Rewriting the Rules of Programmatic: Adam Heimlich on Agentic Bidding, ARTF, and Building Chalice AI

Watch on Youtube

Listen on Spotify

---

Read Full Transcript Bellow:

Brett House (00:01)

Hey, everybody. Welcome back to Signal and Noise. I'm Brett House, joined by my co-host, Rio Longacre. And today we're talking about one of the most interesting founders working in ad tech right now, someone who has seen the industry from both sides. Adam Heimlich spent years inside the agency world, including leadership roles at Verizon Media, working directly in the trenches of programmatic. He saw firsthand how the system actually works, the incentives, the inefficiencies and the limits of the tools media teams ⁓ for the media teams that were using them every day. Eventually, he decided to do something a little more dangerous, something that ⁓ I've done recently and it is terrifying at times is build a new company, launch a new company, ⁓ go out on his own. So he's the co-founder and CEO of Chalice AI, a company for those that don't know, trying to rethink how programmatic buying actually works from agentic AI to new RTB standards through a containerized architecture. And we will talk about what that means that lets advertisers run totally run their own decisioning logic directly in the bidding process. And again, if that's flying over people's heads or in one ear and out the other, we'll talk about it in detail. ⁓ I, you know, my perspective on you, Adam, is you've got a very different way of thinking about programmatic infrastructure. You're certainly a visionary in the industry and you've done your fair share of shitposting.

Rio (01:02)

Yes, I'm sure there many questions about that.

Brett House (01:27)

⁓ and adding it mixed sharp, it's, know, mixing it sharp insight and LinkedIn and X and others. Welcome to the podcast. It's a thrill to have you. And, ⁓ we'd love for you to give a little introduction. I don't know if I did you any justice there or if that was completely biased.

Adam Heimlich (01:35)

Thanks. That was fair. I just got to plug my live Friday show with Gareth Glazer, Ad Tech Ad Talk, since we're all podcasters here. That is a live Q &A where a buy-side optimizer and a sell-side optimizer take questions. Not many questions this week, but sometimes we get into questions like, between him and I, know where the bodies are buried and how things really work.

Rio (01:52)

It's a good one.

Adam Heimlich (02:08)

recommend that show. It's Fridays at noon ET on YouTube. yeah, everything you said was accurate. I'm an older founder. I had a good 30-year media career, 10 pub side as a writer, a journalist, and 20 agency side. Toward the end of that, I became an ad tech consultant as sort of a bridge to founding. And I founded early in the pandemic, so just about six years ago this March.

Brett House (02:13)

There we go, sit, sit.

Adam Heimlich (02:39)

which ⁓ also coincided with my 50th birthday. it felt like between the pandemic and my 50th and everything I'd seen and the lack of freelance work available when the pandemic started, it certainly felt like a golden opportunity. What?

Brett House (02:54)

Midlife crisis? I said a midlife crisis.

Adam Heimlich (02:58)

It was not a midlife crisis. I mean, I hope I live to 100, but it wasn't any kind of crisis. It was like, really want to work. I have ideas. I had the business plan in the drawer and yeah, there was nothing else to do.

Rio (03:13)

Well, Adam, we're thrilled to have you on here. Like when we started this podcast, not too long ago, there's a short list of people we drew up and you you were on that list and glad we could finally make this happen. And yeah, I think one of the reasons why we wanted to have you on here, besides just, think like, love a lot of like things you say, some of which are, think non, I mean, are completely like make so much sense. Some which are maybe a little controversial sometimes, which I love as well. ⁓ I think that you've got a cool story even beyond the tech, like, like what happens when someone who actually like, you know,

Adam Heimlich (03:23)

Relax.

Rio (03:43)

worked in the trenches. You had a big programmatic job at Horizon, which used to be a client of mine, as well as other places. And you spent a bunch of years running and executing campaigns for some major brands. I think it's cool to see that you went from that to actually doing your own thing. I think that's part of the story here. then when I'm looking at what you've done with Chalice, this whole effort to rethink the architecture about taking what was static bidding logic and opaque take platforms with not a lot of transparency and how you're exploring this world where leveraging AI and leveraging, you know, agentic to see how the kid completely changed the bidding process, changed your decisioning logic with this portable containerized approach that I really think is cool. So we're going to dig into that and the idea how like these initiatives you're doing, like the agentic RTB framework, ARTF that's being explored with you along with IAB Tech Lab. I think this really brings out like so interesting questions, if agents are increasingly making media decisions, what does this mean to for advertising and how media is bought and sold? mean, I think it has some really big implications, which are, which are pretty cool with that. Let's kick it off. Some questions. I think Brett, you had the first one.

Brett House (04:57)

Yeah, having these leadership roles at companies like Horizon Media, would that experience teach you about how programmatic actually works versus ⁓ what the industry likes to say about it? So the positioning versus the reality. What'd you learn?

Adam Heimlich (05:13)

Yeah, I mean, because I'm old enough that I saw pre-internet media business, and just because I'm kind of curious and I like to take a step back and have a level of awareness of what I'm in from an imaginary outside position, I kind of saw the incursion of technology into advertising as like an invasion. it wasn't just technology, it was VC too that came into advertising and really changed it. So think having the knowledge of what it was like before allowed me to follow that as a narrative of how, you know, this is how the tech people think about it, this is how the money people think about it, but I'm native to this advertising and media domain. ⁓ And I could trace the fault lines of where they're not communicating or where it's not working. And I think by doing that it put me in a good position to match new technology with old problems. That's what I was always trying to do. I adopted search really early. That's how I got into advertising. And I realized, just by adopting this thing that was lying around, Google AdWords had been out a couple of years before big advertisers used it, you can make a big change just by learning the tech. And it was the same with Facebook ads and...even more so with Programmatic. I got really into this pattern of adopting the technology and how the technology changes the advertising and advertising will shape the technology if you have a voice in how it's made. So I saw a lot of people split off and form service companies around just the fact that they could use the technology. I stayed within the agency world. I was just getting more and more executive around this ability to use the adopted technology to client benefit.

But then I chose AI as my wave. Like I said, I was turning 50 and the AI wave was already gathering. It was clear it was going to be something. I might have underestimated how big it would be. It was a big wave on a longer ride than maybe I thought. But I think that's what equipped me to say, all right, so this round of new technology hitting the old advertising business is the fourth time I've seen it and there are certain patterns around.

Brett House (07:18)

the tsunami of this, right?

Rio (07:21)

Yeah.

Adam Heimlich (07:37)

⁓ adoption and resistance that I don't have to get upset or knee-jerky about anymore because I've seen it.

Rio (07:45)

Was there a particular founder moment of time you were working, let's say with clients, the agency said, thought, okay, this is broken, there's gotta be better way to do this, I have an idea, was it just the AI thing? And you were early to that, which I think, so your timing was good.

Adam Heimlich (07:59)

Yeah, I was aware of customization and backend data use from my early days at Avenue A, which had the Atlas Ad Server in-house and a really good analytics team. And they would go into the event-level data. That's what we called it, event-level data. Now we call them logs. And they would see that simple attribution was like a rule of thumb for which impressions were effective. But if you went into the event-level data and you could start to see which impressions actually had influence, right? They would...they would use causal influence methods. you know, I was very curious about this. you know, I think over the years I've gained some expertise in what was going on, but I was following that narrative. Like, well, this is better, but it's slow and it's hard to sustain and it's expensive for the agency. you know, just see, you my experience of AI is like, wow, this is now really cheap and fast.

Brett House (08:50)

Yeah, yeah, it's almost like you find you get technical, you learn, you learn the systems and then you find the gaps and then there you find the opportunity, right? As sort of a founder and an entrepreneur in a sense, right?

Adam Heimlich (08:58)

Yes, yeah. Yeah, think what the big thing that tech and money people always miss is the whole enterprise market, because it is a minority of the money, and it is complicated in a way that annoys them if you're trying to think of scale. But that's where I always preferred to work. The problems are more interesting. You meet very interesting problem solvers on the brand side, and I found I could get as interested in you know, the problems that CarMax was having with incrementality is when a kid would get interested in a band or a rap group. it was just, to me, it's just like a media problem and solution.

Brett House (09:41)

Yeah. So starting an ad tech company today is a lot different than obviously 2010 or 2015. We're in a much more mature market. There's a lot of skepticism and it's crowded. What convinced you that Chalice was unique that had an opportunity to really make a meaningful change?

Adam Heimlich (10:00)

Well, I did get a lot of encouragement from people who succeeded in the first wave, right? In Chalice's cap table is Tradesk's TD7, and we pitched Jeff Green directly, Brian O'Kelly and Jonah Goodhart, Matt and Art from McCordin, Joe Zawadzki and Eric from Imperium are in our cap table. So all these people seeing the business plan and being like, yeah, this can work, was a huge wave of encouragement. It came like a year later, but I think before that was like, you the first people I showed the business plan to and the first ⁓ data scientists and engineers who were coming on who were very encouraging and saying, yeah, that like what's happening now in cloud and AI services is making this cheaper and lighter and faster than it would have been even, you know, two years ago or even faster than last year. And that dawning realization that the cloud companies had were investing billions, even hundreds of billions in infrastructure in anticipation of companies becoming AI factories was like, great, we're like early in like, know, Databricks and Snowflake, they love us because we've been building an AI factory for six years on their infra, running up bills and you know, creating use cases. that's, that's a huge win on your back that I don't think the first wave had.

Brett House (11:04)

Yep.

Rio (11:17)

You know, it's it's like I said, you were early, which is really cool to the AI story. And there's a lot of different AI companies now. And it's interesting what you said, Brett, to about like the cost and time it takes to have a startup. I think I read a stat the other day that the cost for building a company has gone from, I think, is three, three and a half million down to five or six hundred thousand. Right. With some of the new AI tools⁓ but, but regardless of cost, it still is a big psychological change, right? I imagine going from an elite, going from operator to founder, like, like what, any, any takeaways or things to call off for other people. Cause a lot of people right now are considering that I've noticed so many people like, know, Brett, you know, included who are taking a, taking a plunge. it, cause yeah.

Brett House (11:55)

Or yeah, taking the jump into the abyss.

Adam Heimlich (12:00)

Yeah, I mean, thought, I mean, my first realization was like, ⁓ it's not a certain kind of person. It's just something you do. you do, once you do it, you're a founder and then you figure it out. It's not like anyone. I mean, you meet people who are entrepreneurial in high school and you know they're going to be founders, but that's unusual. And there's no founder type. I met a lot of people who are founding at the same time as me, who I'm still in touch with. And it's a very diverse group you don't have to have the courage all the time. You have to have the courage at the moment of founding. Then just hang on.

Brett House (12:35)

Yeah, yeah, exactly. And then it's like it's like it's the old fuck plan B ⁓ Sort of mentality right there is no plan B like if you really

Adam Heimlich (12:39)

It's like having a kid too, it's like

Rio (12:40)

Ha

Adam Heimlich (12:44)

you're not going back, so it's like, you know, you just have to be.

Rio (12:46)

Burning the boats, right?

Brett House (12:47)

There's no way but forward, right? So, well, I think one of the most interesting things, and I gotta say this, being married, I've got two boys soon to leave the house, one's already left the house, but you built this alongside your partner, Ali Manning. You know, I tried to imagine myself in your shoes and I thought, wow, I don't think that could work on my...

Adam Heimlich (13:02)

Yeah.

Brett House (13:09)

I love my wife. Sorry, I echo if you're hearing. But how did you pull that off? How does that dynamic work and how do you feed off of each other and stay out of each other's way when you need to?

Rio (13:09)

I probably couldn't work with my wife, but...

Adam Heimlich (13:18)

Well, I'm not going to say it works. ⁓It happened that, I mean, me and Ali met at work. We met at Avenue A, Razorfish, and we're all working in backend data for a large banking client. So ⁓ we had, you know, experience of working in the same industry and we coached each other. For many years, she was at Google while I was agency side. So we were sort of dealing with each other's companies and we were both advancing in our careers and helping each other. you know, I knew that we were a team. Chalice was founded in those first six months by me and Ken Rona, who's a data scientist and kind of hacker engineer, a real wizard in the early stage stuff. But by that summer of 2020, we wanted to pitch for funding and make a business plan. Allysing is operations, so she's made business plans for billions of dollars worth of budget I said, will you help us with a business plan? And I think she just got interested. It was the pandemic year. No one really had full-time work and could be that engaged in what had been an office job and then a home job. So I think the more she did it, the more she enjoyed it, though she had been trying to get out of advertising. ⁓ And yeah, it's really rocky. All the reasons people say that it'd be hard for them or that it's a dangerous thing⁓ are true, like, you know, want when your partner is your coach from work, like they're entirely on your side, whereas when you have a work argument, they can't really coach you through it because they might, the physician might be adversarial at work, and feedback, negative feedback from your partner hurts much worse than negative feedback from a partner, but I will say it's gotta be personal.

Brett House (15:03)

Yeah, you're much more likely to take it little personally, right? Yeah, that's to say the least. Yeah. it's only professional. It's only professional, right?

Adam Heimlich (15:10)

But think it works for the company the founders should be essentially in a love relationship. They should be complementary and better together than neither of them are apart, which is close to the definition of love or a marriage, right? It's like...

Brett House (15:25)

Yeah, without the codependency, without the codependency.

Rio (15:28)

Well, I think the best startups, the best startups have that, right? You had like Jobs and Wozniak, had like, know, Gates, Gates and Allen, where you have people who like bring complimentary skills. So yeah, why not? Right.

Adam Heimlich (15:33)

Yeah. Yeah and help shape each other's best ideas and protect the company from each other's worst ideas and all that's good. But it can be a real struggle for us to come together, the decks we've made together, some of them have been struggles and involved like a couple of therapy sessions just to finish the deck. those are our best decks and that's our best work. But we do have a pretty solid department split, so we're not running into each other. She is Operations. Sales, Marketing, Finance, HR, am Product, Strategy, and Partnerships.

Rio (16:14)

Well, from the exterior seems like you're making it work. I follow both of you on, you know, online and I see a, she'll post a picture of we brought the kids to the office because school was closed or some emergency, ⁓ you know, it seems to work. It's pretty cool. So like, like, I'm going to start up life is supposed to be like all consuming. Like, ⁓ how do you balance that? I mean, I can imagine it's a lot, right.

Adam Heimlich (16:23)

Yeah. ⁓Yeah, it's a lot. mean, a lot of people with kids work hard and we were never like, you know, like young founders who were like working ridiculous, you know, 18 hours a week and through the weekends, we never worked like that. We never worked those kind of hours at agency or... Yeah, we've kept some balance, but you know, the reality with little kids is...

Brett House (16:52)

That's good. So you've kept a little bit of a balance, right? You keep, you keep, you're keeping.

Adam Heimlich (16:59)

if you have good health, like it's a few hours. There's an hour or two in the morning and you get them to school, there's an hour or two in the evening and then you put them to bed. So it's not like, it was manageable. making, you know, we had some money and really, really good help and a good neighborhood and some relatives and all that. So I don't want to exaggerate like how crazy it was, but it was many, many years of not really resting, like going from work, you know work to the kids, so the work to the kids and even a vacation being tough, like for many, many years. Those are tough years and it was a tough thing to do being over 50, but it's a lot easier.

Brett House (17:38)

Yeah, yeah, it'll keep you sharp going into your later twilight years, know, more active you're, challenge your brain, right? And I think what I've learned is you gain a little bit of wisdom with age. And I just, I'm a much more, I think I'm much more capable of doing this now than I was in my twenties. Like when Rio and I started something together, you know, I didn't know shit. I just didn't know enough from both organizational management to multifunctional management. I just didn't have enough.

Adam Heimlich (17:40)

That's it. Yes.

Brett House (18:06)

sort of wisdom and knowledge to be able to run that.

Rio (18:07)

Well, plus remember Brett, like, cause we ran this whole like tech company, like, know, was early ad network. We, you know, we had, did a lot of media buying, you know, for, for some clients. And we worked a ton, right? But I think a lot of that was just like hanging out, play after hours, playing ping pong and stuff like that too. So like, think that now, like I agree with you, Brett, probably better at managing my time and better at focusing a hundred percent on stuff. Right. So compartmentalizing. Yeah.

Adam Heimlich (18:20)

Thanks

Brett House (18:21)

Yeah. Yeah, compartmentalize.

Adam Heimlich (18:29)

And team,

Brett House (18:29)

Yeah.

Adam Heimlich (18:29)

know, like the reason it's, big reason these easier now is they have a really good leadership team. And some of these people have been with us for years. Our CTO, Joe Wilson, and head of AI and data science, Thailand Petri joined us in 2021 or 2022. And they, you know, have led and grown their own departments. Those are, of course, the two of the hardest. it's, you know, Rob Mooney, our head of product has also been with us for many years.

Brett House (18:53)

Yeah, so.

Adam Heimlich (18:58)

It's, well, those are, most of the jobs.

Brett House (18:59)

That's great. Well, that's a testament to your leadership skills, to the fact that, ⁓ yeah, I people that stick around, that just shows that loyalty is there for a reason. ⁓ That's great to hear. Yeah, it's become a family in a way. ⁓ All right. Now that we've gotten personal, so yeah, yeah, yeah, let's talk about technology. So you've been deeply involved in the agenticRTB framework, ARTF, for the acronym, if people aren't familiar with it, ⁓ with the IAB Tech Lab.

Adam Heimlich (19:07)

Yeah, we're on a mission. We're definitely on a mission.

Rio (19:16)

Let's go to the pivot to the tech. Yeah.

Brett House (19:27)

And for listeners that may not be in this day to day, what problem? I have my own summary of what I think ARTF is actually trying to solve, but I'd love to hear from you what you guys are trying to solve when it comes to autonomous media buying systems. Yeah.

Rio (19:41)

and how it's different than some of these other approaches we're seeing too.

Adam Heimlich (19:45)

Yeah, let's start with what people know. People know that AI is really powerful, and they know that behind this magical machine is an enormous amount of data processing, right? Everyone knows that, right? They can see that. They read about the costs. ⁓ And in ad tech, there was really no way to bring what we know of as AI, like that sort of magic, into the event level, where the exciting part is the actual bidding. ⁓

Brett House (19:57)

Yep.

Adam Heimlich (20:15)

because of the pipes, right? It's like you're trying to bring a 2025 hot tub into a 1930 house, right? It's like the infrastructure can't carry that. everyone knows it's a lot of data processing, and there are actual data processing limits of the platforms. So that should be simple enough for everyone to understand, right? But now getting a level more complicated is we started...and the idea that we should use every field in the bid request. One of the ways that that pipeline has been narrowed for platforms is they mostly look at the IDs, and not all the IDs, they look at their own IDs, and they especially look at IDs that they could track conversions of, and especially IDs that they've tracked conversions of lately. So that shrinks things down to a much more manageable level for their machine learning. So our first idea is, all right, we're going to work with bigger models. first of all, we're going to look at all IDs and second, we're to look at other fields. And as LLMs get bigger and bigger, we started to get more interested in the language fields, which are mostly thrown away in the old ML pipeline. So there's domains which were turned into categories, so they could be numbers, but we actually started using the words in the domain. And then the words in the subdomain, then the words in the full URL. And then we started working with Deepsea, who scrapes the web we said, all right, let's every time there's a full URL, use all the language on that page and the metadata about the quality of the page. Now we have a much more predictive model and actually fix a big problem in programmatic, which is when you only look at IDs, you don't care at all about the quality of the page. And if you're trying to get the lowest possible CPM, the incentive becomes to find that idea in the cheapest possible page. Right. That's why we have clickbait. That's the reason.

Brett House (21:56)

Yeah. Yeah. Without without any of that context, which is so critical. Yeah.

Adam Heimlich (22:01)

Yeah, so everyone knows clickbait's a problem. Most people don't know it's because DSPs only care about IDs and don't look at the page. So we're like, all right, we're going to have a model of all the pages on the open web and we're going to score them not in general for quality and context, but for how well they match what the advertiser's trying to do or how effective they are for a campaign outcome. And then we're going to try to deploy this. And we had a lot of trouble deploying it because it's big ⁓ and DSPs aren't built for it we had been inserting our models into DSP. And we started talking to SSPs and right around this time in 2024, two years ago, Andrew Casale said to me, we can host one of your models in our server, right? He's the CEO of Index Exchange, which is one of the, he's one of those entrepreneurial guys who started our company in high school. He's been running that exchange ⁓ for most of his life and had recently re-architected it and explained to me about this technology called Kubernetes or edge computing, which is the technology of high frequency trading where you replace a server to server connection where a message comes out of one place and lands in another and they send a message back to doing it in a single server where one of the message systems is in a container, like a virtual desktop, a computer within the computer. And then the message doesn't leave the server. It's a lot more energy efficient. It's a lot faster it's a lot cheaper. That's basically it. That's basically the idea of what became containerized RTB and ARTF. It took a long time to build. We didn't test it until the end of 2024. And then in 2025, we started running it. One of the first campaigns we ran was for Zillow. It did so well, the case study ended up winning the Best in Show at the AdExchanger Awards. We took what Index and Chalice said.

Brett House (23:54)

How did it do well? What were some of the metrics that you were looking at that sort of said, this is it, we nailed it.

Adam Heimlich (23:59)

It was Zillow drives leads, but then they also want to look at the quality of the leads. So it turned that ID looking thing into how many leads are actually going to be useful for Zillow. The people who are actually home shopping are actually buy a house. So we looked in the quality and context of the pages to get much better signal over whether this Zillow lead was a legitimate house shopper. And it did like 40 % better cost per lead basis than a DSP on its own.

Rio (24:28)

So Adam, so it sounds like just to summarize then for, guess, sort of viewers. So it's a technology that's using, like, cause you're not only looking at, like, the standard contextual signals, but you're really like, you're using AI to really understand really deeply based on domains, subdomain, the actual content in the page. Like really what's going on and what this inventory looks like. And then instead of just like, you know, in a normal DSP looking at the audience and trusting the ID, just to say, okay, this is what it is. You're able to better target.

Adam Heimlich (24:42)

Yeah.

Rio (24:55)

the advertising, then it's actually running on the SSP, the supply side, but this is something you're selling then to the brand who's actually buying media, right?

Adam Heimlich (25:03)

Yeah, yes, the ARTF was just a means of deploying a bigger model. And what the model does is predict the cost and value of an ad opportunity or of all the ad opportunities, right? So just bigger model. That was step one. Then we took it to IAB. I was able to convince Anthony, the head of IAB Tech Lab, to make it a standard. The standard committee was very well attended. All the top publishers and Netflix and Amazon wanted to sit in the committee, and their interest is super cool. They're looking for ways to control where their data goes, right? Because these infrastructure requires them to either withhold show-level data from everyone or send it to everyone. This container starts to give them the capability to send it only to people they trust to properly value it, which is the opposite of what happened with newspapers, Nobody properly valued newspapers and they lost all their money. And so continues, right?

Rio (26:00)

That continues, yeah.

Brett House (26:01)

Yep.

Adam Heimlich (26:03)

They're being smarter about it. They're interested in the containerization. But now it is a standard. It's been adopted by OpenX very successfully. Pubmatic is building it now. There'll be probably a couple more SSPs this year. So it's a very successful standard. And the roadmap is very exciting because what started as just a means to deploy a bigger model now becomes a whole other framework for companies to work together. Amazon has launched something really similar called RTP Fabric the idea of a single server environment where the bid request can ping around to different partners who can change the bid request, augment the bid request, add data, and then a custom model can answer that data. the implications are very profound. We're going from a world where there is one set of bid requests and everybody got it to every advertiser might see a bunch of different bid requests depending on their partnerships and that can respond with custom models with much greater ease than ever before.

Brett House (27:05)

So they have a lot more control over the decisioning logic before the bids actually won and the data, right? Which allows them to kind of almost insert for the simple minded, but insert some of the strategic planning, some of the context that only they have as the brand into that play so that they know, which is gonna drive effective outcomes, right? Because it's gonna be more likely to.

Adam Heimlich (27:09)

and the data, the data and the decisioning logic. Yeah. Yeah.

That's right. I'm glad you said that, Brett, because yeah, because I love people say, it really agentic? It doesn't fit the definition. But what you said is what makes it agentic. What makes it agentic is that it will serve the advertisers intent. Right. A lot of people when they talk about agentic forget to ask, who is it? Whose agent is it? And what does it do for them?

Brett House (27:40)

Yeah. Yeah, yeah, yeah, yeah, exactly.

Rio (27:47)

The agents should be represented, the agents like representing the interest of either buyer or seller. think that brings up an important question, Adam. So like what you've created, mean, it's like, bid management might be one of the applications of it, but in theory, you're giving all this incredible contextual information, Brett's your point, the brand can then insert some of what they want to accomplish into this. Like when looking at buy side and sell side agents, could this apply to different types of deals?

Adam over time and different ways of negotiating between buy and sell side.

Adam Heimlich (28:18)

absolutely. And not just by IncelSide, but all the partners in between. And then rearrange them, right? So, ⁓ like an agency roster of partners, you can think of them as now being more or less drop or drag into this container environment. know, the tools haven't been built where they physically drop and drag a tile and it happens in some container, but it might only be a year or two away. And... ⁓each of these, it's a very different way of working than a platform, right? Because everyone is modular, doing their piece in an environment they don't control, and they have to serve the end user intent, which might change. that's what being an agent is about, to me. You're a modular service that changes depending on the end user intent. We're seeing a lot of things called agentic that is just workflow automation, and maybe that is a kind of agent, but for me, it's much more exciting to ask when all the ad tech partners are agentic, it will mean that the end user can arrange them to do exactly what they want.

Brett House (29:22)

And I mean, do you see a world where like, there's all these kind of fragmented decisioning layers, like within RTB. mean, it's like the strategy is happening on the brand side. Then you get the supply path and supply path optimization, and then the bidding requests and brand safety, which is a different vendor. And then budget pacing, which is obviously within the DSP or SSP. Yeah. And then attribution, which is another vendor altogether. Do you think containerization can actually help bring this stuff into one sort of...

Adam Heimlich (29:37)

Yeah, it's slow and efficient. Well, well, Chalice alone, that's Chalice alone's mission is to put all the decisioning logic in one calculation. And we're getting there. We're getting to where we can do it. People don't realize that most DSPs will do ⁓ a value calculation, like, do we want to win this one or not? And then bid and then do a bid shading calculation separately. That's extremely inefficient because the calculation doesn't catch the relationships between predicted value and predicted cost. And those are correlated, right?

Brett House (30:05)

Yeah. Yeah.

Adam Heimlich (30:17)

This has been statistically proven that you shouldn't do those separate. You should do those together. But it's a lot more computationally... ⁓

Brett House (30:22)

Yeah, otherwise you might be buying some really shitty CTV inventory that's appearing on mobile devices and not on televisions, right? Because you've... Yeah.

Adam Heimlich (30:27)

Well, it makes the error very high stakes. if, right, like if everything is based on, like what we do is we do value to cost prediction of everything and then stack rank them by the ratio. So the first things we're going to buy are the value to cost are furthest apart. Then if you're awful little, it's still profitable. When these two things are separate, being awful little can cause you to miss an auction you're really supposed to win.

Rio (30:53)

Who is this approach resonated the most with so far? mean, it sounds for a brand perspective, it sounds pretty appealing. I mean, I also think for people in supply, right? mean, it's because it's like, it's demonstrating like the people, well, let's say the people with high quality inventory, right? This must be great for them. Like what are you hearing from the marketplace as you talk to different people?

Adam Heimlich (31:12)

A lot of excitement from agencies as a way that they can add value again, because it really plays to their strengths of having a full marketplace view ⁓ and be able to offer different partners and different configurations and handle that. I think the agencies that are looking for a way to charge transparency for service advertisers can't do themselves.

Brett House (31:32)

⁓ transparency is a service. TAS, I love it. We have a buddy that's contributing to Signal of Noise, ⁓ Mike, who coined CLAW as a service. So everything seems to be as a service, but transparency is a service. That's an interesting play because that is a big... We talk about it all the time. We Dr. Fu on the podcast. We've had a bunch of people about ⁓ ad fraud and how it continues to be this persistent problem in the industry.

Adam Heimlich (31:43)

Hehehehe. Hehehe. Hehehe. Mm-hmm.

Brett House (32:00)

And it's just a complete black box oftentimes to the advertisers.

Adam Heimlich (32:03)

Yeah, so mean, live ramp sidecar was the first commercial container and adds a live ramp ID to a bid request in one millisecond. And Dr. Fu could have a container that adds his estimation of the percent fraud in one millisecond. And Kristoff from analytics can add his in the same millisecond, right? Because those don't depend on each other. So you could see it's endless. You could have hundreds of them going simultaneously and some of them have to be sequentially like Chalice isn't going to bid until we get all these.

Brett House (32:20)

Yeah.

Adam Heimlich (32:32)

But that's what I mean, like then we could calculate all at once. Foo's estimation, Christoph's estimation, the estimated cost, you what Gareth said about displacement history, like all in one calculation. That's the most computationally efficient, the most energy efficient, and it very much aligns with what we're seeing with AI. The reason these tools are so magical is they're doing tons of computation in the backend altogether, right? The language model.

Rio (32:58)

Yeah, scale and quicker. Yeah, it's amazing.

Brett House (33:01)

Do you find that you're in clients of it? Is this a tough sell? Do you find yourself doing a lot of explaining so that they can sort of wrap their head around or are there people in the industry on the DSP side or otherwise that are sort of pushing against this sort of new approach, which is different than what we've seen before?

Adam Heimlich (33:17)

There's a lot of, I mean, this is what I was talking about adoption. Like there's, cultural barriers to adoption of new technology every time just because it's new and it disturbs workflows. And it's magnified with AI because it's scary. Although people were scared of search social and programmatic too, maybe not as scared. But I'd say we got the early adopters and Chalice was fortunate to get the attention of early adopters and get money flowing and quickly become profitable and now like we know that the middle and late adoption is a heavy service lift, a heavy evangelism lift, a heavy marketing lift. You know, we hope it's not so expensive that we don't need to throw giant parties again, but we will if we have to. We go, we bring a big team, but we haven't rented a big space yet. But it's like I say, I come from agency and this is something we have to tell people need service. Like we have an excellent

Brett House (33:46)

Yeah. just for the fun of it.

Rio (34:01)

I saw you a can last year. I think you're saying.

Adam Heimlich (34:14)

customer service team and we expect we're going to run one 10K campaign and hopefully show them something they've never seen before. It's going to let us get a little by little.

Brett House (34:22)

Winning hearts and minds, right? Yeah. No, you had a post that I read quickly yesterday, I think. I don't know how recent it was, but you said, and I thought this is a little more theoretical, but I thought this was a really good point. I think it ties into the point you're making about people adopting this in kind of moving through this sort of resistance layers of what they know or the workflows they currently work in. And you said, I just want to point out the idea that we all have, ⁓ that we'll have AI agents that can do things for us. that is not consistent with the idea that they'll do exactly what we do without them today, right? It's natural. The first thing people did with movie cameras was film plays before they knew what a movie was. So do you see us, I mean, to me, that's a, do you see that we're in this moment right now where people are almost containerizing, to use the word, their own view of what AI can do? And it's really just, ⁓ it's sort of in a box. It's what we do currently today. It's just,

Adam Heimlich (35:16)

Yeah, it's just human nature. I don't get too upset about it. McLuhan said we drive into the future looking in the rear-view mirror. And that's true, like, you know, most people can't visualize things. Like, I've been deep in this six years, so I could visualize it. I could explain it as well as I can. But the best I could do is show you, like, let me run a campaign of any size and hopefully I could show you something you've never seen before.

Brett House (35:17)

Making it's compounding this. Yeah Yeah. Yeah.

Adam Heimlich (35:41)

I used to get more worked up about it, but it's like if someone's never had home cooking, they've only eaten McDonald's, Wendy's, and Burger King, like you could tell them about it, but you know, until they taste it, they're not gonna believe you.

Rio (35:53)

Yeah. mean, well,

Brett House (35:54)

Yeah. ⁓

Rio (35:55)

but it's interesting. I mean, at the point you made Brett about like when, like when AI is first coming out, like we're just going to try to use it to automate things we already do that maybe we shouldn't even be doing. Right. Or maybe they will look very different once AI is actually applied. I think Jeff Green set up Market Tech. heard a couple of people commenting on the crowd. You're saying how like AI, like programmatic, fewer things lend themselves better to, I'm just paraphrasing it to AI, to AI than programmatic. And well, then I guess, I don't know, like programmatic is a way of automating things with right with ⁓ primarily through real time bidding. So when we actually go agentic trading, the do things look the same? don't know if you have a point of view on that, but I think they might look different.

Adam Heimlich (36:33)

I think there has to be an infrastructure upgrade. Yeah. Scaled SaaS is a plumbing system. The Scaled SaaS in that tech is an outdated plumbing system that needs an upgrade. And it's not crazy. I think Trade Desk has done big parts of this upgrade and ⁓ Index like kind of took a year off from promoting new products while they were internally doing these upgrades, re-architecting and it'll happen. the cultural stuff is moving pretty quick.

Brett House (36:56)

Rearchitecting? Yeah.

Adam Heimlich (37:03)

⁓ But there's something about the mentality of Scaled SaaS that's going to be tougher for the platforms. Like the idea that we're a hub of data and data comes in and we make the AI and sell the product back to you. That doesn't look as great an idea to me as it did 10 years ago. It's because it's not the only way. ⁓ And it's probably not the best for a big advertiser because there's multiple hubs. If there was one, maybe. But I have to send all my data to Meta and to Google and to Amazon and to Trinidad.

Brett House (37:35)

Yeah, multiple point solutions, multiple data warehouses is supposed to managing it all through one system, whatever you are, a Databricks house or Snowflakes house, and then plug in the application layers, the IAL layers, right, into that.

Adam Heimlich (37:38)

It's weird. Sometimes I lead a pitch with saying, we estimated your data at a trillion events and two terabytes a day, and we think it's massively effective for training AI. So whose AI are you training now? Let that sink in. was like, you don't know? I'll tell you.

Rio (38:03)

trying to front your model probably, right? ⁓

Brett House (38:06)

Yeah,

Rio (38:11)

But yeah, but, when looking at like skilled, like skilled sass and the impact here, mean, like, don't, I mean, there are having people saying recently on there's a lot of conversation on X. ⁓ and we'll get into the social thing in a minute. I know this is something I wanted to talk about, but, but, which I still call Twitter most of the time. ⁓ but, but.

Adam Heimlich (38:23)

X with that. Yeah, Twitter. ⁓

Rio (38:31)

But, but, know, most of that tech is

Brett House (38:31)

You're aging yourself.

Rio (38:33)

on there and people, people were like the other day saying, will there even be a, will there even be DSPs in five or 10 years? I think there's going to be disruption, but, you know, looking at Martech and I think Martech's more probably going to be disrupted more than Martech stocks, however, really tanked a I don't see like, like it's an enterprise SaaS going away anytime soon, but I think their ability to sell licenses and the whole, the model they go to market with and bill for may probably be disrupted to a certain extent.

I think ad tech's less vulnerable, if you ask me, because these are big models. They're fast to your point. There's a lot of moving parts. ⁓ But I do see the way people work and these platforms changing.

Adam Heimlich (39:08)

Yeah, people aren't great at second order effects. If the way things play out along the same track, it's going to be so much more competitive. There's going to be so many more companies and the ease of bringing a product to market and making it big, the cost of that is going to go down so much that our whole economy will look very different. So the question to me is, is will that be allowed to happen or a few big companies can be able to shut it down? You know, it really should.

Brett House (39:37)

Yeah, that's that.

Rio (39:37)

hope it's allowed to happen. mean, it it would be a shame if it's not right. I mean, cause you know, the point we made earlier about it's so much cheaper to create a create technology. Like we've had guests in it. Yeah. We've had guests in this pod who are saying how like they like one, one, one dude can like do all of the work with clients, gather requirements, prototype design, vibe code. you know, it's like six or seven roles like people, and they can do things in a couple of weeks that would have taken months before. So it is kind of amazing how quickly things are going. ⁓

Brett House (39:45)

Yeah, product live cell compression. Yeah.

Adam Heimlich (40:06)

And the marketing, you know, Facebook has changed the economy in that anyone who builds any sort of gadget can market it on Meta and find their first, you know, 5,000, 10,000, 100,000 customers for like a few bucks each. That's insane, but it's only Meta and only gadgets and direct to consumer energy drinks and stuff. But that could be the whole economy and a lot more than Meta able to do that. And we'll see a lot more companies in every category. Should.

Brett House (40:39)

Yeah. So should we pivot to Adam the shit poster? ⁓

Rio (40:43)

Yes. Yes, do it. Yeah. We had a couple of questions in there, Adam. So I love to hear your, yeah. I love to hear your thoughts on it. I mean, I love, I love some of the wars you get into on, uh, you with, some of there's, there's a lot of trolling going on. Um, and I'm not saying by you, but in general on, uh, on social media ad tech is, is, is a funny ad tech, you know, on Twitter slash X is a bunch, bunch of people. Um, curious, get your thoughts on it. Um, I know that you have some rules. You told me the other day and we were hanging out at ramp up about like.

Adam Heimlich (40:55)

Yeah.

Rio (41:12)

Who you engage, who you engage with, who you just say, I'm blocking you too bad. Like maybe we start there.

Brett House (41:19)

The rules of engagement for Adam Heimlich on social media.

Adam Heimlich (41:23)

Look, think when I got into the media business in the 90s, it was not that professional. It was not an elite profession. And even advertising back then, it was starting to become professionalized and elite. And the culture of the American elite that you never act out and you never melt down. It never really works for me. It doesn't seem right to me. think it's repressed and it causes problems.

Brett House (41:51)

It's just disingenuous, yeah.

Adam Heimlich (41:52)

It's disingenuous and like, yeah, like a lot of America hates it who aren't involved. And yeah, it's just it's I think it it's sort of an aristocracy creates one thing. And I think I think it creates I think one of the reasons you see like really rich, successful people act out so much as they feel like I've I've held in bad behavior so long and I've made it and I'm going to. Yeah, I deserve this now. Yeah, and.

Rio (41:57)

I hate it too.

Brett House (41:59)

Yeah. Yeah, who's gonna restrain me? Like, who's, like, like, you can't, nobody can fire me, you know?

Adam Heimlich (42:22)

Yeah, that's where I come from. I was raised in New Jersey, New York City family, and I don't think... I don't want to hurt anyone's feelings, but if I'm sort of rough with someone that I think their post is bullshit and they say so, I don't think they should think it's the end of the world, even if they think I'm a dick, which is fine, but I don't think they should cry about it. Or if they do cry about it, they should get over that.

Brett House (42:47)

It's healthy debate, that's what it is. ⁓

Adam Heimlich (42:50)

Yeah, but, ⁓but yeah, so that's, that's it. I'm, know, I'm not always proud of my social media behavior and arguing. Sometimes, sometimes I get edgy, you know, sometimes I'm too angry or edgy or things come out, come out wrong. But most, but my best, I'm trying to be funny. I'm trying to lighten the situation and add some humor and levity to these environments where every people are promoting themselves or, or say, you know, doing their best at saying accurate things, but missing.

Brett House (43:18)

Yeah, yeah, they're either shameless shills or they're just inaccurate and you're just correcting their inaccuracies.

Adam Heimlich (43:21)

Hahaha

Rio (43:23)

Some people take the yeah, they

Adam Heimlich (43:24)

I love that the foot again

Rio (43:26)

they think themselves way too seriously. Some of these people, right? I mean, it's it's.

Adam Heimlich (43:28)

Yeah,

Brett House (43:28)

Yeah.

Adam Heimlich (43:29)

the deflating comments can be really fun. know, there's other people engaged in the same way, but you know, if I make people laugh, sometimes I meet people at conference and they say I posted something that made them laugh and that's good. That's what I feel like. That's good. It's not a bad brand.

Brett House (43:46)

Yeah, well, sometimes sometimes people like us that are from the New York, New Jersey area, we can come across as, know, we show our cards and we play it straight. And yeah, ⁓ there's no obfuscation here. It's just, you know, this is what it is. So I appreciate that.

Adam Heimlich (43:53)

Yeah, it's a rat.

Rio (44:00)

Yeah, Yorkers call the spade a spade. I love that about they're brutally honest, which is good. I being a New Yorker myself. ⁓ But one of your rules, if I remember though, was like, if you tell me don't sell me and someone continues, they're done, right?

Adam Heimlich (44:11)

Yeah, yeah, yeah, I accept most LinkedIn connections, but if they pitch me, I say not interested. Another pitch after that, and mostly it's bots now anyway. I just blocked them, which resulted in me having like thousands of people block them. But there are also a few people who their commentary has annoyed me so much I've blocked them, and I've heard that's really hurt some feelings. I try not to do that. You could unfollow them, you know, in ways, you could get them out of your feed in ways that don't.

Brett House (44:23)

Yeah.

Adam Heimlich (44:40)

they don't see and that hurt your feelings, but there is one

Brett House (44:43)

Yeah.

Rio (44:44)

If someone's going to go to political commentary, like you got to accept that's part of it, right? If you're going to challenge people and argue with them, they can block you, right? mean, I don't know. Go ahead. I cut you off Adam.

Adam Heimlich (44:53)

yeah, political, yeah, but it was actually, there's an industry guy who every use post is like, the startup CEO should, and the startup CEO must, and the communications of the startup CEO, and I was like, I feel like this guy is like yelling at me, like I'm doing everything wrong, I feel like I'm getting constant negative feedback, so I'm just gonna block him.

Brett House (45:07)

Yeah, yeah. Sort of a dogmatic approach, right? Yeah. I mean, yeah, you don't have to say in my opinion before you say everything, but you could also be not so ⁓ prescriptive, right, about these things. ⁓

Adam Heimlich (45:22)

Yeah, and there's a

Rio (45:23)

All right.

Adam Heimlich (45:24)

generational thing too, like being our age and like the whippersnappers, like great, more power to them and you're up and coming in the industry, but sometimes they're...

Brett House (45:35)

Back to the comment we made earlier about wisdom. Knowledge and wisdom are two different things, right? Fundamentally. And it takes experience to know the difference, I think, between the two things.

Adam Heimlich (45:38)

Yes, I'm not sure if that's... Nah! Yeah, I don't think that's bad. When I walked into the newspaper and if I said something I thought I knew that the older people, and I was wrong, the older people would not hesitate to mock and smack me down in the harshest possible way.

Rio (45:56)

Well,

Brett House (45:57)

And put, yeah, put you in your place! Put you in your place!

Rio (45:59)

well, yeah, well, but, that's the generational thing. remember my first couple of jobs being yelled at, being told I was garbage, being like, mean, some of it was really tough, but it did, it did, it did make me like, like you gotta be good at your job. Like I remember first got into consulting, right? I like, Cartners would tell you, this is crap. Like, like go redo this.

Adam Heimlich (46:04)

That's not good.

Rio (46:17)

I can't believe you showed this to me. You're like, wow, like I thought I was pretty smart. And you realize why I didn't know as much as I thought I knew, right? And then this person that's been doing this forever. So I actually think it was good. And I'm not, don't, I wouldn't treat anyone like that, but, it did make me better at my job. So I do understand looking back why they did.

Brett House (46:23)

Yeah, yeah.

Adam Heimlich (46:34)

Yeah, you shouldn't be harsh and hurt people's feelings, but in writing, I mean, that's another difference, right? Because I was a writer first, I see social media as a writing exercise, and some people are thinking like, this is me. Like, this is actually me, and this is me speaking. I don't think that. I think this is your writing persona, and I'm going to respond in my media persona.

Brett House (46:51)

Yeah, yeah, yeah. Yeah, and that's back in comparative literature days. I I studied rhetoric, you know, and that's what that is. That's rhetoric, right? Like I'm using a rhetorical approach that this is how I represent myself professionally on X or whatever it might be, right? It's not who you are at home with your kids.

Adam Heimlich (47:01)

It's rhetoric, Yeah, it's my voice now.

Rio (47:11)

Yeah. Well, yeah, plus

Adam Heimlich (47:13)

Yeah, I think people don't know that.

Rio (47:16)

knowing, yeah, plus it's knowing this is your work too. mean, it's not you, right? To your point, Adam. So I remember when I first started writing professionally, bringing drafts of things to the editor, just crossed out, crossed out, correction, correction. First couple of times it's like, wow, I thought I was a good writer. This kind of sucks. But you realize, okay, this is making me a better writer. It's not personal. So I don't think that's not personally now, right? And I wouldn't, so I don't think people should be.

Adam Heimlich (47:36)

Yeah. I got some cloud feedback the other day that really made me want to punch cloud in the face.

Rio (47:42)

That's awesome ⁓

Adam Heimlich (47:45)

Yes, sir.

Brett House (47:45)

Yeah, so should we go to some of our questions? Yeah, quick hits, right? So which part of the ad tech stack do you think is most, I we've been talking about AI, we've been talking about containerization and your solution, is most at risk being replaced by AI? You DSPs, SSPs, measurement vendors we talked about, incrementality, whatever we want to call that, media agencies. Where do you see the biggest sort of weaknesses in the stack?

Rio (47:49)

Quick hits, let's do the quick hits, yeah.

Adam Heimlich (47:51)

Yeah. Well, I think SSPs are ahead right now in what Index and OpenX and some others are doing will give them a new breath of life. But ever since header bidding came out, I think their position has been a little bit precarious. Like, what really is the function? They're not doing yield management. They don't have original bid requests. Like, what are they really taking money from? know, shifting to saying like we're infrastructure for all kinds of of agentic flows of the bid request is a good one. But

Brett House (48:34)

Yep.

Adam Heimlich (48:42)

How many of those are we really going to need? I actually think the DSPs at the moment are more antiquated and maybe more threatened. ⁓ They have a more or less permanent foothold in being the buy-side source of monetary truth, of budget truth, and pacing, which is really important function that's hard to do at the edge because you need the steady state of the total budget. ⁓ So I think buy-side tech is a little more stable than ⁓ than sell-side tech. I think most measurement and data companies better be ready for a price war. Everything in there is going to be heavily commoditized. They'll all survive and have a function, but it's already the case that people don't really want to pay, don't believe in premium data or premium measurement.

Brett House (49:20)

Yep. Yeah, and a lot of those sort of advanced analytics, and when I came from the new star side of the fence for a little bit, so the market share solution analytic partners, and a lot of these are kind of heavy implementation, low weights, they're heavy, long managed services cycles, right? And I just don't think people are gonna be, yeah, it's too slow, it's too expensive. Yeah, it's consulting with a little bit of tech to kind of, but it's, and it's how.

Rio (49:48)

That model is getting blown up.

Adam Heimlich (49:50)

Yeah, it's like consulting. It's like...Yeah, AI is a big threat.

Rio (49:57)

Yeah, take it, take it six months or eight months in order to like, you know, have people come in and audit all your data and collect it. Yeah. I I, I could do that so much quicker now. So, ⁓ yeah. Yeah. But, but you, like SSPs, mean, I would have, you know, I don't know. I a year ago, I remember there's a debate about, they all, what is it? Are they all resellers? Right. ⁓ so, and, then, but, but I don't know DSPs like with some of the changes are maybe looking a little more vulnerable, but I think that is interesting. ⁓

Adam Heimlich (50:04)

We'll see what happens. Get on the query, yeah.

Brett House (50:05)

Yeah.

Adam Heimlich (50:16)

Yeah.

Rio (50:24)

But looking at, if you had to place a bet, will RTB still be the dominant buying model in 10 years? I mean, I think it's like 60, 65 % right of all, uh, programmatic is RTB right now. mean, like, like, like, like, will that change in the future as we go to agentic? Will there be, mean, obviously there's different types of deal structures, but like, do you think that changes dramatically?

Adam Heimlich (50:44)

Yeah, I think it'll split along Enterprise and SMB. I think Enterprise wants a really big auction, really needs it, and to be able to fiddle around in there and own their models, and that it doesn't matter so much in SMB, and they could do more behind the scenes. think it's kind of like finance. There will be a market that's for big players, for high rollers, like there is in finance, and there'll be more commercial markets for everyone else.

Brett House (51:15)

So what's the worst take you see repeated on LinkedIn talking about things that perturb you? Yeah, worst thing you've seen, either repeated or period about AI and advertising, but just the hype.

Adam Heimlich (51:20)

Yeah

Rio (51:21)

Or even it wasn't repeated, what's the worst take you've seen, period.

Adam Heimlich (51:24)

Hahaha

When they compare platforms and they're like, this one has the data and this one doesn't have the data, that is so stupid and I'm so tired of it. Like this idea that there's some data ownership that's determinative of success in a platform. This got disproven by Verizon and AT &T, and so many companies that tried to go into ad tech and copy Google. I mean, Google has the most data.

Brett House (51:55)

Yeah, where they can build a mode around some sort of data advantage that doesn't really exist because the data is relatively ubiquitous and it lives in big cloud infrastructures, right?

Adam Heimlich (51:58)

Ugh. Yeah, yeah, you hear it over and over in the trade disc criticism. Well, they should buy someone that has data. What is that gonna do? It's gonna piss off everybody in other side of the market.

Brett House (52:11)

Yeah. Yeah, they've got the same data as everybody.

They're saying that's the big problem with the Spotify's and the Amazon Musics of the world, because you're not getting anything unique. Everybody's got exactly the same catalog of music. Right. That's their data. And it's like you got that repeated over and over again. So each one of these services, it's just a race to the bottom from a pricing perspective.

Adam Heimlich (52:24)

Yeah, listen to that. Yeah, data is commoditized, modeling and service are not, and a lot of people have that backwards. You should read the Google trial, they had the most data by a mile, and they had this whole stack, and they still had to break the law.

Brett House (52:46)

Yeah, modeling and service is not ⁓ imminently replaceable. that's, I think that's key. That's AI plus the human kind of orchestration component, right? And conducting component, right?

Adam Heimlich (52:54)

Yeah, yeah, that's what you'll be able to charge for, yeah, I think.

Rio (52:59)

Yeah, well, there's definitely more more demand for specialized advisory and services around. I mean, not only AI, but I think tech in general. think that's the interesting thing. Although AI is disrupting providers of this to a certain extent and their business models is more demand than ever. No disagreement.

Adam Heimlich (53:15)

What gets overlooked a lot is that the adopters will make a lot of money. ⁓ the investor Jerry Newman has made this point about shipping containerization. Shipping containerization is like AI in that it added a massive amount of efficiency throughout the industry and it didn't change who the big players were, just like in AI. Amazon was big before. Amazon's going to be big in AI. Same with Google and Microsoft. But IKEA, who adopted shipping containerization to become a global shipper, Right? Changed everything for them. And Walmart, who just said, we're going to have the cheapest prices on everything no matter what, right? Becomes much bigger and more profitable. So his point was like, for an investor, it's actually really hard to pick winners because AI and service hard as hell to predict and how big do they get anyway. Too late to invest in the models and the adopters is a crap shoot. Like, unless you have inside information, you don't know who's going to be the IKEA or Walmart of AI.

Brett House (54:10)

That sounds like the source of your idea here, right? Were you thinking of the shower? You're like containerization. Think about what they're doing and how do we actually apply that in the tech space? It makes sense if you think about it.

Adam Heimlich (54:13)

Okay. It was very fortuitous. article is called AI Will Not Make You Rich. Brian O'Kelly posted it, and I was like, God, this is like the smartest thing I ever read. I read it a bunch of times. Thank you, Brian, for posting it. Then I talked about it on my show, and I was like, I've been fascinated with this article, and then this man, looked him up, and he was Corp Dev at Omnicom, and then early investor in the trade desk and became an investor. So after the show that I mentioned it, I got an email from Jeff Green with him CC'd It was like, meet Jerry. Great guy, Jeff, Seth. So I got to meet Jerry after studying his article, and I got to talk him for an hour. But incredible thought, and yeah, funny coincidence. The chipping containerization includes that word containerization.

Rio (54:51)

That's cool.

Brett House (54:51)

Nice!

Rio (55:05)

Well, it's crazy to think that people used to ship stuff and just like, guess, manually just stack it in ships, right? In hulls. didn't have these...

Adam Heimlich (55:10)

Yeah, they used to take it out of the box and put it in another box. They used to take it off the ship, out of the box, then in another box and on the truck. And this people come along and said, why don't you just ship it on the truck box or the train box and it reduced shipping costs at least 70%.

Rio (55:28)

Yeah. Well, it's funny, a friend of mine used to, his parents, I guess his family business at the port of Munich, no, no, Port Hamburg was like they would, cause the ships would come in, there'd always, if there'd be damage, there's stuff with some pallets back then. This is pre-containers, right? Like stuff would move around and get damaged if there was a storm. So a lot of these pallets would be right off and they would just buy them in bulk and then sell them through these retail. And we never know what they'd get. So, but the whole business went away when they switched to the container model. My buddy was telling me.

Adam Heimlich (55:52)

Yeah, I mean, that's the cultural adoption. If your job is to take the things off the boat and put them on the truck, you're not going to love this new technology. It took a few years, but once, you know, it's inevitable that it's going to be completely adopted, was his point, because it's so much cheaper and the cultural resistance can't hold out over a critical mass. That's what's going to happen.

Rio (56:03)

They might disrupt you.

Brett House (56:03)

That's great.

Rio (56:19)

So Adam, looking over the next few years, like any big predictions or are there any misconceptions you think people have that'll get cleared up? Like where do you see things headed?

Adam Heimlich (56:28)

I think the small business enterprise split is starting to happen and a lot of people have missed it and they talk about advertising like it's unified across those two and it has been largely because of Google. But that's a really weird thing that the Coca-Cola company uses the same tool as your pizza place. That's not the norm.

Rio (56:47)

Well, well, for social

Brett House (56:49)

Yeah.

Rio (56:50)

and for Google, but how about for like for TV? I think that's interesting. I mean, like, I actually think that the opposite might happen, right? Like some of the, like the generative AI video, video tools might actually help small businesses like enable them. So.

Adam Heimlich (56:57)

Yeah, that's a point.

Brett House (56:59)

Yeah, yeah, because it reduces production cost of the actual commercials themselves. And then when you have programmatic inserts at the local level, right, it does sort of democratize advertising and on television, arguably.

Adam Heimlich (57:05)

Yep. Yeah, I would say it'll be easier to get into the big, it'll be easier to get a ticket into the big company markets than it was before, but that'll be substantially different than the small company markets and for the better because the degree of complexity they need has not been met by Google and Meta.

Brett House (57:24)

Yeah. Got it. So yeah, think that covers everything, right? We're a little bit over an hour. ⁓ Thank you, Adam. This has been ⁓ super fun. ⁓ You're very well read. I can see all the books in the background. ⁓ And for those that didn't know, he's got this green screen, right, that he customized. ⁓ But it was awesome having you on the show. We'll probably see you at Possible. I'm sure you'll be there. Everybody's gonna be there. It's the can of the US now in Miami.

Adam Heimlich (57:34)

Good. This is fun.

Rio (57:38)

was great.

Adam Heimlich (57:41)

It's My library.⁓ yeah.

Brett House (57:58)

April 27th, O'Rio and I will be there for signal of noise, interviewing people.

Rio (58:01)

Yeah, it won't be quite as

Adam Heimlich (58:01)

Yes, let me see that.

Rio (58:03)

hot as can. It'll be great.

Brett House (58:05)

Yeah, and for all of those that made it this far the episode, this was a pretty quick episode. Sometimes we go over an hour and 15, hour and 20 minutes. ⁓ Be sure to recommend Signal & Noise to all your friends and colleagues. You can find us at SignalAndNoise.ai. Yeah, on YouTube, Apple, Ad Tech, Ad Talk as well.

Adam Heimlich (58:10)

Have fun playing Smash.

Rio (58:19)

smash that subscribe button, follow us. Yep.

Adam Heimlich (58:22)

Ad Tech Ad Talk 2

Rio (58:25)

Yeah. And every, every, actually Adam, tech, ad talks every Friday, right? It's live event, right?

Adam Heimlich (58:31)

Yep, Friday's at noon Eastern.

Rio (58:33)

And if people want to find it, how do they do that? Do they follow you on LinkedIn?

Adam Heimlich (58:36)

Read this search site at YouTube, or LinkedIn. My name's Adam Heimlich.

Brett House (58:41)

Yeah, and I think.

Rio (58:41)

Okay, cool. And if anyone wants to get a hold of Chalice, I might contact you about to learn more about this containerized approach or anything.

Adam Heimlich (58:47)

You could write me at adam at chaliceai or linkedin, but don't pitch me

Brett House (58:52)

Yeah, exactly. I connected with Adam, but I did not send him a pitch. knew his reputation preceded him. And I think my homework, Adam, is one of the big takeaways here, just generally your vision, your founder growth here, and your ability to run a business with your wife, I have huge respect for. That's terrific. Yeah, yeah. Is AI Will Not Make You Rich? I think that's an article we should all go out and read. It sounds like it's super interesting.

Rio (58:53)

You you I warned him.

Adam Heimlich (59:09)

She's not far on this show too. It's very different. We'll get a different take.

Brett House (59:21)

So I'm definitely gonna take a read of that over the weekend and we'll talk to you soon. Thank you, Adam.

Adam Heimlich (59:23)

Thank you. All right, now grab them.

Rio (59:28)

Thank you.