Loading summary
Transcript143 lines
- [00:01]
Ari
This podcast is brought to you by Azerian. Azerian is one of the world's leading full stack digital advertising partners for 300 plus agencies and 2,500 plus brands. Their proprietary platform called Azerian Infinity seamlessly integrates the capabilities of owned and operated dsp, ssp, DMP and ad server technology to drive incremental performance, value and efficiency across the full marketing funnel. At Azerian, they offer custom audience activation, leveraging cookie identity data and contextual signals to engage users across digital ctv, OLV dooh audience in game channels everywhere else. Choose between direct or PMP activation, leveraging world class trading teams to deliver top of plan results or alternatively access audiences through a data connected PMP as a market leader in the sell side curation category. That's Azerian, the sponsor of this podcast. This podcast is brought to you by the Current Ever wanted to sit down to a candid conversation with the marketing leaders of the world's biggest brands? The Current podcast is your chance. On the Current podcast you'll find exclusive interviews with the experts and trendsetters who are on the front lines of digital advertising and they always leave the ad tech jargon at the door. Subscribe to the current@www.thecurrent.com or anywhere you get your podcast today. Welcome to Markitecture Podcast. We have an exciting episode today. We have Sir Martin Sorrell. He comes in and talks about AI. He talks about a lot of interesting things. Now we recorded this episode a week ago and as a result he was not able to give us his wisdom about the Omnicom acquisition. So you'll have to stick with Eric and I for the insights about that. Eric, what did you think about the Sir Martin interview?
- [01:57]
Eric
I mean, he's a legend, right? And just, you know, being able to chat with him is a treat in and of itself. His perspective on hold goes is, you know, I think very unique, very valuable and I think a good setup for the news of the week, which we'll talk about after, you know, is able to ask him, I think pretty tactically about AI and you know, how he thinks things shake out over the next year or so. I think that's a must. Listen, one thing about Sir Martin that I've always been really impressed with, and this is like going back from when I first had exploiters to the guy 25 years ago to now is he has the ability to retain numbers and stats and information like no one I've ever seen. Things like market cap, growth rate, gross margin, the rankings. Like I'm just in awe and he just, throughout the conversation, just peppers us with knowledge that if you go and fact check, it is like, always. Right. It's just like the guy's a machine, so it's just awesome having him on.
- [03:02]
Ari
Yeah, you can't forget he got his start as CFO of Saatchi and Saatchi. So he definitely has a financial orientation, but he brings to it the strategic side as well. Yes, we'll talk about that. We'll talk about sora, the new OpenAI release, Lina Khan being out. A lot of interesting stuff going on this week in terms of housekeeping. We have a new Ad Tech Explained article. So if you're not reading adtech Explained, you probably don't understand how the world works. You should go to AdTech Explained.com, subscribe to the newsletter. It's really high quality. It's really good content. And also markitecture will be at ces. We're running a program called Suite Suite. We're recording people in their suites and then putting on YouTube. So you'll see a lot of good stuff there. It's exciting. If you want your suite on suitesuite or if you don't even have a suite, call us. We will help raise your exposure at ces. It's basically MTV Cribs for ces. It's awesome Sounds Suite. And then on Wednesday of CS Week, we have a panel and happy hour in conjunction with Magnite at the Barbershop at the Cosmo. So that's invite only. So reach out to your local markets or birth. Probably not me because I'm kind of useless, but one of the other people that actually gets work done and they may possibly let you in and you can hear our insights and things like that. Okay, great. Let's go for this. So we'll be right back with Sir Martin.
- [04:19]
Eric
Let's go.
- [04:21]
Ari
So, Sir Martin, thank you so much for joining us. We're really excited to have you.
- [04:25]
Sir Martin Sorrell
Pleasure to be here. From Miami.
- [04:27]
Ari
From Miami. You must put in a lot of miles. How many. How many. How many miles do you have?
- [04:32]
Sir Martin Sorrell
No, no, I didn't. No, I don't follow, Margaret. Well, Mosey. Warren Buffett used to call it. He called us playing the indefensible.
- [04:41]
Ari
So we have a lot to talk about with your. I'm really, really interested in your perspective on recent acquisitions in the agency holding company world. There's been a lot. Flywheel was acquired by Omnicom, influential by publicist. And just today there's new news. Breaking news that IPG acquired a company called Intelligence node, which is like retail technology. So what do you think is going on? Why are holding companies acquiring these sort of technology companies, these that have hybrid offerings?
- [05:12]
Sir Martin Sorrell
They're trying to catch up, I guess. So it went, you know, there were two waves that really data and digital media that have swept the industry for the last, I guess 20, 30 years, you could argue. And the holding companies with a traditional model have been playing catch up. You know, when I was at wpp, we started to bring catch up on technology in the 90s with the Internet and then data as well. Publicists made early moves and I think probably have benefited. You mentioned that in relation to influential. But publicists with epsilon and with Sapien made early moves which were initially criticized, I guess, and probably were difficult to integrate. But they seem to have got their act together because now actually they're going to surpass WPP in terms of size. And I mean, interestingly, because of the demise of WPP or the pressure, it's not just in the advertising business, they're being surpassed by publicists. But it looks as if Ipsos does its deal around Canton Media, which is in the press being discussed, or that deal is being discussed, you'll have another French company, Ipsos, in the data industry becoming a preeminent force too, or one of the preeminent forces. So it's interesting what has happened, but I think the basic answer your question is they are traditional models. The older companies are traditional models and they had to adapt as rapidly as possible in a world which is being dominated by digital media and by greater digital media. This year are going to be 700 billion out of a trillion dollars of ad spend. So 70% share. And the biggest media organization in the world is Google at 25% of that market, including traditional. And then you have meta at 15%, Amazon at 6%. I mean, the numbers are easy because it's trillion dollars and those are the big three in the West. And then TikTok is about 40 billion outside China, outside mainland China. ByteDance as it holds 150 billion. So basically the answer to Christian is they're trying to clutch up and invest as rapidly as possible in a data and digital world. And I'm not sure IPG is in the same league as Kuvlis. I mean, you've got sort of two leagues into the holding companies. You've got the Premier League, as I'm, as I'm here in Miami with FIFA, you've got the Premier. I can use a football analogy. And the Premier League is. Publicis is leading the Division, I guess Omnicom is second. I guess of us you put third and then division one or the lower league and you probably put I guess IPG maybe ahead of WPP or WPP and IPG and Dentsu, those would be Division. Division 2 and Division 2 is all suffering in relation or being outcompeted by the Premier League at the moment. Whether that switches or not, I don't know. But the battlefield is around digital and breakable.
- [08:27]
Ari
And how does it affect these companies financial positions to have a mixture of what you call services and technology? Is it accretive to the margins to have these technology businesses?
- [08:39]
Sir Martin Sorrell
Well you say, I mean they're technology services businesses. I mean they're not often, often not creating the technology. I mean the basic platforms, that trillion dollars or that 700 billion in digital. Google is 250 meters 150 and TikTok, as I say it's 40. So 500 billion out of the trillion, 50% is in those, those four platforms which is quite, you know, everybody says the world's got more complex. I think it's got simpler. And these companies are big companies. You know, if you include companies like Apple and Nvidia, they're what, three and a half trillion dollar companies. So they're as big as countries. If you equate market cap to gvp, it was big as Germany and probably will be bigger than Germany in the future. So I think they're countries and these companies are huge. They're the ones that produce the technology. What the agencies do is use their technologies and adaptive for their uses. I mean you didn't mention the media ocean investment that the three Alden funders have made as of this morning. You know, that's the, the old platform, the old media praying and buying platform that the agencies use. Interestingly, public is not included in that, which I thought was interesting. But that's again, you know, a system which was being used. But again they're using other people's technology and they're using it as services. So when you say how does it affect margins? The core margin of the holding companies I guess comes from the media business.
- [10:16]
Ari
Right?
- [10:16]
Sir Martin Sorrell
Right at wpp. I think even though I think on revenue probably this is probably eclipses wpp and I see they're hiring Snoop Dogg to celebrate the French victory. What they're trying to do is really driven by their media planning and buying operation. And I think WPP will still be bigger in terms of billing. So they do about 60 billion a media people forget that the true revenues of the holding Companies are not their net revenues or indeed their gross revenues, it's their billings. And their billings, certainly because the WPP discloses billings or used to, is around 60 billion. So there are big businesses. I mean that 60 billion equates to assumption in terms of, in terms of revenues. And those are the true revenues. And people forget that. I mean it was always amusing to me. Clients would say when I was at wv, you are our biddy supplier. And I would say, I wish we were. And the procurement had given the CEO or the CMO the figures for Billy's, they wouldn't give the figures. So I think the guess, the hefty and the power from the holding companies really comes from media business and they're trying to adapt the technology to that position.
- [11:34]
Ari
What was the insight that started S4 like? What did you see as the opportunity?
- [11:40]
Sir Martin Sorrell
Four things. Digital only, beta driven, going to market as faster, better, cheaper. And now with AI more, which is a risky thing to say that say you're cheaper for the obvious reason, but we thought it was the best way of doing it. And then a unitary brand, those were the four, the four strengths.
- [11:58]
Ari
How do you feel it's playing out? Is it working?
- [12:02]
Sir Martin Sorrell
Digital only when we started was 50, 55% of the market. Now 70. Data is even more important in a AI age. I mean what are the two keys for AI? For the incumbents in AI, data and distribution are key. Faster, better, cheaper is the mantra that people are increasingly using. And with AI more too. And then unitsu brand, everybody's trying to provide end to end solutions and get everybody, you know, even Omnicom, which had, you know, very strong brands or has very strong brands, bvdo, ddb, TBWA is now amalgamating them under, what do they call them? Omnicom Advertising Group OAG. So I.O.A.G. is effectively going to become one agency. I mean they're going to try and try and simplify with the back office. Going back to your question of margins, probably the answer is your question is it's not helping the margins because the way they're going to get margin improvement is by consolidating back office immediate. Ocean is another example of that. Why are they doing that? Because they want simplified back office functions.
- [13:10]
Ari
The MediaOcean thing was also in support of the innovative acquisition because they want these companies to switch not just their billing, which already is locked in, but also the ad serving. That would be my guess about that, yeah.
- [13:22]
Sir Martin Sorrell
But it's simplifying the systems that the companies use.
- [13:26]
Eric
Can we talk about AI? You Sir Martin gave a talk, I think it was probably 18 months ago, and you sort of laid out the framework of how AI could disrupt agencies. And I'm doing this from memory, but it's probably what you said, but effectively it was like, you know, you think about it in terms of, you know, five areas. Creative automation, personalization, media planning, buying measurement and like data sharing. Is that your framework?
- [13:54]
Sir Martin Sorrell
Yeah, yeah. I mean it's visualization and copywriting, hyper personalization at scale, media planning and buying, general efficiency for agency and client and then democratization of knowledge, providing knowledge across an organization and planning.
- [14:10]
Eric
You said it way better than me. Obviously. Now it's probably 12 to 18 months later. What's your view on the impact that AI has had so far on the agency landscape? And where do you think the real impact is going to be in the next maybe 12 to 24 months?
- [14:27]
Sir Martin Sorrell
So I think the first year, let's say it was 23, although Google acquired DeepMind in 2014 and Eric Schmidt told Matt Britton in 2004, the London Olympics, that AI recognized Kots for the first time in the system. But 2023 was, you know, wow, 24 was how. And maybe 25 will be. Now I think those are the. That's. I think Karim Lakhani, the be a business school professor leading DQ supplied the third element, you know, now, and I think that's the issue when you look at use cases, case studies in the industry, in the marketing and McKinsey says the marketing function has been most affected by AI. Of the functions inside companies, marketing and sales has been most effective as we. That's their view. We saw that analysis in Riyadh FII a few weeks ago. If that's the case, and I've got no reason to say that it isn't, we are seeing at a micro level, what we're seeing is workshops, audits, tests, pilots. We're not seeing wholesale transformation. The GM transformation that is taking place with their marketing, to my knowledge, is the single most important transformation. We're seeing sort of not tinkering, but we're seeing experimenting. And that's why I went through those, you know, wow, how. And now all this capacity is being built. I mean, there are, there's a strong point of view and it goes back to your question about technology and adapting traditional models, holding company models to a new world. What do you do? And in listening to Jeff Bezos being interviewed by Andrew Rossokin yesterday at Dealbook, he made the same point that you always sort of underinvest. I think Bill Gates has said the same thing. Sundar said the same thing in relation to Google. When questioned about the investment they're making in AI, he'd always over invest rather than under invest. And I think the natural tendency, and Bezos said this was for people to overestimate the risk of what they're doing, which makes them more cautious and obviously sometimes it doesn't pay off. But he was saying it was an interesting point that, you know, entrepreneurs naturally, well, people generally tend to be risk diverse or overestimate risk and not take enough risk in doing what they're doing. And he was referring to when you raise the first million dollars for Amazon and beyond. It was an interesting conversation in that sense. So I think that's, you know, that's the way I see it playing out. Playing out generally. I don't. What's happening now with AI. Amazon spending supposedly a hundred billion on AI, which Bezos is heavily involved in. He said in the interview that's what it's focused on. Google is sending 15 billion a quarter, so 60 billion a year. And Microsoft is spending the similar amounts. So huge investments in a. Huge investments in capacity. Many people in that council said that they don't think there's an over investment in capacity. What I'm concerned about is the use cases. And I see a hesitation, not a hesitation, an unwillingness to, to jump in. There's a willingness to do experiments, to do costs, to do audits, to do workshops, but to jump in on a whole, you know, if business is good, the mantra is, if it ain't broke, don't fix it. So you tend to be that. That's what I say.
- [18:24]
Ari
We've covered this on this pod a number of times that the various agency holding companies have given out investment figures in AI that are pretty modest, like 20, 30 million dollars a year, with WPP probably the most, like 100 million. But it feels like it's very much focused on creative development and storyboarding and sort of the obvious use cases.
- [18:43]
Sir Martin Sorrell
I don't know where those triggers come from.
- [18:45]
Ari
They've publicly said them. I don't know what. They.
- [18:50]
Sir Martin Sorrell
Hadn'T explained what they are. They have not. They threw out a number and I don't know what it means. Right. I don't know where that 100 million that you quote is. What is it?
- [19:03]
Ari
Yeah, and they're not building our models. They're not.
- [19:06]
Sir Martin Sorrell
Well, is it creating their own models? Is it adapting other people's models? I mean, what is it? I think it's just, I think, you know, it's in the panic to do an investor presentation. Somebody asked somebody to come up with a figure. You know, if people throw up that figure, there should be some sudden to it and they should be called to account for it.
- [19:27]
Ari
Well said. I mean the old saw is that people overestimate the impact of technology in the short term and underestimate it in the long term. If we think about like five, ten years from now, do we think the agencies are going to be radically different in their operations, radically different in their staffing?
- [19:44]
Sir Martin Sorrell
Well, I think so. But you know, I may be wrong, been wrong before. I don't think 250, 300,000 people are going to be doing media planning and buying. I mean OM Gottlieb used to call media planning buying media investment management.
- [20:00]
Ari
Right.
- [20:01]
Sir Martin Sorrell
And he was dead right. If clients are spending a trillion dollars, why are we doing things semi manually, you know, spreadsheets, manual spreadsheets? Do we think blackrock and Blackstone who have a multiples of a trillion, but, but do they do things algorithmically, do they do things manually? They do them algorithmically and the same thing. If 700 billion is being spent on digital ad platforms, that analysis should be a mechanical algorithmic analysis and the agencies should be validating that with plants. If the algorithm spits out a distribution of a certain distribution between the platforms, between the six platforms that I mentioned, the three Western and the three Eastern, that should be checked by the agencies. So what I personally think is going to happen, these platforms where the concentration is and obviously it's subject to very regulation risk, I think they should be self regulated. That's a different point. Unless you give your arm of the other Exxon ATT and break them up. I think that self regulation is what you're doing. I don't think regulatory agencies can go at the same speed, but basically these are going to become bigger platforms. AI is going to make Alphabet Meta, Amazon, Alibaba, Tencent, ByteDance, Microsoft, Apple, Nvidia, Oracle, Salesforce, Adobe, Musk. More important, I'm including OpenAI under Microsoft, Sam Altman and Mistral and Gene42 under sort of Microsoft. Those are going to become even more important and more dominant and clients are going to have to get closer to those platforms and the agencies are increasing. You're going to be validators for them. You're not going to have 250, 300,000 people in media planning. By doing that 60 billion of media at WPP it's going to be a totally different scale of people of human input. The outputs are going to be Even better. And the people who interpret those outputs are going to be even more skilled. But you're not going to rely on the 25 year old media planner or buyer anymore. You're going to rely on an algorithm that tells you. And the other thing that's coming down the pike, you know, you mentioned investments in ID5 and EV Scientific. I think things like that here. We just invested in our venture fund in Runway. We're looking at a Chinese or Asian equivalent called Minimax. These are text to video, text to audio models, which in a way we're already getting to because small and medium sized businesses, which are, let's say 60 to 70% of Alphabet, meta and Amazon business, maybe even more. You know, if you're a small businessman or small businesswoman and you have a. You want the platforms to help you build content, they can. And then if you want to create a media planning and buying program or distribution with pmax or with Advantage plus, you can. So we are, we are only a few steps away. I don't know how long it seemed to take to have a fully integrated model where you have a verbal prompt or a text prompt and you get your content, you get your plan, you get your buy and that will be a fully automated system.
- [23:27]
Ari
Yeah, radical transformation. I want to, I want to talk about one more subject before I let you go, which is the double click Google situation, the antitrust situation. I just talked to Rob Norman this week and he told me a humorous story that he was out at a club at two in the morning in Spain or something. The minute the DoubleClick Google acquisition was announced in 2007, you called him and said, what should we do? Rob had no idea what you were talking about. You said, we cannot afford to not be in that business. That was a secondhand quote he said from you. Can you tell us what you were thinking in 2007 when this happened? I know you were sort of mildly against it. You had public comments that were against the deal.
- [24:12]
Sir Martin Sorrell
What were you thinking in those days? I used to describe it and maybe it's old age that has changed the view. I don't think I used it then. It came from a woman who worked with me, a Harvard Business School graduate. And anyway, she said to me she coined the phrase frenemy.
- [24:30]
Ari
Frenemy, right.
- [24:31]
Sir Martin Sorrell
Yeah. And we, and we used it on Garen and I guess I saw it. Let me just tell you a very frank story and I don't know, I won't say which platform it was and which client it was, but about it must be about 8, 9, 10 years ago, one of our biggest clients of WPP, the non CMO of that client, I called the chief sales officer of one of the biggest platforms and said, you know, can you come and see me? And you can't. You went to see him and he said, why can't we have a direct relationship here? Why do we have to work with wpp? Why can we do it and the hood and sales organization? Well, we're a capital intensive organization, we're not a labor intensive organization. And all I would have to do is we create the agency. There's no point in me doing that. If that conversation took place today, the answer probably would be very different. The head of sales at the platform would say, you know, it's true. I mean, Google has 180,000 employees, despite WPP's travails, has 100,000 still. But look at the relative market caps. You know, you look at 2 trillion versus 10 billion. So it's a different business now. And I think the answer would be yes, we can do that. We'll need the agency to validate. But the role of the agency, so that frenemy word is now not appropriate. I think the, or the, the friend part of it has become far more important than the enemy part of it. So going back to that 2007 whenever it was, I saw that as a competitive threat for the reasons I just laid out now, given the decline of labor intensity and the rise of capital intensity, I mean, what are the two things you need? You need data and distribution. You need compute power and energy. That's in the favor of the big companies. But basically these players are going to get bigger and more important.
- [26:33]
Ari
Do you think Google should be broken up?
- [26:36]
Sir Martin Sorrell
No, I think it's ridiculous. In the world that we're in, where we have US China, relations are bottom and probably going to get worse. Where we have Russia, Ukraine, which may be solved unsatisfactorily. You're not going to curb President Putin's expansionist urge. And where we had Iran, maybe there's better news humming with the Trump administration around Iran and Israel and the Saudis. But in a world which is fractured, having big tech is fundamentally important. Not just from technological advancement, but from a defensive point of view. Forget about offense. From a defensive point of view and what's really interesting to me, and I think this is true, President Xi say he was going to cut up Alibaba into eight pieces or nine pieces. Has an atomic. Why hasn't it happened? I think because they understand that big Tech is important. We see it in Ukraine, we see it in the Middle east with the drone wars, technology and private, private companies. You know, let's take the poor old U.K. you know, we're an impoverished nation and we're spending two and a half percent of GDP on defense. We need to spend more. We haven't. Two and a half percent is not enough. The head of the armed forces, government ministers been on television recently saying it's fundamental to have private public partnerships for defense and indeed for offense and for cyber security and technology. If you break these companies up, you lose. And by the way, we in the uk, we don't have them. We're going to create, going to be relying.
- [28:24]
Ari
You're still trying to break them up.
- [28:26]
Sir Martin Sorrell
We're going to have to rely on Alphabet and Microsoft and Palantir and everybody else to dig us out of the problem that we're in. So I think it's fundamentally important to have big tech. So self regulation is critically important. And these companies, which are big nations, you know, we're just being through that. They are as big as they are countries effectively, and they can make or break precedents. You know, Bezos decides, he owns the Washington Post, decides not to recommend anybody. Why does he do that? He got a very eloquent defense of it. But there are other reasons too. So they're inextricably leaked in and I don't think they can be broken apart. And if you break them up, I think you have, you have a big risk from a defense point.
- [29:15]
Ari
The really interesting point. I think we'll really, I think we'll call it on that. So, Sir Martin, thank you so much for sharing your insights. Strategic view with us is really appreciated.
- [29:26]
Sir Martin Sorrell
Okay, I hope it's good.
- [29:29]
Eric
It was. Thank you very much.
- [29:30]
Sir Martin Sorrell
All right, very good. Thank you guys.
- [29:33]
Ari
This podcast is brought to you by Inmobi. With the average person spending more than five hours a day on their phone, you have to ask yourself, what's your mobile strategy? Inmobi Advertising helps brands and publishers thrive in the modern mobile era with AI powered solutions to acquire, monetize and delight the right users at scale at each step in their journey. From a pioneering DSP built to accelerate app growth in a privacy first world to a leading exchange and ssp. Forrester just named a mobile must have. Inmobi advertising has the tools to capture user attention on their most personal device. Check out their Spotlight series on the advertising forum for key strategies to mobilize in 2025 and visit go.inmobi.com pod that's go.inmobi.com pod to learn more in mobi driving real connections.
- [30:20]
Sir Martin Sorrell
All right.
- [30:21]
Eric
And we are back with news of the week. Big Holdco acquisition and announcement. The announcement of the long awaited announcement of Sora from OpenAI and the Kruger's acquisition of Albertsons was denied. And some other stuff. So lots of stuff to talk about here and I think the first two are interesting because you don't seem to care about either of these things.
- [30:44]
Ari
I don't.
- [30:45]
Eric
We should have a little debate here.
- [30:47]
Ari
It's below my line.
- [30:50]
Eric
Oh God. For those who don't know what Ari is referring to, there's a viral moment I think a year or two ago on the all in pod when one of the co hosts referred to something about a certain ethnic group and it took the Internet press storm and it's still weighing on him to this day.
- [31:12]
Ari
Unlike that person, I'm not sociopathically dismissing the suffering of millions of people. I just don't care about agency holding companies.
- [31:21]
Eric
Well, also suffering people at least maybe not millions, but hundreds of thousands of them.
- [31:27]
Ari
Let's set the stage here.
- [31:30]
Eric
Unfortunately, after the announcement or after the interview rather with Sir Martin, we had an announcement that rocked the agency world. So Omnicom announced that they were buying Interpublic. So let's talk about a couple things here. To set the stage. We'll put the link to the official release in the newsletter. But there's a lot of aggressive language in this. So the first thing was it refers to the combined company not as an agency, not as a holding company, but a marketing and sales company, which I thought was very interesting. Refers to industry leading identity solution. You know, presumably axiom refers to greater capacity for investments and acquisitions. I think nod to where they've been less active than others relative to, you know, WPP and publicists. And also 750 million in annual cost synergies, which would not make me I was an employee at those agencies too comfortable.
- [32:32]
Ari
Yeah. Cost synergies is a synonym for real estate and people.
- [32:37]
Eric
Yeah, exactly. So all right, I want us to do a mental exercise here because you and your newsletter this morning made me laugh out loud. Basically saying like you're a tech guy, you don't care about this. You don't understand why anybody cares about this. I'm oriented similarly to you, but I think the reality is like Ari, if you know, for some reason sub Fortune 500 brand hired us to be CO CMOs of their brand, we would probably call one of the large hold cos first to help Us think through what we should do and plan out our 20, 25.
- [33:12]
Ari
Yeah, yeah, you know, I'm using the.
- [33:14]
Eric
Exactly.
- [33:14]
Ari
I'm using the phrase I don't care very glibly. These companies do work for clients. Clients presumably are happy with that work because they pay them. There are many good people that I know personally who work there. When I say I don't care, unpacking what I mean, I should see a therapist about this caring issue. But it's like, is this a substantial bending point in the story of technology, advertising, the media, et cetera, or is it just tactical? It's just mushing two companies together. Some people lose their jobs. It's a shame. So that's what I need to be convinced to hit my caring threshold, that there's something bigger here, that we would look back and say, oh wow, that changed everything. And I'm sort of at a loss to say that.
- [33:58]
Eric
Good point. So the language in the press release aside, assuming this goes through, this would create the largest holding company by revenue. So it's 25 million, I think, plus or minus, which is larger than WTP number one. And then number two, it would create the largest network expressed as the just number of agencies that exist underneath what would be a mega umbrella.
- [34:24]
Ari
Yeah.
- [34:25]
Eric
So in an age where, and this is going to be the question, in an age where scale matters or maybe doesn't matter, this then points to a formidable opponent for the other hold codes.
- [34:38]
Ari
I agree with that. The inter holding company competition is obviously very affected by this. But this is like four or five supermarkets in the same town just reducing the price of bread or something. It doesn't change the nature of how people eat or how people shop in that metaphor. I'm kind of interested. So here's my take. Big picture, which is we've been covering on this podcast for a while all these acquisitions that these holding companies are doing, which are sort of techie. They're not exactly hardcore tech. They're tech E. They're things like Axiom for data or Flywheel for commerce, et cetera. Your perspective on this. But think about the margins of tech companies. The reason why software has been such an enormous value creator over the last 50 years is because you create the software once you sell it, every time it's near 100% margin every time you sell it. SaaS was supposed to be like that as well, except it turns out you have to hire a million salespeople to sell it. So the margin is not as good as people thought. But that's not the point. The point is, are we seeing this new bifurcation? Maybe there's some large segment of what we call software that doesn't have those large margins and instead is intensely service oriented. And you have a different kind of company that ends up coming out of this, that maybe is not profitable enough to be a standalone public company. But as part of a Holdco where you have lots of people and lots of services, is an attractive return on investment for shareholders.
- [36:05]
Eric
I like that thesis. The question is, are you pointing to that this currently exists within Dercom or Publicis or wpp or that's how this might come to fruition in its more final state?
- [36:21]
Ari
I think a growing portion of the revenue of these companies is from things that we might think of as software. And that seems to be a trend that's accelerating rather than decelerating. And scale matters a lot there because you do have to pay up to buy these software companies and you have to make money on them and have a lot of staff. So maybe this is a hypothesis, maybe we would see over the next 10 years with AI and with software, these companies increasing their margins, having fewer employees, but being just as important because they lean into what I would call low gross margin, valuable software opportunities. I'm just making those up as I go. What do you think?
- [37:03]
Eric
I get it. No, I get it. Right. So a tech enabled services company that helps clients make the decisions on all of the different technologies they might want to use, be orchestrators in some way, maybe own some of those technologies as well. I think I get it. But that's not how they make money today. They make Money today with FTEs and media buying and principal buying, all of these things that are questionable in that sort of future. So I think there's a real transition that needs to happen, a real path.
- [37:42]
Ari
Yeah. The bear case is that these companies continue to make money in ways that are hard to decipher and there's pressure from clients to stop, stop doing that. And there's also a pressure from AI to do things more efficiently. So you can end up on the wrong side of this equation where you have declining growth rates and you're scrambling to try to maintain your profitability.
- [38:02]
Eric
So the bull case is there's a future where scale matters. In a technology led agency form, there's a future where scale does not matter if this is all about media planning and buying, because increasingly the large platforms are taking a disproportionate share meta. And Google TV itself, where scale still matters, is increasingly fragmented. And going to ctv, which is now the domain of programmatic. So it's really about what type of scale is going to matter in the future. And if it's the bull case of tech enabled services, you could imagine these companies pivot and evolve and become important yet again in a new way.
- [38:48]
Ari
Yeah, I think you also made a prediction like earlier this year that we would see a AI first agency, a new one develop. Not that old ones would fire people and use AI, but you could have like the next VaynerMedia could be created by like two people in a room with a lot of AI. And I'm still waiting for that. I think that's a spot on prediction, but maybe we're another six months away from that.
- [39:10]
Eric
Yeah.
- [39:11]
Sir Martin Sorrell
Yeah, perhaps.
- [39:13]
Eric
There were a couple of things posted this week all over the Internet. Mark Penn had a good take on this. We should put it in the, in the newsletter. He's obviously a competitor. He's the CEO of Stagwell, which is a smaller but, you know, still significant holding company. He had a, he had a really good take on this. And I think just kind of question the point that we're landing on here, which is, you know, how important is scale in the, in the future? Right. But this is going to be interesting to watch and will it get passed? Right, that's another big question.
- [39:41]
Ari
Yeah, I mean, it is an interesting question. I imagine the Europeans will have a lot to say about it given how important European business is in the agency world.
- [39:50]
Eric
Yeah, absolutely. Final point here, I guess. Also timely. Terry Kawaja had a great piece in Ad Exchanger a week or two ago on why Publicis wins. And he pointed to three reasons why Publicis has won over the course of the past year or so. Again defined as public market performance. And he said they win on tech, they win on leadership, and they win on smart deal making. And I question if putting Interpublic and Omnicom together satisfy any of those.
- [40:24]
Ari
Yeah, I don't know. Those are really good questions. I'm really not sure.
- [40:29]
Eric
Got it. Okay. All right, well, we'll, we'll continue to watch this one. I think it's also interesting that this week was the announcement and launch of Sora, which I don't know if you trolled me on X, but you were basically like, I don't know what this is. You should, you should talk to me about it.
- [40:48]
Ari
No, I really don't know what it is. I've got so much capacity in my brain. You put one fact in, one fact falls out.
- [40:54]
Eric
Well, I think Sara is super important and I think also as we question what the future of the agency is and you know, what the, what the future of this whole thing is as it relates to AI, it's like one of the most important developments. So SORA is a text to video generator from OpenAI. Everybody that's in front of their computer or on their phone or whatever should just go to sora.coms o r a.com and you can see some of the outputs of Sora in a Explorer page kind of fashion.
- [41:27]
Ari
Sora the explorer, Is that what you're saying?
- [41:29]
Eric
Wow, that was really good.
- [41:33]
Ari
I need a map. Sora, get me a picture of a map.
- [41:37]
Eric
Yeah. So the Explorer page is inspiring in terms of what SORA can create, but unfortunately the demands hitting this page and the services, so much so that nobody's actually been able to log in and play with the thing. But sounds great. Exactly. You could see examples of the work.
- [41:57]
Ari
Okay.
- [41:58]
Eric
But what I read in terms of the. Some of the features that you can use on Sara and if you map it to what, you know, Ad Creative requires in an AI environment, I think it's like pretty awesome. So with sara, a single prompt. So make me a video of Ari doing space exploration in a purple jumpsuit. You will get four video generations. So, you know, it's not just sitting there trying to painstakingly create one video. Sora just creates four for you. Right in the beginning is number one. Number two, you can blend videos. So have you in a purple jumpsuit, have me in samurai armor and have us go on some sort of exploration together. So you can do that. You could build a video clip just from an image and you can actually storyboard. So a video editor that lets you basically generate and lay out clips across a timeline to create complex videos with multiple shots. That is like a commercial content marketing piece, a mini movie. Right. These are going to be short clips. So if we back up and we think about what that will do for people, it's pretty mind blowing.
- [43:14]
Ari
Yeah. No, I'm as awed by these AI demos as anyone else. I use AI in my work pretty regularly. I mean, I write my own newsletter, but just FYI. But I use it for some images and other things. But I'm sitting here in the trough of disillusionment a little bit. I know it's going to change the world, but moving from demos to really important real world use cases isn't happening as fast as I would like. Like that sounds great. When are we going to be able to realistically really use stuff like that to really create 30 second, commercials or music videos or whatever that don't look like they're AI generated because the burden is much, much higher. And what you see is what some people call slop or just like easy to write summaries of things. And stuff like that is happening very rapidly. But the demos are not translating as much as people thought into life changing experiences.
- [44:10]
Eric
Yeah, that's fair. I think there's two, two points there to believe. So number one, you're thinking back to what Sir Martin just said a few minutes ago. 2023 was the year of. Wow. 2024 was the year of how. 2025, the year of now. So if we follow what, you know, Sir Martin was, was saying, we'll probably see it happen in 25. But if also you think about like what maybe the most interesting ad tech use case is, creating ads for ctv. Creating ads for CTV for either small businesses or performance based marketers, where I think the bar for, you know, beautiful tier inducing creative is not there. I think that can be like one of the areas where you start to see this stuff happen in a really interesting light and it's personalized and you can use data. I think that's the thing that I'm excited about seeing and I hope we, hope we see it in 25. I think Sora could be the platform that helps accelerate it.
- [45:06]
Ari
Yeah, I totally agree with you. I mean, let me torture an analogy time. You know how sometimes you buy spices for your kitchen and you get like the turmeric or the cardamom and then you cook whatever, they're great. But then sometimes you can buy like some spice that's pre mixed, like the Indian curry spice and it's got like 17 things in it and then you use it too much and suddenly your family's like, why does everything taste the same? Because you've got this one Indian curry mix and you're not like actually doing the work to figure out the balance of spices. And that's where I feel like we are with AI right now.
- [45:39]
Eric
Yeah, yeah, I think that's fair. I think that's also something that with smart prompting you could address. All right, let's move on. Number three, you put this one in the docket. Leah Khan is out. And what I think is interesting is right after that, the Kroger's acquisition of Albertsons was announced that it was being blocked. Big news. If we go a step further, like what does that mean for the audience of marketexture? There's implications of what could have been or what you know, and maybe the next administration might still be a real significant player in retail media. So the RM's face is interesting. Right. Like we talk about it a lot. It's still like really Amazon is just the behemoth in a way that you know, Google Meta in other parts of digital ads aren't. They're like 70 odd percent of the of the RM's face. Walmart has 8% and then the rest of it is basically a long tail of 16% plus or minus. The Kroger's Albertsons tie up could have created just by virtue of scale of number of stores and number of consumers like at least something that could be competitive with Walmart. And that was going to be interesting but it's on hold.
- [46:59]
Ari
Yeah, I'm not really sure about it. Seems like the next administration is likely to be much more pro business but still anti tech. And so that's an interesting line to kind of balance against. As for this particular deal, I mean I think that it's very politically advantageous to say that you're doing things to keep grocery prices down. Like that is, that's certainly something that both parties and a lot of people in Washington would love to be able to put on their resume. So I think maybe this was the wrong deal at the wrong time.
- [47:34]
Eric
Yeah, that makes a lot of sense. But again I'm just kind of bringing this to what matters for this audience. I would still like to see an RMN with the scale and capacity to you know, compete. Right. With a, with a Walmart. Kroger's has, they've got their version of other which marketing falls into and others are $1 billion business and it's you know, I think growing 20% year over year. So there's, there's a real thing here that you know again depends on with the new administration what happens. We'll see.
- [48:10]
Ari
Like it was just team up on retail media like this.
- [48:13]
Eric
Yeah.
- [48:14]
Ari
You could do a consortium play, you know, a bunch of non Walmart grocers get together, create a retail media system. I don't think that would be blocked. You never know.
- [48:22]
Eric
Yeah, absolutely, absolutely.
- [48:25]
Ari
Without wall we can't have coalition.
- [48:29]
Eric
Some of the best data for sure. All those, all those shorties.
- [48:33]
Ari
What's a shorty in this context?
- [48:34]
Eric
It's a sandwich. It's a short sandwich. It's like six issues rather than.
- [48:38]
Ari
All right, I thought you were talking, you're calling out to your shorties in Philadelphia. That would be a little different.
- [48:43]
Eric
No sir. Who do you think you're talking to?
- [48:45]
Ari
Yeah, Ms. Raunchy is not happy hearing about the shorties.
- [48:49]
Eric
No. Oh, no, Ms. Fred. She knows the shorties. What are we doing? Okay, a couple of the small ones that we might want to touch on. So you put this one here. Loop Me acquired Chart Boost from Zynga. Tell me why this is interesting.
- [49:05]
Ari
So Loop me. We don't talk about that much because they're based in the uk. They're a sort of a rocket fuel, like, you know, buying platform. Kind of competitive to Quantcast, I guess. They claim they have AI. Who knows what they really are up to. But they're of scale. They're not small. Chartboost was sort of an abandoned asset. It was a mobile ad network that would show offer walls to consumers, like, you know, get extra points in the game. If you do this offer, you buy something, a Starbucks gift card, stuff like that. And Zynga bought it. And Zing, I don't think is doing that well. So this deal is sort of just kind of a potpourri of companies that no one thinks about mixing it up and trying to smell good. So I'm interested. I want to see what happens. I don't really know very much. I can't really prognosticate as to what it means. Presumably it means that Loop Me is getting a lot more serious about mobile. They'll have SDKs and a lot of apps, especially gaming apps. Maybe they're using the AI to. To lean into that market and maybe get some of Applovin's juice on their business.
- [50:06]
Eric
Yeah, yeah, that makes sense. Okay, cool. And then I saw an update from Google. Google had some ad manager updates for ctv. These just seem to be somewhat like market parity tools. Things like obviously buyer connections, ad control, transparency, ad delivery optimization. Is there anything here that you think is interesting from a CTV standpoint?
- [50:33]
Ari
Not really. I mean, Google in the ad serving for video world remains pretty distant second to Freewheel, but also nowadays second to companies like Publica. Generally the market is fragmented such that Google does well with companies that are online only and Freewheel does better with companies that are broadcast and Spring Surface do better with OEMs. But that's not entirely true. So Google's been in a niche in this business for a long time. I think they have to keep investing. There's nothing really that new here.
- [51:05]
Eric
But YouTube is like the number one CTV app. Right? So does this have any implications to make YouTube better for buyers? And is that the angle here?
- [51:17]
Ari
Well, this is an ad manager update. Right. So that's sell side. Right. The number one thing Google could do to make YouTube more attractive would be to let the trade desk buy it. But they won't do that and that would help publishers make more money. The other thing they could do is make publisher ad servers work better on YouTube because they have these pretty crappy agreements with Freewheel and springserve. I'm not even sure if Spring serve is allowed, but when you watch the NBC channel on YouTube, you're getting a really even YouTube TV. You're going to see a lot of no ad breaks, missing ads and stuff like that. Because the integration between the two companies is janky and not in real time. That was something I worked on at Comcast, tried to fix. Well, I didn't do anything. The people who work for me did it. I just sat around telling them they were doing a good job. But the companies have tried to fix and I don't think they've made a lot of accomplishments. So there's a lot of low hanging fruit that would make everyone more money and everyone happy if Google would get off their monopolist bullshit. But they smells like a duck, acts like a duck.
- [52:19]
Eric
Got it. That makes sense.
- [52:21]
Ari
We don't have in the agenda the news TikTok. TikTok lost its lawsuit to stop it from being banned. So as of January 19, five weeks from now, the day before the inauguration, TikTok is not allowed to be hosted by any app stores or cloud services in the United States.
- [52:41]
Eric
So what's going to happen? It's going to host phones. Is it going to be like magically deleted from phones? Yeah. Is there going to be billions of dollars of ad dollars and commerce just trying to find a place which I think for this audience might be the most interesting thing. Tell me what's going to happen here, man.
- [52:59]
Ari
I think if I've read correctly, it will not be removed from your phone. That was considered illegal, like a first amendment violation. But Apple and Google Android will no longer be able to host it for uploads or for updates. So you'll have a dead app on your phone that won't be updatable. And then the hosting thing, I think they're hosted with Oracle and it's not clear what's going on with that. I mean it's totally infeasible to move the hosting in a month. So I don't know how they're going to handle that.
- [53:26]
Eric
Okay, so you have the app on your phone but it's literally dead because what use is a social app that isn't giving you something like new every time you update it?
- [53:36]
Ari
Well, I don't know if that matters to content. I think it's about the software updates. Software updates would stop working. The content could keep going as long as it didn't break.
- [53:44]
Eric
Got it.
- [53:45]
Ari
And because you can't sideload on these devices, there's no workaround. Now, I think what happens is that Trump takes the office and then deals start trying to be made. But the problem that Trump has here, to the extent that he cares what the law says, is that there's a law, there is a passed law by Congress. This is not executive action by Biden. It was a passed law. Congress past it, Biden signed it. That says this stuff, like there's no wiggle room. It says what it says. There'll have to be some real maneuvering to get around this.
- [54:17]
Eric
Is this at the top of his list or near the top of his list in terms of importance? I'm not trying to be funny here. Just, you know, again, like they're, they're, they're doing so much in terms of like, you know, talking about what the plan and the roadmap is with Doge and new appointments and everything like that. Like, I haven't heard anything about this from the incoming administration.
- [54:35]
Ari
I think both, both Biden and Trump want this to not be on their plate, but it's there, there's nothing you can do about it. So my guess is it's not on his list, it's not something he cares about, but unfortunately he's going to have to or his appointees and I would think the FTC are going to have to care about this immediately. It's going to be sort of a day one crisis to deal with when every 20 year old in the country comes down to Washington with pitchforks and wants action. I don't know, man. It's interesting. Yeah.
- [55:06]
Eric
And I wonder if, you know, Elon, who seems to have real influence in this administration and, you know, obviously owns X, has any incentive to push this thing back.
- [55:19]
Ari
And he's got business dealings in China, but he also owns X, so he's pretty conflicted. Everyone's conflicted.
- [55:26]
Eric
Yeah. No conflict, no interest.
- [55:27]
Ari
Yeah, exactly.
- [55:29]
Eric
So, hey, before we go, there's like two weeks left of the year, basically at this point. How busy are you right now? And you know, you're getting geared up for 2025 with market share. Like, what's, what's a day in the life?
- [55:42]
Ari
A day in the life. I have a secret project going on which probably many people know about, but I'm going to remain secret and we'll talk about it. Don't talk about the secret project. I have that. As opposed to note on my computer. No secret project talk.
- [55:56]
Eric
People are going to go crazy right now because I don't think anybody's even heard that.
- [56:00]
Ari
You were. A year ago, I was sitting here, I had a different business called Launch Science, which I shut down. And I was sitting here and I was podcasting and doing stuff and I was making, you know, the whole business architecture was a six figure business at best. It was like doing okay. I was like a hobby. I would occasionally slip out and see a movie in the middle of the afternoon. And now I've acquired three companies. We're doing multiple events. We're doing our live event in New York on March 17th. We're at CES, we're going to be at possible. We have four newsletters, three podcasts. We are a real business that is doing extremely well. In addition to my secret project, I'm insanely busy and I'm going to be working through the entire holiday season. So that's how I'm doing. How are you doing? How's the investment world?
- [56:48]
Eric
Well, just a comment on that. That's awesome. Like you could, knowing you, you could not have been happy going to the movies in the middle of the afternoon versus, you know, just cranking the way it appears you are now.
- [57:00]
Ari
Maybe. I like, I like movies. I saw Interstellar this week. It was great.
- [57:05]
Eric
Yeah, you said you teared up, actually.
- [57:06]
Ari
Oh, teared up. I was crying my face off. That's the saddest movie ever. Have you seen it?
- [57:10]
Sir Martin Sorrell
Wow.
- [57:11]
Eric
I don't think so.
- [57:11]
Ari
Oh my God. You have to see it.
- [57:13]
Eric
I got to see it.
- [57:13]
Ari
It's like the saddest science fiction movie ever.
- [57:16]
Eric
Um, I don't like said movies. Uh, yeah, in, in our world, similarly just kind of like running towards the proverbial finish line. It's been a big month with the acquisition of Autogen by Experian. We, we're going to close two new investments this week. One new investment, one. One follow on in an existing portfolio company. So things are still happening and then getting really geared up for, for ces and then we'll, we'll both be down there and the meeting schedule is, is really packed, so. So I'm excited.
- [57:45]
Ari
Yep, it's been a great year. We're going to have some year in review prediction episode in the next couple of weeks, which we did last year too. Maybe we'll grade our predictions from last year, which probably aren't the best, but we'll see what we can do.
- [57:57]
Eric
Speak for yourself.
- [57:58]
Ari
Let's wait to get all warm and fuzzy for the year end episodes.
- [58:02]
Eric
Totally. All right. This has been great. Thank you Ari and we will see you next week everybody. Thank you for subscribing to Markitecture. New interviews are added every week at Markitecture TV and your favorite podcasting app.
- [58:23]
Ari
Thank you for listening to the Markitecture podcast. New episodes come out every Friday and an insightful vendor interview is published each Monday. You can subscribe to our library of hundreds of executive interviews at Markitecture tv. You can also sign up for free for our weekly newsletter with my original strategic insights on the week's news at News Market tv. And if you're feeling social, we operate a vibrant Slack community that you can apply to join@adtechgod.com.