← Marketecture: Get Smart. Fast.
Marketecture: Get Smart. Fast.
Episode 108: Brett Wilson from TubeMogul is back and is investing in AI
00:48:13
Loading summary
Transcript121 lines
- [00:00]
Ari Paparo
This episode of Marketecture is brought to you by Start IO, an industry leading mobile SSP. If you build ad campaigns, Start IO is offering marketer listeners curated evergreen deals with no audience fees. Visit www.start.IO nofees to activate today. That's www.start.IO no welcome to the Market Tech Podcast. I'm Ari Paparo. I am joined today by Eric Franchi and our special guest, Brett Wilson. Brett, many of you may know, was formerly the CEO of TubeMogul, which was one of the success stories in the early days of the video advertising space. And he's now started a new life as a venture capitalist for Swift Ventures, investing in AI. So, Eric, you have a fellow vc. I'm gonna be outnumbered on this conference call.
- [00:59]
Eric Franchi
We got you outnumbered. We're gonna corner you. Brett's great successful ad tech founder. We should, you know, spend some time talking about the whole TubeMogul story, but you know, having somebody with like real domain expertise, real success, who then, you know, took a step back and immersed himself in this world of AI, pre ChatGPT, pre LLM, and now is, you know, coming full circle, you know, looking at investing in advertising and marketing companies, collaborating with us at marketexture Live. I'm very interested in his perspective of he knows this space, he knows AI. He's now got a thesis on how it's all coming together. I think it's going to be a great conversation.
- [01:39]
Ari Paparo
Yeah, I'm looking forward to it also. So I've known Brett for a long time, but I just became sort of reacquainted with him because we decided to bring him in as a co host of our marketexture Live event which is coming up in New York City on March 17th. So Brett has brought a great deal of AI expertise and some guests to our event. So he'll be interviewing Taz Patel of Perplexity, talking about their advertising initiatives. And he, along with Eric are going to be hosting an AI startup Lightning Round. Sort of a, I wouldn't call it a pitch contest, but we're going to have a bunch of early stage AI marketing folks. They're going to give us their best shot and Eric and Brett will listen and make incisive comments and make them cry maybe a little bit.
- [02:27]
Eric Franchi
We won't do that, but we're looking forward to it. Lightning Round. And people are going to see what the bleeding edge of AI innovation is going to be all about. And I think we're going to close off the day with Live pod, live news of the week live Banter, right?
- [02:39]
Ari Paparo
Yeah. Yeah. So then after the startup session, Eric and I will be recording that week's episode of marketector Live, potentially with a special guest, tbd. Hold onto your hat. In the meantime, you should get tickets because they're going fast. It's marketlive.com. all right, so with that, let's get Brett in here and talk to him about AI marketing and investing. All right, and now we have Brett Wilson from Swift Ventures. Brett, thank you so much for joining us.
- [03:09]
Brett Wilson
Thanks for having me, guys. Happy to be here.
- [03:12]
Ari Paparo
Yeah, we're really excited to have you involved with marketexture Live, but also to come here and talk to us about your investing and AI and everything else that you see going on. So, you know, you've had an interesting journey. Why don't you kind of tell us a little bit how you went from operator in ad tech to now, you know, a venture capitalist. I. I think Eric might have a little bit of empathy with your. With your journey, but it'd be interesting to hear.
- [03:36]
Eric Franchi
We've got you surrounded, Ari.
- [03:38]
Ari Paparo
Yeah, one day I'll be, I'll be taking that 2% and working a couple hours a week, but, you know, for now, I still got some podcasts to record.
- [03:47]
Brett Wilson
Eric and I are proof to the saying that you can throw a rock and head of UC these days. And I, I never saw it coming. I mean, I. I vomit in my mouth a little bit when people call me. I always thought of myself as a founder and founders are up here and VCs are down here. And I started this ad tech company called Tubemogul. A lot of people probably don't even know it now, but that was a 12 year adventure and it was super fun and really proud of what we built. And we brought some transparency and the automation to the industry. We built this killer culture, but it was also a slog. We started this company 2007, 2008. So we're running around trying to raise venture capital in the midst of the Great Recession. If you remember what that was like, ad tech comps were down. So we took this company public. On the two weeks of the IPO roadshow, our comps went down like 30%.
- [04:54]
Eric Franchi
Who were the comps back then, Brett?
- [04:56]
Brett Wilson
It was like rocket fuel and tremor. And, you know, on our roadshow, we heard this like 50 times, like, love you guys. I get that you're different that, that you're software, but we've lost money on some combination of, you know, rattle off your public tech company of the era. And if we want Exposure to your sector. We'll just buy Google or Facebook. So we were a good company but a bad public company. I think we had eight beat and raises and more often than not we beat and raise and the stock would get down. So Adobe ended up buying a company, had rebranded our group as the Adobe Ad cloud and hundreds of tube mogulers are still there which is really cool. But at the end of that I just didn't think my body could do it again. And meanwhile, and I'm curious how you got into this Eric, but I was just beginning to invest in founders in kind of a philanthropic way. Just help other people. Like it's so crazy to start a company and I was starting to do some boards like I joined the to be Bored and was just having a lot of fun like helping the founders navigate the perils of kind of venture capital backed companies. And I wasn't sure what I want to do next. I had a year where I was kind of at home like driving my wife nuts and rearranging closets and taking naps in the afternoon. But like making excuses because I was embarrassed when my kids would ask me about it.
- [06:44]
Ari Paparo
I am proudly nap. I have no issues with that. Don't want to interrupt. So then you just strapped on the fleece blazer and became super vc?
- [06:56]
Brett Wilson
More or less. I mean I partnered up with John. He was my co founder at TubeMogula. He was the smart one. He built everything. And we had kind of a thesis that the tools that we all use to kind of deploy AI systems were immature. And this was 2020 like pre all the generative AI madness. So we started a small fund and our marketing for that fund was hilarious. Like we, we basically told lp it's like we've never done this. We don't think we'll lose your money. You know we, we think our company building expertise is going to translate but, but we're not sure. And so we, we moved in some of our best angel investments that we had made that were doing really well at cost just to kind of give us the ability to get our sea legs and make sure we didn't lose people's money. And we bet on 30 companies and we raised a fund too last year. So that's how, how I got into this.
- [08:02]
Ari Paparo
So what are you investing in? Is it all AI? Is it generative AI? Is it all advertising and marketing? What's the overlap there?
- [08:09]
Brett Wilson
Early stage AI companies and I bucket that into infrastructure companies. The tools that you need to deploy an AI system or manage your data an example of a company in that category is Arise AI. This is a fun one because it was started by former tube mogglers. But this company, they help you monitor your AI and ML models and troubleshoot. And that was always a problem that we had. We're doing robotics, so we back companies that build robots for everything from sanding and polishing to cooking and packing. And then applied AI companies like B2B software companies that are challenging some vertical solving some problem with a more AI centric or data centric approach. And we've done everything from health tech to property tech and just did our first martech deal recently as well.
- [09:14]
Eric Franchi
There you go. So AI, we can think of it as a technology shift and you walk through some of these technology applications and then we can think about it as a platform shift. Thanks to OpenAI launching ChatGPT a couple years ago. Like, you know, basically that was a moment that started a new epoch and everything changed. How do you think about this platform shift and where do you begin to think about, like, what are the investable things in a world where people are going to be changing everything. Changing how they search, changing how they shop, changing how they interact, like with everything.
- [09:53]
Brett Wilson
I mean, in marketing, I've had kind of marketing, ad tech, ptsd.
- [10:01]
Ari Paparo
So we all, I think all of the listeners of this podcast feel the same way.
- [10:07]
Eric Franchi
We're like therapy for them every week.
- [10:09]
Ari Paparo
Yeah, that's going to be our business model. Instead of selling ads, we'll have sponsored support groups in different towns.
- [10:14]
Brett Wilson
Brought to you by BetterHelp.
- [10:17]
Ari Paparo
Learn to control the things you can control. Sorry it's a rough trev your PTS day.
- [10:25]
Brett Wilson
It's true. I've learned every founder thinks that their industry was uniquely hard, but ad tech is uniquely hard because we're all in a cage match for the scrap that Facebook and Google kind of don't take. So I had intentionally kind of avoided the industry. But a couple things got me excited about. I was sharing them with Eric earlier this week and he was like, oh yeah, I've been writing about this for months. So I, I'm not sure like how much of a, of a hot take it is, but I think the, the biggest trend that I think is investable is just this notion that the way we all find and discover information is changing. And we're going from this era where you type something in, you get all of these seofied, you know, blue links and you gotta click around to where you're going to an AI, to an LLM that synthesizes everything for you and so in this world where like an LLM becomes the primary interface for information in commerce, I think it just, it may necessitate an entirely new ad tech stack and marketing stack. And already for me like going to Google just feels a little stodgy. And it you think about this world, like what, what could happen? And you could kind of see like the advertising surface area decreasing a little. Like today, digital advertising is primarily about driving people to media property. Increasingly all those things are going to be done in an AI. You're going to have a conversation, you're going to buy something, you're going to book something. So I'm just kind of fascinated with what tools we'll need in that type of world.
- [12:27]
Ari Paparo
One thing I find a little mind bending is that it may be kind of straightforward to think about what the opportunity right in front of you are. Is the AI for creative AI, like SEO for AI, there's million companies doing that.
- [12:41]
Eric Franchi
The technology side, yeah, for sure.
- [12:43]
Ari Paparo
Right. But what if the technology, the consumer behavior just leapfrogs all that and you don't have creatives or you don't have search. Right. What's the point of SEO for AI if there's no search? Right. I just, it seems like a very hazardous place to be investing.
- [12:55]
Brett Wilson
Yeah, I mean AI in general is a hazardous place to be investing. Like we were investing kind of state of the art machine learning companies and a year later you have companies who are like, yeah, we don't do machine learning anymore. We have people that talk to the AI. So I don't know about you, Eric, but we're definitely kind of investing in a disciplined way just because of all the shifting sands. But I think there's a lot of opportunities right now as well.
- [13:26]
Eric Franchi
Yeah, no, it's interesting. All right, Just to your point of, you know, the, the unknown future, investing at the beginning of these, you know, next epochs has always been like scary and messy. Right. Like the beginning of the Internet. Most of the original, like Internet companies, they failed. But then there were a couple that, you know, became the greatest businesses of all time. Right. Like Google and Amazon. So it's like undoubtedly there is going to be a messy period and then a period where there's, you know, going to be monsters. And then I think that also piggybacks onto something you, Brett, which is, you know, for a long time you're fighting over the scraps and I would argue that's changing. But with, you know, things like CTV and retail media, but for sure, like in what we would consider ad tech, you're fighting for scraps. I wonder if you're no longer fighting for scraps because the future is not guaranteed to Google and Meta.
- [14:19]
Brett Wilson
Yeah, it's interesting. I mean, I think right now, for better or for worse, like the, the companies that are building the LLMs that are the best that we all use are for the most part the same old wolf garden companies. I mean, the one thing that kind of gives me hope is OpenAI came out of nowhere. No one saw them taking on Google's dominance. So that's exciting. The open source movement is exciting. So hopefully it's not the same old players, but right now it is. It's like that, that who's on meet meet the new boss, same as the old boss. I mean, that's what it looks like at the moment.
- [15:01]
Eric Franchi
Yeah. Speaking of OpenAI, we're recording this a few days after people in AI were just screaming about this deep SEQ company for this very ad tech centric audience. Could you walk us through what happened with deepseek and why it matters?
- [15:18]
Brett Wilson
Yeah, it's funny, in the AI world that feels like kind of ancient nudes.
- [15:23]
Eric Franchi
Dude, it was five days ago.
- [15:24]
Brett Wilson
Yeah, I mean, I almost thought about it. Basically, a Chinese company launched an open source reasoning model that on a lot of different measures is as capable or more than OpenAI and other models. And they claimed to have done it not using the latest and greatest Nvidia chips and not spending a lot of money on training. And to date the paradigm has been more chips, more data centers, more compute, more energy, more money. And this maybe challenges that paradigm a little bit. So how you reacted to the news really depended on your view of the world. You know, if you were a chip company or maybe if you're, you're meta, you're creating kind of competing models, you were scrambling. But for most of the companies that, that we invest in, you know, they're building on top of models and I'd say they're model agnostic. So a lot of them were scrambling over the weekend to kind of integrate deepseek into their stacks and you know, their, they're building. These are software companies that are building logic into their stacks that kind of use different models and different purposes. So, you know, better, cheaper, faster. That's a continuation of what we've seen and it's largely good thing.
- [16:59]
Ari Paparo
Yeah, I mean if models are super cheap, that's great for everybody. The interesting thing for me is how many VCs and how much conventional wisdom just a year ago was don't be a wrapper company. Oh, they're just a wrapper company. Right. But rapper company in my mind, Tell me if I'm wrong is a synonym for an application someone uses. Right. And I never saw the logic of don't be a wrapper company.
- [17:26]
Brett Wilson
I didn't either. I mean, I've never invested in foundation model companies just because that's always felt like this Game of Thrones with the biggest dogs in the world and it's a little bit of a race to the bottom. Whereas software companies that are just solving some important problem, but with an AI based approach, I mean, call it an AI wrapper or whatnot. I mean, I think that that's where a lot of the excitement is. I mean, the best companies, it's not like they're just building this kind of lightweight UI on top of a model. I mean, they're using their own private data, they're introducing problem or vertical specific workflows. I think that's where a lot of the action is because most VC firms don't. They're not capitalized to invest in these foundation model companies in any event.
- [18:23]
Ari Paparo
Yeah, exactly. So what are you seeing that is interesting in the advertising and marketing field, in the AI and marketing overlap?
- [18:31]
Brett Wilson
You know, I'm just beginning to kind of wade back in those waters cautiously a little bit. I'll take more of your therapy that you offered our first investment and I want to caveat, like I still think you're investing in people at these early stages and you're trying to bet on people that are going to be able to navigate through these technology shifts. And that's really what we do. We bet on people and ideas. But the first bet we made was on an AI SEO company you mentioned earlier. There's a lot of those. The company is called Bluefish AI. The idea of this category and there's four or five companies there is just if you're a brand and you spend all your time to date optimizing your media property and making sure that you rank high, what do you do now? And you almost have to market to these agents and you have to produce content in a way that's best in alliance of how an LLM kind of curates and presents information. So these companies kind of do that. And Alex, the CEO of Bluefish, he had this great line I attributed to him. He said marketing to agents is the new mobile optimized. And I thought that was so, so right on. So that that's kind of the first investment we, we've made in the category. But as I look around at like, you think about different parts of the stack. Take analytics. It, you know, we've all used analytics systems that are kind of rules based and they require a lot of setup and interpretation. I think the AI native analytics product, it's going to alert you proactively instead of you having to set up reports. And I think that goes for CDPs and personalization and so on and so forth. So that's part of why I'm coming to market live. You know, I want to, you know, learn more about what's, what's happening.
- [20:46]
Ari Paparo
And Bluefish will be there as well. Right there. Breakout session. So I'm excited about that. I think of it like, maybe this is a crazy metaphor, but it's kind of like the Star Trek Replicator problem. Like, imagine if we invented the Star Trek Replicator. And for those of you who don't watch Star Trek, you can just ask it for anything you want to eat. How do you get the consumer to say they want a Coke instead of a Cola? That's a marketing problem. Right. I'm sure the Replicator has to pay some royalty to the Coca Cola company if it replicates a Coke. Right. That'd be a legal problem. We'll discuss that on stage. Eric's giving me this look like, how did I get involved in this podcast? What is happening?
- [21:24]
Eric Franchi
Talk about the Swift AI index.
- [21:28]
Brett Wilson
Yeah, so we created this index of 88 equities that are the most like AI of all stocks. And we did it because we had a lot of LPs that are constantly asking, should we buy more? Nvidia, how do we invest in AI? And a lot of the action in AI is in private companies. And it kind of was an interesting question, like, what, what is an AI company? I mean, everybody talks about it. Do you get to be an AI company just because you talked about it? So we built this product and we're always trying to build things with the tools, but we're looking at companies, actual investment in AI, who they're hiring, what percent of their population are AI related roles. So machine learning engineers, data scientists, AI engineers. What's their IP like? What are they contributing to AI related research? And then we've kind of fine tuned an LLM that reads their transcripts and tries to understand how fundamental AI is to the business. And we scored a bunch of companies who Creativist index. And the inspiration was something another VC firm did previously called the Bessemer Cloud Index. When, like cloud was a new thing, we, we thought it would, like, help us, you know, be Smarter, be, be interesting for people. So we launched that a couple months ago and it's, it's gotten a lot of, a lot of attention. So.
- [23:05]
Ari Paparo
And we're, I'm going to guess it's been going up. Has it been going up?
- [23:09]
Brett Wilson
It's been going up. And it, it, when you back test it, it beats the, the nasdaq, the, the S and P. And we're kind of coking around with the idea of maybe turning it into an ETF as well. So it's, it's, it's viable. We're trying to figure out how much work would that be, how distracting would it be relative to our day jobs. But it, it's been a fun intellectual and, and marketing exercise to date.
- [23:36]
Ari Paparo
Yeah, there are these companies that let you create ETFs. There are certain minimums and they expect you to basically sell it. They'll just create it and list it for a certain amount of money. Then you have to do all the hustle.
- [23:47]
Brett Wilson
We dove into that world. They're called white label ETF providers. And it's cool. We're veering far from AI marketing now, but because they deal with the SEC and all the regulatory and compliance stuff and you just kind of contribute your ip. But it's what you said. You have to spend time getting people to buy it. And sales and distribution of an ETF isn't something that we do.
- [24:17]
Ari Paparo
I was thinking about creating a market index. It was just going to have the trade desk and Wawa in it and ride it to the moon.
- [24:25]
Brett Wilson
Amazing.
- [24:26]
Eric Franchi
Brett, you were very early to point out when you were, you know, active in this market as a, as an operator, Google as a walled garden. And I don't know how much you've paid attention to the latest, especially over the course of the past year. Ari's been covering this better than anyone. You know, there's a lot of scrutiny around Google. Are you surprised? Do you have any predictions? Have you not been paying attention? And you know, I just want to take a, take a pass on it. That's fine too.
- [24:55]
Brett Wilson
I feel like we were vindicated. Ten years later, we did this. People always say don't poke the bear, but we definitely poked Google. And we did this worldwide digital and outdoor campaign. It's called Independence Matters. Just kind of poking fun at Google in a humorous way about their misaligned incentives and conflicts of interest. And for a while it was really fun. And we would like, walk into your room with an advertiser and they would be like, we really love what you're saying we connect with your values. And then Google cut us off from FedEx YouTube inventory on the day of earnings and we, and I was like, oh, maybe we shouldn't have poked the bear. But yeah, I mean, I've always thought that they had kind of very monopolistic tendencies and, you know, stifle a lot of innovation, particularly in our space. So it was great to see that the government start to maybe do something about it. We'll see what actually happens.
- [26:03]
Ari Paparo
Yeah, we'll see. They're still not looking at video, unfortunately. They're only looking at banner ads. But that's just a matter of how the legal side shakes out, I guess. Last question, which is you're coming to New York helping us host marketexture Live. What are you hoping to get out of that? What are you hoping to see?
- [26:21]
Brett Wilson
Yeah, I want to connect with old friends and we're almost staged in a fireside chat with Perplexity. So share a little bit about what we're learning with where AI meets marketing and just talk to marketers and hear what kind of the problems du jour are and have some fun with you guys too.
- [26:44]
Ari Paparo
It should be fun.
- [26:45]
Eric Franchi
I'm going to make a production prep. I think that you're going to come back, you're going to see your old friends, you're going to get excited, you're going to see how big this market is becoming and you're going to get the itch to start another company in ad tech and incubate it. And I don't know what the percentage of odds are, but I think the odds are greater than zero. I've seen it happen several times to exited founders. We've taken a little break, come back and then they're like, huh, Do I.
- [27:15]
Brett Wilson
Have a blank check from Aperium Ventures if I do that?
- [27:18]
Eric Franchi
Are you the CEO?
- [27:20]
Brett Wilson
Of course.
- [27:21]
Eric Franchi
There's a non zero chance.
- [27:23]
Ari Paparo
Yeah. You get investment from Eric. I put in about $50 and call myself an investor. And then you get like to come back on our podcast a bunch and hype up what you're doing. It's a complete package deal.
- [27:35]
Brett Wilson
Depends on how many drinks I have at markets.
- [27:40]
Ari Paparo
We're still working on the green beer, my planners because it's on St. Patrick's oh, I didn't. We're working on it. No promises. We might have to settle for an Appletini or some other green related thing.
- [27:53]
Brett Wilson
We're excited. Thank you for letting me be involved.
- [27:58]
Ari Paparo
Yeah, I'm really excited about it. All right, let's take a quick break and Go through the news of the week. This episode of Architecture is brought to you by Start IO, an industry leading mobile SSP. If you build ad campaigns, Start IO is offering architecture listeners curated evergreen deals with no audience fees. Visit www.startionofees to activate today. That's www.start.ionofees.
- [28:29]
Eric Franchi
All right, we are back with the news of the week. We got a bunch to go over today. Nielsen is back in the news or continuation of the news from last week. There's some like wonky. Mmm. Stuff happening in the world that I think is super interesting. Meta earnings, some stuff or some stuff didn't happen at IB Alm. And we'll see if we can talk a little bit more about Deepseak. So let's get into it. All right. Nielsen is going to stop selling panel only data later this year and they're only going to sell hybrid measurement with data from smart TV makers and other digital sources.
- [29:02]
Ari Paparo
Yeah, so I'm your measurement guy. You said last week, oh my God. The important point here is that this was widely misreported. So a lot of people posted on X and other where other places that Nielsen was abandoning its panel. That is not true. They're just not selling it separately. You can't buy panel only ratings like you used to be able to. They're only selling the hybrid ratings. And I think that's natural because the panel ratings are becoming less relevant because of the math involved. You need to have a lot of people watching for the panel to get accurate ratings. Every year fewer and fewer shows can get the accuracy that is needed. So they're kind of burning the boats and going for hybrid. Last week we reported that hybrid got MRC certified. So I don't think this is nearly as big news news as some people are making it.
- [29:50]
Eric Franchi
Is it disruptive though to like the just the stale process of TV that runs on panel or do people have enough time to figure this thing out, adapt, you know, get their. Get their numbers at parity?
- [30:02]
Ari Paparo
Yeah, it's a good question. I'm not sure how different the panel only versus the hybrid numbers are going to be. The problem you have as a product person at a place like Nielsen, because I did that briefly, is that you may roll out something that's better, but the disruption, the discontinuity in the numbers is really a problem. People want everything to be super smooth, but you can't improve things and make them smooth. So I don't have an answer to your question, but it's an important question is how will the transition for anyone who currently uses panel B. And that in the past has not been a great transition for Nielsen clients. So I hope they work in a seamless way this time.
- [30:42]
Eric Franchi
Can you believe that people are still freaking talking about Nielsen seven years later after you left Adtech?
- [30:48]
Brett Wilson
That's what I was about to say. It's like you're rip and winkle, but you come back and nothing has changed. Like, you know, we're in this digital world and we're still talking about this relic of panels. So yeah, it's sort of unbelievable.
- [31:05]
Ari Paparo
We should solve it with AI.
- [31:07]
Eric Franchi
Exactly.
- [31:08]
Brett Wilson
There you go.
- [31:09]
Eric Franchi
There's your first idea, bud. All right, cool, let's move on. Google launched Meridian. What is Meridian? Meridian is an open sourced MMM solution, media mix modeling. It's downloadable on GitHub. It's got a partner program of agencies and performance agencies. So it incorporates incrementality experiments, reach and frequency analysis, media data. I was including media data and this got, I think some people talking because it's an open source, AKA no cost project that does what companies charge thousands of dollars to do. All right, do you think this is a threat to MMM companies or this is, you know, again, what if Google enters your category and you need to be a smart kind of specialist operator?
- [31:57]
Ari Paparo
Yeah, and I put in the notes that Meta did something very similar recently also Meta has something called Project Robin, which is an open source framework for mmm. So I think both these companies are of the opinion that with enough measurement you're actually going to increase spend on their platform, not decrease it, which might be original thought because so much of it currently goes there. And MMM solutions are usually proprietary and old and operated by companies like Nielsen. So I think the tech giants are doing a little jujitsu here and taking the lack of loss of identity and using it to their own benefit because statistical methods may show that they generate more conversions than directly attributable methods that people consider the gold standard. Interestingly, we have next week. I don't know if you know this yet, Eric, but we have Ron Jacobson from Rocketbots on, good friend of mine, former apnexian, who's an expert in all things measurement and we can really interrogate him about this stuff.
- [32:56]
Eric Franchi
Good. Yeah, look forward to it. There was someone, it was Eric Tilbury on X. He threw out there that he thinks that the opposite will happen in terms of people's analysis and spending. And you know, they're going to see kind of like what Tim from, you know, our at home guy said, you know, they'll See like the true performance of things like out of home or mean your favorite channel that you might, you know, have a, have a position in and you know, things will get more balance. But I'd imagine that Google wouldn't roll this thing out if they thought there was a shot in hell of that happening.
- [33:27]
Ari Paparo
Another friend of the pod, our mobile guy, Eric Seufert, had an excellent blog post this week about. He basically said that Wanamaker's 50% of my advertising is wasted is not correct. It's that 50% of my spend is measured wrong, which is like quite a little bon mot, as they say. But effectively, I think we're in an environment where so much is difficult to measure that statistical methods might be the best you can do.
- [33:53]
Brett Wilson
I do think that when I lock advertising. Mmm, you're hiring a small shop with these PhDs and super black boxy kind of formula and I think the text changing. We're now seeing LLMs for time series data where you may be able to do with numbers and forecast what we do with language. And so I do think whatever the future is, it's going to be a lot more automated than it is now. I don't trust Google. I mean, I'd love to see independent measurement providers, but it's probably the right trend here.
- [34:34]
Eric Franchi
All right, cool. Let's move on. So speaking of Meta, Meta absolutely crushed earnings. So their Advertising revenue is 40.46.8 billion in Q4, the majority of total revenue, which was 48.4 billion, up 21% year over year. So they continue to grow quite nicely. So a couple things on the AI front here. Brad, love to get your POV. So more than 4 million advertisers are using at least one of Meta's gen AI creative tools. That's up from a million. These are big numbers, but up from a million. So 4x6 months ago is 0.1.2 advantage plus which is their set it, forget it. Programmatic algorithm network product has surpassed a $20 billion annual run rate, growing 70% year over year in Q4. And they're going to turn it on by default for certain types of campaigns. And then there's some interesting stuff there. It's kind of technical about AI being used to experiment around increasing ad load, where across the various meta properties they've got the opportunity to stuff some more ads in. So they actually had an impress increase as well. Brett, do you think that Meta is just like far ahead of everyone just based on this type of stuff or are you seeing like the other Big platforms doing this type of work. I mean, this is like really, really impressive.
- [35:59]
Brett Wilson
Yeah, I don't think anybody's close. They have this money printing machine and when you take the data that they have and their AI capabilities, I mean, they're the closest to this visions that we've had for a long time where ad campaigns just run on autopilot. I mean, maybe that happens a little bit on the media side, but what they're doing is combining creative and media together where you have this kind of rumple peel, set it and forget it at campaign, which is what especially small businesses want. So it's pretty damn impressive.
- [36:38]
Ari Paparo
Does Meta ever past Google as the biggest advertising platform in the world? I think round numbers, Meta is at 200 and Google's at 3. Somewhere in the 300 to 350 range, I think so.
- [36:51]
Brett Wilson
I think Google's franchise is under more direct frontal assault with these new AI search engines. I could totally see that happen.
- [37:02]
Eric Franchi
Yeah, I think it makes sense, right? Because you know, what is Google Media? Google Media X, you know, network stuff is search and it's YouTube and search is declining. YouTube is growing. But Meta has a portfolio of apps, B, the ability to do, you know, kind of like interesting things from an ad load perspective on said apps, and then C, they're rolling out more products like threads, which they can begin to monetize. So Ari, I think absolutely the chances there to Meta, you know, surpasses them at some point in the next, I don't know, call four years, get Zuck.
- [37:41]
Ari Paparo
Another chain you'll have to add when Zuck's got two chains.
- [37:45]
Brett Wilson
Well, no, you got to add them to your. Your 2 equity ETF. All right.
- [37:50]
Ari Paparo
Yeah, it'll be Meta Trade Desk. Wawa.
- [37:52]
Brett Wilson
Yeah, done.
- [37:53]
Eric Franchi
And I love it. Wait, by the way, okay, so Ibalm, you skipped. I skipped Ibalm. Or you were someplace.
- [38:03]
Ari Paparo
Thought I was in Sundance for a couple days, which was a little unusual. And I wrote about it in my newsletter. It was super fun. I saw movies made by brands, which is a category that you may not expect to see at Sundance, but the nice people at brand storytelling run effectively a parallel film festival for brands. And it was mind blowing and really cool.
- [38:24]
Eric Franchi
Was it happening during the same time as the Sundance film?
- [38:27]
Ari Paparo
Yeah, it's an officially sanctioned Sundance eventually.
- [38:31]
Eric Franchi
Super cool.
- [38:31]
Ari Paparo
Yeah, it is cool.
- [38:32]
Eric Franchi
Did you, did you see anybody famous?
- [38:34]
Ari Paparo
No, no.
- [38:37]
Brett Wilson
Was there a media link party?
- [38:39]
Ari Paparo
I'm sure there might have been. I wasn't invited though.
- [38:42]
Eric Franchi
Now at alm, there were famous people there. I saw pictures of Ed Norton. I saw pictures of Benji and Joe Madden milling around Palm Springs with a bunch of ad tech people wearing lanyards. So it was a much different experience apparently.
- [38:57]
Ari Paparo
Yeah, Tech God was there. So you may have seen him, but you don't know who he is because he doesn't put that on his badge. So some random person you saw last week might have been ad tech God.
- [39:05]
Eric Franchi
If you're right, yeah, now you see him, now you don't. ALM traditionally has been a big kickoff to the year and it seems like there wasn't too much in terms of announcements, news, anything coming out of it. There were basically two things I picked up on the news. I don't know if you guys have anything more, more to add. So first thing is there was I guess a discussion with Google around Cookie, the, you know, the cookies going away or the process for that and they basically said nothing. They said that it will be a one time global user prompt, you know, somewhat like we saw with, with Apple when they rode that out and the industry will have quote unquote months of notice. So literally they said nothing.
- [39:50]
Ari Paparo
You know, I mean the Google people, they live in this world where they are like why don't people like us, we give them great stuff, search and Gmail and then they do this sort of thing. So like, I mean I want to like Google. I use Google every day. We're on Google Meets all day, right? But then they do this. They're like getting rid of the cookies or not. They're not telling anyone why. They're maybe working with regulators, they're maybe not. They, they're driving us crazy, man. This hot and cold stuff, I can't take it anymore.
- [40:19]
Eric Franchi
The other thing that was interesting was, I guess there was a fireside with, with Jeff Green. He talked about a bunch of things including them since our acquisition. But he guess he either like challenged Google to or predicted that they will exit the open web to focus on YouTube just because YouTube is, you know, the future and you know, the open web stuff isn't necessarily the best thing for them outside of all the headaches that they get. What do you think of that?
- [40:48]
Ari Paparo
Well, I think directionally most people probably agree with that. Whether they have a heartbreak and officially exit the open web, it's a bit more speculative. We'll have to find out one day when he shows up in this podcast.
- [41:01]
Brett Wilson
Maybe Jeff is projecting a little. He's like, I'll take care of the open web for you guys.
- [41:06]
Ari Paparo
I got you a little hopefulness there. If Google exited the Open web. And the only thing you could use DV360 for was to buy YouTube. There'd really be a question about whether DV360 should exist, which would be really good for Jeff and for Trade Desk. Right. And the market sector etf. We would have a huge boost.
- [41:26]
Eric Franchi
Yes, we would. All right, back to Deepseek. Because it was such big news this week and there was actually like. So Brett, your post, it wasn't from an advertising perspective. It was just like sort of explaining Deep Seek for the rest of us. It was great. It was the best thing I read. People, people should check it out. We'll put a link to it in the, in the newsletter. But there were a bunch of hot takes on, on ads. So Bach weighed in, Chris Vanderhook weighed in, REU weighed in. And everybody seems to think that, you know, this is, you know, somewhat proof that A, the value is in the application layer that like we were saying. And then B, this just like shows that there's real opportunity A for you know, ad tech in general. And then B, if you know, you have the big companies a little bit on the back foot with respect to models, like, that's generally a good thing too.
- [42:16]
Ari Paparo
Yeah, My take on this, which is that, you know, coming from ad tech where every penny counts and like, you know, I know at Beeswax when we, every month we'd get our AWS bill and freak out and then like have emergency meetings and try to figure, figure out how to reduce a bit here and there and save $10,000 per bit per month. I think that sort of ethos is what this Chinese hedge fund had that maybe doesn't exist in the DNA of people who work all day at Google, OpenAI and other companies that are raising $50 billion rounds. So that's one way to look at this is that the reason why there was this bloated amount of money being spent on the models was because the kinds of people and the kinds of institutions that are building the model were the kind really care about day to day money. And this Chinese group did that may not turn out to be true. No one really knows the details of how they built this model. But I definitely think there's a cultural difference here.
- [43:14]
Eric Franchi
And Brett, there's like security concerns. A bunch of people look through the TOS and it says deepsea collects your IP address, keystroke patterns, device information and stores it in China, where presumably all that data is vulnerable to arbitrary requisition from the Chinese government. How much of a concern is that? Or is that just like not your game and where you're focused?
- [43:38]
Brett Wilson
I think for most of the companies that I deal with, they're downloading the model weights of the model and they're using it in their own US based infrastructure. So there's no security concerns at all. I guess that perspectives from US users downloading the app and I guess I've always thought that the security concerns of TikTok and things like that are just a little bit overblown. It's hard to see how foreign enemy, you know, really uses that data to challenge us. So I think it's topical because of the America versus China narrative right now. I think it's topical because we all have Nvidia, a lot of capital gains in our portfolio and there's a massive amount of financial media. But it's not that it's not a big deal, but I'm not sure that it was worthy of stock market tanked and three days of nonstop news either.
- [44:36]
Eric Franchi
That makes a lot of sense.
- [44:38]
Ari Paparo
I think it's also interesting that we keep bringing up some new app that's stealing data from American consumers or a security risk. And a lot of this problem would be solved if Congress got off its ass and passed a national privacy bill or if the anti TikTok bill had had more general feelings about apps from China as opposed to just picking a single one or a single company and scapegoating it. It'd be nice if we had that sort of infrastructure for our security and our protection. Can you picture a bunch of Chinese communist military folks with pictures of us up on the screen being like, how do we manipulate this impressive podcast that the advertising community turns against, you know, Western capitalism? What can we do? Where's the Wawa angle? Well, you know, that's picturing.
- [45:24]
Eric Franchi
That's exactly what, what, what they're doing. Brett, where can we find you on TikTok?
- [45:29]
Brett Wilson
I'm not really on TikTok, so I'm on a. I'm a blue sky and I'm on LinkedIn.
- [45:37]
Ari Paparo
Do you get any action on Blue Sky? Because I get nothing. It's just like I try, I post a lot, nothing happens.
- [45:43]
Brett Wilson
I've got a lot of followers. I don't know why, but I think I'm part of one of their packages.
- [45:49]
Ari Paparo
But yeah, me too.
- [45:50]
Brett Wilson
My wife made me join and I thought it'd be good to have kind of a counterbalancing social networks.
- [45:56]
Ari Paparo
I think it's important to give a little plug here, which is the marketecture has a really good TikTok channel that doesn't involve me. We have this woman, Caitlin, who does our TikToks one a week news of the Week and it's better than our News of the week on this podcast. Eric, I just have to tell you that it's like really good. So if you're into TikTok or Reels, you should follow Market.
- [46:16]
Eric Franchi
Brett, this has been awesome. We look forward to hanging out and mixing it up at marketecture Live and we appreciate your time, man.
- [46:24]
Brett Wilson
This was so fun, you guys. Thank you for letting me be a part of it.
- [46:27]
Ari Paparo
Absolutely. Thanks for being here.
- [46:28]
Eric Franchi
See you next week, everybody. Thank you for subscribing to Market. New interviews are added every week at Marketing and your favorite podcasting app.
- [46:46]
Ari Paparo
Thank you for listening to the marketecture 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 marketecture tv. You can also sign up for free for our weekly newsletter with my original strategic insights on the weekly 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.