Inside the CTV Contextual Advertising Toolset
CTV contextual advertising, like many things in life, is all about making good decisions, and making informed decisions based on a wealth of data means leveraging the right tools—often AI-driven—to gather and distill and interpret that data. Sometimes developing sound contextual media plans involves working with in-house tech and other times it means working with third-party tools, as Team Whistle (a DAZN company) president Joe Caporoso and Intersection CEO Chris Grosso explain in this discussion with SVTA subject matter expert Bhavesh Upadhyaya in this clip from Streaming Media Connect 2026.
The Coordination of Multiple Teams
Upadhyaya begins the conversation with Caporoso. “What are some of the tools that you guys use to help … understand the live events that are coming through and understand … the audience that’s watching as well too for the right ads?” he asks.
Caporoso discusses Team Whistle’s approach over the past couple of years. It has a team that has built “some proprietary technology around the Looker data and the audience data that we ingest. They coordinate with our primary distribution team and leads in terms of how we look at that data. And then it’s sort of distributed in real time to the production team, the programming team, and the advertising team in terms of what we’re seeing on a daily basis,” he says. The company has built out these internal approaches to match “with how we’ve sort of adapted some of these different feeds and how that information’s coming in. We’ve also been doing some different preliminary research on certain companies like Screens that give us opportunities to move around some different things from the cloud in terms of how we position some side-by-side advertisements."
Caporoso provides the example of his company’s global deal with NFL GamePass. “We are figuring out different ways how we fill the spots over the local advertising, which we don’t run as we ingest that feed. We have to fill it with either our own content that’s customised or localised advertising that makes more sense,” he shares. “And a big part of what we’ve done around that—beyond using some of the traditional AI editing tools to make that more doable, like a WSC where we could churn that out at a higher volume—is also, again, just having a data team sort of attached to the tech and product team to review everything in real time. And we just pop that into our internal CRM and we treat it like sales data, we treat it like social media data, and we try to react to it as quickly as we can overall.”
Tech and Programming Working Together
It has been encouraging to see that the tech side of the business isn’t siloed, Caporoso notes. It’s done hand-in-hand with programming, which he says incorporates not only the on-platform programming team, but also the advertising and social teams. “A huge part of what we have to do with so many live events and streams going at once is just monitor and visualise all chatter going on around because the feed will go out periodically—things happen—or there’s little dials that we have to turn on the audio and ensuring that the ad is coming up and dropping at the exact right time on the free tier compared to the pay tier, compared to [for example,] what’s allowed in Germany compared to what’s allowed in Saudi Arabia,” he explains.
In summary, “it’s been a mix of internal proprietary tools, researching newer companies that are bringing very unique tech to it, like Screens. And then also just again, using all the editing data that we have internally and expanding our relationship with companies that allows us to make a high volume of content that we could put in the different ad breaks that we need to fill in so many different countries, because we’re not just running the standardised national advertising when we ingest the feed like NFL GamePass or any of our other global channel market partnerships.”
Two Ways AI Helps With Intersection’s Work
Upadhyaya asks Grosso to describe his perspective: “You’re totally out of home, so you’ve got a different tool set compared to the programmatic ads, et cetera, inside here. What tools do you guys use to help pull it through? And are you guys using generative AI or any other kind of artificial [intelligence], any type of machine learning, to get more information and insight into the data?”
Grosso notes that similar to Caporoso’s company, “one of our biggest areas is creative optimisation, and that hits in a couple different [ways]. The first is in the digital out-of-home space, there’s a large degree of fragmentation and many different screen sizes even within our network. So being able to use the generative AI tools just to resize creative, make sure the DPI is all right, make sure you get the right resolution,” is helpful. He discusses the example of a New York City ad campaign looking different from the creative side depending on whether it’s based in Soho or Flushing. “The ability to generate dozens of different types of creative around the same campaign in a way that you can never do with humans has been a big, big area for us,” Grosso says in praise of AI.
The second way optimisation is affecting his work is in media planning, he says, “because every individual screen has its own set of metadata around location, [showing] what types of audiences are around that screen. And it historically was pretty challenging to put together sophisticated media plans around that. With the AI tools, you can start doing that and start transacting that.” He continues, “We’ve made all of our inventories available programmatically, so you can go on The Trade Desk and buy our digital inventory just like you buy any other programmatic, so you can deal with that. But increasingly on the direct side, we need to start building much more sophisticated media plans.” Grosso notes that it’s been important to make sure that “the data is readily available and being able to be accessed digitally by third parties.”
Predictions for AI Use
Grosso predicts that in the next 6–12 months, people “in the out-of-home industry are going to increasingly start putting MCPs [model context protocols] around their order management systems so that other third parties who want to use an AI tool can kind of like plug in and do the same things that we’re trying to do internally.” He continues, “All the transactional pieces of the ad-sale business are going to move very much to AI-enabled platforms in the next couple of years. And really, what will end up happening is the salespeople [will] have to be hyper-focused on creativity and relationships, because the transactional back and forth, all that is going to be highly automated even on the direct side of the business.”
Related Articles
Matching ads to programming context and audience interests is theoretically what CTV and OTT can do better than traditional linear television, but it requires gathering data and leveraging it effectively to make sound contextual calls. Philo Head of Advertising Partnerships Aulden Kaye Yi and Roku Head of Supply Side Ad Platform Charlie Goodman discuss what works on their platforms and how their teams approach contextual advertising in this conversation with SVTA Subject Matter Expert Bhavesh Upadhyaya at Streaming Media Connect 2026.
10 Mar 2026