Q&A with Shoaib Mansur, Head of Data Science @ ElementalTV

If I had a $ for every time I’ve written or talked about AI in the past 6 months, I’d probably be retiring early for a low-profile life in the Bahamas. 

With all the hysteria around it though, it’s important to cut through the noise and break down how AI/ML and advanced data science techniques come into play on the CTV side of things.

We sat down with Shoaib, Head of Data Science and Monetization @ ElementalTV to discuss how he and his team work their AI/ML magic on the daily to make sure every CTV campaign lands, every impression is worth the $, and every ad, relevant.

Q. Tell us a little about your role at ElementalTV? How do you and your team work your magic on the daily?

I lead the Data Science team at ElementalTV and we like to believe that there would be no ElementalTV or VISTA without me and my data comrades. JK. 
As the lead data guy at ElementalTV, I work around the clock (mostly) to execute data-powered strategies for furthering our business objectives. On a tactical level, this looks a lot like distilling valuable insights from data, constructing predictive models and integrating them into our wide range of CTV solutions. 
Your value in the CTV industry today lies in the robustness of your data and the capabilities of your AI/ML models. Going by that standard, I and my team try to make sure ElementalTV stays somewhere at the top of the CTV food chain.

Q. ElementalTV recently launched VISTA - an audience activation platform. Can you help us understand what the VISTA advantage is and how you’d pit it against other similar platforms?

VISTA might have launched only a few weeks ago, but our platform and tech have been continuously evolving and expanding for the past 4 years. From being an OTT/CTV ad server in 2019 knee deep into programmatic, to niching down to advanced CTV yield management for publishing giants, ElementalTV has really seen the streaming tv space metamorphose into what it is today first-hand.

As a company we have always heavily leaned on data, automation and a smart algorithmic approach to solving problems. Continuous learning and innovation are at the heart of everything that we do, which allows us to be at the forefront of industry trends. And the great thing about working with people at ElementalTV is that even if some of us are not hardcore data nerds, all our decisions are grounded in data so we don’t really give anyone a choice!

Our data obsession led us to create the powerhouse that VISTA is. Being in the ad tech space for so long helped us identify actual challenges and gaps. One pressing issue was that of unfilled campaigns and unrealized revenue opportunities due to a lack of deep audience insights and a mechanism to harness all that data. VISTA addresses this by amplifying the frequency of ad campaigns as well as through increasing reach via lookalike modeling. 

Q. Would you like to shed some more light on the concept of lookalike modeling?

The CTV industry is growing at a skyrocketing pace and precision targeting has become the linchpin for campaign success. Lookalike modeling is a sophisticated targeting technique that harnesses the power of AI/ML to identify and engage audiences mirroring the characteristics of high-value customers. It goes beyond traditional demographic targeting, unlocking new dimensions of audience expansion and engagement.
By scrutinizing data facets such as behavioral patterns, preferences, intent, semantics etc advertisers can gain profound insights into the traits that define their ideal audience. Armed with this knowledge, machine learning algorithms work in tandem to discern intricate patterns within a broader audience pool.

Q. There’s a lot of conversation around contextual targeting and AI’s foray into making it even more intelligent and relevant. Can you highlight how VISTA utilizes the latest in targeting tech?

The VISTA revolution has only just started. There’s a mammoth revenue opportunity here for CTV publishers, provided they utilize the vast amount of untapped audience data. 
Our team is hard at work capturing contextual and other relevant signals from existing internal data streams, including programmatic pipelines, as well as externally from across the web. Owing to our extensive experience in the industry, we have in-depth domain knowledge to guide the process better than most other players in the space. Our in-house AI/ML models are continuously fed with new signals to bring in any possible improvements, a process that helps us uncover new insights and build a smarter curation technique that gives our publisher partners an edge when they are approaching advertisers. In a world where AI is making frame by frame content analysis possible, VISTA makes sure you’re always ahead of the CTV curve. 

Q. What’s a challenge you commonly notice with other CTV platforms claiming to provide premium inventory pools and state of the art tech for audience activation?

While there’s nothing wrong with jumping on the promising CTV opportunity, as a technology partner though there’s quite a few problems that may arise if you don’t have the proprietary tech to scale demand. With ElementalTV’s vast network of key publishing partners and advanced ML models we always have our finger on the pulse of the latest innovations in the space. Our dedicated data and engineering teams are diligently working to continuously improve and expand our ad monetization solutions unique to individual use cases. 

Q. Do you have any final thoughts you’d like to share?

Yes! There’s so much more to discover AND learn about AI/ML x CTV and we’re beyond excited to be one of the early adopters. If you want to discover VISTA in greater detail and how it’s transforming the landscape for CTV publishers, feel free to hit us up @ https://www.elementaltv.com/contact-us.