TIP841: Palantir – Palantir is Cheaper than I Thought! w/ Daniel Mahncke & Shawn O’Malley

Daniel Mahncke and Shawn O’Malley take a deep dive into Palantir (NASDAQ: PLTR), the data and AI platform behind everything from battlefield targeting systems to Airbus’s A350 production line, with customers including the U. S. Army, the NHS, Ferrari, and Airbus. Palantir is growing revenue at over 90%, with net dollar retention at 157% and adjusted operating margins above 60% — a Rule of 40 score of 155%, roughly double what the best software companies in the world achieve. It also trades at around 60 times sales, which is why The Economist called it possibly the most overvalued firm of all time. Daniel and Shawn discuss whether the ontology layer is a sustainable moat now that Microsoft and Google have launched competing products, what Alex Karp and Peter Thiel’s politics mean for a company selling to Western governments, and why twenty years of losses turned into the fastest-accelerating software business on the market. In the end, Daniel values the business and decides whether PLTR deserves a spot in The Intrinsic Value Portfolio. IN THIS EPISODE YOU’LL LEARN: (00:00:00) Intro (00:02:29) How Palantir’s Ontology actually works (00:16:26) How AI turned around Palantir’s business (00:20:59) Why Palantir has the best margins in the industry (00:34:22) Why no one seems to be able to copy Palantir’s Ontology (00:38:01) How Palantir grows its customer base (01:07:36) Palantir's valuation discussion (01:11:37) How much Palantir stock is actually worth (01:12:52) Whether PLTR will be added to The Intrinsic Value Portfolio Disclaimer: Slight discrepancies in the timestamps may occur due to podcast platform differences. BOOKS AND RESOURCES Join the exclusive ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠The Intrinsic Value Mastermind Community⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Track ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠The Intrinsic Value Portfolio⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Learn more about how to join us in NYC for our ⁠⁠⁠⁠⁠⁠Intrinsic Value Conference⁠⁠⁠⁠⁠⁠. Portfolio Review ⁠⁠⁠⁠⁠Submit Tool⁠⁠⁠⁠⁠. Satish Terala’s AI Masterclass. Future Investing’s Interview with a Palantir FDE. Check out our previous Intrinsic Value breakdowns: Constellation Software, Dell, Alphabet, Constellation Software Spinoffs. Related ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠books⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ mentioned in the podcast. Ad-free episodes on our ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Premium Feed⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. NEW TO THE SHOW? Get smarter about valuing businesses through ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠The Intrinsic Value Newsletter⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Check out ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠The Investor’s Podcast Starter Packs⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Follow our official social media accounts: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠X⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ | ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠LinkedIn⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ | ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Facebook⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Try our tool for picking stock winners and managing our portfolios: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠TIP Finance⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Enjoy exclusive perks from our ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠favorite Apps and Services⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Learn how to better start, manage, and grow your business with the ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠best business podcasts⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. SPONSORS Support our free podcast by supporting our ⁠⁠⁠⁠⁠⁠sponsors⁠⁠⁠⁠⁠⁠: Scribe Plus500 Netsuite Plaud References to any third-party products, services, or advertisers do not constitute endorsements, and The Investor’s Podcast Network is not responsible for any claims made by them. Support our show by becoming a premium member! https://theinvestorspodcastnetwork.supportingcast.fm

• 0:00Speaker 00

  You're listening to TIP.

• 0:03Speaker 01

  Welcome back to the Investors Podcast, episode 841. And the last time the two of us talked, we discussed our biggest winners and our biggest losers out of all the companies that we pitched in the last one and a half years. And I thought that was a lot of fun.

• 0:19Speaker 02

  It was a nice change of pace, but today we're getting back to what we do best, and that's flipping over more rocks, looking for opportunities. And you've got a stock pick today that listeners will probably have very mixed opinions about. And there's others who have heard a lot about it, but I'm guessing really are going to be excited to better understand what the business is all about.

• 0:39Speaker 01

  It is a divisive company, I would say, but also a very popular company, especially in the investment circles. And I know you and Kyle looked at SpaceX not too long ago, and surprise, surprise, that company is far away from being a value play and therefore not the type of company that we want to invest in. And I thought it might be the same for today's pitch, which maybe we should say it is Palantir. And please don't get me wrong, Palantir is not a value play. But when I started my research, I did it thinking in the back of my mind, at least, I might need to abort and look for another company because there's just no sense in pitching it. But that wasn't the case. I actually found myself pretty excited about the opportunity. Just a year ago, Palantir traded at twice the multiple with half the growth rate, and that has completely shifted now. And again, that doesn't make

• 1:23Speaker 01

  it a value play. But for everyone who loves high quality companies, as most of us do, and is a generally curious person, I think you should listen to this. I think you will be positively surprised by today's pitch.

• 1:34Speaker 02

  Yeah, maybe I'll be one of them. Just like you, I never looked at Palantir thinking it was way too expensive anyway. But if you're intrigued by it, then I'm ready to get into it.

• 1:45Speaker 00

  Since 2014, with more than 200 million downloads, we have interviewed the world's best investors, studied deeply the principles of value investing, and uncovered many compelling investment opportunities. We focus on understanding businesses and intrinsic value, investing accordingly, and sharing everything we learn with you. This show is not investment advice. It's intended for informational and entertainment purposes only. All opinions expressed by hosts and guests are solely their own, and they may have investments in the securities discussed. Now for your hosts, Sean O'Malley and Daniel Manca.

• 2:31Speaker 02

  All right. So where do you want to start? Personally, I think I could use a brief explanation of what exactly Palantir does. I can't really say I have a good idea of what they do. It's a very enigmatic company.

• 2:41Speaker 01

  Yeah, before looking at this for the episode, I honestly didn't really know either, but it's quite fascinating. So to give a first quick example, let's imagine any large organization, right? It could be an airline, it could be a hospital, it could be a car manufacturer, or the army, which is actually a big client, but it could be anything else too. So there's always a lot of information in all of these organizations, right? But it's scattered across dozens of separate systems and departments that barely ever talk to each other, at least not efficiently. And I certainly can say that's the case in Germany, but I think it's also the case in the US. So let's just stick with the example of an airline for now. Assume you have an airplane that runs into technical problems in, let's say, you know, Frankfurt Airport, which likely happens again in two weeks' time when I'm

• 3:23Speaker 01

  on my way to our intrinsic value conference in New York City. That's traditionally how it works with my travel. But then the question for the airport is, what happens now? Well, the maintenance team would know which part has failed and how long the repair would take. A different system would then know, you know, what's about the crew and how many duty hours have they left. A third one, which would know what passengers are on board. Then a fourth would know which of those passengers have connecting flights that they need to get. And a fifth one would know what those replacement seats are worth. So you basically have five different systems and teams that need to communicate and coordinate. And the way this is actually still getting resolved today in 2026 is by just phone calling each other.

• 4:03Speaker 02

  And all the while, Daniel, you're just sitting there inside of that plane stressing about whether you're going to make it to the conference in time and hopefully not finding yourself waiting on four canceled flights, like what happened when we went to Omaha last time for the Berkshire Hathaway shareholder meeting. But it's a good thing you'll be arriving in New York a couple of days early this time.

• 4:23Speaker 01

  Yeah, I'm not taking any chances. By the way, if you're asking yourself, what is this conference in New York City? Well, as Sean, you put it rightfully before, you labeled it as the value investing conference or event of the year. And that means on Saturday, the 19th, we will host an intrinsic value conference in Midtown Manhattan in September, and we will have speakers from our mastermind community. And of course, you, Kyle, and I will also be on stage. And if you think that sounds great, and you want tickets, I will link to those in the show notes. Otherwise, the easiest way to join for free is to be part of our mastermind community. And I will also have a link to the application process for that in the show notes as well.

• 5:05Speaker 02

  Yeah, if you want a ticket, you can just go to intrinsicvalueconference.com. That's another way to do it. But all right, let's get back to Palantir. So what do they actually do? And what do they sell?

• 5:18Speaker 01

  Yeah, it's kind of complicated. So what Palantir sells is a software layer that sits on top of all the five systems that we talked before, right? And then, you know, to just say with the airline, they would tell you where's the airplane, where's the part that you need, where's the crew, where the passengers, and the connections between all of those. And they call it things. So all of these things are now connected on what you sort of can imagine as a map. And crucially, the software also knows who is allowed to do certain things. So rebook a passenger, assign a different crew, rather than aircraft somewhere else. So you can imagine it as looking at this map that knows everything that's happening in real time. That's what, you know, Palantir calls the ontology. And you might think, well, isn't this just a database? Because that's what came to my

• 6:01Speaker 01

  mind first. But the difference is that what Palantir builds is not only facts. So you know, the plane's tail number, for example, or passenger seat, but also the relationship and the actions between all of these things that I mentioned. So Palantir's system knows that passenger X is on this particular airplane, which needs this particular part, which is fitted by this particular crew. That's how you can imagine it. And it also knows that rebooking is an action that exists. So who is permitted to approve it? They know that. And what else has to happen when it does? So the combination of all of these things is what Palantir calls an ontology. And the ontology is what we'll talk about today quite often. So again, if you just have to have something in mind, think of a map that

• 6:44Speaker 01

  has everything that you could imagine on it and basically knows who is who and what is somebody allowed to do. So it's easier to communicate.

• 6:52Speaker 02

  Yeah. So basically, you have this all knowing map or ontology layer, as they refer to it. And that always basically knows what's going on and what has to happen next and who can make that happen. And so, gosh, it kind of reminds me of something that would be part of a Hollywood movie, like some action movie where the lead roles are out there in the field fighting the bad guys. And then there's like this control center that always knows exactly what's going on and is like feeding them advice. That's what I think of when I think of Palantir.

• 7:20Speaker 01

  That's probably not that far off, especially when you think about the surveillance and defense work. And I'll get back to that and explain it in more detail and how it all works. But I think we should also take a quick detour and talk about the history of the company, because that's quite interesting as well. So Palantir was founded in 2003 by Peter Thiel, Alex Karp, who is also the current CEO, Joey Loonsdale, Stephen Cohen, and Nathan Gaddings. And the idea actually goes back to one of our favorite companies of all time, PayPal. And I say that somewhat ironically, because we lost some money on that bet. Obviously, Peter Thiel also founded that company, PayPal, and he repeatedly had problems with fraud. As you can imagine, it's a payment provider. There are a lot of people who want to take advantage of that. And the fix he eventually came up with was

• 8:04Speaker 01

  that, you know, you could have a mix of software, which could sort of surface suspicious patterns by going through a ton of data. And then you have a judgment call of a human that is still at the end. And at least then, the most accurate thing that you could do. And then something happened in the US. And there was 9/11. And it became clear that US agencies generally had all the information they needed to stop the attack from happening. But the information was scattered across different agencies that couldn't bring it all together in time. Again, going back to this ontology idea, they had all the information, but it was not on one map. They didn't know who can communicate to who. And that was the idea. And that's why and also how Peter Thiel got the idea for Palantir, was he thought, what

• 8:46Speaker 01

  if you could build a system that connects all of these fragmented, seemingly unrelated data points to map out hidden criminal behavior, similar to what he did with PayPal, just on a whole nother scale?

• 8:58Speaker 02

  Do you know why Peter Thiel isn't the CEO of Palantir? Or why he didn't want to be CEO? Because as far as I know, he chose Alex Karp early on to run the business. And he's taken on more of a fundraiser and advisory role. But he also owns the largest individual stake in Palantir at about 4%. And Alex Karp holds 2.5%. So he has the most skin in the game of anyone.

• 9:22Speaker 01

  I think we all don't fully know what goes on in Peter Thiel's mind. But I would say there's something to say about that being a strategic choice as well. That Peter Thiel might not be the best person for the CEO job. He had the vision, certainly, but he might not been the guy who wants to be the CEO of the company. Actually, if you think about PayPal, he wasn't the CEO there either for a very long time. And he just generally tends to work more on building new stuff and maybe gaining influence in both the tech founder world, but then also in politics, which seems to be more interesting to him nowadays. And I would say that's at least how it seems to me. And I don't know as well as you what Peter Thiel's reputation is in the US. But I can tell you that it's not that great here in Germany. And I'm sure there's more of a spotlight on him since he was born in Germany. That could be one of the reasons.

• 10:06Speaker 01

  But I think getting international deals with governments might just be easier, especially considering the sensitive information needed if Peter Thiel is not the CEO. Although I gotta say, Alex Karp is probably on his best way of getting an equally bad reputation. And again, he might be viewed more critically here in Germany too, because he actually lived here and he studied here for over a decade. Actually, fun fact, he got his PhD at the Goethe University, which is also where I studied. And his doctoral advisor was the famous German philosopher Jürgen Habermas.

• 10:36Speaker 02

  Wow, so Stanford first and then go to university. I didn't realize how strong that German connection is that runs through Palantir. And for Peter Thiel, gosh, I mean, he's controversial, of course. Rightly or wrongly, he gets caught up in a lot of conspiracy theories. And I have to admit, he does give off like supervillain vibes, which doesn't pair very well at all for a company that is really generally working to increase government and corporate surveillance in some sense. But whenever I think about Palantir, I don't know, again, like the empire from Star Wars comes to mind. You know, like the bad guys, Palantir and Peter Thiel. They probably both need a lot of PR help. Some sort of rebranding would go a long way.

• 11:19Speaker 02

  But again, it also goes without saying that Peter is hugely respected in the investment community and in Silicon Valley. And his book Zero to One is one of my favorite investing books ever.

• 11:30Speaker 01

  And if you read that book and then you look at Palantir, I think it gives you an idea of what the philosophy behind founding that company was. And besides Peter Thiel, who bankrolled an initial cost of $30 million for this company, in QTEL, the CIA's venture arm, also put in $2 million, which doesn't sound like a lot when you compare to Palantir's size today, which is depending on the day, between $350 and $400 billion. But it was less about the $2 million and more about the access that you get to the government sector if you're sort of sponsored by the venture arm of the CIA. They actually introduced Palantir to working for most of these intelligence agencies, but also analysts. And for a good two years, Palantir basically built the product, their only product, just for this one client.

• 12:14Speaker 02

  You said it casually there, but there's a venture arm to the CIA. That's something I'm going to be researching more after this episode. That sounds really interesting to learn more about what kind of venture bets the CIA makes. But you said two years, is that how long it takes to onboard a customer for Palantir?

• 12:31Speaker 01

  Well, it used to, and that's actually quite interesting because the long onboarding time is why Palantir actually looked like a failed company just a couple of years ago. So first of all, why does it take so long to onboard a customer to Palantir? As we said earlier, the ontology layer is basically a digital twin of a customer's entire organization. So you need to understand every single little detail of a corporation. If that's a huge one, that takes a lot of time. And the setup was that Palantir had two kinds of engineers. So there were product engineers who built the actual software. And then there were so-called forward deployed engineers or FDEs, who flew out to the customer and worked from the customer's offices for at least three to four days a week. And that could easily take a long time.

• 13:14Speaker 01

  I mean, not necessarily two years anymore, but it can still take months until they've actually finished their job. And on the ground work is what costs so much time. I've gone through quite a lot of reports from ex-Palantir employees, where they walk through their experiences, basically. And many of them mention what we said above, which is that before AI, Palantir was even seen as a fake SaaS company. Because basically, many people thought of them just like enterprise consultants and not much more. Because they travel to the client, they work with them, and they deploy some software. How much or how difficult can that actually be? And that's because a lot of what they do is on the ground work, just figuring out how things work, how things work together and what works and what doesn't inside the company and all that sort of stuff.

• 13:55Speaker 02

  The boots on the ground approach reminds me of CoStar, which is a SaaS company that we have in our intrinsic value portfolio. And for decades, though, they sent people to commercial properties to take photos and map it all. And they still do that. And they have armies of people that are creating this basically unique data that nobody else has because they're going out and physically documenting all the details of these properties. And that might not be the most efficient way to do it. It's not the fastest, but it definitely builds a moat because most of the competitors are not going to go that far into the weeds. And it would be hard, it would just take them many years to recreate something like the database that CoStar has. And so it seems like Palantir has maybe done something similar.

• 14:40Speaker 01

  Actually also thought of CoStar when I heard about this. But obviously, what Palantir does is 100 times more difficult. You know, no disrespect to CoStar, it's a holding on our portfolio, but it's way more complicated. I mean, the biggest problem is that the data, it needs to create these digital twins. And highly capable and actionable databases is scattered across the entire company. And often the data is only halfway digitized too. So I don't know, think about handwritten documents that were scanned to a PDF. But as you would know, that still makes it unsearchable, right? It's digitized, but you cannot just search it up, which makes things just way more complicated. That's just one out of 100 possible problems. Again, when I went through these ex-employee articles of what they encountered there, it was just, sometimes it was almost unimaginable.

• 15:22Speaker 01

  And another very interesting part is that many ex-employees described internal politics as one of the toughest hurdles. So before Palantir comes in, everybody is doing their own thing. And because of inertia and perhaps also the fear that you just won't be needed any longer when Palantir's software actually exists, many times the middle management more or less refuses to hand out the actually important data. So again, the stories you can read about that are actually quite funny. And I might link to some of them in the show notes. But of course, they can blatantly refuse to give out the data when the company has actually hired Palantir to work with them. But they will just come up with these little excuses, something like, this specific database can't be accessed because, you know, security protocols. Or maybe they just literally keep delaying the meetings and cancel on calls.

• 16:07Speaker 01

  That's the stuff that people did in the past. So in short, I think it just turned out that a lot of the work has gone into figuring out internal politics. And then especially in high-tech industries or national security, there are actual concerns about how to best give Palantir the data they need without risking any espionage or just data leaks or whatever.

• 16:28Speaker 02

  It's interesting because it sounds like it's just more difficult than one would assume to actually get the data to even start the process of this ontology layer and mapping the entire business. And what's really interesting to me when I look at the numbers for Palantir is that there was actually a period of time here where growth slowed down a lot. And that was pre-AI, I guess, pre-ChatGPT at least. And we just talked off the record about why that happened. And you said that the entire process of onboarding and using Palantir was so complex that it really hindered their growth. And so if you look at the growth chart now of Palantir's revenue, you basically saw growth decelerate from 2019 through about halfway through 2023,

• 17:10Speaker 02

  where year-over-year revenue growth was as low as 12%. But then from there on, it's accelerated dramatically like a rocket ship. And we're talking about 100% year-over-year growth, doubling revenues every year.

• 17:23Speaker 01

  That's pretty insane. And I think to explain what happened, we should start talking about the four platforms that Palantir has. So they're calling them Gotham, Foundry, Apollo, and AIP. And Gotham was the first platform that Palantir built. And it took about five years of consistent updating and improving. It was part of the first project that they did for the government agencies. And it was built for, as I said, intelligence generally. So that work started in 2003 and was based on solving the problems primarily leading to 9/11. And Gotham is basically the part that creates the digital twins for defense agencies. You know, while Foundry is primarily a platform that was built in way later, 2016, so 13 years later, actually. And they build digital twins for corporate operations.

• 18:07Speaker 01

  And both of them sit on, and now it gets a bit complicated, both of them sit on Apollo. And Apollo is just sort of a deployment engine. So whenever a new update drops, Apollo is shipping it to Gotham, Foundry, and also AIP. And you might think, okay, well, why do you need a deployment engine? Is it so difficult to just get an update? But if you think of all of the silos that the data is sitting in, you know, on the one hand, you have intelligence agencies, on the other hand, you have huge corporations. It's not just one click of, hey, we have an update, you can click here and you get it. It's way more complicated than that. You know, it's getting an update into what is siloed databases. You know, there are a lot of security behind that. So that's why you need an entire deployment engine to some extent.

• 18:47Speaker 01

  The real game changer was AIP, which is a very smart name for the AI platform, AIP. That was an abbreviation that I think most of us could also come up with. But it actually changed the game for how Palantir approached customers and also how efficiently the software can be run on each of those platforms.

• 19:05Speaker 02

  Yeah, one of the things they did is they started to run bootcamps, right? They basically invited a bunch of CEOs and CIOs from potential clients to showcase what Palantir can do. It's obviously the potential clients had to bring maybe some of their data, probably not sensitive data. And then in a workshop over three to five days, Palantir showed them some ways and exactly how they could improve their business and solve some problems for them. And so I think the powerful thing about this is that every company thinks it's different from all the others, right? And to some extent, that is true. But because of that, even if they generally know how strong Palantir is, they still think it might not work for them. They might be hesitant. And these bootcamps are really set up to prove that Palantir can, in fact, really help almost any type of company.

• 19:50Speaker 01

  It's important to say that Palantir paid for those workshops out of their own pockets. So companies that attended took on the risk, except for, as you said, handing over data. But, you know, again, that was most likely not the most sensitive data in the company. I mean, perhaps we're talking about some inventory data or something similar with just the goal of showcasing how Palantir can make the business more efficient. And you could probably say that this was the turning point that you alluded to when you talked about the numbers and growth going down to as much as 12% in 2023, because AIP made using and also onboarding so much easier that Palantir suddenly exploded. And what onboarding is an important part was not so much the customer count that grew. In fact, customer account growth has actually come down from about 50% in 2024 to about half of that lately.

• 20:35Speaker 01

  But the money existing customers spend with Palantir increased significantly. So net dollar retention, which is the metric that we're talking about here, is basically measuring how much existing customers spend with you. And that metric increased from a low of about 100% in 2023, which basically means they spent as much this year as they have last year to almost 160% in the last quarter. So for every dollar spent last year, a customer now spends a dollar and 60 cents. Let's take a quick break and hear from today's sponsors.

• 21:05Speaker 02

  Early in my career, I was shipping away at student loans and it felt like it would never end. I'd make a payment, check the balance, and it barely moved. What finally changed my mindset was seeing a real finish line, a date I could actually circle on the calendar. That's what Monarch's debt pay down simulator gives you. It uses your real balances and payments to show your exact debt free date. Then you can run different scenarios side by side, like adding an extra $100 a month, and instantly see how much sooner you'd be done. Watching that timeline shrink is really motivating and you're the one in control of the plan. Plus, a quick check in on your phone or laptop tells you exactly where your money stands.

• 21:46Speaker 02

  It's so important to manage your finances and I believe Monarch can help you in doing so. Write your own money story with Monarch. Use code WSB at monarch.com to get your first year of Monarch Core half off at just $50. That's 50% off your first year at monarch.com with code WSB. Interested in futures trading and looking for a place to get started? With Plus 500 Futures, you can explore some of the world's most active markets, all from one easy to use platform. Trade futures on popular assets like the S and P 500, NASDAQ, oil, gold, Bitcoin, Forex, and more. Whether markets are moving in crypto, commodities, or indices, there's always something to follow.

• 22:31Speaker 02

  Plus 500 Futures is designed to make trading straightforward with a platform you can access wherever you are, right from your phone. And you can get started with a minimum deposit of just $10. Not ready to trade with real funds? Try the free unlimited demo account first and explore the platform without putting real money at risk. Backed by nearly 20 years of experience and trust by millions of customers worldwide, Plus 500 gives you the tools to stay connected to the futures markets. Visit us.plus500.com to learn more. Trading in futures involves risk of loss and is not suitable for everyone. Not all applicants will qualify. Plus 500, it's trading with a plus. You don't need me to tell you that rent and mortgages are only getting more expensive.

• 23:15Speaker 02

  That's why Built created three credit cards designed to offer the richest rewards on rent, mortgage, and everything else you spend on. Whether you rent or own, you can earn up to 1.25X points on every housing payment without the transaction fee, no matter where you live. Points can be redeemed for flights, hotels, Lyft rides, a future housing payment, and much more. Turn one of the most unavoidable payments in your life into the most rewarding. Find the card that's right for you and apply at joinbuilt.com/tip. That's J O I N B I L T dot com slash T I P. Use our link so they know we sent you.

• 23:58Speaker 02

  Terms apply, subject to approval and eligibility. Built cards are issued by Column NA, member FDIC pursuant to license from Mastercard International, Incorporated.

• 24:09Speaker 01

  All right, back to the show.

• 24:11Speaker 02

  Incredible thing with Palantir is that all this growth, which is pretty much unheard of for a company of this size, did not come with any margin compression. It didn't take any major marketing expenses or anything else that they really needed to invest in to scale the business dramatically to accelerate this growth. And profit margins expanded alongside their top line growth metrics, which was, of course, an explosive recipe for shareholder value creation. And so back in 2023, when growth slowed down pretty materially, the company was barely breaking even with a net profit margin of, you know, just about 3% to 4%, which is pretty modest. Today, though, net profit margins are 60% or almost there. And so you have this double engine of incredible growth on the top line,

• 24:57Speaker 02

  paired with this incredible margin profile and a huge, I mean, a tremendous amount of operating leverage. I think that's the biggest swing in margin profitability that I've seen from revenue scaling from maybe any company besides Uber, which is one of our favorites.

• 25:13Speaker 01

  Well, there's not one episode where we don't have to talk about the margin inflection of Uber, but it's actually quite impressive, especially when you think about modeling these companies, right? We always think about, well, five years out, where do I see the margin? And you want to have those companies that surprise you to the upside on exactly that metric, right? We just did an episode on all of our winners. One thing that we saw with many of them is that the margin goals that we set for three or four years time, they basically hit next quarter, which is, you know, they just completely changed how they think about capex, spending and all of that. And also, you know, that's, as you said, like a recipe for just success. And Palantir showed both of that to an incredible extent. And one thing you can do for especially software companies is look at the so-called rule of 40. That's often used again for software companies, but also some other companies. We technically can say, well, you can generate high margins, right?

• 25:58Speaker 01

  We're not talking about, let's say Dell or these infrastructure companies. So what you basically do with this is you add up a company's growth rate, the top line, the revenue, and the profit margin. So let's say you have a company that has 20% top line growth and a 20% margin. That gives you 40 in total. So, you know, this would be a company that's hitting the rule of 40. And that's considered a pretty healthy company and growth at a good margin. Now for Palantir, that number is not 40%, it's 155%, 65% in profit margins and 90% top line growth. And if you believe CARP and, you know, the guidance, this will get even better in the next quarters. And I cannot say it enough, this is an insane number if you look at it the first time.

• 26:41Speaker 01

  That's also the case because Palantir only goes for big customers with significant upsell potential, which is, as I mentioned before, where the overall customer count of only 1,000 seems quite low. I mean, Palantir's average deal size is in the millions. And just this quarter, they closed 220 deals of at least a million dollars. Close to 100 of those was at least $5 million, and more than 70 deals were at least $10 million. So most enterprise deals that we know from companies like Salesforce, for example, which is also a company that we covered here on the show, they're in the, you know, five to six figure range, maybe a seven figure deal every now and then, but Palantir is only going for this big one. They're not even trying to win those smaller ones, which is why the customer account generally is not accelerating as fast as the volume.

• 27:26Speaker 01

  So gaining 42 new customers might not seem that impressive, but closing 220 million dollar deals, that's pretty impressive. And that's why net dollar retention is so high and why Palantir is growing like a weed.

• 27:39Speaker 02

  It's absolutely insane what they're doing. Words really don't do it justice. And that's why the market has, I think rightly, put absurd multiples on Palantir. Because the multiples look absurd because they're based on trailing earnings and they're not accounting for or not reflecting the future growth that it seems like is very possible for Palantir. And so you had the stock at times trading at like 100 times sales, which again, sounds ridiculous. And we'll talk about valuation later. But when you're growing at this incredible rate and there's such a large addressable market for you to continue to roll these services into, it starts to feel a whole lot less crazy for the stock be trading at 100 times trailing sales, emphasis on trailing.

• 28:24Speaker 01

  We will talk about it later and I don't want to spoil anything, but I can say that Palantir is actually as attractively valued as ever, even though the multiples look, you know, kind of scary at first. And again, I really came into this with all sorts of biases. There is more to it than, you know, you would think looking at a headline P, as you just said. And I gotta say, still, I'm a bit mad because when I first looked at Palantir, which has been about a couple of weeks ago, I thought this looks like a pretty good opportunity. And then earnings came up and suddenly the stock went up 40%. And I gotta say, that was a bummer, you know? We still sound pretty confident here and I think many people are surprised if they listen to this episode and the value guys suddenly like Palantir. But again, when we go to the valuation and also, you know, all this stuff now that comes to

• 29:04Speaker 01

  just the qualitative factors of the company, I feel like people will understand why we like the idea so much right now.

• 29:10Speaker 02

  Yeah, it's hard to say that the valuation is as attractive when it's jumped up 40%, but also it's a great testament to the fact that Palantir is just continuing to perform and absolutely blow out the numbers of any kind of estimates that people think of are possible. And so judged by CARP's growth outlook, even after that jump, there might still be an opportunity there. So, you know, talking about the latest earnings release, my question for you is, does Palantir report which platform is driving the most profit or is growing the fastest? Is there any insights that we get into that?

• 29:46Speaker 01

  Unfortunately, no. We only get to see the split between commercial and government revenues, as well as a breakdown by geography. And you can generally say that government and commercial growth are pretty much in line, which is something that I like to see because it also reduces Palantir's dependency on government contracts, which, at least in my opinion, can be a bit more volatile field depending on who's in the White House and also internationally, and that's a similar, more volatile field to play in. And then in the last few quarters, commercial has actually outpaced government, but this has shifted in the past, so perhaps we'll see that shift again. Although I would expect commercial to grow more in the long run, just given that the universe of customers is obviously larger and Palantir has improved significantly, which is what we discussed, in the onboarding of customers.

• 30:30Speaker 01

  So I would be surprised if we don't see that outpacing government in the long run at least. And also looking at the EBIT of both sectors, so the earnings before interest and taxes, which is a profit proxy, you can also see that Palantir has a similar margin for both government and commercial. So it's not like one of them is significantly more or less profitable than the other.

• 30:50Speaker 02

  There is a significant difference here between growth in the US and the international markets, and the US growth has more than nine Xed since 2023, which is so absurd. And it's up 115% year over year this past quarter, while international growth has definitely been much more volatile and is currently sitting at 30%, which for any other company on the planet would sound very impressive, but for Palantir, it does not sound very impressive. And so my question, I guess, is, is that primarily because of Palantir's poor reputation internationally? Or what really is the reason for that? Is there any kind of explanation there? I've read that Palantir systems were critical to helping the US government, for example, track down Bin Laden back in the day.

• 31:33Speaker 02

  So if anything, I'm actually kind of surprised that the company's even allowed to work with international clients and even commercial clients, based on the company's importance to the Pentagon and access to national security data.

• 31:45Speaker 01

  We'll get to that later, but it's actually quite fascinating story of, for example, why Palantir is allowed to work with, you know, European customers and also the corporate space. And generally, I would say it's a mix of things. So there's certainly a European resistance to being reliant on American companies, especially when it's about data management on the level that Palantir practices it. I mean, both France and the UK have recently ended deals with Palantir to go for local solutions that I think do not yet exist, but they want to build them. We'll see how that plays out. But there are also, you know, sales and market dynamics that we have seen in other markets, like, you know, cloud computing or even now AI, where adoption just takes longer in Europe. And given how closely Palantir has to work with its clients, you can probably also just,

• 32:28Speaker 01

  you know, not hop on a Zoom call and then onboard, you know, the client and the new software. So you need boots on the ground. And while Palantir has offices in most major cities in Europe, I would assume there's much less capacity there compared to the US. And I think CARP's relationship, and that's what I'm, you know, talking about earlier, with other countries is also a bit difficult. I think he lately said in an interview that he just likes to support Europe's institutions, otherwise he wouldn't even make any business there because, quote, "the growth sucks." So, I don't know, think about that whatever you want, but I guess the fact that Europe has been growing much slower than the US is another major reason for why US numbers for Palantir are just better. And by the way, on the last earnings call, because we talked about guidance, for example,

• 33:09Speaker 01

  here, Alex Karp said that he will grow the business at a rate equal or above what the US commercial business is doing for the next 18 months. And just for context, that would imply a growth rate of about 150%. And even if he only reaches, you know, let's say 100%, that would take Palantir to $17 or $18 billion in revenue at the end of 2027, which is insane.

• 33:33Speaker 02

  Even if I don't exactly understand what Palantir does or what makes them so special compared to other consulting and data businesses, the financial fundamentals and the business itself are looking really, really good. But again, I'm still asking myself this question of why is there no one that seems to be able to copy what Palantir is doing? I mean, what about OpenAI and Anthropic or the big software companies you mentioned earlier, Salesforce, ServiceNow, or even Microsoft? Couldn't they do what Palantir has done with their suite of products and especially the connections into the enterprise universe that they already have? You would think they might be able to build on that and create some kind of ontology similar to Palantir.

• 34:15Speaker 02

  And so we've mentioned many times in recent episodes that distribution seems to be one of the most important advantages in the age of AI. And that means, you know, already having a widespread customer base. And as outsiders, without any insights into Palantir, how are we supposed to have conviction in their technological advantages? How do we even begin to rate the strength of their moat?

• 34:39Speaker 01

  Yeah, this will be probably the most important point of this episode, and it's not easy, primarily because a lot of Palantir's advantage seems to come down to execution, which is always a bit difficult to talk about and judge if you, you know, as you said on the outside. But I will try my best. And probably the first thing I have to talk about, it's the switching cost mode. You know, we talked about how Palantir creates the ontology and how long all of that can take. And well, once you go through that multi-month or even year-long process, I think you will think twice about switching and doing it all over again. And since Palantir has by far been a first mover here, right? There is no competition that has come up even remotely in the last two years. They just built a massive client base before anyone else can start. And especially if you think about who are those customers?

• 35:23Speaker 01

  It's not only 1,000 customers that you have, but you have the biggest corporations and you have the government. So those are contracts that if they are multi-year long, I think it's just incredibly difficult to get into that sort of space and compete and to just zoom in on that process again. That alone can probably not be done by most of the companies. We talked a bit about it before our call here, but as you can imagine, when listening to an Alex Karp interview for just two minutes, this company has a very unique culture. And I mean, I think we have a pretty unique culture here at TIP, but it's something entirely different when your company is worth hundreds of billions of dollars. There's just not really a hierarchy, as it seems, and people can do what they believe creates most value at the time. And of course, there are limits to that, as always, but that's how it generally seems to work.

• 36:07Speaker 01

  And then when you look at it from just the product side, it basically goes against all industry wisdom, right? I mean, the way enterprise customers used to buy software tools was to go to many different companies and then pick the best-in-class solutions for each of the use cases that they look for a product for. And then a company might use Slack because they prefer it over Teams, just as we do at TIP. They might use Excel because they prefer it over Google Sheets, which is what we don't do at TIP. And then Zoom because they prefer it over Google Meet. And, you know, maybe Salesforce is the company that they have for their databases because they prefer it over whatever competitors are out there, right? So that's how you can think about it in the past, and Palantir sort of changed that dynamic.

• 36:48Speaker 02

  So you usually have companies that build maybe one or two best-in-class tools and sell them to as many different enterprises as possible. And so that's lower value per transaction than what Palantir does, but it is much higher volume. And so it's sort of like Palantir is a more comprehensive AI-native version of Salesforce, almost, at least in terms of importance to enterprise productivity.

• 37:10Speaker 01

  Yeah, I guess that's fair.

• 37:12Speaker 02

  And then you also have a company like Microsoft, and then you have Google too, to some extent, that put out many tools where only one or two are, again, actually best-in-class, but the others are good enough to be sold alongside them as a bundle. And so I guess, wouldn't it be possible for those companies to level up their entire bundle and connect it with AI, similar to what API has done for Palantir, and then basically offer a similar form of ontology? And it sounds like Palantir developed a really specialized machine learning technology set well ahead of the AI craze of the past few years and primarily targeted it at the government and military. But now growth is exploding as they've rolled out this model to commercial customers.

• 37:55Speaker 02

  And so I just can't imagine, though, that companies like Salesforce, IBM, Google, Microsoft, many of the biggest names in tech, I can't imagine that they're just going to take this lying down. I've seen estimates that Palantir could have as much as a $2 trillion addressable market. And if that is even remotely true, it's going to surely have some serious competition from some really powerful companies chasing after those same dollars.

• 38:22Speaker 01

  I think it's safe to say that Palantir has not yet been tested by competition, but I also think that it's easier said than done to switch from building tools that are just good enough to tools that are actually best in class. And not only one tool, we're not talking about Slack, but a tool as comprehensive as what Palantir built in. We said in both of our Google and especially our Microsoft episode, that's what they do, you know, good enough tools that they can just put into the ecosystem and it works because they sell it to these enterprise customers. And I should also say that Palantir is not taking their market. So Google can sell their bundle just like they did in the past and Microsoft can do the same. So they're not necessarily competing. So you don't have to invest just as a defense mechanism, basically. And you often hear me say that I don't like to bet on companies that compete with the likes of Google, Microsoft, Amazon.

• 39:07Speaker 01

  But I think in this case, I don't know, I think they're way too bloated and just not flexible enough to compete with a company like Palantir on that ontology front. I mean, I told you before that if you ask yourself, which of these companies should have actually sent out FTEs, these Forward Deployed Engineers, to enterprise customers to set up their tools? I think none of them would have done that. All of them would have settled for keeping the engineers in-house and deploy them for their software solutions where they make 60% margins, right? And I mean, Microsoft is trying to do that right now, the, you know, billing and ontology layer. But I don't know, looking at it, and by the way, they call it fabric IQ, I would call it an attempt at best. I think it makes sense for them just to better connect all of the different products they have, but I don't see it rivaling Palantir anytime soon.

• 39:50Speaker 01

  And this goes back to the sort of good enough, you know, you're good enough to connect all your tools. But this is so far away from what Palantir is building for the highest class customers that I think it's a reach to say that this will compete with Palantir anytime soon. And I think the one company where I can't really figure out whether they would become a bigger competitor eventually is ServiceNow, because it's the one company that we discuss here now that I haven't yet deep dived. So my understanding of ServiceNow is significantly below the level of the other companies. From what I understand though, ServiceNow understands itself as, you know, a SaaS company that becomes AI native, right? And they actually have a lot to lose if they would have, you know, this $2 trillion market that they are just not participating in.

• 40:32Speaker 01

  And it's very good at workflows that are based on, I think, what they call predefined relationship data. So how can I explain that? I would say that predefined relationship data is you already know what you're allowed to do, and now you just put it in a sequence that you now know, hey, this thing is allowed to give access to this thing. And it makes sense to now put this thing just after that. So you sort of get this linear line of what are you allowed to do? What Palantir is doing, it creates those relationships. You know, it's saying, we know that there are these things, you know, X, Y, Z, and they have permission to do certain things. Now, how can we connect that to the entire rest of the map so that it makes sense? It's just hard to imagine, but way more complicated than what any of these SaaS companies are currently doing.

• 41:16Speaker 01

  And I think that's just a complex way of saying ServiceNow is good at managing data that is already set in place, where Palantir is basically taking raw data from all sorts of places and then creates a relationship in all of that itself.

• 41:29Speaker 02

  Well, maybe I'm naive, but what about companies that maybe can use Claude, Co-Work, and ChatGPT, Codex, and these different AI native tools to actually recreate this ontology in-house, right? Maybe AI enables them to map things in a way that only Palantir was able to do, but now that technology has been more democratized. And I guess building here doesn't exactly mean doing all the coding work, but I just mean sort of what you described in the beginning, finding a way to group all sorts of best-in-class services and redefine what the best relationship between them would be and connect them in different ways.

• 42:10Speaker 02

  And so essentially you would just be rethinking how things are connected inside your organization. And maybe you won't be able to do it as well as Palantir, but for a fraction of the cost and with different AI tools, perhaps it's good enough.

• 42:24Speaker 01

  And actually this time, good enough would actually be good enough because I think if you get a tool from Microsoft that they think is good enough, it's not actually good enough for your corporation. But if you build it in-house, you know, you know, what level you need to reach that actually makes sense for you. And just for context, in case anyone is asking themselves, well, if that would be possible, wouldn't they have done so already? I think the reason it might be more popular today is that Palantir showed companies just how inefficient they actually are with their data management. And before, it just wasn't as obvious. I mean, everybody sort of knows that there are things that are, you know, not the most efficient, but it always feels kind of unchangeable, right? And now that AI is there and seeing what AIP is doing for Palantir, I think that just makes them realize there's so much more that they can get out of their data and so much

• 43:09Speaker 01

  more efficiency gains that they could get. And also, you know, cost savings, which is one of the biggest things that customers of Palantir are actually reporting about. And all that said, I think in-house solutions are to some extent wishful thinking. And we already went over why, basically, when I explained how Palantir is going into these companies with the engineers and how they implement their products. I mean, there are dozens of different incentives for people working in the same company in different departments, right? So again, that's why I talked about, hey, Palantir is going out there and befriending, for lack of a better word, the C-suite first and then the project workers instead of the middle management, because that's where a lot of the resistance for new stuff actually comes up. And then if you would think about who is building these in-house solutions, well, it almost has to be the middle management, right?

• 43:53Speaker 01

  The CEO can maybe say, that's what we want to do, but then the middle management has to take that idea and deploy it in their own department. And I don't think that will happen. It hasn't happened in the last decade. I think it's still unlikely.

• 44:04Speaker 02

  I can see why that worked really well at a smaller scale, but it doesn't really sound scalable to me. And I wonder if something is lost in translation as they change their playbook to accommodate more types of enterprises. And so, I mean, you can't befriend the C-suite of every single customer you have if you want to be a trillion dollar company. And to your point though, I think it does make sense as a middle manager, you don't want to rule over a smaller kingdom or suddenly have someone from the outside coming in, changing up your entire department and telling you what to do. And so as the CEO or CIO, you care more about the overall performance of the company. And as a project worker, you just want the best tool at hand so you can do your job as efficiently as possible.

• 44:47Speaker 01

  I think they noticed that it's not that scalable in 2023, when you know the growth rate has gone down to 12%. And that's why they started the workshops, you know, where they said, well, if we go out to all of those companies and we befriend the C-suite, that takes a lot of time, you know, time that is inefficiently spent to some extent. So why do we not just give them proof of concept? We invite them after Palantir was better known in the enterprise space. Now they are coming over, so they already sort of in the realm of becoming customer of Palantir. Now they're in the workshop, now Palantir is showcasing their product, and that's how you, you know, scale significantly better. I think I heard that they had over 1,000 workshops already. And mind you, there are CEOs and CIOs sitting in all of them. So you can sort of see the scale of this new approach.

• 45:30Speaker 02

  What about Anthropic and OpenAI, just to bring them up again? I mean, if there's any company that should have the best AI capabilities to get this done, it would be them. I know Anthropic in particular is focused on taking more of a B2B approach, whereas OpenAI has hundreds of millions of users and that's been sort of actually a problem for them because it's so compute intensive to create responses to very basic queries from everyday people. But they very much want to move into providing more enterprise-based solutions. And so I guess with that said, I'm wondering now, what model does Palantir even use? Do they have their own LLM? Are they using OpenAI and Anthropic themselves?

• 46:14Speaker 02

  And, you know, I wonder what the competitive dynamics between them are and how that could evolve over time.

• 46:19Speaker 01

  Yeah, that's a good question because it's actually an important point. I mean, Palantir does not have its own model. It uses a variety of models out there and it basically depends on the customer, which model they want to use, and maybe also they change between them. So Palantir itself offers, you know, Claude, ChatGPT, but also Gemini, Mistral, Lama, and basically whatever model is out there. And I would say the magic of Palantir really comes from the context that the ontology creates, you know? Again, imagine you have this map and before, if you were, you know, an employer of a company who uses this, you have to go manually through all of these data points, through all of the things that are possible within this map. Now you're just typing in a prompt, you're saying, hey, I'm looking for getting this aircraft to another tower, right? And then just typing it in there and then whatever model you choose, it's doing all the work for you.

• 47:04Speaker 01

  So it just simplifies things and that's how I would think about it. So the magic of Palantir really comes from the context that the ontology creates. So I think we actually have a video on our YouTube channel that is called A Masterclass on AI by one of our Mastermind members who understands LLMs and the entire space much better than we do. And he actually has an argument for why some models feel so much stronger than others, although they're kind of similar in strength. And I would say we just show it here because he makes the point pretty good.

• 47:32Speaker 03

  The common word in parlance for us is what is called as a harness. So we call this as an agent harness, right? And agent harness is access to the data that's sitting on your desktops or some particular ecosystem where the data lives. In this case, in some cases, for example, as you know, Jeff, you were asking about what happens in Palantir, right? The LLM is still the actual reasoning and planning part of it, right? But the agent harness now involves data that comes from your databases. The ontology now becomes a thing that the agent, that you can provide to the agent and say, here's the ontology of how my data sets look like, right? Here's where I have my inventory data, here's where I have my supplier information, here's where I have my X, Y, Z information. And Palantir simply builds an ontology on top of it. Now the agent can look at that data and then reason about it right now, right?

• 48:17Speaker 03

  That's why all the LLM can look at the data and reason it. The whole thing is what is called as an agent at this point.

• 48:22Speaker 01

  So just as he said, you know, the context is very important. So Claude feels much more powerful than, for example, ChatGPT, just because they do a way better job of giving context to the model. And it's kind of similar with Palantir. Its ontology is basically feeding context. So that is data governance rules, guardrails to whichever LLM is used. And thus the LLM stops hallucinating and they make the model much more powerful because it's basically a custom tailored solution for the use case at hand.

• 48:52Speaker 02

  I think that makes sense, but at some point after OpenAI and Anthropic have IPO'd, they're going to be trying to justify these trillion dollar valuations. I would think it would be very tempting for them to start a business where they basically offer their service to customers in a similar way to what Palantir has done and maybe creating a similar ontology. I know we keep using that word a lot today and really taking advantage of their own models and the context they have and the fact that they do already have, let's say, a lot of existing enterprise relationships and basically leveling those up. So how do you think about that? Is that something that is a real threat to Palantir?

• 49:29Speaker 01

  I guess it's a problem in theory, yes, but I'm also asking myself, why should any business want that over Palantir? I mean, after all, you would now have pretty much the same product, assuming OpenAI and Anthropic are actually able to build such an ontology layer, but you're limited to just one model instead of being model agnostic. And if you think about, you know, a corporation and what that means, basically all the pricing power then sits, for example, with Anthropic, where now if, you know, the Anthropic model is too expensive, well, you can still use Palantir and you just use ChatGPT. And if ChatGPT is too expensive, you just use Mistral. So I think it's just a way more diversified way. You could still say, especially looking at Palantir's margins, where the pricing power

• 50:09Speaker 01

  sits with Palantir right now, but probably we will see that come down over time, especially with, you know, maybe Anthropic or OpenAI will actually start and put up a competitive business. I still think that Palantir would win, but certainly we haven't seen the margins being tested by competition. So that's definitely something we have to look out for. Generally, Alex Karp is frequently shooting against Anthropic and OpenAI and basically warning about those companies, you know, and how you shouldn't give them your IP and your business model because these frontier models sooner or later will just take over your business. And obviously Alex Karp has a major incentive to fuel fear among corporations because those are his potential customers. But I would also say that doesn't mean he's wrong. I

• 50:53Speaker 01

  mean, whatever you put into ChatGPT or Claude, they can take it to train their model. And who knows what the actual policies are for saving your data. I know that I personally sometimes ask it, hey, by the way, if I send you this PDF, will you save it? Or is it just gone when I delete the chat? They never give you a full answer to what they actually do with the data. So I can say that I take the liberty of not caring too much, but most of the time it's, you know, not that important information. And probably it's a bit naive on my point, but if you're a company and you know, you have important IP and you have a business model to protect, you should care much more about your privacy and safety than I do.

• 51:28Speaker 02

  Let's take a quick break and hear from today's sponsors. You know that sinking feeling when a meeting ends, everyone's nodding along, and two days later you realize that nobody wrote down who was doing what. That's why I use Granola. Granola is an AI-powered notepad that makes your meeting notes useful. It captures what happens in your meetings and it turns it into clean, structured notes with decisions and action items pulled out and made easy to find. On days with back-to-back calls like guest prep and team check-ins, community calls, I used to spend the gaps cleaning up messy notes and trying to remember who owned what. Now I stay in the conversation and weeks later, everything is searchable in one place. Once you try it on a first

• 52:12Speaker 02

  meeting, it's hard to go without. You can try it totally free. Head to granola.ai/wsb. That's granola.ai/wsb to get your time back. Go to granola.ai/wsb and try it on your next call. Interested in futures trading and looking for a place to get started? With Plus500 Futures, you can explore some of the world's most active markets, all from one easy-to-use platform. Trade futures on popular assets like the S&P 500, Nasdaq, oil, gold, Bitcoin, Forex, and more. Whether markets are moving in crypto, commodities, or indices, there's always something to follow. Plus500 Futures is designed to make trading straightforward

• 52:54Speaker 02

  with a platform you can access wherever you are, right from your phone. And you can get started with a minimum deposit of just $10. Not ready to trade with real funds? Try the free unlimited demo account first and explore the platform without putting real money at risk. Backed by nearly 20 years of experience and trust by millions of customers worldwide, Plus500 gives you the tools to stay connected to the futures markets. Visit us.plus500.com to learn more. Trading in futures involves risk of loss and is not suitable for everyone. Not all applicants will qualify. Plus500, it's trading with a plus. This September, I was in New York for the Intrinsic Value Conference. The talks were great, but what I kept thinking about is what happened around them. On Friday night,

• 53:39Speaker 02

  about 20 of us crowded around a long table at a family-style Italian spot near Times Square. Saturday, we did it again. And Sunday morning, a group of us were still debating Socrates over breakfast before everyone headed home. Many of the people at those tables had only known each other from our calls in the Mastermind community. That is the community I host with Kyle Grieve and Daniel Manca. Each week, we get together on live calls to dig into markets and individual businesses. We bring in fund managers, operators, and authors as guest speakers. And a few times a year, like New York this fall and Omaha every spring, we meet in person. If you want to think in decades alongside people who

• 54:21Speaker 02

  do the same, apply for free at theinvestorspodcast.com/mastermind. If your application stands out, I'll be the one to reach out to you personally. All right, back to the show. I guess the question is whether you can trust Palantir and aka Karp and Peter Thiel with your data. And perhaps we can talk more about the management team than in general. Without getting too political, Karp and Thiel are both very controversial people, as we've alluded to. And you mentioned earlier how that can actually impact, for example, the international process.

• 54:53Speaker 01

  First of all, I should say that I don't know if you can trust them, but Palantir's customers keep all the rights to their data, of course. So while Palantir, you know, gains a lot of insight into how the business operates and works, that's a bit different from just using Claude or ChatGPT where they more or less own everything that you feed to them. But getting to the management part, as mentioned before, Alex Karp has been running the business as CEO for 23 years now. Stephen Cohen, who built the first prototype back when he was still a student at Stanford, is still part of the board. Thiel is also on the board, but the entire founding team, you could say, is involved, but not as actively as Alex Karp. So if you now think about just one person and Palantir, it would be Karp and to some extent still Thiel, just because you feel like the vision might be his. And

• 55:38Speaker 01

  just for context, Karp and Thiel met at Harvard, the law school of Stanford. So Karp then went to, again, I mentioned before, my alma mater in Frankfurt. And then he actually learned German and lived there for about a decade, which is significantly longer than I thought. And I think he mentioned that he was primarily interested in the Frankfurt School of Thought, which you told me you did some research on that after I told you about it. And it's basically a philosophy that analyzed how modern capitalism, mass culture, which was also a big thing, and fascism shape human psychology and social control. And Karp, just like Thiel, was extremely interested in just, I think you could say, understanding fascism. And I'm not quite sure whether his interpretation of the Frankfurt School actually

• 56:22Speaker 01

  aligns with the original idea, but it is quite important to understand how he thinks about it, just for, you know, getting the vision of why Palantir exists and acts the way that it does.

• 56:33Speaker 02

  This has got to be the most philosophical we've ever gotten during a company analysis.

• 56:38Speaker 01

  Well, I gotta say, I keep it brief on the details. I tend to get lost in that stuff just because I'm with slight interest in it. But I also got to admit that it has been a while since I actually read a lot of the theories of the Frankfurt School, basically a couple of years ago when I was still studying there. But I would say it's also not too important to actually get the details. The main idea that I found so interesting is that Karp is thinking that American thinking, whatever that means, cures nations of fascism and that it's a sort of moral obligation for Silicon Valley to return to its roots. And, you know, that basically is the US defense system. And that goes back to your point earlier in terms of, why does the CIA have a venture arm? Well, in the early days of, you know, the Silicon Valley, most of the companies were founded, especially these early

• 57:22Speaker 01

  semiconductor companies and the computing companies, just because the Pentagon paid for them. And then somewhere in the last 30 years, that's Karp's theory, the industry decided that working on national security is sort of distasteful and that they should redirect the best engineering talent in the world towards consumer apps and advertising. And one of his examples was Google, which, you know, is the biggest company in our portfolio. So we're happy they did this sort of switch. But his example was Google walking away from a Pentagon contract in 2018 after employee protests. And he claims that this has been a mistake for both sides, actually. So tech lost any sense of what he calls the larger purpose. And the government lost access to the best engineers and the best software

• 58:05Speaker 01

  engineers in the world. And I think he partly believes that this was part of the critical theory that came out of Germany and Europe, where philosophers were more or less critical of technology and innovation because, well, one of the most advanced nations in the world still started the Second World War. And Karp's things, and this is the last point I will mention here, that this defeated mindset that the critical theory had in many aspects also came to the US, although they won the war. And anyway, long story short, Palantir is now supposed to bring back, you know, this American tech to its roots, to the defense of America and the West and their allies generally against attacks from the outside.

• 58:45Speaker 02

  It seems like a very ideologically driven company, which is sort of weird. We don't usually get into this terrain in our episodes. And I suppose when you take that context, it makes sense why Karp would do business in Europe. He's not really thinking about the economics, right? He said that growth sucks, but he is seeing it as part of a mission to support the West, whatever that means.

• 59:08Speaker 01

  I guess so. I think it's kind of up to you whether you believe all of that. I mean, Palantir is still generating quite a lot of money in Europe. So I don't know, I doubt that altruism is the primary motive for making business there.

• 59:21Speaker 02

  I suppose Peter Thiel has the same or some sort of similar mission.

• 59:27Speaker 01

  More or less, I would say. I don't know whether they agree on everything politically, but that's also not really the topic here. I mean, we only discuss all of this because their worldview is important when you build a company that is taking on some of the most sensitive defense contracts in the world. And right now, I think I could say that few investors care about Palantir's reputation outside of the investment world. But when people are actually afraid to invest in Meta because of its public perception, a company that is building an app called Migration OS for ICE definitely comes with much more reputational and regulatory risks, if you ask me. And in several countries, courts have actually ruled that the way Palantir software connects and automates data is not in line with law.

• 1:00:10Speaker 01

  So there's a lot of potential for problems internationally, probably also in the US sooner or later.

• 1:00:16Speaker 02

  There gotta be two angles to investing in Palantir. So a personal one where you have to decide whether investing in the company aligns with your values. And then there's a question of how this reputation that Palantir has can actually affect the business and cash flows. And it's similar to thinking about Tesla and the political impact that Elon Musk has had on that business. And so as you hinted at, France's internal security directorate replaced Palantir. Germany's domestic intelligence service appears to be likely to do the same. You had the mayor of London blocking a police contract that would have been worth 50 million pounds. So those are some lost revenues for Palantir already. But to be fair, as we've gone through the numbers earlier, this is just a drop in the

• 1:01:01Speaker 02

  ocean compared to the business that Palantir now generates.

• 1:01:04Speaker 01

  And fact of the matter is that Europe is spending a lot on defense right now. And if they can't build a viable alternative to Palantir themselves, I believe that sooner or later, I would at least be surprised not to see deals with governments in Europe more. And I don't know, I'm a bit skeptical that, you know, I think Germany and France are sort of building a system on AI platform to challenge Palantir, knowing about our data laws. I'm not quite sure if that will turn out successful.

• 1:01:31Speaker 02

  Moving on from politics and philosophy, let me ask you a question about what feels like a lighter topic for investors, which is how does the capital allocation look at Palantir?

• 1:01:42Speaker 01

  Well, Palantir has a major cash pile. They hold more than $9 billion in cash and there's no debt on the balance sheet, not long-term, not short-term. So they could return all that cash to shareholders if they wanted to, but I gotta say, that doesn't seem to be on the plan, at least not anytime soon. They don't do any M and A either, and they terminated the share repurchase program that they had in place in January of this year. And I actually think that's a smart decision because I would rather have the opportunity for more opportunistic buybacks in the future instead of buying back stock when it's trading at triple-digit sales multiples as it has in the past.

• 1:02:16Speaker 02

  Yeah, the problem of having no buybacks is that stock-based compensation is running at 13% of revenue. So shareholders are getting diluted at a pretty significant rate. And while the trend is moving in the right direction, I think that's mostly because revenue is increasing so much. Two years ago, the stock-based comp was as high as 20% of revenue. And so that is a real hit to shareholders and shareholder value. Their basically slice of the pie against their own consent is being shrunk down. And so as the pie keeps growing bigger and bigger, if you have a smaller slice, you know, financially it can still work out, but you are being diluted.

• 1:02:58Speaker 02

  And, you know, like I said, I'm sure they didn't care that much about this solution when the stock was going from $20 to $170. But as we're looking forward and thinking about our decision today, it's definitely a headwind. And as you said, on the other hand, I'm not sure how excited I would be to see them buying back stock at 150x earnings or 100x sales, you know, overpaying to retire shares. If you believe that Palantir is overvalued, that can be quite destructive to shareholder value, even if it's somewhat mentally reassuring to see, hey, there's buybacks going toward offsetting dilution. Again, like I said, if you're dramatically overpaying well above intrinsic value, you

• 1:03:40Speaker 02

  are destroying value for shareholders, which is, you know, a little abstract, a little bit academic, but it is, I think there's truth to that idea.

• 1:03:49Speaker 01

  And by the way, if you're asking yourself, well, aren't you guys bullish on Palantir? You're talking about that for 60 minutes now. So wouldn't you think that it's attractive to buy back shares at today's prices? I should really clarify again that last year and the year before that, the story was completely different. So Palantir actually changed compared to last year because growth has accelerated so much and the valuation has come down that I do believe it's attractive here and especially it has been before the earnings release. Just a year ago or two years ago, I considered it rightfully so, I believe, to be one of the most expensive companies in the market and it probably would have been a good shot. But what I ask myself in terms of capital allocation is why does Palantir not invest more money in their workforce, especially given that they always say they have so much more demand for their products and they keep generating more revenue per user?

• 1:04:33Speaker 01

  So I just feel that an investment in the workforce would seem like quite a good investment. What Palantir does instead, and I talked to you before our recording about this, I don't fully understand it, is that it partners with the consultancy firm Accenture. And basically they use Accenture employees as forward deployed engineers. So Accenture trains its own consultants on foundry and NIP. So they obviously have a better idea of what Palantir is than the average Accenture consultant. But still, those people then do the deployment work at the customer side. And Palantir sells the software license. So on paper, it makes sense because they keep the high margin revenue. But I feel like that doesn't fully make sense because you feel like Palantir is only hiring

• 1:05:16Speaker 01

  these highly talented people because a lot of the work that they do is going to the company, figuring out how it works, doing all the politics and then getting the ontology. And now it seems like you can just get 1,000, you know, nothing against people who work at Accenture, please don't misunderstand that, but you can just get people from Accenture who don't understand Palantir to the same level than Palantir engineers and they do the work for you. And again, it's not a small amount of people. For context, Palantir has about 4,500 employees. Accenture has close to 800,000 and they already have more than 1,000 foundry specialists.

• 1:05:52Speaker 02

  I'm stunned that there are companies out there employing 800,000 people. That I can't even fathom that. I don't know where you even begin in overseeing that many employees. And so it probably makes a ton of sense why Accenture would be looking for outside help or why Palantir would be looking to tap into their network of employees. But, you know, how about we transition to talking about valuation now? I'm very curious to hear how you think about this as a value investor, but also as someone who seems quite excited by Palantir's technology and the growth. It is really riveting. And so admittedly, I would have dismissed Palantir as being a hopelessly overvalued meme stock.

• 1:06:34Speaker 02

  But you already told me before recording that there is maybe the chance for legitimately attractive entry points in this stock, especially with how volatile it is.

• 1:06:44Speaker 01

  I was surprised myself, you know, as I say today, multiple times in the episode. And obviously, again, you don't have a value stock here. I'm fully aware of that. But you really got to put the valuation into perspective, considering the incredible growth. And if Karp is actually right about being able to grow Palantir overall at the current rate of the US business until the end of 2027, well, then the price to sales ratio would decline from where it is today, which is about 60 to about 20. And then it pretty much all depends on, you know, the growth rate in 2028 and 2029. And of course, this is speculative, but Karp has always delivered up until now on the guidance that he gave. And of course, you could argue this is not only a guidance, but something that he said in an interview, but I would just assume, I will take it for granted here that this

• 1:07:28Speaker 01

  is what he actually aims to achieve. And analysts have massively underestimated Palantir again and again. And I should say that if you think a world price to sales is not the most useful metrics, generally I agree with you, but I think it makes a point here quite well because Palantir has incredibly high margins. So when the average enterprise software company is trading at about, you know, seven times sales on roughly a 20% operating margin, that's a 35x on operating profit. Now, if Palantir on its assumed 2027 numbers would be about 22x sales, but it has a 60% margin, that would also result in 35x operating profit. So the massive growth and the margins make Palantir relatively, at least to other software companies, look more expensive today than it actually might be.

• 1:08:12Speaker 02

  Yeah, that just shows the power of having high operating leverage and higher profit margins. And again, like you said, it's not a value play, but it's also not as ridiculously priced as I would have thought.

• 1:08:26Speaker 01

  I think the only thing we can do to convince people here is actually jump into the model. So I'll build two base cases this time, one with the numbers that Karp predicts until the end of 2027 and one with the analyst's estimates. So of course, it's a five-year model, so everything after 2027 is still obviously my own estimate, but still in the first base case, using Karp's growth rate of about 150% for next year, IDK growth after that personally to 60% and then 15 to 20 percentage points per year lower. So we would end up with, you know, 28% growth in 2031. You know, just summarizing that because it's a lot of numbers, the revenue K in this model would be in the high 50s for the next five years. For comparison though, the second base case built up on analyst's estimates works with

• 1:09:11Speaker 01

  a revenue K of only 22%. So I keep margins in both cases relatively stable, which is about 60% operating margin. You could say, well, if there's more competition, so let's say a topic gets into the market, Microsoft gets into the market, the margin will be pressured. It's possible, but at this point, it's completely speculative whether that would actually happen. So let's just assume a 60% margin here. If you then go with the analyst's estimates and you apply a 30x exit multiple and a 10% discount rate, the 20% margin of safety, which is what we usually do, you would get a fair value of $90. If you trust Karp's estimates and slightly higher growth afterward compared to analysts, and when I say slightly higher, we're talking a difference about 10 percentage points per year. So you could argue it's significantly higher.

• 1:09:56Speaker 01

  The fair value is at about $240 a year, so a significant difference and way more than the stock is currently trading at.

• 1:10:03Speaker 02

  Looking at your model, you also adjust for dilution with share count growth of 1% annually, even though we talked about stock-based comp being in the double-digit percentages. And so you don't have buybacks or dividends either in your model. So it seems like you don't think Palantir will be paying cash to shareholders or doing any kind of capital returns.

• 1:10:22Speaker 01

  Yes, I think you will not see any buybacks or dividends anytime soon. By the way, I've also done the model with more dilution, so 1.5%, 2.5% a year. Just because of the growth, it doesn't change your return massively, especially in COPS scenario. If you think just about the analyst estimates, there's more of a change. And again, you can just download the model. We'll have a link to it in the show notes. And also, we also have a newsletter about Palantir coming out today. We'll also link to that one in the show notes. And in that newsletter, we'll go through the model in more detail, and you can download it. And if you do that, you can adjust all of my assumptions in the way that you want. But yes, to your point, I think of a dilution of about 1% to 1.5% a year, which is in line with the last couple of years, I should say that, and no buybacks and no dividends.

• 1:11:08Speaker 02

  I gotta say, you got me excited here. I didn't expect to end up liking Palantir at all, but I'm just so intrigued. It's maybe one of the most interesting companies that we've come across from both a financial perspective and also a non-financial perspective, when you get into the philosophy and the people involved with the business and the types of projects that they worked on, right? We're talking about a company that helped track down Bin Laden. So it's just really unique. And yeah, after today's episode, I can definitely understand why you have felt some excitement too about looking at Palantir. So with all that said, I assume that you are going to recommend not investing in it, nonetheless.

• 1:11:49Speaker 01

  Well, I was closer to wanting to invest before earnings, I gotta say that. But even then, and this sort of goes back to the learnings that we discussed in our Biggest Loser episodes, Palantir's outcome is almost completely dependent on its top line growth. And one thing that I like to do when I look at companies that are growing rapidly, and everything looks great at the surface, is that I come up with a scenario in my head, which basically goes like this. So I would assume the next earnings report is coming out, and then growth is cut in half from what people are expecting. So let's assume your COPS outcome, 150% growth. Now suddenly, you only get 75%. Would I think I'm confident enough to be able to figure out and explain why the company grew significantly slower than people anticipated? I think, for

• 1:12:33Speaker 01

  example, if you would have looked at Lululemon's growth slowing down, it wouldn't have been that difficult to figure out why. In the end, they sell less clothing, and the reasons for that are quite straightforward. Either you have customers going to a competitor, or there's a recession and every brand has problems selling their clothing. If you think about payment companies, it's a bit less straightforward. It's also about competition or macro, but it's not as easy to track which one is actually the answer, because getting the data is a bit more difficult, but still you can do it. And then there's a company like The Trade Desk, where I have absolutely no clue why the top line growth is declining. And I also have no clue where the bottom would be. And I think if I'm being honest with myself, Palantir has to fall into that category. I think when the business is doing

• 1:13:17Speaker 01

  great and it's growing, everyone has this illusion of knowing and understanding why it's growing so great. But if that was suddenly changing, I'm just not sure how many would still claim to understand the business at all. I think the entire story here comes down to, does it make sense for me? And do you fully understand the technology behind it? And the honest answer is no. And just to make it clear, I feel like you don't need to understand everything. If you, for example, work in this sector and you have a much better understanding of how the industry generally works, I think that's not the case for you and I, Sean. I think the two of us would probably be better off sitting on the sidelines, just admiring what Palantir is doing and not investing today.

• 1:13:57Speaker 02

  Surprise, surprise. The value investors here, two guys who are very inspired by Buffett and Munger, didn't want to invest in Palantir. I don't think that's a shock. But also, that doesn't mean Palantir won't continue to do well. If anything, my gut feeling is the company probably will do very well, and probably the stock will too. I don't know about from current prices, but if there's a sell-off in it, that could be a really interesting entry point for folks who are willing to speculate more. And as you said, if you don't really understand what drives the business, then to me, you're not investing. To be an investor, that really means you have to think like an owner and feel like you understand

• 1:14:40Speaker 02

  all of the variables affecting the business. And so if you're just buying a stock and you don't have that owner mindset and that understanding of what is driving things, well then you're just speculating. And speculating can be profitable, or it can be quite costly. But our aim, first and foremost, is to be disciplined investors. And so, and if we're going to adhere to that, then yeah, we can't invest in something like Palantir, even if it's very exciting.

• 1:15:08Speaker 01

  At least not today. I think it's one of those where, for example, with The Trade Desk, I just felt like it's very difficult to understand, and maybe there's also not a reason to do so, because the stock is not, or the company is not doing that well. I think if Palantir keeps shoving along like this, and we have some people, especially in our massive, my community, that will most likely understand this business much better than the two of us do. So I could see myself with a couple of talks, understanding the business a bit better, seeing even more proof of what they do, even better understanding what the ontology means, and whether that is actually copyable for other competitors. I think all of that means I could have another watch list and at the right price, and especially the right understanding of me and the company, I could see us investing in it generally. I wouldn't say that it's a stock that I would never put money in. It's primarily about

• 1:15:52Speaker 01

  how well do I understand the stock right now? And also, where's the trading? And even at 130 bucks, which was the price before earnings, I think was not cheap, but it was kind of reasonable if you believe in the growth. I think today, a lot of that opportunity is basically gone. And if you see the stock below $100 and nothing changes, count me in, and I would be way more interested to, to kind of say it with Buffett, in terms of your philosophy of speculation and investing. Buffett said, quote, "The line separating investment and speculation, which is never bright and clear, becomes blurred further when most market participants have recently enjoyed triumphs. Nothing's today's rationality like large doses of effortless money." And I would say that you have a lot of people

• 1:16:35Speaker 01

  in Palantir who made a lot of money, and I'm glad for all of them, but many of them might not understand the stock to at least the extent that the two of us would want to understand the stock when we invest. Again, it's up 2,400% from the 2023 lows. So it's only natural that some people are okay with some blind spots in their thesis. And with that, I hope you guys enjoyed today's episode as much as we did. See you all in the next one.