← Marketecture: Get Smart. Fast.
Marketecture: Get Smart. Fast.
Episode 160: Matthew Egol on Why We Need an AI-Specific Industry Association
00:55:24
Loading summary
Transcript146 lines
- [00:00]
Experian Narrator
New Year, New Me. Cute, but how about New Year, New Money? With Experian, you can actually take control of your finances. Check your FICO score, find ways to save and get matched with credit card offers, giving you time to power through those New Year's goals. You know you're gonna crush start the year off right. Download the Experian app based on fico's great model, offers an approval not guaranteed. Eligibility requirements and terms apply, subject to credit check, which may impact your credit scores. Offers not available in all states. See experian.com for details.
- [00:28]
Ari Paparo
Experian.
- [00:31]
Indeed Sponsor Voice
This episode is brought to you by Indeed. Stop waiting around for the perfect candidate. Instead, use Indeed sponsored Jobs to find the right people with the right skills fast. It's a simple way to make sure your listing is the first candidate. C According to Indeed data, sponsored jobs have four times more applicants than non sponsored jobs. So go build your dream team today with Indeed. Get a $75 sponsor job credit@ Indeed.com podcast terms and conditions apply.
- [01:08]
Ari Paparo
Welcome to the marketexture podcast. This is Ari Paparo. I'm here with Eric Franchi and our guest this week is Matt Eagle, who is a longtime friend of mine. I actually went to business school with him back when I still was young and sprightly and wasn't even involved in.com stuff yet. And he is a longtime consultant, founder and CEO of journeyspark, which is a consulting company. He's an author of a book called CX and Culture Connection. And he is now the reason we're having him on is that he is now the one of the leads@agenticadvertising.org, the new nonprofit that is spearheading AD CP. So Eric, how you been? What do you think about this whole like multiple nonprofits proliferation thing?
- [01:55]
Eric Franchi
There is the question, is there a need for another trade arc? So that's the question I want to get into with with Matt today. I feel like I know what his answer will be with respect to, you know, what the what, what's going on in the space and the, and the need for it. But I think that's, that's interesting. And then Matt is at the forefront of all of these conversations. So I just can't wait to get his perspective on all things AI, even down to all the stuff in the, in the refresh with a lot of the news around super bowl and all these AI announcements.
- [02:26]
Ari Paparo
Yeah, yeah. Did you watch Super Bowl?
- [02:28]
Eric Franchi
I assume, of course we're going to get into it. We'll talk about our favorite ads.
- [02:31]
Ari Paparo
Who's Your team. Are you a Jets fan or Giants fan?
- [02:35]
Eric Franchi
Giants.
- [02:36]
Ari Paparo
Giants. I think one of the one thing that was annoying about the super bowl was Bon Jovi introducing the Patriots. That really didn't strike me as okay. That kind of crossed the line for me.
- [02:47]
Eric Franchi
Yeah, that's culturally unacceptable in New Jersey. I don't know where that was coming from.
- [02:54]
Ari Paparo
It's okay to like drink Dunkin Donuts. We don't have to be like super prejudiced against that region of the country. But the Patriots, like, was like Ben Affleck not available. Like, why Jon Bon Jovi? I didn't. I don't appreciate that. Anyway. Hey, how many days till.
- [03:12]
Eric Franchi
Markets? Extra three.
- [03:14]
Ari Paparo
Oh my God, I have to write a deck. Oh my God, it's 30 days. So like, you know, I have. So we all have our homework here. So I have to create a deck because I'm the mc, I'm the intro. I gotta like bring it. Eric, you got to select some startups, but I think we're pretty close on that. Oh yeah. And then listeners, you know, you have to do you have to buy some tickets. So like we're 30 days out. Bad. What are you doing? Like, I keep getting these texts from people like, hey, should I come? Yeah, you should come. Or like, hey, get a free ticket. Like, no, you can't get a free ticket. Just buy a ticket. So the code is ARIPOD30 and you get 30% off and that. And the website is marketecture live.com and. And we're announcing new speakers all the time. So some new announcements. I'm not even sure these are on the website yet, but we're rehashing some of our most popular speakers from previous events. We have Chris Kane and Micah Sullivan are going to rerun what they called the debate about inventory quality, which was definitely one of the most popular thing from our first architecture Live. And then Rob Leathern is coming on. He's an expert in policy and things like that and he's going to talk about the fact that your house is inf with bots and they're doing things you're not going to be happy about. Also one of the most popular segments from the last market at your Live. So I'm excited to bring in some of these deep thinkers who have been on the show. You know, they're, they're friends of the pod, as they say. So again, registermarketecturelive.com and use code AIRIPOD30 to get a discount. Hope to see you then. Okay with that. Let's jump in With Matt Eagle and agenticadvertising.org, aDCP and all that good stuff. All right, welcome, Matt Eagle. You seem to be coming to us from some very consulting like background, some nameless, faceless hotel ballroom or something. What's this background you're showing here?
- [05:07]
Matt Eagle
I am in the Wynn Hotel in Vegas, Back in Vegas and CES for the Medallia Experience Conference.
- [05:14]
Ari Paparo
All right, that sounds pretty exciting. Not on my calendar. So, Matt, you and I go way back, right? Like we went to business school together. We were both at Columbia in the 90s, so I always thought of you as like kind of the ultimate management consultant guy like you, you really dove in. You're at Booze Booz Allen for quite some time, right? How's it been? How have you seen like the world change?
- [05:40]
Matt Eagle
You know, when you and I went to Columbia, the Internet was still a puppy. It was, it was a new, new, new thing back in 97 and 98. And you know, we've seen a lot of change over the years. The media business has been moved from more of a pure entertainment to more of a technology data driven business. And we've also seen brands move from buying ads to building experiences. So advertising is still really important, but brands are publishers and brands are investing heavily in, you know, how does advertising connect with customer experience, whether it's E commerce or loyalty or mobile or what have you. So that's been a fun ride for me as I started on the media side in the media practice at Booze, which is actually where Randall Rothenberg and I met when he was the CMO of Booze and I was, I ended up leading the digital retail team and then helped launch Booze Digital. And now we're back together at Journey Spark doing work with agenticadvertising.org and other clients. And it's been a fun ride together.
- [06:43]
Ari Paparo
Yeah, I don't want to spend too much time down memory lane, but it's kind of a hard, hard to imagine, but in the late 90s in, in Columbia Business School, the Internet people were this small minority and we had to sit through these classes where they kept talking about how awesome like the classifieds were and you know, whether telecom regulation would be the next big important thing in, in digital media. And I remember even sitting in a professor telling us, telling the whole class that like digital music would never work. Right. It was just too much bandwid.
- [07:13]
Matt Eagle
Yeah, we had to memorize a lot of acronyms in telecom. But you know what, there are a lot of changes that happen and I actually think there's some parallels to what's going on. We'll get more into ADCP and agentic advertising, but with standards adoption fall into different patterns and I think what's happening now actually has more in common with like HTML and things that are more cutting across many industries and drive transformation and new things emerging as opposed to narrow with an industry like the Swift banking payment system.
- [07:48]
Ari Paparo
Yeah, yeah. So that gets us to our topic at hand. So we've been talking a lot about ADCP on this podcast. Maybe too much. I've been told too much by some listeners. But I was happy to see your name and Randall Rothenberg put out as leading a new non profit organization called agenticadvertising.org Tell me about it, what is it, how'd you get involved, etc.
- [08:13]
Matt Eagle
Well, so agenticadvertising.org or AAO, the brands like Alpha Alpha Omega, you know, it's actually a different, different kind of association. We actually think of it more like a professional association than a traditional trade association which represents the commercial interests of buyers or sellers. AO is a community of builders which cuts across brands, agencies, publishers and tech. And it's focused on the development and adoption of open standards, education and professional certification. So unlike trade groups like the ANA or forays that represent buyers and sellers, AO is modeled more after the IEEE or IAPP and privacy. Its goal is to advance the discipline and the competence of the people practicing it. So it's also into the professional development side and certification, not just the standards. All the standards are really where it grew out of an open source development community. But this is a much broader movement across industries and across roles in organizations than just the tech side.
- [09:22]
Ari Paparo
Do you think there's going to be like accreditation of some kind in the future?
- [09:26]
Matt Eagle
Absolutely. That's where this is headed, like an ANSI certification. When we, We've done over 100 interviews with executives across industries as well as a survey of the 1300 people who participated in that October Zoom launch for ADCP. And we, we actually heard one 80% of people say there needs to be a new organization that, that and this is a broader tent than, than than participate in other organizations. A a lot of individual practitioners and not just the big companies really embracing the openness of what we didn't fully realize with the open Internet. So this is a bigger tent bringing people together. And when you look at what makes successful changes occur, there's actually a lot of the more change that occurs in the way work happens, not just the technology, the more you need the professional certification side. Which is like a big part of the success of organizations like ieee. And the individuals are the ones who govern and own the professional certification side, whereas the corporations drive the adoption of the standards. So this is a hybrid model that needs both corporations and individuals.
- [10:49]
Ari Paparo
Is the certification, if you have one, going to be limited to humans?
- [10:54]
Matt Eagle
That's a great question. We see dual certification because we're trying to build trust, Ari. And the trust needs to be both of people and of the claims of the organizations. So we see the member directory as needing to have validity, where the claims of companies needs to be certified. So that's part like the Good Housekeeping seal of approval. So we see the corporate certification as well as individual certification.
- [11:23]
Ari Paparo
Right. And BOT certification. Like if my Claude BOT is better at adopts than I am.
- [11:28]
Matt Eagle
Well, what we would. We're not thinking of quite the way you are. Although now you're making me think like about. But like, you know, is this bot actually certified to do the way an individual would be? We were thinking it more like if the standard says you need to do these things around brand safety or identity or things like that, does it actually conform to the standard?
- [11:49]
Ari Paparo
All right, so you said 80% of the people surveyed wanted a new organization. There are two other organizations that are directly relevant here. There's like 20 other organizations in total, but there's two that are really.
- [12:05]
Matt Eagle
There's actually more than 3,000 nonprofits serving the marketing and media industry. So there's a few that come to mind first. But if you scroll down the search results, there's a lot of other ones.
- [12:16]
Ari Paparo
Terry, get us a nonprofit Lum Escape. I want to, I want to see it. That would be the lowest ROI activity Terry and his minions could possibly do.
- [12:26]
Eric Franchi
All right, but before you go on, Matt, were you surprised at that number? I was surprised when you said 80%. Just given the, you know, all the air that gets sucked up by maybe the, the two or, or three majors and then the 3,000. It's like, what does the world need? Yet another one. But I think in this moment what the world needs is yet another one. Huh?
- [12:45]
Matt Eagle
Well, we, you know, we, we basically thought this needs to be more than an open source development, you know, GitHub group and you know, and just the technical standards because you want to drive the practice around it. So that one of the first questions is, well, should this live somewhere else, right. And build on the great work that everyone else is already doing? I don't want to take away from all the great work that should continue to. And our vision for AO is actually highly collaborative. Look what we're doing with Prebid is actually an example of collaboration versus competition and we want to see the same thing with ana, with iab, with with the consumer brand association with other things. Now what's different here is because it's across many industries and because this is actually broader than we don't think of this as programmatic plus we actually don't think of this as just like adding Agentic to programmatic because if you look at the value chain. I'm going to talk two things here. One's the breadth of the value chain. Most of the early activity and value of Agentic is actually in discovery and planning which is left of transacting and it's going to extend over to measurement which is allows more of a stateful instead of stateless as memory. It has the understands the provenance. Ben Massey said this really well at Triton that programmatic is like RTB is like the high speed rail and Agentic and ADCP is more like the nervous system that sits above it. So this is a broader across the value chain and it's not competing with rtb. It's actually a broader set of inventory that's relevant here and a broader set of activities value chain. So we want it to work together but we don't see it as a replacement for in competition with rtb. RTB will continue.
- [14:49]
Ari Paparo
So, so let's take the topic of pre bid first guys. There was this headline. I'll read it. The headline says Pre bid takes over ADCP's code for creating sell side AI agents. Please parse that headline for me with regard to two, a couple things. Number one, why isn't, why does your organization exist when pre bid has an organization that presumably could do something similar? Two, what does it mean to take over your code? And three, why is it only sell side agents in that headline?
- [15:21]
Matt Eagle
Just so everyone knows the context. Your AO transferred its sales agent repository to pre bid and pre bid's gonna let you know. This will allow us to leverage the pre bid developer community and the thousands of publishers they work with for ongoing management of the sales agent development. And the pre bid sales agent will continue to work under the AD context protocol adcp, you know, governance framework. So pre bid is managing the code allowing AO to focus on the market like the education, the certification, adoption of use cases and the broader governance framework. So they're complementary. You know, without the relationship with pre bid we'd be doing a lot more hands on keyboard development and really pushing into that area which would be needlessly duplicating work. And there's thousands of publishers that we're tapping into to more rapidly get use cases out into the market. So it's a real win win for everybody.
- [16:21]
Ari Paparo
Is there, is there a separation between the standard and the code that we should understand? Like is your organization going to maintain the standard for NCP and Pre B? That is the sample code.
- [16:33]
Matt Eagle
Exactly.
- [16:33]
Ari Paparo
Okay, right.
- [16:35]
Matt Eagle
So you, I mean, and you know this, there's a distinction between protocols and standards.
- [16:39]
Ari Paparo
Yeah, sure.
- [16:40]
Matt Eagle
Right. So like we're more focused on the protocol and then maintaining the integrity of the standard, but not the, in this case we don't want to get into the code writing. We want to tap into the energy of the prebid community.
- [16:51]
Ari Paparo
This is a little bit like the way the IAB maintains OpenRTB protocol or standard either way. Sorry. And then, and then prebid implements it in code. Is that, that's a similar analogy.
- [17:02]
Matt Eagle
Exactly.
- [17:03]
Ari Paparo
Okay, so then the last part of the question was they specifically are sell side. That's what pre bid does. Does that mean that the buy side code still is with you and you're looking for someone to take it over?
- [17:13]
Matt Eagle
Well, we, I'll say definitely stays with us. And, and it's possible that we will forge other collaboration opportunities that have, you know, when we leverage this experience. There's many different types of agents now just if we run in the aperture for a second, this is all about agentic commerce and marketing and customer experience. So this is a broad lens. We're not talking about sales and service agentic, we're really focusing on marketing here. We're not trying to look at everything but there's connectivity to the customer experience side around leveraging data signals and being able to optimize the advertising, leveraging the full set of data signals and the rest of the experience and having content experiences be seamless across the whole omnichannel experience. So there's shopper agents, you know, like, you know, Sparky at Walmart needs to talk to, you know, or you're going to have new, you know, agents in the LLMs talking to retailers. Already a huge amount of web traffic is being, you know, staying within the LLMs rather than actually it's not becoming web traffic. Right. So when you design websites in the future, you need to think about like human centered design takes on a whole new meaning when like half the people there are bots, you know, so that's agentic. There'll be creative bots, there'll be data signal bots. So it's not just buying that we're talking about here. All these bots need to be interoperable and work together. And that's why when we talk about the breadth of the value chain and there being many industries involved and not just buyers and sellers, we're all builders.
- [18:51]
Ari Paparo
Right, but are you saying that that bigger picture is part of agentic advertising or.
- [18:57]
Matt Eagle
Or yes.
- [18:57]
Ari Paparo
Where. Where do you draw the line? You said marketing is the line.
- [19:00]
Matt Eagle
Yeah, like, we debated, like, even calling it agentic. Advertising.org is actually constraining.
- [19:06]
Ari Paparo
Okay.
- [19:06]
Matt Eagle
You know, this is really marketing, but we want to focus on how advertising connects to the rest of the marketing. So it's a paid media and how paid connects to owned and earned, rather than trying to really drive it holistically. And all have owned and all have earned.
- [19:22]
Ari Paparo
Got it. Okay, let's talk about the beef with the IB tec. What's going on? I. There's definitely beef. Like, you know, I feel like the. I. I haven't really dove in, but it feels like the IB is saying we don't really need new protocols, new organizations. We were fine as is. You should just use our existing documentation. Please tell me I'm wrong. They're wrong, you're wrong. Who's wrong?
- [19:49]
Matt Eagle
Well, we. We actually don't think there needs to be or really is a beef there. There have been a lot of conversations, but we really want to focus on collaboration. Like I mentioned, we see AAO as not for buyers or sellers, but all builders. And it's a big neutral tent, like we said.
- [20:14]
Ari Paparo
Right.
- [20:15]
Matt Eagle
It's really around humans to govern autonomous systems responsibly. So given this, we want to work together, we want to collaborate, but there's different folks involved in ao. If you think of a Venn diagram of who's involved in each, there's overlap. But the AO tent includes many more participants that are part of iev. It's also a broader lens of where we're focused across the value chain and a broader set of inventory that's relevant here. So we want to focus our collaboration with not just IB, but others where that intersection is strong, but not. The members actually spoke when they. When we. When we. When we set this up. Because one of the first questions, should this be separate? And they said yes, many of whom are in the iab.
- [21:08]
Ari Paparo
Why do they think it should be separate? What's the short answer for why it should be separate?
- [21:12]
Matt Eagle
Avoid the perception of capture, openness, a broader set of participants driving the standards.
- [21:20]
Eric Franchi
So.
- [21:20]
Matt Eagle
So not saying IB is captured, but we want to make sure that the agenda, what pursuing for AO is, is designed from the beginning to be not representing sellers. We want to make sure it's representing everyone and that it's representing a broad set of interests, not just a narrow group of sellers, which is what a trade association does.
- [21:45]
Ari Paparo
Yeah. I mean, the ib, historically, in the dawn of Programmatic, the biggest criticism of the IB was that it was a seller organ. They didn't really represent buyers of Programmatic very well. They. They've. They've changed a lot and they're now there. You could almost make the argument the IB is defending Programmatic against Agentic to some extent. Yeah.
- [22:06]
Eric Franchi
Now people complain about something else.
- [22:09]
Matt Eagle
Well, the. I think you're right that the whole launch of Tech Lab was a vehicle to broaden the participation of the membership, but there's still a narrower set of members in IEB in terms of size of companies and the orientation of the players. And I think the collaboration with IB will be very important. So will with the ana. I think that is very, very important here. And I think the IB gets a lot of attention right now, but it's settling down. And a lot of that. Let's. You know, I like to let people like Ben comment on that or other members, rather than, you know, you know, get into the whole LinkedIn back and forth, because I think it's not productive.
- [22:54]
Ari Paparo
Gotcha. So in terms of where this is, this movement is as of, you know, February 2026, what I had heard people. When I talked to people at the IAB leadership meeting a couple of weeks ago, what I heard was something like, okay, we've done a couple of test campaigns. We had. We had the CEO of PubMatic on the show talking about how they did the first ever test campaign. There's a lot of interest in doing test campaigns on the buy side especially, but there's really no volume, no revenue, and we shouldn't expect it anytime in 2026. That's kind of the consensus opinion I heard. What are you seeing in terms of, like, actual activity for using these things?
- [23:37]
Matt Eagle
I think if you kind of step back and look at this in the broader context of the adoption of a new standard and has the potential to be very disruptive. We're actually where you'd expect to be one quarter into the adoption cycle. That's all we are, you know, one quarter in which is a very thin but meaningful, you know, set of things happening. So initially, people were testing to make sure, like, technical feasibility. Does it work? Does it do what it says? Is this scalable and We've seen some initial tests of buyer and seller with like Probatic and Triton and Scope three and Yahoo and others. And now Salesforce is getting in involved now too to help leverage their data. So we're seeing activity occur but it's still thin and it's still focused in the planning and discovery portions of the value chain. It will expand. You'll see more on the creative side in the future. We see a lot of people start to talk and wade in like their brands getting involved. And there'll be more announcements in the coming weeks about some major brands that have joined. Three of the big agency holding companies have joined ao. So there's going to be a lot more experimentation. I think it'll accelerate over the course of the year.
- [24:58]
Ari Paparo
Last question. Give me an update on governance. So you and Randall are interim heads and then there's going to be an election of a board or something like that. I read on your website. But do you just walk the audience through that?
- [25:10]
Matt Eagle
Yeah. So Randall and I are both partners at Journey Spark Consulting and we're, we're acting as advisors to current board which is a handful of folks. But the, the first elected board and will be around an event we're going to do in May in New York that'll be invite only around 120 to 150 people. And the first board dinner will be around that event in New York. We're looking at around 40 to 50 board members split equally across brand, publisher, agency and tech. So we want that to be what represented by design to be a mixture of those players. We call them the primary categories of different size companies where one company equals one vote. Right. So if you have a thousand subsidiaries or one, you're one vote. Right. If you're a million employee, if you're 200,000 employees or 2,000 employees, you're one vote.
- [26:13]
Ari Paparo
Do the bots get to vote?
- [26:15]
Matt Eagle
The bots will not get to vote, but individuals will be involved more on the education certification side. So individuals will vote on professional certification and corporations will vote on the standards adoption.
- [26:31]
Ari Paparo
I'm starting a new org called adclawed.org that's bot only. No human beings. They could just talk about their ad ops problems and complain about all the confusing standards and nonprofits in the industry.
- [26:44]
Matt Eagle
When bots hang out in a room, do they still use jargon?
- [26:48]
Ari Paparo
I want the bots to come up with a new acronym. That would be, that would be the sign of sentience. All right. This was awesome. Thank you for taking us through this. I know it's complicated. We're going to take a quick break. Could come back with a little bit of a delayed takes on the super bowl, more AI news and some earnings. So we'll be right back. This podcast is brought to you by Media Ocean. Everyone knows the Lumascape.
- [27:14]
Matt Eagle
Now.
- [27:14]
Ari Paparo
Behold the Media Oceanscape. A singular graphic showing all the companies they've come together to form the Media Ocean of today. 28 acquisitions in 59 years. Yes, you heard that right. Founded in. In 1967, they are the OG of ad tech. Next year, MediaOcean will be celebrating its 60th anniversary, and they're showing no signs of slowing down with the acquisition of Innovid and launch of a new AI orchestrator. To learn more, go to mediaocean.comscape that's mediaocean.comscape.
- [27:55]
Eric Franchi
All right, we're back, everybody, with the refresh and we have Matt on. So there's gonna be a lot of AI just putting it out there. Just be prepared.
- [28:04]
Ari Paparo
How's the differ? It's always about that.
- [28:07]
Eric Franchi
I know. Let's. Let's talk about the Super Bowl. We didn't have the bet of how many AI company ads there would be, but you think thought there would be 10 or so. And you were right. There were nine. Nine companies either explicitly AI companies or, you know, like talking about their AI capabilities. So well done, Ari.
- [28:32]
Ari Paparo
Yeah, I kept a little notebook by the side while I was watching just to record this. There. There are definitely some. Some borderline cases like the Alexa ad or the ring ad, but I went with how many are literally just selling prompts. And I got nine. I'll list them. Wix, OpenAI, Claude, Copilot, Gen, Gemini, Alexa, AI.com and Base44. That was my account.
- [28:54]
Eric Franchi
Yeah. Oh, God. So we should talk about the super bowl for a couple of minutes. So first was AI.com. so somebody bought. So the founder, as I understand the founder of crypto.com bought AI.com, you know, wanted to launch a new thing. I think he bought the domain for $70 million and, like, didn't buy the. The traffic bursting capabilities, so he basically crashed the site. So you know what, it was a viral moment for him that was memorable. Before we get into maybe everybody's favorite ad, just to underscore what you were talking about, that this was the AI Super Bowl, I caught a link in Fast Company that there were 15 out of the total 66 commercials. So it's roughly 23% featuring AI in some way. Either it's promoting an agentic product or using AI in the ad itself. So absolutely. This was the AI Super Bowl.
- [29:51]
Ari Paparo
Yeah. I think it cuts both ways because people's immediate reaction is like, oh, this reminds me of the crypto super bowl, which was however many years ago, which was right before a lot of the crypto winter. And it reminds people of the dot com Super bowl of 2000, when everyone pets.com went out of business like four weeks after their super bowl ad ran. I think the continuing theme here is that AI has a really bad PR image right now. The normies do not like AI. The people like us are obsessed with it, and the normies hate it. And I'm not sure these ads did a lot to solve that problem.
- [30:28]
Eric Franchi
Yeah, let's talk about that before we get into our favorite ones. So leading up to it all, anybody? It's so funny. Like, this feels like it was months ago, but this was just last week. Leading up to it all, anybody could talk about what the anthropic ad that was dunking on OpenAI, not by name, about running ads on the LLM and anthropic, saying they would never do that. And you know, just having these, these, these ads that dropped. People were so fired up about this. They were, I mean, and everybody loved it. And it turns out it was like a nothing burger. There's some conflicting data on this, but there was, you know, some data that, that made it around that normies didn't love the anthropic ad and they were actually kind of weirded out or they didn't get it and they were wondering what the hell was going on, which I thought was pretty fascinating. What did you guys. And maybe you were with some people. What did you guys think when the anthropic ad finally came out?
- [31:23]
Ari Paparo
If I was a little anticlimactic, like it really works on the first view. And that's always one of the big questions a lot of these agencies have is whether to pre release the ads or not. And it seems like it really cuts both ways. And in this case, maybe it took a lot of the emphasis out. Also there was some reporting or I think it's true that they toned down the tagline at the end. Yeah, yeah, yeah. So I do.
- [31:48]
Matt Eagle
Yeah, I think the. That ad lost some impact because it was. Didn't have. Was a lot of words, a lot of talking, and it didn't have as much emotional appeal. So it wasn't very memorable.
- [32:01]
Eric Franchi
Yeah, well said. I was surprised by the OpenAI ad. I thought the OpenAI ad for Codex was really good. We'll put a link to it. I think the tagline was, you can just build things. It was inspiring. It was fun. Like, it drove emotion. So I thought OpenAI was better.
- [32:18]
Ari Paparo
I like the OpenAI ad. I'm not sure how many people had moved to do anything, but, you know, but it was. It was nice. I think the winner was Gemini. The Gemini ad was brilliant, with the mom showing the kid how their room will be decorated.
- [32:32]
Eric Franchi
Rate it.
- [32:33]
Ari Paparo
Excellent. Really good.
- [32:35]
Eric Franchi
Yeah. What was. Okay. AI Ads aside, we'll start with you, Matt. What was your favorite super bowl commercial?
- [32:43]
Matt Eagle
It was the dancing robot with the vodka.
- [32:52]
Ari Paparo
Oh, my God. Go on.
- [32:55]
Matt Eagle
It was just. I thought it was cheeky. I thought it was clever. It was irreverent. It was fun.
- [32:59]
Eric Franchi
You got. You got a runner up the.
- [33:01]
Matt Eagle
On the Budweiser ads. Can art. I like the. The horse with the bird with the eagle. I thought that was really cool. Two Americana images. It was a maybe thing of Snoopy and Woodstock.
- [33:12]
Eric Franchi
All right, what about you?
- [33:14]
Ari Paparo
Oh, my God. The Budweiser ad with the eagle. I. When it first came on, I said to my wife, like, oh, God, I can't take the horse. The horse schmaltz is, like, what I called it, but it wasn't that bad. It was kind of funny, so I'll give them some points on that. So, like, I really did like. I think the Gemini ad was maybe the best out of the whole. The whole game, you know, and there are definitely, you know, there are definitely some weak ones. I I As a. As a, you know, watcher of ads, I. The biggest thing I hate is the surprise brand at the end, where you're 25 seconds into a story and you don't know what the brand is. I think it's just a waste of money. Like, those ads just don't work using comedy or drama or whatever for this giant setup. And then it's like insurance, you know, I think it was Rocket Insurance had one of those. I can't even recall the ad anymore, but they. It just doesn't work. This bad use of money. You got to have your logo at the start. A good. A good example is, I think Dove soap had an ad that was like, you know, continuing their support of women. But what they did, and I'd have to find the actual creative is like, they showed Dove at the very start. They're like, dove brings you blah. And then they showed the emotional story of the teenage girls, and then they closed it with Dove. That's just so much more effective.
- [34:30]
Eric Franchi
Yeah, that was an excellent ad. I Too. Just, you know, pulling on the heartstrings. I thought Budweiser was, was good. I love the little, little baby chick into, into Bald Eagle. I thought Duncan was good. Just, you know, packed with the celebrities and just being like very, very, very self deprecating. And then the, the one that was, it was. I don't think it was a psa, but it almost was like a psa. The tight end for the non invasive prostate exam. I thought that one was really good and maybe just because, you know, whatever demographics, but I thought that one. Do you remember that one?
- [35:02]
Ari Paparo
I thought it was funny that they didn't emphasize the double entendre of tight end. Like they were just. They just threw it. They just threw it in. Tight end. We just picked that for a prostate exam without like mentioning why again, if.
- [35:15]
Eric Franchi
You'Re in the demo, you, you got the whole thing. We can move on. Anyway. All right, let's talk about AI again. This was a big week. Ads launched on ChatGPT. I think all three of us probably pay for ChatGPT. So we didn't see it, but ads are out there. They launched an ad pilot program. Seemingly all the hold codes are participating. Omnicom had 30 brands alone, paying a minimum of 200k for this. So this thing is out there. There was an opinion piece yesterday, I don't know if you guys saw it, title in New York Times. OpenAI is making the mistakes Facebook made. I quit. From an ex Open AI employee basically saying that this person believes they're gonna veer. It's only a matter of time before they veer away from their principles that they outlined and start making things like hyper targeting evasive and creepy.
- [36:11]
Matt Eagle
What do you guys think?
- [36:11]
Ari Paparo
Think quitting big open open labs is like the thing to do. These people quit so often. Like, this is a big week for quitters people. People stay at these jobs for like a month and they make a billion dollars and then they retire. There's that guy. There's some guy who like moved to England to write poetry because he makes so much money on anthropic.
- [36:30]
Eric Franchi
He's a senior anthropic guy.
- [36:32]
Ari Paparo
Yeah, yeah, yeah. It's, it's. Hey, it's good. Good gig if you can get it, I guess.
- [36:38]
Eric Franchi
Big week for quitters.
- [36:38]
Matt Eagle
It's a refresh on the, it's a refresh on the whole zero moment of truth thing, right? So like, you know, the stuff shifts further. The digital shelf gets getting closer to the couch, right? And so now what we're seeing is, you know, reimagining advertising much more, you know, in the, you know, in the search and discovery phase, reimagining the experience itself. That's actually one of the reasons I'm so excited to be involved in AAO actually is it's being developed to work across all surfaces, across all types of experience. It's about innovating the experience, not just automating what we do today. And I think this is a great. There's going to be a lot of innovation and changing the way the shopper journey works and being able to better serve the shopper. That's what this is doing. So I think it's a good step.
- [37:29]
Eric Franchi
I was waiting for this moment for us to be able to talk about this stuff with you. Thank you. Matt, are the LLMs involved with AAO yet?
- [37:38]
Matt Eagle
We do have one of them that's on the verge of joining, but I can't announce the names yet. But yes, one of them is coming in and we expect others to follow.
- [37:48]
Eric Franchi
That is awesome because I think that's the thing that has thus far been missing with all this collaboration and talk, which is the LLM just being leaned in on the advertising front.
- [37:59]
Matt Eagle
That's really good news and that's important and so is big brands.
- [38:04]
Eric Franchi
Right.
- [38:04]
Matt Eagle
So I think, you know, like, you know, we have one that's in our kitchen cabinet, we call it hasn't gone public with their involvement yet, but that'll come shortly. And then we have several other brands. So we want to have 10 brands on the board. Right. And right now, looking at our pipeline, we have around that many we're talking to. So hopefully we'll have more news to share shortly.
- [38:26]
Eric Franchi
You have so many things you can talk about. Somewhat frustrating.
- [38:29]
Matt Eagle
I wish I could share more, but they would probably be upset if I put their logo out there before they've actually signed the papers.
- [38:35]
Ari Paparo
I'm just going to ask Claude what Matt won't tell us.
- [38:41]
Eric Franchi
What do you final question on this front. Do you have any thoughts on what ads on LLMs end up looking like? Everybody's going crazy saying they're going to be commerce ads. They're going to be just kind of native contextual. Like because you're so close to this stuff, you're the person I, I really want to weigh in here.
- [39:05]
Matt Eagle
So. So like think of it how retail media is more search than display. This is similar, right? Like a lot of what the power of search is that people reveal their intent. It's the ultimate contextual advertising. Right. And behavioral contextual together. Right. So this is like it's native. It's in, you're directly involved in what they're doing. They don't, you know, they're, they're in shopping mode rather than, you know, they're, they're, they're actually entering and revealing what they're doing and you're solving a problem. It's like you're solving, you're solving what they're doing problems. The wrong word. You're creating a solution for what they're trying to do right there. So it's not just you know, an ad, it's actually more directly relevant to what they're doing. So I think it's, it's a great evolution of search. That's why obviously Google is like all in and like with Gemini because like their core business is impacted here. But I think you'll see a lot, a lot of retail media impacted by this play. You know what's going on here as well.
- [40:04]
Eric Franchi
Ari, have a POV on this.
- [40:06]
Ari Paparo
Yeah, I, I think it is much more like search but it's richer because, because first of all you have more words. You know, search terms are 2, 3 words and prompts are 20, 30. You also have sort of this history concept where I'm continually asking more and more questions to narrow things out. I honestly don't know if Google uses immediately prior search terms in their algorithm. I don't think they do. Which kind of makes you think a little bit about how valuable this could be. My big hope, and I've said this on the pod before, I've even said this to some of the people who work at these companies, is that they don't assume it's gotta be the same ad format as search. You know, short text links. I want to see some images. Give me some images. I'm a human being. Picture tells a thousand words. Let's get some images.
- [40:53]
Matt Eagle
This, let's think for a minute the way the human brain works, the human brain is a memory making machine that we think actually 95% of human thought is more emotional than rational. And it's actually image based, not word based. So we think in metaphors and there's actually 20 to 30 universally relevant metaphors. Like think of like, like a good neighbor all you know, State Farm is there. The metaphor is neighborhood. Right. And so really good advertising often taps into our metaphors. Coca Cola Hilltop, the iconic ad is a connector. Lou Carbone taught me about this. He's one of the, he invented the phrase experience management and he very much about the importance of emotion and memory. So really good advertising taps into Emotion taps into imagery. The brain works that way. AI is increasingly about images as opposed to language. There's actually a risk here that if we focus so much on text analytics and bots mining text, we actually lose what drives the human being in emotion. So if bots are going to a site analyzing text, then that's the 5% of the iceberg that's above the water rather than 95% below it, which is emotion.
- [42:13]
Ari Paparo
Right. And this is the story I've played out in digital advertising to date is effectively that Meta is the top of the funnel and the middle of the funnel and, and Google's the bottom and that reflects in the ad formats and the targeting types and really everything else. So this is unknown really what, what AI based ads will work best with. I think it could, I think it should work in the entire funnel. Right?
- [42:39]
Eric Franchi
Absolutely. You just made me think of that stat from last year that was like eye popping which was Meta's reels are now on a run rate that's larger than US television. And it's all AI by the way. Right. From an art, from a targeting optimization perspective. So you can, you can see how this stuff can work. There's still some stuff to talk about. So Geo Matt, is this a category that we're going to have some, some board representation on? Geo aeo.
- [43:04]
Matt Eagle
You got me there. So like I, I, I think that we hadn't thought about a fifth category like that. Like we have talked about. Do we need people who are like luminaries like from like thought leaders or other association executives or things like that? What you're describing for me feels like it falls into the tech bucket like the, of our four primary categories rather than like, like for example we're not breaking out LLMs from other ad tech, you know, and giving them their own categories so that Geo would fall there in my view.
- [43:35]
Eric Franchi
Got it. That makes sense. Geosa. I mean it's a category. There's like probably a dozen companies, they've all raised now just with the brand light raised of this week, 20, $30 million and you know it's a, it's, it's, it's growing fast and I think it's, it's having a moment. There was an announcement this week that I thought was pretty cool. I don't know if you guys had a chance to take a look at it. So Evertune, which was one of the first, it's a team that has some experience from ttd. They launched what is effectively like answer retargeting. It's like A pretty cool way to use some of that data in a programmatic way. And they're doing it in concert with obviously TTD and index. I thought that was neat.
- [44:19]
Ari Paparo
Yeah. I read this press release and I scratched my head and said that sounds cool. I can't really figure out how it works, but seems pretty cool. Are they cooking people who are at second tier AI usage?
- [44:35]
Eric Franchi
There's light on that.
- [44:36]
Ari Paparo
Yeah. I think this category, I'm fascinated with it. It's already pivoted. The O is not appropriate in the acronym. They're not not optimizing anymore because basically the first pass at optimizing turned out to be kind of like gray hat, black hat SEO where the way to optimize your brand was to post a whole bunch of fake crap on Reddit and no one really wants to do that. So all these companies have pivoted and now they're basically analytics. It's like, what are the LLMs think about your brand? Is the question they're answering, which is really valuable and really complicated.
- [45:11]
Matt Eagle
Yeah.
- [45:11]
Eric Franchi
So important. We made an unannounced investment in this category and that's exactly what the company's focused on.
- [45:19]
Matt Eagle
I think this is a huge. It's not just the search but the whole social influence social optimization play. You see firms like Bazaar Voice making a big push around how to think about influence and social proof and getting the AI and how does an AI experience and look for social proof. So I think, you know, search and social have it. Are being hit. Are going to be changed through this. Yeah, absolutely.
- [45:47]
Eric Franchi
Final, final point on another emerging category that, that we're kind of obsessed with this idea of content marketplaces. There was an announcement. I guess it was an announcement. It was, you know, a scoop that Amazon is reportedly reportedly planning to launch one for publishers this year. So I think it was last week we had Microsoft, this week it's Amazon. This seems to be a real thing.
- [46:11]
Ari Paparo
It is really the talk of publishers right now. I had a publisher meeting, I won't say with who, and they were. They're very much looking at this and they said a couple things which I thought were takeaways or interesting, which was one, yes, we want the money and they should pay us for our content. Okay, big surprise there. Two, there's a huge difference between a deeply reported investigative journalism report that could cost tens of thousands of dollars to produce a certain, a single effectively essay about the topic versus throw away Stringer reporting on sports and finance. And Mark, it's not clear how marketplace deals with that. That's Kind of like point number two and point number three, which I thought was fascinating was, and it was a big publisher. They said we don't have the rights to our own content all the time. We may write an article, have a stringer write an article, we own the text, and then we put a Getty image as the main image. We do not have the right to resell that image. Or we may have a guest op ed. A lot of op EDS are guest posts or have, you know, rights sharing with the author, you know. And so there's actually a really deep rights problem when it comes to licensing content in an automated fashion.
- [47:28]
Eric Franchi
What a headache.
- [47:31]
Ari Paparo
I also had a separate conversation with, I think I can say it was with John Roberts from People, because he's kind of one of the intellectual leaders of this whole movement. And he was just basically saying the math is what are you making on ads on an RPM basis? Revenue per page, not per cpm. Right. And versus what's, what are you gonna be able to license it for? Because effectively you're replacing a human being with a bot in this case. And, and then there's other things to consider. Like you don't have to pay to get the traffic. Whereas humans, you have a marketing budget to get people to the traffic and stuff such. But as a first pass, that's, that's the way you should look at it.
- [48:11]
Eric Franchi
Yeah, that makes perfect sense. And with traffic down, you know, we keep hearing this 30% plus or minus stat being thrown around like it's not an insignificant amount of money you need to make up.
- [48:22]
Matt Eagle
We went through a trend years ago where the cost of content was, was out of whack with the revenue per content, which created a real challenge for content publishers that had lots and lots of islands of content that they couldn't afford to. They weren't able to monetize a lot of the content they were producing. I think, you know, for a lot of brands or publishers, that one CPG I worked with actually we quantified, it spends 300,000 man hours a year creating content. So there's enormous amount of investment in content that goes out there. One of the underappreciated things with Agentic is it will dramatically lower the cost of complexity of creating content and experimentation. When you look at advertising, DCO is largely a limited number of versions still because the brands and the agencies are worried about creating hundreds of versions of the content and experimenting with it. But now what you're going to see is the ability to create lots and lots of versions and experiment and A much larger scale goes up. Plus you're going to have the data signal to be able to optimize which content is used against which audiences in a way that's aligned to what's happening elsewhere in the customer experience. And so this, this, by the way, when you do this, you get a 50 to 150% improvement in AD effectiveness. So. So like this is not an efficiency play. This is an effectiveness play.
- [49:48]
Ari Paparo
Yeah. As a separate related note, you know, Nikita, I forget his last name. This the head of product.
- [49:54]
Eric Franchi
Product for.
- [49:54]
Ari Paparo
Yeah, Beer for the head of product for X. He posted yesterday something kind of like he must have been having a bad day. He said something like six months from now, 99% of all content in all channels will be fake.
- [50:06]
Matt Eagle
We call it Randy and I call it crap with a I, C R A I P instead of slop. Like, you know, is it creativity where creativity is spelled C R E parentheses AI and parentheses tivity. Right. So is it creativity or crap and slop? And there's too much, you know, stuff that is. Obviously people have learned to do performance marketing with YouTube and things like that. Too much crap out there. But there's going to be a lot of creativity where people embrace it and figure out what's working. And so there's. I actually think this is a. We think it's a creative opportunity. That's a creative revolution. That, that's why, you know. But Jim Lusinski at Northwestern, you know, refers to it as moving from authorship to codesmanship, from authorship to orchestration, where creativity is, is merged with science. We came up with some Personas which we published about recently in our subs. Winning, Winning experience where one of them is the molecular gastronomist, which is basically cooking with chemistry. You know, art and science together.
- [51:11]
Eric Franchi
That's super cool. Stay on X for, for one, for one second. There's so like they're pushing articles. There's so much slop with these articles all day. But there, there was that one that was viral. Something is happening. Did you guys read that?
- [51:27]
Ari Paparo
I did.
- [51:29]
Eric Franchi
Matt Schmerr. Yeah.
- [51:31]
Ari Paparo
Yeah. It kind of goes back to a little bit what I was saying about the super bowl, which is the majority of people have no Idea how much AI has progressed in the past 12 months. 18 months. I was talking to my wife about it. The ordinary non technical people still believe it's a great replacement for Search, but it hallucinates a lot and you can't count on it and it can't really replace you in your job. And they're Totally wrong. Wrong. And this is what the article was saying, which is basically, it's ALREADY Better than 99% of programmers. It's better than probably 90% of lawyers. It's better than 99% of customer service agents. It's better than 99 percent of account managers. Like it hit that level just in the last 60 days or so. And people need to update their preconceptions about what AI can do because you don't have to.
- [52:21]
Matt Eagle
I'll give you two examples. Real, real client examples. Don't hate me for not using names, but I will tell you enough to. You could probably guess, but. So like working with one travel client we were looking at, we built some Personas of who they were targeting and we essentially my point here is that the human's still in the loop, but you can do like in days what used to take months. So we created lots and lots of versions of the ad in days where we took their existing content and insights and ran basically prompted the AI to. In this case, we use wevo AI because it's been trained on millions of user tests. And we essentially ran the creative and it actually comes up with alternative copy and images that you can do and then the human beings iterate on them. So we're able to. And then we ran those through Facebook, through meta ads and we got a 50% lift compared to DCO. So the control was DCO and we got a 50% lift using the ads that were AI generated with human editing afterwards. So it's an iterative process. That's one example. Another is a one of a top five E commerce client. And what they're doing is they're trying to dramatically improve the workflow for creating content, copywriting, image resizing, image design translations. Like you can dramatically turbocharge the amount the workflow. That's what we're excited about is the discovery and the creative and the planning and then the measurement, not just the executing the transactions. The programmatic is a thin slice of the value chain. It's an important one, but it's actually the efficiency side of it is not what gets me jazzed. It's the effectiveness side that gets me jazzed.
- [54:12]
Eric Franchi
Totally agree. Hey, I think we're at time. You want to touch on one more thing or call it our.
- [54:18]
Ari Paparo
No, let's call it. This was a great conversation. Matt Eagle, thank you so much. You're here from Journey Spark consulting company. Also an author and podcaster and now involved with agentic advertising.org thanks for being here.
- [54:33]
Matt Eagle
Been a pleasure. Thanks for having me.
- [54:35]
Eric Franchi
Thanks, Matt. We'll see you next week, everybody. Thank you for subscribing to Market.
- [54:40]
Ari Paparo
New interviews are added every week at marketing and your favorite podcasting app, Sam.