TIVP031: The Trade Desk (TTD): Heir to the Programmatic Advertising Throne? w/ Shawn O’Malley & Daniel Mahnke

29:07So it's a different play on the thesis we discussed with Roku where streaming platforms are all pushing pretty aggressively to reach scale with their user bases. And then at the same time, you just have structurally more advertising dollars moving from cable to the internet every year. And to illustrate that, connected TV advertising, so that is the ads that are shown when streaming on a smart TV, is about half of the Trade Desk sales currently. So they are very closely tied to the same trends that underlined that Roku pitch we went to recently, and actually are at least currently a much more profitable and healthy business than Roku, which is why it caught my attention.

29:49The numbers look fantastic. I mean, the results have been not only inspiring thus far, but it also seems that they can keep going like that. Revenue has compounded at 34 % a year on average since 2016 and with 27 % free cash flow margins, which is also, you know, fantastic result. And according to a Morningstar report, the total addressable market for digital advertising is about$700 billion right now, but it will likely grow to$1.5 trillion by 2034. Right now, I think the trade desk accounts for something like 2 % of the market share. And it seems like it has been capturing market share in the last years and may do so faster if the regulatory enforcements against Elphabit actually proceed as expected.

30:34And that's just a huge opportunity to grow even 10 years out. So you've captured a chunk of my thesis perfectly, Daniel. My idea is that if the trade desk can expand its market share to touch something like 5 % of digital advertising over the next decade, well, how would it imply that they could continue to compound revenues at least 18 % a year for the foreseeable future before running out of runway? And that is what gets me so excited about the company's potential. And the opportunity there is even bigger if OpenPass adoption continues to accelerate. That's totally understandable. But before we get too ahead of ourselves, one of the more important terms we probably need to understand here at the core of what Trade Desk does and represents is this concept known as the open internet.

31:26And we've already talked about it and perhaps it's clear to many, but just to make sure we don't lose anyone, can you just define for us what exactly that term describes? the open internet refers to the parts of the internet that are accessible to advertisers through interoperable programmatic ad platforms and which are not controlled end-to-end by a single company like meta or google and when you scroll through things to watch on your roku tv that is sort of another way to access again what you would call the open internet and meanwhile facebook and instagram on the other hand are not the open internet they are these walled gardens So they're self-contained ecosystems.

32:05And that is really not what we're talking about with the trade desk. The open internet is more about independent, decentralized publishers and content. News outlets, podcast companies like ourselves, streaming apps. Those can all make up the open internet in different ways, especially with respect to advertising. And in a walled garden like Facebook, meta completely controls the sale of advertising inventory to prospective advertisers, whereas the open internet is maybe a bit more democratic. And in the context of advertising, walled gardens refer to the platforms that are really in charge of what content users see, which companies can place ads, which user segments are served the ads, what users can do after seeing the ads in terms of clicking a link and being driven to a page from the advertiser and how this goldmine of data is collected, shared, and then re-monetized.

33:02And so alternatively, any place that can tie into cross-platform programmatic advertising really is what you would call part of the open internet. You might think Netflix, for example, is a walled garden, but actually it is more of the open internet since they enable programmatic ads. They used to be a walled garden that was closed off and didn't even have any advertising, But now they sell their advertising inventory through Microsoft. And then on the buy side, they work with the trade desk. That's very interesting because what's kind of floating around in my head right now is the question of how much of the overall digital ad market is open and how much of it is a walled garden.

33:41And beyond that, whether there's a trend to what one of them growing faster than the other. I'm not sure if you have any data on that. So I did do some digging myself and it turns out that my intuition is kind of right that walled gardens are actually the bigger part of the market. Depending on the reports you look at, they're supposed to control 60 to 80 % of the digital ad spend. And that's interesting, but there are also signs that this is kind of reversing, especially due to regulation. And perhaps that's also where Netflix comes in because, I mean, you would think that they have everything to do it internally, right?

34:14Most importantly, the data and also the technological know-how. So I guess my question is, why has Netflix sort of outsourced this and embraced a more open internet approach? Is it part of the trend that even for companies like Netflix, it makes more sense now to not go the walled garden approach? It's a really, really interesting question. And the answer from what I know is that companies like Alphabet, Amazon and Meta have just simply built up the infrastructure and scale to manage their walled gardens over many, many years. And these are immensely complex algorithms with very powerful servers behind them requiring tons of data to be competitive if you're truly taking a walled garden approach.

34:59And so Netflix, despite how big it is, isn't quite on that level. And they don't have the diversity of advertising inventory that I think would make for a good ad exchange. So they just opted to more quickly monetize their user base by taking the programmatic approach and relying on the trade desk. as opposed to trying to build all of that advertising infrastructure themselves. So there's not a huge idea behind it. It's more just a pragmatic move to save on costs. I would probably say so. And I think it's an implicit acknowledgement that they just don't have a scale, like I said, in data that these other players have.

35:36I mean, think about how much more information Google has on you than Netflix. Netflix doesn't need to fully open the floodgates. I think it can remain selectively open and offer some of its premium inventory to high-end advertisers through curated buy-side partners like TTD, which gives it the benefits of openness in terms of demand and the tools they can offer without totally giving up control of their platform and the data they have on customers. Yeah, I think that makes total sense. I wasn't aware of just how complex these algorithms need to be. And at least until I heard of the 20 % takeaway that TTD takes, I figured there have to be some serious value add for companies to pay that.

36:22So now we've covered what TradeDesk does, their unique positioning in the industry, what the open internet means, and we also talked about OpenPath. My understanding is that other big variable in this story is something called UID2. So how about we double click on that? Well, so it's going to get even more complicated and wonky here, but let's do it. UID2 is this new open source identity framework developed to operate on the open internet. And it bypasses the need for the legacy system called third party cookies in many cases, which at this point, I think we've all probably heard of. Most people probably don't know what cookies are, other than that they're kind of tracking you.

37:07But they have very much made up the backbone for a lot of how digital advertising has worked in the 21st century. And now we're at this point where cookies are increasingly being phased out. And UID2 is Trade Desk's idea for what can replace cookies. And now that they've made UID2 open source and handed it over to an independent body, they have separated themselves enough that it could actually be adopted widely without this obvious conflict of interest. And the main goals are to preserve the value of relevant advertising opportunities while providing consumers with more transparency and control over their personal data, if that makes sense.

37:51So you might kind of think of it as an upgraded version of third-party cookies that are more privacy conscious, but still balance the needs of advertisers. Sean, in case you want to dig deeper into the technology here, I will take the fall. And I will say that I'm one of those people who don't really know what cookies do beyond tracking the details of my behavior on a site and kind of using that to build a profile on me for ad purposes. So perhaps you can dig a bit deeper on that. Okay, let's get into the nitty gritty then for a moment. The way UID2 works is by transforming encrypted email addresses or phone numbers provided by users into a pseudonymous advertising identifier, aka a UID2, that is designed not to directly identify the individual personally.

38:44And according to the company, they believe that UID2 has reached a critical mass of adoption and continues to gain strong momentum among publishers, data partners, and advertisers. And some of those notable adopters are ones we've discussed with OpenPath, like Disney, Warner Brothers, Spotify, Roku, and Fox, as well as DirecTV, SiriusXM, and iHeartMedia. And basically, if it's widely adopted, UID2 could reduce the cost and complexity of aggregating consumer behavior data. It basically, long story short, more effective advertisements at a lower cost for ad buyers. But of course, that is a big if.

39:26And if agencies and brands can better target audiences across the open internet due to UID2, especially on connected TV and streaming platforms where authenticated logins are common, that is going to be good for the trade desk business long term, even if they don't directly oversee UID2 anymore. And more precise targeting drives higher ROIs for advertisers, lead to more spending on TTT's platform, where it takes that 20 % take rate we've discussed. And the way it kind of works in practice is that UID2 could link a streaming ad shown on a Roku TV to a website visit from someone who sees that ad and then pulls up the page on their phone.

40:11That is the kind of data and tracking we're talking about and why it matters to advertisers. Okay, so another way to think about this is that platforms like Meta and Google have access to vast amounts of proprietary user data, but they tightly control how advertisers can use it. And in contrast, UID2 helps kind of level the playing field by enabling user level targeting and measurement across the open web, making it a viable alternative to these world gardens, but at the same time reducing reliance on third party cookies, which world gardens don't use anyway. But that brings me to the next question for you, Sean.

40:54What are the main competitors to UIID2? And what could stand in the way of widespread adoption and indirectly strengthening the Trade Desk's position? I think the team at the Trade Desk would very much agree with your assessment there because it sort of paints them as being this Robinhood-like figure. It's sort of a noble hero protecting privacy and empowering advertisers across open web. And that is a simplification, but directionally, I think that is maybe the right way to think about it. But in terms of UID2, Alphabet continues to be a competitor even here. They obviously have their tentacles around everything, but Alphabet has something called the Google Privacy Sandbox, which on the one hand is even more privacy focused to satisfy regulators, which makes sense.

41:44Google's gotten all this negative attention on them, so they're kind of overcorrecting. But then on the other hand, that comes at the cost of cross-site tracking and attribution, which makes these advertising tools less useful to probably most advertisers. And so this is primarily useful for tracking identities when searching a computer browser. Whereas like we've talked about with UID2, you can actually track identity and engagement across a smart TV, email and phones. And so they almost serve totally different purposes to an extent. UID 2 is good for premium news or TV streaming where people tend to have to log into a service and unconnected at TV at least.

42:25Cookies were never really used here anyways. But actually, Google has, to make this all more complicated, kicked the can down the road a bit and recently did a U-turn on its decision to abandon cookies. So the whole future of their privacy sandbox is entirely being called into question now after Google has made some promises like cookies would no longer be needed, and yet they don't seem to be able to fully move away from them. To my knowledge, the entire sandbox project was actually stopped just about two months ago. I guess that kind of just shows how hard it is to balance privacy with effective targeting.

43:05And even the regulators didn't really like it. I mean, you would think they are a fan of more privacy, and I'm sure they are, but they didn't like the idea of Google once again being the dominant driver behind innovation in that space and basically consolidate its dominance in the online advertising market instead of weakening it. But is Google even the most serious competitor in that segment or other competitors as well? How does the landscape overall look like? Yeah, I should say there is another product called Ramp ID from a company called LiveRamp, But it is more of a closed ecosystem than UID2, which is very much an open source project at this point.

43:46And there are probably four or five others I could name around the world that are popular in different areas or that are relevant within different niches and may ultimately be used in combination to complement each other. But given Alphabet's control of search browsers and search itself globally, they still very much shape everything that happens in digital advertising. And we'll have a lot of say over privacy and tracking protocols going forward. But as far as the open Internet is concerned, UID2 is pretty well positioned, in my opinion, to redefine the status quo here. And the way to think of it is that UID2 is the trade desk offensive weapon to make advertising across the open internet more competitive and maybe even superior to advertising in the walled gardens of any of these big tech giants.

44:41And if it doesn't gain wide adoption, ad dollars will stay trapped or maybe even shift back to these walled gardens, which would, of course, hurt TTD's margins in competitive mode. So for as intangible and maybe complicated as these privacy protocols seem, I know it doesn't make for very good podcasting to linger on them for too long. But the reason we're doing so, I think, is just to emphasize how important they are to understand. UID2 is the epitome of the open internet revolution and potentially the future of advertising outside of platforms controlled by companies like Amazon and Meta. I think it's really not an easy topic to understand it to kind of wrap your head around.

45:27We even talked before the call and said that it's not even entirely clear how UID2 will be rolled out over the entire open internet. And who actually has the biggest leverage when it comes to implementing it all over the place. So is it consumers who, you know, have to either put down their email address or kind of accept their terms and conditions pretty much like it is now with cookies? or are we talking about the publishers, regulators or the advertising agencies since they ultimately decide where all the ad dollars go? I think all of this is still kind of in the open and it will be interesting to see if they all can make it happen together.

46:05But now just keep everyone on the same page so we don't get lost in this rabbit hole of privacy and advertising tracking. Can you help us take a step back and maybe get to the investing case? So perhaps the bull case versus the bear case at a very high level for the trade desk. I think it's a good call to take a moment here and take a step back, as you said. And for starters, as an unbiased advocate for advertiser budgets, TTD, again, has arguably a structural advantage over these less transparent ad giants like Google Meta and Amazon. And that is, in a nutshell, a big part of the bull thesis.

46:45On top of just maybe generally believing that digital advertising growth will continue to outpace GDP, it'll become more and more effective, and it'll suck up dollars that are currently being spent on things like cable TV ads, especially as live sports content increasingly moves to streaming since TTD has partnerships with both companies like Netflix and ESPN. ESPN. And then I think you could get really excited about the proprietary algorithms and things like OpenPath and UID2 as almost perfect solutions for the needs of digital advertisers coming from the trade desk, especially for those advertisers that are interested in running ads programmatically across the open internet.

47:28So they are uniquely focused on what is in advertisers' best interests, which I do think gives them some type of competitive advantage. For as profitable as it has been for Alphabet to run a two-sided monopoly over digital advertising exchanges, that just is not sustainable because there are these conflicts of interest that are just ripe for exploitation that regulators were never going to tolerate indefinitely. And just lastly, if the firm can get widespread adoption of its solutions that simplify the advertising supply chain, it could, you know, if you really buy into the bull thesis here, usher in this new golden age of programmatic advertising with the trade desk being at the center of all of it.

48:14I've long been asking myself who even is still out there watching, you know, legacy TV, as I sometimes call it. So I think it's more than realistic that we see way more ad dollars coming into this market and then a year ahead. And probably the trade desk is the one company that will benefit the most just by looking at their market share data of connected TVs right now, which as you said, is about half of the business. But now we've talked about the bull case. How about you tell us a bit about the bear case? Well, the bear case is pretty straightforward, right? And most advertising dollars are routed to the ad giants.

48:48Firms like Google and Meta have a ton of power and really no incentive to work with TTD. And convincing publishers to adopt TTD solutions like OpenPath will require considerable technological buy-in. That is just, it's far from guaranteed to happen. The lack of industry adoption of UID2 would keep TTD's data aggregation and mapping costs high relatively and just further disadvantaging them against these big tech giants. And on top of that, I should say that with connected TV advertising, about 75 % of ad spend goes toward directly negotiated transactions. And only 25 % of smart TV budgets go toward the type of programmatic advertising where TTD operates.

49:35Meaning these big events that get live streamed on streaming platforms could go for directly negotiated ad deals that bypass middlemen like TTD. Point being, NFL games on a streaming platform are not guaranteed to be available to programmatic advertisers, even if it's occurring in a domain that the trade desk normally works in, if that makes sense. And on top of that, there's the risk that streaming players try to cut out trade desk more structurally. Amazon, for example, is using its own DSP for its ad tier subscriptions to Prime Video. So there is a risk that other large players like Netflix, Disney, and Spotify move away from what is beneficial temporarily to partner with TTD to eventually follow the Amazon playbook where they try to take on the role of being an ad broker in their own advertising ecosystems.

50:33And basically, the Netflixes and Disneys and Spotifys of the world essentially becoming walled gardens, to put it simply. And yet, the counter argument to that is that Amazon has been developing its ad platform for quite some time, many years, and has many partnerships already because of its e-commerce advertising offering, while these other streaming players just might find it a lot harder to develop their own ad tech stack and then actually build relationships with agencies and brands at scale to attract enough advertising dollars to make those investments in ad infrastructure potentially make sense, which is why, again, I think they kind of cop out to saying, let's just work with the trade desk.

51:22That makes a lot of sense. But saying that 75 % of the ad spend of connected TVs goes to self-negotiated ads, of course, that's not a good sign. And it's kind of the counter argument to my bull argument I gave just two minutes ago, where I said there will be a lot of ad dollars coming into the market from all these, you know, legacy TVs. But maybe to just linger on the bear case a bit further. One other thing I would mention are the recent shareholder lawsuits. I think in 2021, the board approved this massive comp package for CEO of Jeff Green, which was actually one of the biggest pay packages in Silicon Valley history from what I've heard.

52:01And even though the lawsuits were most recently dismissed, this still feels undesirable for long-term shareholders. Stock-based comp more than doubled as a percentage of sales because of it. And on top of that, there was another lawsuit just this year over the company potentially misleading investors about the potential of its new AI-powered system called Cocai, while they were actively having execution challenges while rolling it out. To what extent do these lawsuits give you pause? I'm not going to defend it, but I'll maybe give a little more context on the pay package. So green received about 19.2 million shares worth of stock options with an estimated value of a lot,$828 million, something like that.

52:52But that money would only become exercisable in eight tranches over a 10-year period, assuming the company meets certain goals for the stock price. So if it hits$90 per share,$150 per share, all the way up to$340 per share. And so he's actually had some chances to exercise these options and he has held off on doing so. But yeah, if the stock ever reaches the high end of that range,$340 a share, green could receive more than a$5 billion payout if he exercised those options. And that would be very dilutive. But also from a shareholder perspective, if we go to$340 a share, we're talking about a multi-bagger from current prices.

53:36So some dilution for the CEO is maybe not the end of the world at that point. And accounting for this pay package though, and the value of the options has had a pretty outsized effect on the company's results. And almost alone explains much of the drop in operating margins of, you know, they had 17 % operating margins in 2020 to basically 10 % or less over the next three years before finally rebounding last year. And that's really not what I expected when I first looked at this company. I was thinking, okay, you know, this is a, as we talked about an asset light platform, as it scales, as revenues go up, operating margins should go up.

54:17And really what I realized was that this big pay package is what has, I don't want to say artificially brought operating margins down, but it has distorted what is otherwise going on with the business. So there was definitely an upfront effect to what was meant to be this 10-year mega grant. And by the latter half of this decade, the incremental expense from that award will either be fully recognized or be much smaller relative to TTD's larger revenue base. So a smaller percentage of sales. And that makes it maybe a little more digestible. And honestly, we're not talking about Elon Musk here who made headlines for what I think was one of the most egregious stock-based comp packages in corporate history when shareholders approved this$45 billion package for him, just to basically incentivize them to continue focusing on Tesla with everything else he has going on.

55:11And my maybe naive impression is that Jeff Green is a very, very different type of person. Unlike Musk and many other CEOs, Jeff Green has signed the giving pledge. So he's giving away 90 % of his wealth to charity. And actually, he's a pretty principled person. He exited the Mormon church because he believed that they were not doing enough globally to advance equality. so we're talking about someone who has been described by the wall street journal as an aggressive philanthropist and somehow knowing that green is this charitable person driven by more principled purposes in life makes me feel a little bit better about why he was awarded the pay package that he was especially when you contrast that with these other big comp packages that CEOs have gotten, like with Elon Musk, as I said, where Tesla shareholders are almost literally bribing him for his attention.

56:11He does all these other things. And so you're paying him to massively prioritize Tesla. And I don't think that is what has happened here with Jeff Green. And of course, it bears mentioning. And I wish that the package was not as big as it was. So it's something of a yellow flag. But after looking into it more, for as bad as it looks on paper, I don't think it's quite as bad as it seems. I think it's fair to say that the context here helps a bit. And I also rather see a founder CEO receiving such a package than, for example, an outsider CEO, just because they have the track record to back up what they did for the company.

56:54And still, as a shareholder, I would want what's best for the company today and not pay extra for past achievements. And a comp package that causes a 40 % decline in margins just has a bitter aftertaste for me personally. But anyway, what can you tell us about the second point? And that was the drama surrounding Kokai. Yeah, so let's start with what Kokai is. Kokai is a new platform from TTD to help advertisers maximize the effectiveness of their programmatic ad campaign. So it's an AI-driven optimization tool. And there's been a lot of excitement about it, as is true with anytime people mention AI today.

57:35But obviously, there's been some disappointment that triggered these lawsuits. People were hoping that it would dramatically improve the returns on ad spend that advertisers saw and help direct them on how to best use their ad dollars. This all came to light a few months ago when Trade Desk reported earnings for Q4 2024 and the stock pretty promptly fell 30 % after the company missed its own estimate for the first time in over eight years. So that is to say for 33 quarters in a row, they delivered on or beat expectations. And so a lot of that had to do with KOKI beyond just the fact that they missed earnings.

58:15So the rollout was slower than expected. And it kind of seems like advertisers haven't bought into it as quickly as anticipated. And I actually spent some time on a subreddit devoted to programmatic advertising. And while there wasn't a lot of love for Kokai, just to put it nicely, it sounds like they released the product way too early, rushed the timeline, and just shipped what was simply an unfinished product that they're now trying to fix. and that has damaged some relationships and trust in the industry. And I don't think the consequences are long-term, but they certainly could be. And still rolling out a major AI platform is inherently complex and a missed revenue target while disappointing doesn't automatically prove fraud.

58:59So we'll see how the case goes, but I'm not sure yet if this is more about TTT's management or just about AI more generally and how there's now some pushback from investors and legal groups to force management teams across the world to kind of rein in expectations about what AI can actually do, which I think is totally fair. But I also wouldn't say co-guys all hype either. According to the company, on average, clients have seen a 42 % reduction in cost per unique reach with 24 % lower cost per conversion and 20 % lower cost per acquisition. And so to any marketers out there, the numbers speak for themselves.

59:41They will know that that is really good stuff. I think you don't even need to be a marketer to understand that those numbers look really good. And I would also be surprised if this actually turns out to be fraudulent in any way. And as you said, I think the expectations for AI are just becoming more and more unrealistic at this point. I mentioned this in our portfolio review just a couple of weeks ago that investors as well as users expect AI to change everything and today. And I do think AI will change how we interact with many, many things, especially in the digital world. But turning to the old saying, people overestimate what they can do in a year and underestimate what they can do in 10, I guess the same is somewhat true for the expectations for AI, where people overestimate what AI can do in a year and perhaps underestimate what it can do in 10.

1:00:31So that might be what happened for Kokai as well. Maybe I'll just read some comments from TTD's CEO in response to the disappointing Q4 results. So just reading his quotes here, he said, For Q4, the reality is that we stumbled due to a series of small execution missteps while simultaneously preparing for the future. If this were a sporting event, we would still have a championship caliber team, But in this particular game, we turn the ball over too many times. That said, we see a larger and faster growing market than we originally expected, which is why we have been making changes and will continue to do so.

1:01:14And so Jeff Green continues here by saying, first, we did the largest reorganization in company history in December. And while we often make structural changes at the end of the year to improve our business, this one was bigger than usual. For most people in the company, we provided a much clearer view of their roles and responsibilities. And for most, it also meant a change in reporting structure. So he rounds all that out by saying, in Q4, there were a series of decisions that we could have made to enhance the short-term performance of the company and neglect the long-term. We consistently choose to focus on the long-term opportunity and maximize our market share over the long term, as I believe that is in the best interest of all of our stakeholders.

1:01:58We are keeping our focus on the massive TAM in long-term opportunity. You know what these sport metaphors remind me of? They kind of give me flashbacks to Nike's earnings report and what Elliot Hill said, explaining the quarter. I think sport seems to just be a good way to explain hiccups in the own plan. But jokes aside, Trade Desk is not looking. even remotely as bad as Nike's quarter. So I never cared much about a single quarter anyway. I think for them, it's more of a hiccup and it's definitely not thesis changing, at least from what I can see in the numbers. And in this case, I also buy into the story that they kind of rushed the process and that led to mistakes showing up in the product performance.

1:02:42And if this isn't turning into a trend, which I don't think it did following the quarters after this one, I don't think there's too much to worry about. the stock rebounded pretty dramatically after their subsequent q1 2025 earnings report and they did crush sales and earnings expectations so that does make the q4 stuff seem like kind of this one-off outlier and and some of the other highlights were that apparently two-thirds of their clients have now adopted coca and open path is is continuing to drive some great outcomes the new york post for example reportedly saw its programmatic ad revenues double due to using open path.

1:03:20So I see a lot to be excited about without dwelling too much on the initial COCAI rollout and the earnings miss and some of these negative headlines that I think could distract from the bigger picture here. Yeah, it definitely sounds like they're not going into the right direction again. So maybe to transition a bit, John, we can't have this conversation today without discussing the latest legal rulings against Alphabet more in depth, since this will probably very much ripple back to affect the trade desk as well, which is something that we've also been tracking because Alphabet is one, and I repeat myself, of our biggest portfolio companies.

1:04:01In fact, it is the biggest. So of course, our eyeballs are on that company specifically. And as just some context for the audience, On April 17, a US district judge found Google had engaged in anti-competitive acts to acquire and maintain monopoly power in the publisher ad server and ad exchange markets for display advertising. But stop short of concluding that Google's ad buying tools hold monopoly power. But still, the opinion found that Google's advertising exchange, which is called AdEx, was granted a first look at all requests for bids on inventory from Google Ad Manager services, which unfairly limited exchange competition.

1:04:44In other words, that basically just means in two of the three pillars of programmatic advertising as an ad exchange and in representing publishers, Google acted like a monopoly. But for supporting ad buyers, Google's actions are concerning, but don't meet the monopoly threshold. So we don't know what will come next yet. But the Department of Justice could force Google to divest from adx, which would allow for more competition in ad exchanges. And I would think more of a level playing field among ad exchanges should enhance the Trade Dex programmatic advertising business. But you would know that better than me, Sean.

1:05:21So do you think that's what actually happens or could it turn out another way? I think you're on to something. And that is exactly what the thesis is here, that the programmatic advertising market will become fair in a way that it hasn't been for years, while revenues move away from Alphabet's Google network segment toward companies like Trade Desk, Pubmatic, and Magnite. And it's pretty clear for publishers and advertisers to now see from the court rulings that Google has been something of a bad actor in the digital ad exchange space. And so that has fueled TTD's business to grow to 30 % clip per year, while Alphabet's Google ad network revenue has actually been declining for two years after many, many years of steady growth.

1:06:06And basically, the court ruled that Google illegally monopolized two of three critical layers of the open web advertising market, the publisher ad server market and the ad exchange market. that DOJ now is seeking what they call structural remedies. And so they want to rectify the effects of these monopolies, the ill effects of these monopolies, I should say. So they have asked the court to compel Google to divest major parts of its ad tech business, including the AdX Exchange, and then its double-click for Publishers Unit, which is now known as Google Ad Manager. and the idea would be to force Alphabet to sell off these units and dismantle Google's two-sided grip by separating the buy side and sell side operations into different companies and again this is where I should emphasize the beauty of the trade desk and being a neutral party devoted only to the buy side of programmatic advertising they just do not have the same entanglements that make everything so messy.

1:07:17And do you think that will actually happen? I mean, we've seen cases and hearings basically on all the big tech giants in the US again and again. Nothing ever really happened. But in this case, it just feels like they cannot go back. Like now they have said that Google needs to split up these parts. So I think there's a high likelihood of it actually happening. Is that what you see as well? Maybe it's a cop out, but we don't know yet. I'm not an antitrust lawyer, but the status quo is clearly being rewritten in real time. And what I know for sure is that we won't return to a world where Alphabet can just continue to run both sides of the programmatic advertising market to the detriment of really everybody else involved, including advertisers and publishers.

1:08:03And Google has, of course, vowed to appeal any adverse ruling. And they're trying to argue that they face competition more broadly in digital advertising from Meta and Amazon, which just feels like such a classic example from Peter Thiel's book Zero to One on what to look for when companies have true market power. and before regulators, they try to redefine the area they compete in as being much bigger. So rather than specifically competing in programmatic advertising across websites where they clearly dominate, they try to zoom out and say they really compete in the world of all digital advertising, which is a much bigger market.

1:08:42And so they want to downplay their dominance. And that's actually how you kind of know it's a really dominant business. And unsurprisingly, the trade desk CEO, Jeff Green, has been vocally supportive of antitrust action against Alphabet. And he has even called for Google to exit the open web ad business entirely or be broken up by regulators. And his argument is that if Google's dominance has been due to vertical integration and that forcing Google to quit at least one of those jobs, either on the buy side, sell side, or as an exchange would remove these unfair advantages. And in his view, Google might choose to spin off the exchange part of its business, aka AdEx, since it's less lucrative than its own ad sales on search and YouTube.

1:09:29And if that happens, advertisers would no longer be restricted to Google's exchange. And the trade desk could capture spending that might otherwise have been locked into Google's ecosystem. I think it's not hard to see why Jeff Green is an advocate of that. But we've covered a lot now from how programmatic advertising works to UID2, OpenPath, the shareholder lawsuits against the TradeDesk, and also the regulatory reinforcements against Google. So now comes the part where I ask you how it all comes together to shape your view of TTD's valuation. There is no way around it. TradeDesk is richly valued.

1:10:11The market knows they are a great business that is positioned to become an even better business and that they have historically been operationally excellent. They have executed very, very well. And the founder, Jeff Green, quite literally was part of this small cohort of people who flipped the advertising world upside down and helped move ad deals away from being negotiated in person at fancy restaurants in New York to being a digitally native process where ad traders can bid on impressions across the open internet. He really made advertising digital. Back in the day, digital ad campaigns were exclusively negotiated directly by people.

1:10:52So there has been a real paradigm shift in that automation and Trade Desk has been at the center of it. But yeah, point being, the market knows how great of a business this is, which is sort of unfortunate for us who would like to own it, I think, at 88 times last year's earnings and 40 times earnings estimates over the next 12 months. It's hard to look at this and say, oh, this is screamingly cheap without being at least very, very optimistic about their growth prospects. So then I guess the question is, at what level do you start find the stock interesting or attractively priced? For me, that number is probably at around$50 per share or lower, which would require something like a 30 % correction from current prices at the time of recording.

1:11:43and as part of that in modeling the company i did a lot of work to understand their margin potential how profitable the business can be at scale because i already know this business has a ton of room to keep growing sales as digital advertising increases and as new things like uid2 help make ad spend on the open internet competitive with the walled gardens with the point being i think the top line can easily compound at 15 to 20 percent a year for a while into the future. I don't have any hesitancy kind of estimating that, underwriting that. But given how much their operating margins have fluctuated, the key variable in my mind for understanding this company is how much operating leverage there actually is and where margins can plausibly be in five years.

1:12:30And just in the last five years, like I said, operating margins have fallen from 17 % to 10 % into 7%, back to 10%, and now back to 17%. So they've been all over the place. And Before that, operating margins were actually more consistently in the 20 % range. And so really, much of that has been distorted by Jeff Green's pay package that we talked about earlier. So when you normalize that and look at the margins the company has been able to achieve previously, while also accounting for the benefits of further scale for a software-driven business like this, I do not think it's totally crazy to think the business could hit a 25 % operating margin a few years down the line.

1:13:10And so that expectation is sort of the basis for my valuation. If margins are half that, like the past few years, well, this company would just have dramatically less earnings power and value than I currently think is possible. And some analysts actually think they can hit 30 % operating margins, but I find that to be a little bit too optimistic, given that the Google network segment at Alphabet doesn't even have margins that high, and they're running this two-sided monopoly. So it's hard to make an apples to apples comparison for TTD, though, since many of its peers are much smaller or much bigger, like with Google.

1:13:47So it's hard to find good peer comps. Pubmatic, for example, on the publisher side has some pretty uninspiring operating margins, but they're also one 70th of TTD's market cap. So they are not at the same scale either. And as we talked about, there's different economics between the buy side and the sell side. And the same with Magnite, which is also on the sell side working with publishers. And they're a company that only has 10 % operating margins, but their market cap is also only $2.5 billion. So this is also not a huge business at its current scale. And on the other hand, Applovin is a much bigger company than the Trade Desk operating in kind of a somewhat similar area, but instead focused on advertising in mobile apps specifically.

1:14:30And their operating profit margins are actually well north of 40%. So that might make you think that 25 % estimate for Trade Desk is too conservative at scale. So after thinking about the operating margins, looking at the range of comps for this company, thinking about what the company has done historically, estimating their margins by 2029, and then trying to assume what I think would be a reasonable exit multiple for a company of this quality with these kind of growth prospects of maybe about 30 or 31 times operating profits, I think a reasonable fair price to target buying the stock at is between$45 and$50 per share.

1:15:08And given how far above that we are at current prices, I'm not recommending we add it to the portfolio. I mean, a 25 % margin assumption seems realistic without being overly conservative, I think. And AdBloven's margins are significantly higher, but that's also because the mobile ad market operates on a bit of a different model with much tighter vertical integration. And then you can add the scale to that. So I think that explains a lot of it. Now, could the trade desk eventually exceed 25 % margins? Sure. But if 20 % top line growth and a 25 % margin only, quote unquote, imply a fair value of just below$50, then I think it makes sense to wait for the stock to come closer to that range.

1:15:56Especially considering that I think for both of us, at least I assume that the underlying tech behind the trade desk remains at least somewhat of a black box. that doesn't disqualify it as part of a diversified portfolio, which we are building, but it does raise the bar in terms of valuation. And when you look at the stock chart, we've had several opportunities over the last few years to pick this stock up at$45 or less. That doesn't guarantee we'll see that price again. But just like with Amazon, volatility makes it much more likely that we eventually get a shot at picking it up. So what do you think, Sean, to kind of put a close at it?

1:16:37tdd is just a bit difficult for me to get comfortable with i'll be fully honest i was expecting for it to be easier for me to wrap my head around the trade desk after already having gone deep on alphabet and roku but ad tech is just so very messy and complicated i just feel like the ad tech industry probably goes in the too hard pile and the trade desk as a pure play on demand side programmatic advertising, you really need to understand the digital advertising ecosystem fully and technically kind of know what's going on behind the scenes to appreciate what makes the trade does valuable and what can make it less competitive in the future.

1:17:19So this is one of those where having some industry experience would be very helpful. And you might be like, OK, well, how can you own Google or consider Roku since they're both also closely tied to digital advertising? And the difference is a few things. Firstly, Alphabet is a ridiculously diversified business that's dominating across a number of areas. So I don't feel like I have to be an expert on a specific area like cloud computing to appreciate the quality of what's going on there. And even with Roku, which I'm not as bullish on, but still they're installed in tens of millions of homes and control how people access streaming content.

1:17:58And I've used Roku devices, so it's more tangible. I can understand the value of their positioning better without necessarily needing to deeply understand the relative technical advantages of their products and services. I know I've used a Roku, and I know I've used an Amazon Fire TV, and I think the Roku is probably better, right? There's these kind of intuitive things. But with a trade desk, they operate as a pure play in a specific niche in B2B ad tech. And I don't work in B2B ad tech, obviously. So there's just no way, There's no B2C component to what they do, nor is there a lot of diversification to their business for better or worse.

1:18:34So there's just no way for me to really know what's going on without having directly interacted with the services, without having worked in the industry, without having spent 100 hours talking to people who work in the industry. It's just one of those places where a lot, a lot of work is needed to be done to appreciate how deep or shallow their moats are. when you and me were at the berkshire's annual meeting this year buffett once said that charlie always wanted to know and understand how and why things work and buffett was more or less okay with knowing okay they do work and i found that very interesting because we always talk about you know circles of competencies and all of that truly understanding a business but as investors we don't need to understand everything if we can reasonably assess that something works and we don't always need to understand the nitty-gritty details.

1:19:26However, that's where the margin of safety, at least for me, comes into play so that when I'm aware that I don't fully understand everything, then I have to apply a higher margin of safety to the investment. And we did that for Roku and we're doing it now for TTD. And that's why we decided to not add it at these prices. And with Google, as you said, the difference is that one, we got a price that we find very attractive, including our margin of safety assessment and two we both feel like we understand most of the company and its dynamics because we used it for decades and I think that just helps a lot and we cannot say that about TTD and that just makes it so much more complex I feel like too much of an outside observer to add to say what is going to happen with OpenPath and UID2 and how that will be adopted or to try and predict you know What Alphabet can do or won't do to protect their Google network sales from being lost to competitors.

1:20:26So the setup has been really interesting. Interesting enough to catch my attention. But again, there are too many technical points, I think, that we need to understand. And maybe this is just me trying to be a bit pragmatic and honest about the difficulty of research in this one. But simultaneously, there's this weird thing where I feel pretty confident that if you said, hey, your life depends on guessing this right. Do you think Trade Desk will be a dramatically bigger or smaller business in five or 10 years from now? I would say I'm pretty sure it's going to be dramatically bigger. But that's just not how I want to make investment decisions.

1:21:02I think we want to be a little more thoughtful and more thorough about feeling like we know what really drives the business. And if not, at least having a very, very attractive price to entry to kind of hedge some of that uncertainty. So long story short, I am happy to pass on TTD and just watch it and learn more about it. But man, how funny would it be if the trade desk were to acquire Roku at a premium and then directly leverage Roku's wide reach to further submit their positioning in programmatic advertising? And so, I mean, a lot of the a number of analysts have pitched the idea that the trade desk should acquire Roku, which is what drove me to want to learn more about the trade desk after having dug into Roku for a pitch two weeks ago.

1:21:52and we've run out of time to talk about it today, but TradeDesk has recently ventured outside of its core competency to launch a smart TV operating system, which would actually position them to more directly compete with Roku, but also it would make sense for them to want to buy Roku and integrate them into it. So I don't know. It would be ironic for me to pitch these two companies, pass on both of them, and then to have some acquisition goes through that ends up at least being a big payday for one of them. so maybe we shouldn't get caught up in what ifs Daniel why don't you give us your hints for your pitch next week you know last time I pitched PayPal and I kind of want to stay in the same space once more just because I like the idea of researching more than just one business in an industry to get a better feeling for the competitive landscape so this time the company is significantly smaller and also a bit more niche, focusing more on peer-to-peer than PayPal probably does right now.

1:22:51And it did take a hit due to the newly proposed big, beautiful bill. And perhaps that's even too much already. So I think I'll keep it at that and let our community guess it. All right, folks. Well, as always, we will leave you with a timely quote. This week's is from Henry Ford, who says, stopping advertising to save money is like stopping your watch to save time. The trade desk is very much tied to the reality that so long as the global economy is growing, so too will advertising, which is an essential part of business generally. And the trade desk is an essential partner for many digital advertisers.

1:23:30That is how I think about them at a high level. So that's it for now. Thank you all for tuning in. We will see you again next week.

1:23:43Thank you.