TIVP060: Constellation Software (CSU): Historic Drawdown, Historic Buying Opportunity w/ Daniel Mahncke & Shawn O’Malley

17:01Daniel Mahncke:Clearly there's enough proof of concept for this business model that we shouldn't be questioning. But when you compare that to how Berkshire works, and also just this general idea of having one genius capital allocator on top, it is surprising to me that such a decentralized organization can make it work. And I mean, Mark Leonard would really, he would honestly barely know a lot of the people down the chain of command who are making capital allocation decisions effectively on his behalf. And so that requires a tremendous amount of trust in the organization you've built.

17:36Shawn O’Malley:Yeah, it shows that he's significantly smarter than I am. Because if you had asked me whether this business model could work before all of these highly successful programmatic acquirers started, I would have said no. It feels so counterintuitive to me that you have this genius capital allocator on top. But instead of letting him make the decisions, you basically push the decision-making process down further and further. And, you know, Mark Leonard just puts deep trust into the idea that bureaucracy kills the company and that leveraging human scale, for the lack of a better word, is the superior approach.

18:09Shawn O’Malley:And teams tend to work best when they are about five to 40 people. That's when there's high trust between the group members and, you know, people feel like they have an actual impact on the outcome of a situation. And that's what you need to not only attract, but also retain highly talented and motivated people. So whenever a business unit gets too big, you want to split it up to keep, you know, operating at the highest efficiency possible. And incentives are another incredibly important part of the story because CSI gives every single operation clear hurdle rates for the acquisitions. And what matters here are mostly the IRRs.

18:41Shawn O’Malley:So CSI doesn't care about how much one of the companies they want to acquire is growing organically when they acquire a company. So in 2015, in one of the shareholder letters that are, by the way, a phenomenal resource for everyone, and you guys should read it, Mark Lehner pointed out that when they graphed acquisition IRRs against post-acquisition organic growth, there was pretty little correlation. And the more obvious drivers for IRR, so the internal rate of returns basically, were things like purchase price multiples and post-acquisition EBITDA margins. So essentially, it was more important to acquire companies cheaply and have enough runway for margin expansion.

19:17Shawn O’Malley:And that's one of the things, or maybe the major thing that we've seen with TransTime and their success as well.

19:22Daniel Mahncke:And following that playbook, how does the ROIC look for CSI? How has it worked out for them?

19:29Shawn O’Malley:If we look back 20 or 30 years, returns on invested capital have been constantly in the 30s. And in the last 10 years, you've seen it coming a bit closer to the mid-20s, which, in my opinion, is still pretty respectable. And it's also interesting what Mark Leonard said about returns on invested capital on individual companies that they acquire. So obviously, it is one of the most important metrics that you look at and they aim for these 20 to 25 % opportunities. But he also mentioned how RIC can become misleading when at some point the purchase price is already earned back. So after that, you would get these astronomically high returns on investor capital that don't really give you the full picture.

20:06Shawn O’Malley:So maybe it makes sense to just go through a brief example of how that works. Imagine you buy a VMS business making$2 million in EBIT for, you know,$10 million purchase price. After a normal tax rate of, let's say, 25%, you would have notepad, so net operating profit after tax, of$1.5 million. So to get your one-year return on investor capital, you divide the notepad by the investor capital, which in this case is basically your purchasing price of$10 million, and that would give you a reasonable return on investor capital of 15%. But after seven years, you would have earned$10.5 million in total.

20:42Shawn O’Malley:So$500 ,000 more than the original purchase price. And since these businesses are asset-light, tangible invested capital can become a very tiny position in the years after. So if you calculate invested capital with only the cash that is invested in any given year after you earned back the purchasing price, well then returns on invested capital just look huge. So let's say your invested capital in year eight is$200 ,000. Then your ROIC would be$1.5 million dollars, the money that you earned in that year, divided by$200 ,000, and that would give you an RIC of 750%.

21:16Daniel Mahncke:I think we should emphasize this is a very, very high class problem to have, right? In theory, with businesses that demand more capital, right? So what does that mean? They demand more reinvestment to generate those incremental growth and profits over time. When you have these really asset-like software businesses, it doesn't take a lot of additional You're not building new railroads and factories from scratch. You're just paying software engineers. And so eventually you can have these ballooning operating profits relative to a very small base of capital that was initially invested. And so that's kind of what we're talking about.

21:50Daniel Mahncke:And on the one hand, like I said, very much a high class problem. Our profits are growing so quickly and we don't have to reinvest a lot. And now our returns on capital metrics are getting distorted to the upside. That's a great problem to have. But in terms of aligning incentives based on ROIC, that's where it gets tricky. And so I was just saying, I mean, none of this is rocket science, but I do think it can take a little bit of devoted effort to wrap your head around ROIC math. I mean, if that sounded like a lot of gibberish that Daniel was saying, I don't blame you. Even for me, it's a little hard to follow when you're just listening to people talk about it.

22:21Daniel Mahncke:I will say, though, John Huber, who's one of our favorite investors, a study of Sabre Capital Management, he did a really great job in a series of blog posts back in 2016. explaining basically how to think about ROIC with some really tangible examples. So I would encourage anyone wanting to go deeper into ROIC and ROIIC returns on incremental invested capital to go look those posts up. And I think Daniel will have those included in the show notes.

22:48Shawn O’Malley:Sean and I cover a new company on this show every single week. And the research alone can easily take 40 hours. When you work at that pace, the hardest part isn't reading filings or building models. it's finding new ideas that are actually worth going deep on. If you work in investing, you might have run into this problem before too. That's exactly why Sean and I built the Intrinsic Value Community, which has honestly become a favorite place to discover new investing ideas. A meaningful share of our members are professional investors, people running hedge funds, family offices, or working as portfolio managers.

23:25Shawn O’Malley:But we also have industry experts, CEOs, software engineers and even actual rocket scientists who are as passionate and well-read about equity investing as anyone. Each week we host at least one live call. Stock pitches, portfolio analysis, investment strategy, everything that helps members find new ideas and become better investors. We also host expert guest speakers, people who founded and run their own investing firms, forums, market-beating portfolio managers, prominent CEOs and even famous authors. Of course, we have a library with recordings of every single call we have ever hosted, so you never miss anything if you cannot attend live.

24:07Shawn O’Malley:One of the things out of most is the chat forums for sharing write-ups, vetting ideas and reflecting on changes in markets. And at least twice a year, we bring the community from the virtual world into the real one, including private dinners in Omaha for Berkshire weekend and meetups in New York City to connect, explore, and most importantly, talk stocks. Our last cohort brought together 20 incredibly thoughtful people, some of the sharpest investors that Sean and I have ever met. And if you want the chance to learn alongside people like that, you should join our waitlist at theinvestorspodcast.com slash intrinsic value community.

24:48Shawn O’Malley:That's theinvestorspodcast.com slash intrinsic value community.

24:55Daniel Mahncke:Just like everybody else, there was a time when I was a beginner investor, and I have to say, investing is particularly filled with jargon that can just make it so difficult for new investors to understand what is going on. But it's never too late to get smarter about stock investing from the ground up. At The Investors Podcast Network, we've made a habit of studying the world's best investors And now I'm just selling those learnings into a simple course for you or for anyone in your life who you might want to share the gift of knowledge with. With my How to Get Started with Stocks course, you can master the principles of excellent lifelong investing with valuable insights for both beginners and pros.

25:28Daniel Mahncke:The course covers 10 different sections, beginning with the basics of what a stock actually is and how the stock markets work, to strategies to optimize your retirement savings, how to pick great companies for the long term, what to look for in ETFs, and how to monitor your investments. well, so much more. To begin getting smarter about investing, just visit theinvestorspodcast.com slash get started with stocks. That's theinvestorspodcast.com slash get started with stocks. And for a limited time, you can use code STOCKS15 for a 15 % discount at checkout.

26:03Shawn O’Malley:Yeah, I'll definitely put a link to them. And it's definitely a recommendation that we can both give you. And getting back to the problem that all of this causes is basically around the incentive structure because you have to pay out bonuses on these outrageously high numbers of return on invested capital. And I gotta say I found it incredibly insightful to read about how Leonard thought about fixing this problem. He said that since ROIC is also one of the big drivers of our incentive compensation program, we care about this increasingly high ROIC issue. So when ROIC is very high, bonuses start to consume a disproportionate and inappropriate amount of pre-bonus net income.

26:37Shawn O’Malley:We've actually run into this situation a couple of times, and you can either change the plan, cap the bonuses, or ask the managers to keep their profits and redeploy them in acquisitions or initiatives. We dislike changing bonus plans because it literally takes years for trust to rebuild to the point where managers are willing to trade off short-term profitability and bonuses for higher long-term profitability. We saw this in spades when our major investors put CSI up for sale in 2011. He also said RIC increased sharply, acquisitions slowed dramatically, and initiative spending dropped, faced with the prospect of new owners intent on changing the bonus program and borrowing mountains of debt to acquire the business.

27:17Shawn O’Malley:Our managers reacted, as you would expect, maximizing short-term profitability and bonuses at the cost of longer-term growth and profitability. Then he also said that, you know, capping bonuses isn't great because the last thing you want to do is having an incentive system that pushes employees to basically look for these gray areas to maximize their bonuses and not to earn less than they did the year before. And I find that a very fascinating insight because you always talk about working with people who have high integrity, a lot of integrity when you work with them, and that's definitely the way to go.

27:48Shawn O’Malley:However, Leonard also points out that even if you work with people that have high integrity, they can be pushed into the wrong direction if the incentives that you said are not aligned.

27:58Daniel Mahncke:I think you touched on something really interesting there with the quote from Leonard mentioning the sale in 2011. And I actually wasn't really all that familiar with that. So maybe you can elaborate for me and for the audience about what happened in that episode.

28:11Shawn O’Malley:The company wasn't really put up for sale, but CSI had some early investors, primarily private equity investors, that wanted to cash out. So therefore, the board initiated what is called a strategic review of what to do next. And often that means you put the company up for sale. Mark Leonard, that's important to say, never wanted to sell the company. So again, his main problem with venture capital was that he could never build something for the long run. And fortunately, the company, meaning Constellation, did not end up being for sale. But I think just that experience of potentially losing your company fundamentally changed how Leonard viewed the company's stock price.

28:45Shawn O’Malley:So if it were dropped too low, you might have people come in who force the sale of the company and you lose, you know, your baby basically. but if it's too high you might have employees or parts of the management team selling the stock or want to cash out and you also to some extent attract maybe the wrong investor base so people who start investing simply because the stock has gone up and perhaps and i'm just speculating here that's why the stock is going down and basically has this historic drawdown right now in the last couple of months because i've heard very few investors long-term investors that have sold that position but i've had many people who only had the stock for a short period of time who sold out.

29:21Shawn O’Malley:And to me, it's just incredible. The narrative shift on CSI seems kind of insane. A couple of months ago, I still heard that you would need to pay a premium for this stock because it's so great. And then the same people now are the ones that sell their shares and wouldn't even want to buy it back at half the price.

29:37Daniel Mahncke:So what you're describing here with a company where the stock price falls so low that the founder can lose their baby, as you said, that is probably every founder's worst nightmare is having their company ripped away from them by external investors who come in and take over the share count. And this is why we see this a lot in the companies we cover. There's these dual share class structures where the founders of the company or the management team get basically special voting rights. They get more clout with each share that they have. So instead of one vote per share. And that comes with its own costs because you're concentrating power in a handful of people as opposed to really a more democratized approach.

30:25Daniel Mahncke:But also it hedges against this outside takeover risk. And then on the other side of it, I remember learning from Buffett about how he talks about you get the shareholder base that you deserve. And the way you communicate and educate your shareholder base, the expectations you set, that all attracts certain types of investors. And to some extent, if your stock does well enough, you can't prevent momentum chasers from piling in, but you can kind of set the culture. And obviously, Berkshire has done an incredible, incredible job creating this culture around the people who own the shares. But just kind of going back here with Mark Leonard, it sounds like he took what would be the third route, meaning that he let the capital stay in the operating subsidiaries for them to keep reinvesting into their own businesses.

31:12Daniel Mahncke:And so instead of changing incentives, CSI sometimes tells its operating companies, you can't upstream all that cash to headquarters. You have to reinvest it in acquisitions or long-term initiatives around your own business. And so over time, this reduces the risk of having distorted ROIC numbers, as we were talking about, for bonus compensation, because you have more acquisitions being made with reasonable ROICs rather than the artificially inflated figures from the old acquisitions kind of continuing to impact things that have already been paid back in the purchasing price. And so, I mean, this stuff all sounds wonky, but I can't emphasize how important these little wrinkles are to the business of being a programmatic acquirer.

31:57Daniel Mahncke:Making a business out of acquiring other businesses is very much more art than science, even if you have a programmatic playbook. And the details of it are literally everything. You need to have a perfectly balanced, well-oiled corporate machine where profits are sent upstream to the allocators at the top, like Buffett and Leonard, to make decisions with at optimal times and then retained in the subsidiaries when you have sufficiently attractive opportunities remaining on the horizon within the scope of those operating businesses. And so incentives to managers are how you shape and encourage the appropriate allocations of capital to be able to occur with a high degree of trust rather than needing to have the allocators at the top bogged down by keeping up with their previous acquisitions because that would, well, and that would inhibit their ability to make more acquisitions going forward, which is what drives these types of businesses and their growth.

32:51Daniel Mahncke:So if Leonard or Buffett can't trust that subsidiary businesses are making the correct capital allocation decisions, then they have to focus on auditing every decision that's made instead of prioritizing their time toward the most important decisions at the top of the entire company. And when you run that across a thousand different businesses, you're spending all your time auditing previous decisions and not making new decisions that grow the business. So again, that means fewer acquisitions and perhaps a lot of accruing missed opportunities, which would mean diminishing returns over time for shareholders.

33:27Daniel Mahncke:So it is a matter of necessity for programmatic or serial acquirers to have this downstream trust and be able to know that people are incentivized with the right KPI structure. I mean, that's just the best way to do that. And there's another growth metric that's constantly referred to in the letters and filings at Constellation. And that's ROIC, returns on invested capital, plus organic net revenue growth. So why don't you tell us about that?

33:56Shawn O’Malley:Yeah, that's what's called the combined ratio. And I apologize in advance if the next minute or two again get a bit technical, but I'll try to go slow. So when I first saw this metric, and that's kind of why it's interesting, I didn't fully understand what it was supposed to tell me. If you have a traditional capital intensive business, you cannot simply add organic growth to the RIC because growing revenues usually requires incremental capital, such as buying more factories or buying more machinery or even getting new employees, which then obviously increases the denominator in the RIC. formula.

34:27Shawn O’Malley:And therefore, the cost of that growth is already reflected in the capital requirements. But it is different for CSI because, again, it operates these software businesses that are very capital-light. So because the growth does not require a corresponding increase in the invested capital denominator, the organic net revenue growth is kind of viewed as a distinct additive value driver alongside the return on the existing capital base. And just in case that was too much for a podcast format here. Everything you need to know and understand is that the metric is not double counting the organic net revenue growth, even though it might seem like that at first glance.

35:03Daniel Mahncke:And I think at this point, Constellation focuses more on free cash flow and adjusted net income numbers anyway, right?

35:11Shawn O’Malley:Yeah, when CSI got a bit more competition and also just grew in size, they started paying higher multiples and used some debt for acquisitions as well. And under those new circumstances, the combined ratio just wasn't the best proxy anymore.

35:26Daniel Mahncke:Talking about the M &A process, I want to zoom in more on what companies they acquire exactly. I mean, are we talking about a similar profile to Chapters Group, which is a small portfolio holding of ours, but one of the companies we're more excited to own because of its potential? There's sort of like a smaller German version of Constellation in a way. So what distinguishes these two businesses and how do they differ?

35:53Shawn O’Malley:Constellation is obviously more diversified. It just has more variety in the companies than it owns. And we were talking about over a thousand companies. So each platform has a sort of specialty that they focus on. And the best known platform is likely to be Topicus, which is, like I said before, the Europe-focused part of Constellation. So they are more focused on geography. Industry-wise, they cover everything from government software to education to legal and financial services and all those sorts of things. And our colleague, Clay Fink, has actually talked about his bullish take on Topicus, and in particular, a few times when We Study Billionaires.

36:28Shawn O’Malley:And you also have Lumen, which is another one of the spinoffs of Constellation. And it's focused primarily on the media and the telecom vertical. So long story short, all of these platforms are somewhat specialized in what they do, either by geography or by industry. But of course, you also have a huge overlap between them. I mean, if you own a thousand companies and all of them are in the VMS space and in these small niches, I think that's kind of unavoidable.

36:55Daniel Mahncke:When you hear that they already own 1 ,000 plus companies, you probably get the idea that they're running out of acquisition targets, right? But the estimated number of VMS companies actually left for them to acquire exceeds six figures. So in theory, there's a very long runway left. And currently, CSI is acquiring about 100 companies a year for context.

37:20Shawn O’Malley:Yeah, don't expect them to run into major problems anytime soon. Although there has been an instance a couple of years ago that I would say spooked investors a little. In 2019, Mark Leonard suddenly paid a$20 dividend, bringing the dividend yield of the company to almost 3%. And I say spooked because as an investor in a company that is delivering mid-20 % returns on investor capital, you don't want them to pay a dividend to you. I mean, where else are you supposed to make such returns? So you want CSI to keep the money and then invest it in new VMS companies on your behalf. And if they don't find anything with returns of, let's say, 20 plus percent, they should just go for slightly lower returns, say 15%.

38:00Shawn O’Malley:And I don't know, that's still a great return and it's hard to find in these markets.

38:03Daniel Mahncke:We can attest to that, right? I mean, we try to find these guys opportunities every week here on the show that people have followed along with. And there just aren't too many companies out there who can deliver those kinds of returns that aren't already priced at a level that more than accounts for that potential.

38:23Shawn O’Malley:And especially at the risk profile that you would get at Constellation being a highly diversified company with a phenomenal track record. And I would say the problem with accepting the lower hurdle rate is that it reduces the overall returns of the portfolio significantly. You might ask why, but the reason is that if you need a 20 % return to invest in a company, then you pay close attention to the price. Again, the acquisition price was identified as one of the major factors for achieving that high of a return, which won't come to you as a surprise because we all here are value investors and we pay close attention to what we pay for an asset.

38:58Shawn O’Malley:Now, the problem is, and we're kind of going back to the topic of incentives here, that if you know you are allowed to make a deal at a 15 % hurdle rate, you will get into the habit of paying higher prices. So even deals that you might have gotten at a 20 % return just through, you know, some negotiating, you won't get now because you would rather take the safe route and sign the deal at a 15 % return instead of losing the deal entirely while negotiating for the 20%. And there's even a name for that effect and it's called the magnetism effect. So to solve this, Mark Leonard ended up settling for kind of a middle ground where they keep their 20 plus percent hurdle rates for smaller acquisitions, let's say$5 million,$10 million, but they will lower it on the big ones.

39:39Shawn O’Malley:And another thing they did, which is on the margin, is that they also started to use debt to fund the acquisitions, which obviously helps because the cost of debt is significantly lower than the cost of equity, especially for a stock-like constellation.

39:51Daniel Mahncke:It's a bit of a niche financial topic, but just to your last point there, I mean, generally, it can make sense to finance acquisitions with debt rather than raising cash through selling new shares of stock, right? That's what equity financing is. Because in theory, at least, equity investors demand a higher rate of return than lenders. That's what you're taught in a finance degree program in college. So if you're choosing between$100 million in debt or in equity, you would typically choose debt. And it does get a little weird because if a business is trading at a serious premium to its intrinsic value, though, then it can actually be more beneficial to sell stock for cash rather than relying on debt.

40:31Daniel Mahncke:But that's another rabbit hole to go down for maybe another day. We'll get back to the M &A topic and the moat around Constellation's VMS business here, especially with so many threats looming from AI offering the ability to say, program an app automatically for a business that does what some narrow software tool does and then cut out the middleman. Concern is that as people like to say on X or FinTwit, whatever you want to call it, companies will vibe code their way out of a number of formerly essential subscriptions that they paid for bringing those tools or capabilities in-house thanks to the programming abilities offered to them by llms like clod so basically just building the tools they would otherwise pay for themselves with the help of ai and you know before our call here i also just looked through some of Constellation's numbers again.

41:24Daniel Mahncke:And I came across their share count. It really does stand out. I can't not comment on it. We've looked at 60 companies now on this show. And if I had to bet, Constellation is the only one that has had absolutely zero dilution since IPO. Not just the last five years, not the last decade, since IPO. Not a single share. They IPO'd with 21.2 million shares. And guess what? They still have exactly 21.2 million shares. And that's after going from owning 45 companies, as we've said a few times now, over a thousand. So they acquired 1000 plus companies without issuing a single share to fund that growth.

42:08Daniel Mahncke:And it's just outstanding, assuming that debt financing actually did always come at the lowest cost. But still, it's really uncommon to see this kind of thing. And ultimately, it's a good signal of a corporate culture that is very mindful of its shareholders and what's in their best interest.

42:24Shawn O’Malley:It's incredible. And as you said, it's kind of more a signal than necessarily being the best decision. Because if you have your stock trading at 50x multiples, there's also an argument to make that you can issue some shares and it might even be not the most illogical idea to finance some of the acquisitions that you do. But for those who listened to our episode on chapters, you might have heard that chapter CEO Jan Hendrik Mohr said that dilution is kind of part of the process at the beginning of the life cycle you will have some share dilutions which we definitely have seen in that stock and then later you will buy back these shares again and I agree with him on that position personally but you know you might look at Constellation then you tell me Daniel clearly Constellation was able to do it without any dilution at all and that's true but I would argue that there's probably a big difference between starting with no companies at all and just some little starter capital, basically, as it has been the case for chapters, and starting with still 45 companies when going public, as it has been the case with Constellation.

43:21Shawn O’Malley:And nevertheless, reading about how Mark Lennart just thinks about incentives and capital allocation is nothing short of impressive. 10 years ago, in 2015, he even told the company's board that he didn't want to get paid anymore. No salary, no bonus, nothing. The only way he could make money was by the appreciation of the stock price. And he went even further. He also said that he only then started flying business class because he would pay for it out of his own pocket without billing his travels to the company. Till then, he had flown economy because technically, Constellation was paying for those flights and he didn't want shareholders to pay for his business class seats.

43:58Shawn O’Malley:I don't know. I just think that says so much about him. Of course, it's an anecdote, but I don't want to know how many CEOs would think, gosh, I make shareholders 20 % a year. If they had those shares long enough, I made them millionaires, literally. the least they can do is pay for my business class seats.

44:14Daniel Mahncke:It's more symbolic than anything, of course, because I'm sure his travel expenses ultimately would not move the needle meaningfully. But still, it's a very powerful message across every layer of the company. Immediately, lower level managers think, hey, how can I justify XYZ expense if our CEO doesn't even charge the company for flying business class? And I would bet that has been a very effective, but also very subtle way to push managers in the direction of making the best decisions for shareholders' interests, ultimately, just by being very mindful about spending. And so you will rarely find CEOs as aligned with shareholders as Leonard.

44:57Daniel Mahncke:But that's also why his early retirement in September of last year came as somewhat of a shock to the shareholder base and has penalized the stock in some ways. I think you hinted at this in the beginning, but I understand why you didn't want to frame it as one of the reasons why the stock is tanking. Because in the end, the reasons had really nothing to do with the company, right? Mark Leonard needed to step down due to health reasons. That doesn't fundamentally change the health or quality of the underlying businesses.

45:28Shawn O’Malley:I mean, first and foremost, we just hope that he will come out on top of this and as healthy as possible. I guess it didn't help the situation that he stepped down only three days after Constellation's call about, you know, the company's AI situation. And I actually remember that when Constellation announced Mark Leonard would step down, that was right about the time when we were at the TIP summit in Montana. And I know I was on the hike talking to Clay and, you know, asking him about his opinion on Mark Leonard stepping down and whether that would have any effect on his view on Constellation.

46:00Shawn O’Malley:As you know, he's invested in the company for quite a while now. And back then, Clay wasn't concerned at all. and I'm pretty confident he's still not concerned yet either. Obviously, you don't like the CEO of 30 years to step down, but we should forget that this is one of the most decentralized organizations I've ever seen at that size. So you can compare it to Berkshire, where you can certainly make the argument that both Buffett and Munger, two exceptional allocators, maybe the best we've seen in our lifetime, not only built the company, but they also caught all the important shots when they were still able to.

46:31Shawn O’Malley:And that has not been the case with Constellation and Mark Leonard.

46:35Daniel Mahncke:Mark Leonard is a fantastic capital allocator, but CSI was built to minimize key man risk as much as possible. So the new CEO is Mark Miller. He was the CEO at the time, and he has been with the company for more than three decades at this point, too. He founded a company that became Constellation's first acquisition in 1995, which is a really cool little tidbit to know. And so it can't get much better than that. It would probably be too much to call him a co-founder, but it's as close to being one as it can probably get.

47:09Shawn O’Malley:My funny side note is that talking about Miller, I couldn't find a single picture of the two of them together. Mark Millard and Mark Lennard. And they've worked together for 30 years and there's just not a single picture you can find online. It kind of reminds me of these male friendships. So two guys that are very good friends, but they simply don't make any pictures together. Well, anyway, as soon as he took on the CEO role, he bought another$5 million in stock, which I think is, you know, a sign of him having skin in the game. Now, we have to keep perspective here as he already owned$750 million worth of shares.

47:43Shawn O’Malley:So$5 million is, you know, a small part and barely moving the needle. Still, I think it's a pretty good sign. Generally, insider ownership is pretty high for Constellation. Not the highest I've ever seen in one of the companies that we covered, but still. I think Mark Leonard owns close to 2 % of the company. Mark Miller owns about 1%. And then you have a couple of directors also owning hundreds of millions of dollars in shares. And there's also this kind of special rule that executives and key employees are required to use about 75 % of their after-tax annual bonuses, which are usually paid in cash, to buy CSI shares on the open market.

48:17Shawn O’Malley:And once they are bought, they must be held for at least four years.

48:20Daniel Mahncke:I probably prefer something closer to the Berkshire model, where bonuses are paid in cash to minimize solution. like with Constellation, but then managers are encouraged and not required to buy shares in the business, or at least that's my understanding. It's a great way for Buffett, or now Abel at the top, Greg Abel, to see into the quality of the decisions being made downstream in a way when you set up this structure. Assuming Berkshire remains an excellent investment going forward, and the people managing the different subsidiaries are well positioned to recognize that, then they should naturally want to allocate a significant portion of their bonuses to buying shares on their own without any prompting, right?

48:59Daniel Mahncke:That kind of would be how you measure the quality of the decision making that they're doing. And so it looks a lot better when you have managers wanting to organically buy shares on their own, which is not to say that the folks at CSI aren't inclined to do that, but it is a little different when you're mandated to do so. But still, I'm happy to see the people running the business have significant skin in the game, acquired with their own cash. That is what matters most, much more than having shares of stock gifted to you. And gosh, these RSU packages that I always find a way to bring up, and I didn't leave it out this episode, where you just get handed shares of stock simply based on the amount of time you spent with a company, not even based on your performance, how well you did, how much value you created, nothing.

49:45Daniel Mahncke:So we always do not enjoy seeing those. But with that kind of tangent aside, How about we talk about the elephant in the room, and that is AI. I feel like since we started this podcast over a year ago at this point, at least every other company we've evaluated has required us to talk about AI in depth at some point. And this time, there is just no way around it because it has been such a major driver of the stocks underperformance in the last six months or so. And so to many, Constellation feels like probably the epitome of a company that should be disrupted by AI. So how do you think about that?

50:29Shawn O’Malley:It's safe to say we started this podcast in the age of AI, value investing in the age of AI. That's another name we could have used. And I mean, I can't help but think that we're either completely right about our AI thesis or monumentally wrong. But jokes aside, I think markets are just a bit nervous. I think they want management to just clearly communicate the impact of AI. And, you know, if management is honest, they will likely tell you that they don't know. And that's what Mark Leonard more or less did in Constellation's AI call. And first of all, the fact that they hosted a call in the first place already kind of shows you that there is a lot of at least outside pressure on the company.

51:07Shawn O’Malley:I don't think Mark Leonard would have done that otherwise, especially knowing that only three days later, he would tell the word that he has stepped down as CEO. I think there would have been, you know, different opportunities to spend his time. And this might indicate that they did see the need to talk about it. And that said, Mark Leonard, he basically opened the call with a short anecdote. He talked about how Jeffrey Hinton, who was a Nobel Prize winner for his research on AI, predicted in 2016 that AI would quickly replace radiologists and advise people to stop training for that profession.

51:41Shawn O’Malley:Now it's a decade later, and the number of radiologists in the US has actually gone from 26 ,000 to 30 ,000. And you know, that's a larger increase than population growth, which basically means that the number of radiologists per capita has actually increased. So he wasn't wrong about his AI thesis and basically saying AI will come into the field of radiology. But instead of relying on AI only, it just improved the radiologist's work and therefore they had a higher success rate, which in turn made the field even more important and you have more people training in that profession.

52:13Daniel Mahncke:It reminds me a lot of what NVIDIA's CEO, Jinseng Huang, keeps saying about software programmers. AI might take over the job of writing code, but programmers are fundamentally problem solvers rather than just people who literally write code. And so you'll always need people who can solve problems. I don't think that's going away, especially when more and more AI use also means more code produced, which again means more problems that need to be worked on being solved or at least audited. And so I guess part of the problem for Constellation is that the software companies they own are, to maybe put it nicely, not very complex.

52:51Daniel Mahncke:So let me ask the question of all questions here. can the majority of their software solutions just be vibe-coded as people like to say

53:04Shawn O’Malley:well this is the part where i'll probably use the term mission critical a dozen times again a few at least but before we get to that let's maybe quickly just break down how constellations revenue streams actually look like so to get a bit of a band understanding how precisely they even make money so they're basically for revenue buckets you're talking licensing professional services, hardware, and maintenance. And maintenance is by far the biggest part. It's about 75 % of revenue. And when we talk about chapters, one of the companies we mentioned was, you know, a software for public transport and the system that basically keeps the buses here in Hamburg running and, you know, being on time.

53:42Shawn O’Malley:And Constellation owns these businesses as well. And as you can imagine, those type of software businesses basically need 24-7 uptime all year around. And so the city is basically paying an ongoing fee for support, fixing bugs, regulatory updates, security patches, and maybe even some new versions that are coming out. So it's practically recurring revenue in nature. And either you pay or in this case, basically your city stands still. And then you have professional services, which are the second biggest part with a little less than 20 % of sales, so significantly smaller than maintenance. And you can imagine this to basically be the implementation of it.

54:19Shawn O’Malley:So let's say we use a different example this time. Imagine you have a school district, which buys a student information system and then it needs to be configured. So we're talking data migration from the old system to the new one. We are talking training staff, custom templates, all the sort of stuff that you basically need before you can not only deliver the system, but actually run it on a daily basis. And then you have licensing and hardware, which are both very small parts of the business.

54:44Daniel Mahncke:And licensing could, for example, mean that a new customer buys software for the first time and therefore has to pay a one-time fee to use it. Is that the way you're thinking about it?

54:55Shawn O’Malley:Yeah, that's kind of how it works. I mean, you could also imagine that there's a long-time customer who wants to add a new module to an existing product and then you just have to pay a one-time fee to unlock it. And just to get back to the question of AI, as you can imagine, when maintenance is almost 80 % revenue, Constellation is quite sticky in both the businesses that they own and the software that they are basically selling to customers. And we shouldn't forget either that many of its customers operate in fields where it's simply unnecessary to constantly be on the leading edge of tech. I mean, I've seen pictures of some Constellation software products that make you think we are still living really in the 90s.

55:35Shawn O’Malley:But honestly, that doesn't matter because, you know, when that's a system that's working for employees, then upgrading is not an end in itself. But I still wanted to dig a bit deeper into the AI question here because I feel like that's the main narrative. But we're talking about a thousand companies and you obviously cannot look into all of those different companies. When I talked about chapters, I could make an argument based on the exact companies that chapters owns. And I would love to do the same here, but we just don't have the time to go through all of the thousands of companies that CSI owns.

56:05Shawn O’Malley:And we also don't have the information on them. And that basically leads us to being forced to take more of a high-level view of what is happening. And I think the first thing that I want to mention goes to your NVIDIA point, which is that code might get commoditized, but you would still need problem solvers. And in CSI's case, your mod isn't necessarily that you have the best programmers or the best software. It's that you have the most insight into customer workflows and customer integrations. And I think that is what creates these huge switching costs, or at least has over the last couple of decades for CSI.

56:36Shawn O’Malley:And it's not about whether there could be better code. I mean, again, looking at some of these websites, better code, I'm pretty sure, was already available in 2010.

56:45Daniel Mahncke:I think the impact of career risk shouldn't be underestimated here either. It's an argument you made when you pitched TransTime. And imagine you're responsible for all the IT in a local hospital. I mean, do you want to make the call to switch from software that has been working for 20 years to new AI startup? I mean, would you put yourself in a situation where you can really only lose. Everybody expects that it works out because it has to. So you're not going to get any applause for facilitating this change. But if it doesn't work out, then people's lives could be at risk. And you're sure to lose your job.

57:20Daniel Mahncke:If you're the guy who pushed moving to some AI tool that is a critical function for something like a hospital or even for city buses, I mean, that's just, I wouldn't want to be in that position.

57:31Shawn O’Malley:Yeah, I don't want to be in that position either. I think we're on the same page here, but I also kind of feel like I want to play some devil's advocate because the last thing I want to do is miss something because I just like the narrative too much and avoid looking further. And when I think about software for hospitals, for example, that's not a niche market. That's actually quite a big market, at least big enough to argue that it's a vertical that was maybe previously too small to attract funded startups. But now with AI reducing the cost to build software, you might suddenly get new entrants.

58:02Shawn O’Malley:And another thing could be that, you know, customer expectations shift now that everybody is talking about AI. So it's kind of like a self-fulfilling prophecy. And everybody's talking about how much cheaper software could be. And, you know, even incumbents that aren't displaced, immediately at least, your customers start expecting more improvements or worst case, they just want to pay a lower prices. and if you can't deliver on that they actually do look for alternatives at some point and on that note for many companies Constellation software solution is a very small cost factor you're talking about only 1 % of revenue most of the time that's actually what Mark Leonard said in one of his letters so I would personally argue that's just not worth the risk of switching because you're at this intersection basically where the software is too critical to cut but it's just too cheap to aggressively go about and replacing it.

58:51Shawn O’Malley:I mean, there's always opportunity cause of, you know, hiring someone basically to just look for new software to use. And I think the main bare argument is that, you know, CSI could get hurt by seed compression, which is similar to what we heard about Adobe 2. And I think it does make sense when, you know, someone argues that AI agents could come in and, you know, do the job of two or three employees. And that would basically mean that there are fewer seeds that then Constellation can sell its software to.

59:18Daniel Mahncke:And just to clarify for listeners, sometimes software bundles get paid for per seat, right, on a per seat basis. And a common example of that would be something like maybe a student information system used by a school district or some sort of municipal permitting system used by a city. And in those cases, the vendor would probably charge based on the number of named users who sign up to have access. So maybe there's 50 seats for administrators and staff who enter data and run reports or handle cases, or maybe for, you know, a hundred thousand students in a school district, that would be kind of how to think about it.

59:56Shawn O’Malley:And this is basically the potential risk that I have entertained for some time. It's not so much about AI just replacing the underlying software, but actually just reducing the seed count. So let's say five of those 50 seeds would be replaced by an agent. Well, that's an immediate 10 % decrease in revenue for CSI. So I think I considered this a threat. And then I did some digging because obviously I need to figure out, you know, how much of that is actually reality. And I think I've overcome that fear a bit, at least. If you think about it, this is not a risk for, you know, the small operations or the smallest clients of CSI.

1:00:28Shawn O’Malley:If you run a four-person operation where you basically have one founder, you have, let's say, two engineers and then one support person, you won't cut any seats using AI agents. And probably the same is true for most companies, even 10 times the size. but where seed reduction could be a bigger problem are basically the the big public sector organizations but but you know then again i look at the numbers and i realize well they grow they just keep growing they don't usually cut jobs this goes back to incentives i guess where if you work for the government in a managerial position would you rather have five people that report to you or would you rather have 10 people and perhaps even more important if you already have 10 people reporting to you do you cut that to five because you think well there could be an iagent doing much of that work.

1:01:12Shawn O’Malley:I don't think that's how it is. I mean, I don't know how it is in the US, but I don't see public offices actually reducing headcounts in Germany, which is, you know, in part why we own chapters.

1:01:22Daniel Mahncke:The phenomenon you're referring to, it sounds so simple and it's something that Buffett has talked about many times. And it's that nobody wants to rule over a smaller kingdom. When you zoom in on the specific person actually making these decisions like to expand the company or not, even if it's likely destructive for shareholders to do so, things like that person's ego and their desire to feel important can be real factors in what actually happens. And for people in those roles in particular, having responsibilities taken away from them, right? Overseeing fewer people, even if it's what's best for intrinsic value is not particularly comforting to their own pride.

1:02:03Daniel Mahncke:And okay, so I I would say those are the bare arguments about AI that might hurt Constellation software by either increasing churn or reducing their pricing power and their opportunity to upsell products if there's just more alternatives out there. Because AI products or AI agents could take over some of that functions. But what about the bull case here for AI, for Constellation? I mean, is there even a bull case? I mean, if we assume that the bear case is probably less likely to occur than maybe the market currently expects, then I would guess there are ways for Constellation to benefit from actually implementing AI themselves, at least internally or in the companies they acquire.

1:02:46Shawn O’Malley:I think it's not as obvious as some other companies, Adobe, for example, where you can clearly say, well, if there's more content that could be edited, for example, Adobe is definitely benefiting for that. If we assume that Adobe is still the tool to use for most of the creative professionals working in, let's say, Hollywood or bigger corporations. For Constellation, I think I would mainly be about margin expansion. So if you can reduce, you know, the most people-heavy costs like support and professional services, that would be a huge money saver. I mean, right now, by far the biggest or the highest cost is staff.

1:03:20Shawn O’Malley:Almost 70 % of CSI's expenses are staff. And if you now imagine that, you know, documentation tasks or maybe the implementation of new models can get automated, that could immediately save you a ton of money. You can also imagine that maybe CSI can use AI to create new upsell modules at, you know, a much lower cost than they did before. And And I don't know, perhaps on the acquisition side, you could just cover and keep track of many, many more companies than before. When we talk about 100 acquisitions a year and a VMS TAM, so total addressable market of, let's say, just 100 ,000 companies, you can imagine how much work it takes to keep track of as many of those companies as possible.

1:04:02Shawn O’Malley:And as far as I understand, they really have those lists of all of these companies and they're constantly reaching out to them, mostly once a year. and those are tasks that you want to keep the personal touch of course with the CEO of the company that you potentially buy but there's probably a lot of things that you can automate and that just saves not so much money but just a lot of time.

1:04:22Daniel Mahncke:I think the entire AI story around cost solution is happening on the margins. I mean yes you might lose some customers yes you might have less pricing power with certain customers on the margins but the trade-off is that you can use AI to become more efficient save costs, and potentially even increase acquisition speed. And so we don't know how it'll play out. But to me, it does sound more likely that AI could actually, in a way, help Constellation become an even better company and not necessarily a worse one. Or at least that would be the bull argument. So in the call that they did on AI, Mark Leonard put it this way.

1:04:59Daniel Mahncke:Here's a quote. I believe that vertical market software is the distillation of a conversation between the vendor and the customer that has gone on frequently for a couple of decades. And you distill those practices down into the algorithms, software, data, and reports, and it captures so much about the business. And being able to examine that in a new way because of AI creates new opportunity to modify, change, and suggest new approaches. So yes, I'm hopeful that that unique and proprietary information will be of value.

1:05:30Shawn O’Malley:If you invest in Constellation, you're obviously you're also just, you know, taking a bet on the management team. Of course, I would have liked to have Mark Leonard as, you know, the CEO still, but I have no doubt that Mark Miller will do an equally good job. And, you know, when you have people as smart and experienced on top as they are in Constellation's case, there's also a certain level of trust that you as a shareholder would have toward them handling and assessing the situation right away. And yet again, he just mentioned in the quote that he said, I'm hopeful that basically the company will be better.

1:06:01Shawn O’Malley:And of course, if you're a shareholder, you would like to hear Mark Leonard just say, we will come out of this on top of all the AI narrative and be a better company. And not hopefully, but no, you can't get everything that you want.

1:06:12Daniel Mahncke:On that point, Mark Leonard has said repeatedly in recent years that CSI could also invest in businesses beyond vertical market software. And this was being said even before the AI narrative script markets too. So I mean, what do you think about that? And is that even more of an option now with AI being in the picture?

1:06:31Shawn O’Malley:It's quite interesting because it was brought up in Constellation's 2021 shareholder letter, but the last shareholder letter before that was actually the 2017 letter. So the board asked Leonard to stop writing them as more copycats have emerged. And he gave so much insight into his thought process and also how the company works that they thought it was just better to stop the writing of these shareholder letters in general. But he was basically saying that the original mission of finding and acquiring great VMS companies has been delegated down to, as we know, the operating groups as the organization has scaled.

1:07:05Shawn O’Malley:And you now have about 100 capital allocators in the company. So his job and the job of the head office in general is to quote, be a good steward of investors capital. And there were two ways to go about that. First, you set up a small team at the head office focused on large scale VMS deals. So instead of $3 million deals. We're talking about$300 million deals. The second part is to look for investment opportunities outside of the vertical market software universe. So you shouldn't forget that Mark Leonard was and is a genius capital allocator first. And I think he just saw the tremendous potential in VMS due to, you know, the stick revenue, the high switching costs and the high margins and all that.

1:07:47Shawn O’Malley:But he's fully aware that you can make money in other businesses too. Having said that, it does seem a bit weird that you basically stop writing these shareholder letters for a couple of years. And then one year you still think it's kind of needed that you put out another one. And it's basically one where you talk about not investing or at least entertaining the opportunity of investing into other industries as well.

1:08:09Daniel Mahncke:I think the exact words in the letter were referring to investments outside of VMS was that they will require highly contrarian thinking and are likely to be uncomfortable in the early going. Hopefully, though, we have built enough credibility to warrant your patience as we explore new and underappreciated sectors. So that was just a quote from Linert's writing.

1:08:32Shawn O’Malley:Maybe I'm, you know, over-interpreting it, but if there's no need, because you still have a lot of runway left in VMS companies, why would you even go the way that you would describe as, you know, highly uncomfortable? And that's kind of what I'm worried about. Also, now that Mark Leonard is gone, I mean, he definitely had that trust and the credibility with the shareholder base, and he was a value investor more than anything else. So I would have had no doubt, you know, that he could be successful in other parts of the market as well, but likely it won't be him making these types of decisions in the future now.

1:09:02Shawn O’Malley:And while I fully trust Mark Miller with the VMS part of the business, especially because as we said, it's highly decentralized anyway, the idea of expanding into other verticals is a decision which I think is more dependent on, you know, the top management team. And in 2022, Leonard mentioned an opportunity in the thermal oil sector. And back then it was a bit difficult, as many of you probably know, for those companies to get financed because energy prices came down a lot. And it would have been a deal worth about a billion dollars, so even significantly larger than most of the VMS deals you've seen in the past.

1:09:35Daniel Mahncke:I guess that's what he meant by highly contrarian.

1:09:40Shawn O’Malley:Yeah, I think that was an idea, you know, certainly far outside of the usual circle of competence and certainly very value-oriented, but it didn't even end up going through. So it's nothing that you as a shareholder would need to worry about now. And, you know, since then, there have been no major acquisitions of companies and industry outside of the VMS sector. So one sector that has been more of a target in recent years is payments. But, you know, most of CSI's operating groups have made investments in that space. I would say, though, it's complementary to VMS. So I wouldn't say this has been a 180 on their M &A strategy.

1:10:13Shawn O’Malley:It's more so that it's kind of part of the strategy that they've basically gone for, you know, the last decades.

1:10:19Daniel Mahncke:Before we get to the valuation, which is everybody's favorite part, I just want to talk about capital allocation a little longer. We own quite a lot of companies that buy back their own stock. Constellation has not done this in the past. With a 50 % drawdown now, the valuation is very different than what it once was. I mean, do you expect that to change?

1:10:39Shawn O’Malley:Mark Leonard wasn't a fan of buybacks for really the longest time, and at least not in Constellation's case. in part, I think that's because he doesn't think it's fair towards shareholders since, you know, they have less information about the company than him. And to some extent, that would basically just be a signal to shareholders and they would kind of trust your process. If you buy back shares, you would have to think, okay, well, that's a positive sign. But then again, you don't really know why he does so. So you're basically just relying and putting faith into his decision-making as a CEO.

1:11:10Shawn O’Malley:And I can understand how he's kind of thinking about this in, you know, in terms that makes it seem unfair for him to buy shares while the ordinary shareholder wouldn't have the same information as him. And then you obviously have the valuation too. I mean, earlier in 2025, he said that you would need to buy Constellation at a quarter of the price to make 20 to 25 % annual returns. And to be honest, this is, you know, why it didn't feel like pitching Constellation earlier. However, we are talking about 25%. That's double our hurdle rate. And we are more than halfway down in terms of the stock price now.

1:11:44Shawn O’Malley:So he said that when the stock price was about 5 ,000 Canadian dollars, now it's half of that. So, you know, if we trust him, we should get at least high teen returns going forward. And I don't know if that's enough for him to decide that buying back stock is a better use of capital than investing back into the business. But, you know, they did scale back the dividends with the shift to large VMS deals and at least potentially deals outside of VMS too.

1:12:09Daniel Mahncke:I guess that's a perfect transition over to the valuation discussion. What do you say about that? Is the 50 % drop enough for us to get excited about Constellation then? Or do we wait for Mark Leonard's advice of needing a 75 % drop for the shares to be really attractive?

1:12:28Shawn O’Malley:And first I just gotta say it's it's fascinating how narratives change just with the stock price. Adobe has basically been on a cold streak since early 2024. And, you know, since then there have been rumors about AI destroying the business, which just hasn't happened yet. And for Constellation, though, I didn't come across any such comments until late 2025. But as soon as the stock started dropping, it was just a spiral of bad news about how AI will disrupt the business. And, you know, one of Constellation's spinoffs, Topicals, still had a 70 % rally while the AI narrative was already hurting several SaaS businesses.

1:13:01Shawn O’Malley:So it took a while, but with a bit of time, CSI, Topicus, and the other suspects all got hit by the AI narrative. And anyway, I was actually really curious what type of assumptions you would need to make for this to be an attractive investment at reasonable multiples. And I know this sounds like I'm kind of making up numbers just to justify the current price, but that's not what I plan on doing. It's more thinking about it in, you know, a reverse DCF type of way. How much growth is the market currently pricing in for Constellation? Almost every picture here right now is so much about just the fact that the stock dropped.

1:13:38Shawn O’Malley:And that certainly makes it interesting to look at, you know, it's part of why I pitched it here today, but a stock can be down 50 % and still will be overvalued. Now, I don't want to go through the assumptions for each of the four business units, so licensing, services, hardware, and maintenance. First of all, only maintenance really matters, but also because, you know, we don't have much insight into how exactly they will grow. Organic growth is about 4%, and the rest will then be, you know, about acquisitions. And in some years, you will have more. In others, you will have less. In the past few years, for example, CSI deployed around$1.6 to$1.9 billion in acquisitions and will be slightly less in 2025.

1:14:17Shawn O’Malley:And nevertheless, I expect CSI to still achieve a CAGR of about 12 to 13 % top line growth in the next five years. And in terms of margins, I only expect a very slight increase from 17 % in 2025 to 18 % in 2030. And for context, I'm talking about the margins of the free cash flow to shareholders adjusted for something which is called IRGA liability.

1:14:42Daniel Mahncke:On behalf of the audience, I think I'm going to have to ask you to explain what those things mean.

1:14:48Shawn O’Malley:Yeah, so Constellation reports a number called free cash flow to shareholders. And the idea is to get a number that would be left over basically for shareholders if you assume that CSI would just stop buying companies for a moment. So it basically takes free cash flow and then it subtracts the cash that is needed to keep the business going. So things like ongoing investments, working capital needs, and CapEx. So you could think of it kind of like an owner's cash flow, similar to Buffett's owner's earnings.

1:15:17Daniel Mahncke:The takeaway then to me is that you expect the margin on that cash flow to go up slightly in the next couple of years?

1:15:24Shawn O’Malley:Yeah, again, in my model, it's only 1 % point, driven by a more mixed shift toward more maintenance revenue through basically just a growing install base each year. And perhaps maybe the implementation of AI into workflows, although I didn't want to put that into the model with, you know, high expectations for margins because it's just so uncertain. And if you want to paint a more bullish picture, you know, you can calculate with some more margin expansion than I did. If I put a 25 multiple on those cash flows and then discounted at 8%, which is what we usually do, I get a fair value of almost$3 ,000 USD in this case.

1:15:57Shawn O’Malley:And I got to say, usually we applied margin of safety of either 10 % or 20%. In this case, I did go with 20 % just because of all the uncertainty. around AI and how it still could affect the business. And after doing that, it gives us a fair value estimate of about$2 ,300 to$2 ,400, resulting in an IRR somewhere around 12 % to 13%.

1:16:18Daniel Mahncke:A 25 times multiple is not all that aggressive, especially when you consider how the business has been valued in the past with a premium around Leonard's capital allocation skills. But now with him gone, probably the higher ranges of that kind of historical valuation, I would say that's unlikely. 25 times is not modest either, though. If you take the inverse of that multiple, so it's referred to as the earnings yield or the free cash flow yield, whatever you want to call it, by dividing one over 25, then you get a yield of 4%. So assuming no underlying growth, you would expect a 4 % annual return on your original investment.

1:16:56Daniel Mahncke:And the only reason I say that is just to emphasize that this is not an inherently cheap valuation on its own. If you can buy a business at 10 times earnings, your earnings yield is 10%. So you know with no growth, you can earn a very satisfactory return while having further upside if the business does grow. And so anyways, a 25 times multiple is much, much more reasonable than where the stock has previously traded. But still, it's just a bit away from being obviously cheap, unless you're very, very bullish on the company's ability to keep growing. And And I'm not sure anyone should be that bullish because the company has been open about looking beyond vertical market software, which is their core competency, implying there's not a ton of runway ahead in that core strategy.

1:17:42Daniel Mahncke:At least that's my kind of second order thinking on it. And so when you're implicitly betting on their ability to pivot without having Mark Leonard's stewardship anymore, all the while you have this major uncertainty swirling around in terms of how LLMs will impact their subsidiaries. it's not entirely clear to me that the market is dramatically overreacting.

1:18:04Shawn O’Malley:I think you can definitely say that, you know, CSI traded at multiples that are potentially too high, especially now that you have AI coming into the market. I personally wouldn't expect them to slow down growth significantly, especially since we haven't seen any acquisitions outside VMS after that announcement. And to me, it's one of the reasons where I just got to say Mark Leonard is just, he's an honest guy. Like even if he knows the market doesn't like what he says, he will probably say it. And, you know, part of that is also saying at some point we might actually need to invest outside of VMS.

1:18:35Shawn O’Malley:I wouldn't say that's as negative of a sign as I would think it is when other CEOs would say the same thing. So I think part of the reason why I didn't do a bear and a bull case this time is because I don't really see a need for that. If you believe that AI will disrupt CSI's business, then you shouldn't invest in it in the first place. And we are definitely not yet at a valuation, like you said, well, this becomes a deep value play at all. And if you think this was, you know, too conservative in my assumptions, you can just go to our weekly newsletter, which is also free, by the way, and, you know, download the model and then put in your own assumptions.

1:19:08Shawn O’Malley:So the main drivers that I would recommend you to change here are the margin expansion and or a different multiple that you can use. So of course, you could also say they, you know, need an extra margin of safety. I think, you know, 20 % is pretty fair. And I just got to say generally, I'm torn on the investment decision here. I think it's a great company at a fair price and in manga address, so I'm inclined to add it. However, I'm also a fan of smaller companies that can grow faster and basically have fewer problems to allocate capital than CSI likely has going forward. And in my research for CSI, I came across multiple publicly traded companies that are either CSI copycats or even officially part of their portfolio because they are owned by one of those operating units.

1:19:52Shawn O’Malley:And basically, they are small, publicly traded companies. And I can't help but think that our money would probably be better invested in them. We can't have a full episode on each of them because probably it would be too boring. Again, they do exactly what CSI does. And to some extent, we already covered that with the chapters and just in another geography and at a smaller scale. If you like the idea, though, and listeners are interested, I could imagine making an episode where we go through three or four of them and then we just pick the best one. Or maybe you just say, Daniel, don't overcomplicate it.

1:20:26Shawn O’Malley:We can just buy Constellation.

1:20:28Daniel Mahncke:Sometimes we kick the can of companies because we need the price to come down before we can buy it. And in this case, I'm not sure that's how I'm thinking about it. Of course, a lower price would be great. But to me, it's a question of whether we understand the business well enough looking forward and really its prospects to make a decision at any price point. And honestly, I would actually enjoy maybe spending a few episodes digging into some of these spinoffs or copycats. And if we come on the other side of that feeling inspired to own the OG, then maybe that's what we do. And I would say, you know, we're cautiously optimistic on Constellation software, but probably need to do more work to feel comfortable pulling the trigger.

1:21:09Daniel Mahncke:So I mean, I don't know. How does that sound to you? Is that reasonable, Daniel?

1:21:13Shawn O’Malley:I think I'm a bit more than just cautiously optimistic, but I think we both agree on the fact that CSI is one of those companies where the possible downside of AI is larger than the upside, and that does not mean that it's likely that AI will hurt the business. But we prefer situations where there's a bit of an asymmetry for the upside to us. And for me, this is not so much a decision against CSI or Constellation and more decision for covering smaller VMS requirements. And, you know, there's a saying that you shouldn't listen to what people say, but just look at what they do. And you can apply that to CSI in two ways.

1:21:48Shawn O’Malley:You could listen to them saying that there are still tens of thousands of potential targets out there. But then you got to ask yourself why they look for opportunities outside of VMS. You could also, you know, to kind of give the counter argument, look at it from the other side and say, well, they said they might look outside of VMS, but they haven't really followed up with any meaningful acquisition. So in the end, it comes down to, you know, the size of Constellation, at least for me. I think we can find fantastically run copycats with much more room to grow and the optionality to shift their focus just due to their smaller size.

1:22:22Shawn O’Malley:And if AI should be more disruptive than we think today, I think it makes sense to own just more agile companies. And also we said, and, you know, one of the goals that we set in our intrinsic value community at the end of 2025 that we want to look into smaller companies too. So I think this is actually a good chance to do exactly that.

1:22:41Daniel Mahncke:Well, I think it's safe to say we do already have exposure to Berkshire and also Transdime and Chapters. So nobody can accuse us of not allocating a chunk of our portfolio to these acquirers. And yeah, if Berkshire is on one end of the spectrum of having maybe the least flexibility in what they're able to do and Chapters is on the other end of the spectrum, constellation is probably a lot closer to berkshire and yeah i don't know there's probably some question to be had of just how um how much incremental value is added by putting an allocation to constellation in the portfolio relative to our already existing stake in berkshire maybe we would sell part of the berkshire stake to fund the constellation software stake i wouldn't be opposed to that.

1:23:28Daniel Mahncke:But I wouldn't necessarily want to allocate fresh cash to it to give it, to increase our overall exposure again to more of these serial acquisition companies. I mean, they're great businesses, but I'm not sure that I'm inspired enough to want to make them as a kind of an aggregate, a very significant allocation to the portfolio.

1:23:52Shawn O’Malley:I would say if you go through the shareholder letters again, they are pretty inspiring. So maybe that helps. But also just for the audience, we constantly host meetings, Sean and I, where we just go over our portfolio, we go over our waitlist, and we're constantly talking about the companies that are on top of our waitlist. And if we potentially want to add them and, you know, change them for a position we currently have. So more than ever, positions that are on our waitlist, we will not forget about them. You know, we can likely add them at a later point. So yeah, I would say I'm just glad I finally got over my stubbornness and just had the chance to look at Constellation.

1:24:26Shawn O’Malley:It was long overdue. If you consider yourself a value investor, you should have looked at this business at some point. But I would say it's time to move on. So that means for you to give us some hints about which stock you will pitch to us next week.

1:24:39Daniel Mahncke:So next week will not be a serial acquirer. And actually it's the opposite in terms of having a preference for building their own businesses rather than buying. But that did just recently change, though. This company that I'm talking about made some splashy headlines with their plans to do one of the biggest acquisitions of the decade that is now pending approval and regulatory review. And so I think it's safe to say that the fate of Hollywood is on the line. And I'm sure that's already given away too much. But yeah, you'll have to tune in next week to see exactly what we're talking about.

1:25:10Shawn O’Malley:Yeah, I think some listeners might have an idea of what company we're talking about. All right, let me end it with a quote today by, of course, Mark Leonard. You have to do it. Everything else wouldn't be fair to him. And he said, over the long term, stock returns will be determined largely by which capital allocation decisions the CEO makes. Two companies with identical operating results and different approaches to allocating capital will derive to very different long term outcomes for shareholders. And I think the fact that he's, you know, one of the best to ever do it turned Constellation into what it is today.

1:25:42Shawn O’Malley:and I'm confident it will only get better with time. We're not yet betting on it though. And with that said, see you all next week, next Sunday. Have a good one. Thanks for listening to TIP. Follow the Intrinsic Value Podcast on your favorite podcast app and visit theinvestorspodcast.com for show notes and educational resources. This podcast is for informational and entertainment purposes only and does not provide financial, investment, tax or legal advice. The content is impersonal and does not consider your objectives, financial situation or needs. Investing involves risk, including possible loss of principle, and past performance is not a guarantee of future results.

1:26:18Shawn O’Malley:Listeners should do their own research and consult a qualified professional before making any financial decisions. Nothing on this show is a recommendation or solicitation to buy or sell any security or other financial product. Hosts, guests, and the Investors Podcast Network may hold positions in securities discussed and may change those positions at any time without notice. References to any third-party products, services, or advertisers do not constitute endorsements, and the Investors Podcast Network is not responsible for any claims made by them. Copyright by the Investors Podcast Network.

1:26:46Shawn O’Malley:All rights reserved.

1:27:09Thank you.