Locality's Tom Wolfe On Why Top-Down National TV Strategies Fail In Local Markets

0:04Inside the Stack is a custom podcast series from AdExchanger, where we interview leaders in ad tech.

0:18This episode of Inside the Stack is brought to you by Locality. Locality makes local TV advertising simple by bringing together premium broadcast and streaming inventory in one place. Supply path optimization has become industry shorthand for efficiency. Fewer intermediaries, lower fees, and a cleaner route to inventory. But a shorter path isn't inherently a better path. What matters is the value it unlocks on the other side. Hi, I'm Lynn D. Johnson, Senior Director of Event Programming and Community at AdExchanger. In this episode of Inside the Stack, I'll be talking with Tom Wolfe, SVP of Programmatic at Locality.

1:13We'll explore how direct access can create a stronger foundation for data, audience intelligence, and smarter organization. Also, why that becomes especially important in local advertising, where consumer behavior and campaign performance can vary significantly market by market. hi tom welcome to this episode of inside the stack how are you today oh i'm sorry how are you i'm wonderful i'm i'm doing great and ready to um dive into this conversation about spo um and and just understanding how buyers should be thinking about buying right as media is becoming more and more fragmented. So let's first learn a little bit about you and what you're doing at Locality.

2:15So I know that you've spent three decades launching and leading business units at some of the biggest names in media and tech, Roku, Comcast, TiVo, Yumi. Then you broke that pattern and you co-founded your own analytics company called AIP. Now you're at locality and taking on one of the, I think, one of the industry's thorniest problems, right, is local TV fragmentation. So first, what is it about this specific problem at this moment that pulled you into locality and into working in this arena. Yeah.

3:04Tom Wolfe: For sure. And you remind me of something a former colleague said to me once. If it was easy, everyone would be doing it, so maybe I'd just like challenges. But I started my career in local cable advertising in the New York DMA, and it had never been far from my mind. Every step of my career, local TV fragmentation has kept coming up. Inventory is scattered across markets, publishers, and platformers. Data isn't consistent. The buying process is manual. India Local still represents an enormous amount of total ad spend in this country, a large share. I've seen and worked for nationally focused companies who are attempting to drive local activity, but they often lack the resources to focus from the ground up.

3:48Tom Wolfe: Many of them have realized they need companies like Locality to help. We've built infrastructure around solving this fragmentation challenge. I was drawn to Locality by the quality of the people, and I was convinced by the strength of our publisher relationships, our data capabilities through the Locality audience engine, and commitment to technological excellence in our platforms such as LocalX and Darwin. So my job is to accelerate what's already working, build out the programmatic layer so that local can transact the way modern advertising demands. That's a fun challenge to tackle. Yeah, definitely.

4:24You know, because it's not only fragmentation, there's a lot of convergence because local also now has to have streaming. Right. So and buyers expect what's happening and streaming to be the same as what's happening in local. Right. Right. So we'll talk about that. We'll dig into that some. But first, I want to talk to you about, I mentioned SPO in my opening question to you. Let's talk about supply path optimization for a minute. You know, the way I think of supply path optimization, it's about efficiency. It's about cutting out the middleman. It's about cutting out the fees, cutting out the waste.

5:07Do you think the industry is solving the wrong problem here? What?

5:12Tom Wolfe: It's a great question. Traditional supply path optimization has focused heavily on economics and efficiency. Fewer hops, lower fees, greater transparency. These things all matter. But a shorter path isn't inherently a better path. The goal should be optimizing for value. It's not just a fee or can I cut costs. Many intermediaries offer value. Some don't. Those bad intermediaries insert themselves into the transaction without really improving the outcome. And in a large ecosystem like ours, those companies can pop up frequently. Every additional hop also can affect the signals traveling alongside the inventory.

5:55Tom Wolfe: So data can be removed, altered, hashed, or inconsistently passed as it moves through the ecosystem. So the cost of fragmented supply isn't just the tech tax. It's also the potential loss of data quality and signal fidelity, which impact outcomes. So the question advertisers should be asking isn't simply how many intermediaries can I remove. It's what value am I getting from the path through which I'm buying. And ultimately, SPO should be about preserving the value of what is being transacted. Premium inventory, quality data, the signals needed to drive better outcomes. And that can get lost with buzzwords as they come up.

6:33Tom Wolfe: So bypass optimization inherently is a noble cause if we remember to focus on the value and the outcomes. Value and outcomes. That's what every marketer should be focusing on, I believe. But yeah, like you said, people try to take shortcuts and then they end up with, as you stated, loss of good data, right? And sometimes buying in places where they didn't expect to buy also, right? Yeah. So let's talk about this a little more. Let's break it down a little more. Right. So a buyer has to direct supply as a direct line to supply. He or she have the direct line to supply. What are they actually getting when they have that direct line that they might not get through a fragmented path?

7:24for sure.

7:25Tom Wolfe: The biggest difference is visibility. A more direct connection gives the advertiser a clear review of the inventory and those signals associated with it. I think of the telephone game, right? Is the message ever the same at the finish as it was at the start? Well, if you have a lot of intermediaries, the data is never going to be the same at the finish as it was at the start. And so the outcomes might not be the same. When you're closer to where the impression originates, you have more consistent access to quality signals around content, device identity, exposure. In our case, this means having access within the publisher's ad serving environment rather than receiving the inventory and its associated signals after they've traveled through multiple systems.

8:04Tom Wolfe: But that proximity gives our advertising partners a stronger data foundation in addition to the inventory quality and trust. So value expands when the inventory is coupled with these premier bid stream signals. beyond inventory accessibility, what can we understand in action because of the data that comes with it? So when those signals, those data signals are combined with strong identity intelligence, you can define audiences more precisely, make better media decisions, and you can optimize toward actual outcomes. So I don't mean to harp on it, but outcomes are only as strong as the data coming in, right?

8:41Tom Wolfe: We ran a campaign for a leading home improvement retailer in which we tested a locally informed custom audience segment from our audience engine against a traditional third party home improvement category segment and we drove 20 greater visits than the lift than the traditional segment so that to us proved the data path optimization is equally important to supply path optimization both in the signals received in the in the bitstream to how you associate them with the identity graph to the audience segments themselves and this quality of the data path influences the quality of outcomes as much as the supply path does.

9:17Tom Wolfe: Would that make sense? It does, but break it down to me like I'm five years old. Data path optimization. It's a new one. Yes. I've heard another DPO before, but I think that was a different D. Yeah, there was one. So supply path optimization, if you think about it, you have the source of L. associated with the content. If you look at the data ecosystem, it's filled with companies who try to acquire consumer signals and they trade them back and forth and they buy and sell data to each other a lot. So they're looking at data segments that are called together from multiple different sources, which may not necessarily be bad.

10:03Tom Wolfe: But in those transactions, you don't know what signals are getting passed back and forth because each data provider wants to maintain its value proposition. So from the ground up, those signals are direct. And that just means that the buyer can know, hey, I know where this user is coming from. It starts with the signals you get in the inventory stream. It couples with the other signals that you use to run a campaign. Okay, all right, it makes sense to me now. um but in all of this I still feel that local adds a layer of complexity like nothing else that we know in media so I'm a buyer and I'm working on my media plan I have my buying strategy.

10:59I built it for national scale. But what can happen once I try to apply that across individualized local markets, maybe hundreds of local markets? Are there things that break? What happens if I'm thinking, here's my national strategy, and yeah, it's going to work? Like, what really happens?

11:25Tom Wolfe: A lot of national advertising is built from the top down. National audiences are strategy, scale down into individual geographies, as you just said. But consumer behavior doesn't scale down neatly. Even if you theoretically recognize user ideas to apply to a segment, what's true nationally may not be true in Philadelphia, Seattle, or Dothan, Alabama. Plus, audience behavior may differ meaningfully even within individual markets. So local requires a bottoms-up approach. You start with what you know about the market, the audience, and the behaviors happening there, and then you build and scale up from that intelligence.

12:02Tom Wolfe: Technology has made it relatively easy to penetrate a market from a national seat. A DSP can target a DMA, a zip code, or audience within a geo. But market representation does not necessarily equal market expertise. And just because you can reach people within a market doesn't mean you don't understand where the opportunity lies or how local consumers behave or how those differences should influence the media strategy. So take a retailer with stores concentrated on one side of a market and auto dealer, right? A national audience filtered down to the DMA may technically reach the right demographic as indexed by a national segment, but it may not reflect that retailer's actual customer behavior.

12:41Tom Wolfe: So an auto intender in Philadelphia may behave vastly differently than an auto intender in Dothan, Alabama. Two interstate highways, I-76 and I-95, cut through Philadelphia. The closest interstate to Dothan, I-10, is 32 miles away. Philadelphia still has some brick streets. Dothan still has several dirt and gravel roads. They're different markets, different intentions. Further, those dirt roads are concentrated in the rural portion of the region, and so the auto-intenders on one side of Dothan may behave very differently or have very different needs than the auto-intenders on the other side. And you only get that if you spend the time in the markets and work with the advertisers in the markets.

13:20Tom Wolfe: So market level intelligence helps account for differences in audience behavior, viewership, geography, business context, rather than assuming the same strategy will perform the same way everywhere. And too many companies fall into that trap. So you can still achieve national scale. The difference is how you get there. Are you pushing it down from the top, from the national level down, or are you building it from the bottom up? and that's where we specialize. Right. It sounds like building from the bottom up is the right approach because you have a more diverse read on the data, on the user, right?

13:57So that just, yeah. Not diverse. That's not the word I'm looking for. It's maybe more reliable of the identity. It could be a little bit of both.

14:08Tom Wolfe: I mean, people are different, right? And their needs are different across markets and across communities. So you do need both of it. And that's not to say that there are places where national makes sense for big national advertisers, top 50 national advertisers who really need to penetrate everyone. You know, there may be some more commonality between them. But when it comes to the types of advertisers that we see in the local, regional, mid-market, those differences matter. Okay. I want to hit on this market-level data a little bit more, and then I want to talk to you about AI, because how can we not have a discussion without talking about AI today?

14:54better access, better market level data. How does the campaign optimization change when you have that? And what adjustments or optimizations does a market have to make while it's running? Or will they be able to make instead of waiting to, because like a lot of times people went like, oh, look what happened during the campaign because they didn't have the right data in real time. So what happens when they have better access to better market level data? Like what would they be able to do?

15:32Tom Wolfe: This is definitely related to the AI side of things as well, right? Because there's so much potential to optimize faster with AI. But even without that, depending on the capabilities of your platform. So we've got Darwin and LocalX platforms, and we plan our campaigns, and we run and optimize our campaigns within there. And it's a combination of human intelligence and technology to make sure that you're taking those data signals and evaluate them on a regular basis to make sure you're optimizing in and out. I don't think this is an unusual capability for some platforms where you can do that. But when you're talking about local, now you're talking about scaling down to many markets.

16:20Tom Wolfe: You have to plan for those markets differently and you have to execute differently. And, of course, you have to measure and stay on top and optimize on a more granular basis. Okay. All right. So let's dig into the AI part because everywhere you look, every conference you go to, every pitch you hear from every vendor, everything is AI, AI, AI. It's in every deck. What's really separating one platform's decisioning capabilities from another one's? Yeah. I love an industry in which once an announcement is made, everybody can do it all of a sudden, right? We see that. AI itself is not the differentiator anymore.

17:12Tom Wolfe: Machine learning and automation have been part of ad tech for years. And increasingly, as you said, everyone has similar AI capabilities. I like to say if you've been in ad tech and are only just now starting to use AI, then you're probably not very good at ad tech. You might want to reconsider your career choices here because it's been around for a long time. We just called it machine learning. Now we've seen advancements in the assisted side of AI where it's natural language models. And so it feels a little bit different and everybody's jumped onto that. But when you dig into what's actually happening behind the scenes, the optimization isn't really that different than it has been in the past.

17:46Tom Wolfe: I'm fascinated by the way companies talk about AI. I am reminded of the launch of Windows 95, which some of your viewers may remember. Imagine a company at the time saying, our competitive advantage is that we're really great at Excel. That wouldn't have meant much. Because everyone had access to the same tool. So the advantage comes in what you put into it, how you use it, decisions you make based on the insights you can gain. And AI is basically the same. It's a tool. The same rules apply. Simply saying you use it doesn't tell me anything about whether you're going to produce better outcomes at all.

18:22The different ways AI can create value, there's the assisted AI, which makes campaign setup, analysis, workflows faster, easier.

18:29Tom Wolfe: And then the agentic side is when you go further in connecting systems and helping campaigns produce and optimize better outcomes, right? Yep. If two platforms are using the same or similar AI models, but one has stronger data signals going in, deeper intelligence, then that model is more likely at the outset to come up and make stronger decisions. The quality of the data matters. Better inputs give AI, better understanding of the audience, the market, and what's actually happening within the campaign. think about how you use ai on a data basis you're querying it all the time and trying to figure out prompting if you will to find out the next piece of information that demonstrates that human experience matters greatly as well it should not be undervalued knowing what questions to ask what data matters for your use case how to interpret the output that still requires expertise and human intelligence you still need to apply the data the right thing so good ai paired with bad data or poor assumptions just helps you make bad decisions faster.

19:33Tom Wolfe: AI, as it becomes ubiquitous, the competitive advantage actually sticks away from who has it and toward who is using it the best, has the best data and access, and can really power outcomes that matter for the clients. Yes, that is so true. Garbage in, garbage out, right? So you still have to have that better, that reliable data. Um, uh, so I think I know what it like if I were writing an article about this conversation we just had, I would pull out some keywords and I think I know what the keywords would be here, but I want to give you an opportunity. Usually when I wrap up inside the stack, I ask my guest for three words, uh, that they think describe their sector of the industry right now.

20:23So I think for you, it's about like local and converged TV. What are your top three words and why? And then I'm going to tell you if I was right or not.

20:34Tom Wolfe: So, yeah, being the big fan inside the stack that I am, I've been anticipating this question. And it is both easy and hard. You want to be clever and all that. So I think I want to define industry as sort of a lot of things could be programmatic. It could be CTV, it could be local, it could be data, it could be I, you're right. I'm thinking about all these things. So if I tackle from the angle of convergence, there's a couple of thoughts that have come up. So the first word is maturing. Traditional TV companies are finally fully vested in streaming. As somebody who went through it as both a local ad seller, a distributor of cable TV content, working at Roku and consulting with companies like Vizio and others, it it's amazing to see this evolution but it's taken some time but these big companies when they get there they usually do it right so the traditional tv companies are vested in streaming they're pushing programmatic platforms as a result to respect the structure they've employed in their direct advertiser relationships for decades right so i believe quality is truly becoming important and as many players you have in the ecosystem ultimately some advertisers may say ah it's waste i'll deal i don't you know i don't care about that extra stuff but most of them are going to say i want to know that i'm running in a great place with great data and great performance and it's quality is not just a buzzword but it's real signifier success supply path optimization is a meaningful first step in this journey and you know if if what i think happens with data is going to happen and data path optimization uh will be next so first word is maturing.

22:19Tom Wolfe: Second word, and this one is, I bounced back and forth with the marketing team, but I'm going to use recovering, right? So it may not be the most app, but it applies in a couple of ways, right? I think the biggest is AI. AI had everyone panicking a year ago. A lot of solid people lost jobs as companies grappled with the impact of AI. The dust is not fully settled, but we're starting to see AI not necessarily or solely as a human replacement, but as that massively impactful efficiency tool, which we happen to be using to drive results in our tools for our clients and for our partners. And I think as individuals we use as we plan our businesses and go about day-to-day tasks.

23:05Tom Wolfe: And so I think we're in the middle of a recovery where the panic over AI is starting to dissipate a little bit, at least I hope so. So and that relates to the secondly, I think there's there's been this concern over performance throughout the funnel. Right. CTV, you know, can can get proper attribution for the outcomes it drives? Well, having been in this industry for a long time, we know that CTV drives outcomes. Right. We've always known the power of sight, sound, motion and emotion. We think there's a place also for the walled gardens to attribute where it's appropriate. So, you know, we are starting to recover from that shock of like, wait a second, why are these things so separate?

23:48Tom Wolfe: And I think marketers are becoming smarter about how to look at the whole funnel, the impact of TV, the impact of data and things like that. So recovering is the second word. And you can tell me if you think I should have a better word. Nope. I think that's a great word. Maturing, recovering. And now what's your third? The third is collaborative. and as someone who was born in traditional TV, although we never thought of cable at the time as traditional. We thought that was the next big thing. But we focused on our content, we focused on our reach, and we focused on our Nielsen performance. We had maybe three vendors.

24:25Tom Wolfe: In addition to Nielsen, you might have an MRI or a Polk or an Arbitron. But today, as these sectors converge, the programmatic component is driving partnerships and interoperability. Big buzzword. Most companies now view other companies as partners or, at worst, coopetition. But our industry does an excellent job collaborating, trying to rise the tide rather than just lift one boat. There are going to be walled gardens that are very focused on lifting a boat, but they're going to have ancillary impact that's beneficial to the industry as well. And I think, generally speaking, these sectors collaborate and the companies within these sectors collaborate very well.

25:04Tom Wolfe: So the third word is collaborate. Collaborative. I think that's very important for the future of this industry overall. You went in a totally different direction than I could have ever guessed. And I love it. I'm here for it. Tom, I really want to thank you for visiting with us on Inside the Stack today. Any last words for our listeners? Anything you want them to take away that we may not have covered? Yeah. I think we covered most of my comfort zone right now. And I really appreciate the opportunity to come and speak with you. I am excited about the opportunity that Locality is bringing to the market.

25:43Tom Wolfe: And we couldn't do it without partners across the ecosystem. And that includes AdExchanger. And we thank you very much.

25:57This episode of Inside the Stack was brought to you by Locality. Locality makes local TV advertising simple by bringing together premium broadcast and streaming inventory in one place.