Connected TV may be the hottest thing in media right now, Horizon Media EVP/Chief Investment Officer Dave Campanelli explains both why this is and why linear TV is not dead in this interview, with WARC US Commissioning Editor Cathy Taylor.

Connected TV

This article is part of a series of articles from the WARC Guide to connected TV.

Connected TV (CTV) is in a really buzzworthy moment right now. How well does that match to the reality of how important this is?

It's immensely important and it's here to stay. It will replace linear TV as the primary way people consume their premium video in the relatively near future. Linear, is still larger in totality, but certainly for younger audiences, it's not anymore. CTV is the way that we're going to get our premium video. It’s a very good mix of some of the best things about linear, some of the best about digital. It has the targeting capability of digital, but it has the ability to deliver to a TV screen, which is people's preferred way to consume long-form premium content. And we know through history, our ads resonate best in long-form premium content.

But also, one of the things about CTV that gets very, very complicated, is it's not just one thing. There's non-commercial streaming with Netflix, there's the hybrid, ad-supported SVOD services like Paramount+ or Hulu. There's AVOD services, which are the on-demand, fully ad-supported services – there's the FAST (free ad-supported streaming TV) services which mimic the cable bundle or cable subscription. I think the most overlooked part of it is the vMPDs multichannel video programming distributors), Hulu Live, Sling – now DirecTV Stream are, very much overlooked, but are replacing traditional cable in a lot of ways. It's a big ecosystem that's not just one size fits all, but it means a lot of things when we say CTV.

When you're working with clients, what is your first piece of advice?

The first thing we typically start to share with clients are the consumption trends – here's time spent with linear TV, here's time spent with CTV, and see how it's changed over years. If you look in totality, it's roughly 50/50 right now in terms of time spent, and if you get younger demos, it's already skewed towards the streaming side. So, that's the first thing – to set the table on both sides of it, because we hear a lot about, “linear is dead.” People are still watching linear TV. They tend to be older, but they are still watching linear TV. And it's 50% of this whole pie. Our budgets need to represent that, depending on your age segment and who you're going after.

The second thing we share is a streaming or CTV landscape slide. But it is important to delineate – here's what linear TV is and what cable companies are. And here are vMVPDs, that mirror what linear cable providers do. And here's AVOD, commercialized, not commercialized, and here are FAST channels. It’s to really explain the landscape.

How do you see this evolving both in terms of the percentage of inventory and percentage of money that's going to go to CTV? There's usually a lag between the eyeballs and the ad dollars.

The leaders in the CTV space are non-ad-supported, so total time spent with video might be 50/50, but time spent against ad-supported video is not 50/50, because if we take out Netflix, take out Amazon Prime – and Disney+, those are three big players that are currently non-commercialized. With Netflix’s announcement last week, and Disney+ having limited commercialization now, that dynamic will change. But for now, what we're watching in the CTV space, as Peacock grows, as Paramount+ grows, as Pluto grows, as AVOD and FAST services grow, is it's taking time away from Netflix, taking time away from Disney+, even Amazon Prime. And you can't forget about YouTube. It’s a big player in the CTV space which is ad-supported.

As we have tracked the non ad-supported share of the pie, our projections showed a landscape with less ad supported video than today, due to the shift to CTV and the big players having no commercials. However, our new projections taking into account a modest amount of Netflix ad support and growth from other players, we are now projecting there to be MORE ad-supported video than today. That is a positive development for advertisers and spending trends.

What’s the share of inventory and of share dollars in connected tv versus linear?

Those are all the things that we're paying attention to – how do you divvy up a budget between linear and CTV? – More broadly, digital video of any kind? There was a pretty significant shift in last year’s upfront towards CTV. Within these media company’s portfolios – so from NBC linear properties to their digital, in specific Peacock, from CBS and Viacom networks to a Paramount+ and Pluto there was a pretty significant shift. It was probably – going into last year, roughly 80% linear/20% premium CTV of those properties. Last year was probably more like 70/30 after coming out of the upfront, and I would anticipate another 10-percentage point shift this year to a 60/40 level which will roughly mirror consumption.

Obviously, there are a lot of different ways you can buy CTV inventory. How do you advise on that, because it’s an unusual part of the market in that way?

We try and limit the overlap of buying the same inventory or same properties through different avenues. Going back a couple years, particularly the traditional media companies would make their inventory available in a lot of different ways. They are now pulling their arms tighter around it – so it's not available in open exchanges, or through third parties. So, one, from a negotiation standpoint, we still have massive upfront negotiations, and now the CTV piece is a large part of those conversations. We're looking at an NBC, a Disney, a Paramount Global, as a single entity across all their different properties – and that's how we're negotiating. Open marketplace programmatic becomes more of efficiency play versus the likes of Peacock and Pararmount+. And in the current process we break it out that way.

The device, being Roku or an actual TV set, versus the rest is a little bit tricky as well. The one downside to those is that you're only getting their households. So Peacock – it's available basically on any device. If I'm buying just Samsung, I'm only on Samsung TV, so I can't reach anywhere else. I can theoretically be in any household who signs up, but with the devices, you're limited to their footprint only, which is a consideration you have to take into account. All of that also leads to the challenge of frequency management of occurrence across publisher, across device and that's the next big thing to really sort out because there are some controls within a publisher and you put frequency caps on. And theoretically they're abiding by it. That is really on the agency side/the DSP side – to help monitor that cross-pollination there and how much duplication of frequency you're getting across publishers.

If there are frequency caps there, I'd love to know what they are.

It’s consistently a problem and it's within the publisher. Are they frequency capping the way that they should? When you're buying directly, you're relying on the publisher for all of that data. So, we don't implement a cap, but all the reporting – we’re relying on the publisher to abide by it and validate it, which is why we’re moving towards using programmatic pipes to monitor that. It's not buying it programmatically – and that’s an important distinction – but we are monitoring programmatically and taking those direct deals, and moving them over to programmatic guaranteed deals, buying them the same way. But at least it's running through the agency, the DSP pipes, so that we can monitor and see what that looks like on a frequency by household basis and start wrap some control around that.

Getting back to the different groups of inventory – the smart TVs, the Rokus, the networks – the way the business seems to have traditionally worked, everything always ladders up to the same players – NBC or Disney or what have you. And now in the upfront, there are the linear buys, and the connected TV buys you can do through the same players. Do you anticipate that the industry is going to continue to default to existing relationships because it’s just this sort of ease of doing business?

Yes and no, I mean it's been one of the misperceptions for years that TV is losing to digital. Most of the TV networks own digital properties as well or acquire them. Those bigger players are going to be just fine. But I do think we're moving towards less and less independent networks. They're going to have a harder and harder time surviving in a world where people are linear cord cutting. And even if they're moving to vMVPDs replacing linear cable, those tend to be much smaller bundles and less channels. But at the same time, you're seeing Google/YouTube, being a player in the CTV space. Amazon, clearly becoming a huge player in the CTV space and then slicing off rights – for things like airing NFL Thursday night games. It's interesting to see the Amazons of the world and to maybe a little bit lesser extent, the YouTubes, operate the way that TV does with a more traditional upfront sell.

But then the other side, you see the traditional media companies going more digital and acting more digital. And if you've seen any of the press recently about Disney's clean room and NBC’s clean room – those are not conversations we had with TV networks five years ago. So if there's going to be fewer bigger players, some of the cable TV stuff is getting replaced by the Amazons, YouTubes of the world. They're both kind of stealing from each other in terms of how they go to market and how we negotiate and how we operate together.

There seems to be a lot of excitement around this CTV, partly because we're seeing cookie deprecation – and so suddenly, this comes along and “solves” it. So how much does it solve?

I think that – “picking up the slack” is a better way to describe solve for it because CTV never had cookies, right? It wasn't built that way. The negative aspect of CTV is – you could know about the household, you know about the people in the household, but you don't know who's watching the TV set at any given time. We have household IP address information, and you can make assumptions based on the type of content that's being consumed. It's better in some ways and less than others because you can't necessarily get down to the user. It's at the household level. It doesn't have the challenges of cookies disappearing, but it also has its own gaps as well.

Do you see the objectives of CTV being at all different? How do you – from an objectives/KPI standpoint, look at this inventory versus looking at linear inventory?

We approach the CTV space from two angles: either it's linear replacement, or evolution. If you're buying it more for broader reach, you may layer in some targeting but generally speaking, you're buying it for linear GRP replacement – and that's demo-based or more broadly-based. That's really about content first in audience side.

And on the other side, it's more audience first, where we've identified a precise target we want to reach, and we can reach them anywhere in CTV. We don't care if it's Hulu or YouTube or any other publisher programmatically in the open ecosystem. As long as we're reaching that person, we don't care what kind of context it’s in. We apply targeting to our GRP replacement buys, and do often, and the other side – the audience first-side – content still matters to some extent not nearly as much as when you're buying replacement TV, replacement GRPs.

If you had three or four “watch outs” for marketers, what would they be?

We talked about frequency management as an important part within publisher and across publishers and devices. That's a huge puzzle to figure out – how we monitor that, how you control for it. That's the first thing. Second thing – one of the challenges of CTV is the entry point – the CPM is higher on average than linear TV. And it's argued it's more targeted – and there are lower commercial loads, which definitely has a value. This has justified a higher CPM. The return on investment we've seen in CTV – holds up pretty well. So as you're planning, you have to ensure that what you're buying in CTV is as effective as the lower-priced inventory that you're getting in cable – that's, the second piece. And the third piece is going to be about programmatic. There are a lot of ways into CTV from a programmatic standpoint. Then it's measurement too, standardizing some sort of measurement.

Is there any standardization of measurement that's coming down the pike?

We need to make advancements because the measurement right now is not good at all. But we're not going to get to a Nielsen – one currency world in the CTV space – we're moving away from that in the linear space. It will be a multi-currency space to some extent, and we'll have to be comfortable with it. But we do need more standardization. The players emerging on the linear side of that are Nielsen alternatives, and as Nielsen evolves to Nielsen ONE – they'll all have in the future some capability across the same screen – across linear versus CTV measurement.

Any last thoughts on what people should be paying attention to?

One thing I touched on it a little bit earlier – why I think the MVPD space is overlooked. It's less as a direct advertising medium because we don't usually buy the MVPDs direct. Essentially, it’s the same way a cable operator operates – it's a bunch of cable channels that you're watching live TV on. The national commercials that we buy from ESPN and A&E, etc., are airing on Hulu Live the same way they are on Comcast and DirecTV and Time Warner.

I get frustrated in the “linear TVs dying” conversation and look how many households cable companies are losing. Hulu Live, is six million households, and Sling and YouTube are right around there. There aren't many cable providers that are six million households. There's Dish and DirectTV. But there are 100 1-2 million household cable companies out there – and they're viable businesses. When we look at cable, it's no longer just linear, it's the whole ecosystem, including MVPD. You can sign up for a cable subscription, a satellite subscription or you can sign up for the MPVD – it's basically the same experience, just getting it via the internet instead of a satellite on your roof. I just think that piece often gets overlooked in the in the CTV space.

So, maybe the most important question: What percentage of your clients are buying CTV?

Of those that have historically bought linear or are buying video in general, I think 100%. I don't think we have anyone who's not buying it in some way, shape, or form.