $5 billion by 2020. Hulu alone reached 25 million subscribers in 2018 and saw ad revenue grow by 45%, and as we spend more and more time accessing streaming content through platforms like aforementioned, Roku, and YouTube we can expect those numbers to continue to climb.
How does this relate to fraud? Easy.
To provide a more seamless experience for the OTT end user (and as a way to circumvent ad blockers) advertisers are relying more and more on something called Server Side Ad Insertion or SSAI. And while it does offer additional benefits it is also extremely susceptible to fraud.
Fraud is perpetrated in OTT in a number of ways, one being that fraudsters are hiding behind whitelisted IP addresses where the proxy servers facilitate the SSAI.
What is SSAI?
SSAI, also known as, Dynamic Ad Insertion (DAI) or “ad stitching” is a technology that allows publishers to serve ads from the server-side instead of from the browser; it essentially combines the ad and the video into a single format providing for seamless delivery. Proxy servers in data centers do the “stitching” on the backend.
The benefits of SSAI are that since the ad is being “stitched” into the video within the Content Management System (CMS), the end user doesn’t experience buffer issues that arise from the ad and video being served from two separate servers. Additionally, ad blockers can’t stop the ad from playing since it is embedded into the video being provided by the publisher.
It is important to remember that fraudsters don’t just compete against fraud detection companies, they also compete against other fraudsters. So when a niche market starts booming, like OTT, fraudsters are quick to move into the new territory to secure an advantage.
The only way to detect fraud in OTT is by collecting multiple data touchpoints and triangulating them to spot discrepancies and inconsistencies. The fraudsters aren’t going to give away their secrets. The more rigorous we as an industry are in both the data collection and post impression analyses … the more difficult it is for them to get away with it.
The devil is in the details. That is why we pay meticulous attention to those pesky little things.
Machine learning and artificial intelligence have the potential to scale and bring additional value to our fight against fraud, but it can only get us halfway there. There is still no substitute for real human expert analyses — algorithms have to be written, machines must be trained, the data analyzed — otherwise, you are making business decisions from compromised data ….. And that’s not good for anyone.
OTT is expected to continue to grow and we can expect the fraud to keep pace. Don’t settle for a subpar pre-bid only solution. If you are going to beat the fraudsters you are going to need more than that.
Want to get in touch? Email us at firstname.lastname@example.org!