TIVP062: Duolingo (DUOL): Generational Opportunity or fad? w/ Daniel Mahncke & Shawn O’Malley

19:48Daniel Mahncke:I think I did do an AI call with Lily at one point. And Lily is, you know, it's an animated character. And I think she's got like purple hair and she's like kind of disinterested in you. and maybe depressed. In terms of character development, again, Duolingo has done about as good of a job as you can get. For just a language learning app, there are people out there who will immediately know the green owl we're talking about and Lily and the different characters in the Duolingo universe, almost like they're Disney or Nintendo characters, except this is a language learning mobile app. So I think they've done an incredible job with that.

20:27Daniel Mahncke:And so the next question I want to ask you that was basically what does the conversion of free users to the paid tier look like right now? Because that seems like really the key question to me for how the business is doing and something we'd want to watch going forward.

20:43Shawn O’Malley:Currently about 9 % of the monthly active users are paid subscribers, which is an absolute number, something like 11.5 million people. And while growth is naturally slowing down, paid subscriber growth is still over 30 % a year, which is a phenomenally strong number. However, there's something interesting in the data. Duolingo typically shows paid subscribers relative to, again, monthly active users. So if paid subscribers are growing faster than monthly active users, paid penetration will expand. And that's what we have seen in the last three years, when the paid subscriber conversion rate climbed from 7 % to 9%.

21:19Shawn O’Malley:The more important metric than monthly active users, though, is daily active users as we said already and daily active users has been outgrowing paid subscribers so if we use daily active users instead of monthly actives as the nominator for calculating paid conversion the trend is actually declining and that isn't necessarily alarming because there's often a time lag between someone becoming a daily active user and then deciding to actually upgrade to a paid tier especially when user growth re-accelerates which has been the case in 2022 and 2023 and basically you have then flooding new cohorts with basically new members, right?

21:55Shawn O’Malley:And I still think it's an important metric to monitor because in theory, as daily active user growth normalizes and cohorts mature, paid subscriber growth should re-accelerate relative to daily actives. And if not, that could mean that, you know, either churn rates go up or conversion is worse with the newer cohorts. As you know, my co-host Sean and I are obsessed with analyzing companies. But you probably have noticed from personal experience that talking stocks is not everyone's favorite hobby. And I'm reminded of that every time I bring up investing at dinner or when I'm out with friends, they tolerate it for about 10 minutes.

22:31Shawn O’Malley:But then I get this look, the one that says, we get it, you love stocks, but this is not the place. So Sean and I thought, why not build that place? And we did it. It's called the Intrinsic Value Community. Our members range from pilots and firefighters to lawyers and engineers, but also hedge fund managers, actual rocket scientists and CEOs. And despite those different backgrounds, what connects all of us is the passion for value investing and continuous learning. And each week we host live calls, covering everything from vetting the group's best stock pitches to analyzing portfolios, investing case studies and conversations with expert guest speakers who are either prominent portfolio managers, CEOs or authors.

23:11Shawn O’Malley:And the best thing is that if you ever miss a call, We have a library of recordings for watching back every single call we've ever hosted. And if you prefer reading over watching, well, then we have dedicated spaces in the community to share write-ups, discuss investing ideas, or just your thoughts on the general market. And multiple times 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 explore, hang out, and most importantly, talk stocks. Our last cohort of members brought together 20 incredibly thoughtful people, some of the sharpest investors that Sean and I have ever met.

23:50Shawn O’Malley:And if you want the chance to learn alongside people like that, you should join our waitlist at theinvestorspodcast.com slash intrinsic value community. That's theinvestorspodcast.com slash intrinsic value community.

24:06Daniel Mahncke:Naturally, increasing that conversion rate is the company's primary fixation because it's the best way to not only increase revenue and profits, right? It's the most direct way to do that. But also when they make those conversions, they're also really materially improving the quality of their business or the quality of their earnings, which is kind of a fuzzy concept. But basically the idea is that ads are decent business, but recurring subscription revenues are much, much more reliable and therefore higher quality. And especially for an app, Yeah, I mean, it's just much more valuable to have people paying you every month where there's a degree of predictability as opposed to kind of the cyclical whims of the advertising market.

24:47Shawn O’Malley:Yes, in the long run, that's definitely the way Duolingo wants to monetize the platform. And Duolingo also earns money through ads, as you just mentioned, but also an official English test. But even combined, that's less than 20 % of revenue. So it's really about getting people into the paid subscription tiers. And the CEO, Luis von An, does not get tired of mentioning, though, that at least right now at the current state, they run hundreds of different A-B tests to find out what best works for the app, both in terms of improving conversion, but also just to increase daily active users. And the rationale behind that is, you know, that there's this massive opportunity that Luis sees for Duolingo in the age of AI.

25:27Shawn O’Malley:He seems confident that Duolingo will at some day have a billion users. And, you know, converting more members to the pay tier and growing daily active users are two goals that are somewhat at odds with each other. If you, for example, include more ads or limit the number of lessons that you can do without paying, you will obviously see an increase in the conversion rate to the pay tiers because some users will finally decide to upgrade. But you will also see a decrease in daily active users because some will get annoyed by the changes by getting more ads after each lesson and then say, you know what, I'm out, I'm not using Duolingo anymore.

26:02Daniel Mahncke:To anyone who's used Duolingo, I don't think it's a surprise to hear about those A-B tests. I mean, the app is just so almost perfectly optimized to maximize the dopamine hit that you get every time you use it, which is why we say it feels like a game where you don't always realize that you're learning a language, at least supposedly learning a language. And to some extent, I think it reminds me of how Robinhood kind of operates. And if you compare Robinhood to traditional stockbrokers like Charles Schwab or Fidelity, I mean, gosh, you'll notice how the entire app is just programmed completely differently to keep you engaged and to trigger these emotional reactions that drive you to invest, which was very much not how Schwab, you know, Schwab is like very boring and kind of, you know, sophisticated and has an air of, you know, I guess, like I said, just sophistication to it that doesn't lend itself to the degeneracy that I think Robin Hood is not afraid to indulge in.

27:01Daniel Mahncke:And so with Duolingo, again, like I said, more than anything, it just feels like you're playing a game.

27:07Shawn O’Malley:Well, it's fair to say that Duolingo is a game with learning capabilities and stand off the other way around. I mean, the A-B tests, they run basically can range from, you know, everything like simply changing the colors of icons or buttons to more significant structural changes, such as replacing the tree style home screen with a linear path to better guide learners. And one of the big changes I've noticed since I last used the app is that you no longer lose hearts when you make a mistake, which is also what you alluded to in the beginning. back when I used the app in the free tier, you had five hearts and when they were gone, you couldn't do another lesson.

27:42Shawn O’Malley:Today, Duolingo has what they call energy bars. So as a free user, you have 25 of those. Now for each question, one will disappear. So at first glance, you can do 25 questions a day. However, when you get a streak of questions, you can also get energy back as well. And at first, I think that sounds like a pretty good idea because instead of punishing you for getting something wrong, which is, you know, the case when you lose a heart, when you make a mistake, you're now getting something. You're basically getting rewarded for correct answers, which is usually a much better way to learn anything.

28:14Shawn O’Malley:However, there has been significant pushback because it turned out that in practice, energy goes out much quicker than hearts did. With the old system, you could, in theory, make as many lessons as you want, as long as you don't make any mistakes. And this is no longer the case. Turns out three lessons are about as much as you can get out of the energy bars, not depending on whether you are good and you don't make any mistakes or whether you're very bad and make a ton of mistakes. And this is a pretty good example of, once again, the balance you got to strike between higher conversions to the paid tier and more active users.

28:48Shawn O’Malley:And if energy is used up more quickly than hearts, obviously that improves the conversion to the paid tier. However, it will also reduce the number of daily active users.

28:58Daniel Mahncke:So when the CEO says that he now has shifted the focus from increasing paid conversion rates to growing daily active users, he tends to make decisions based on what increases retention times on the app for free users instead of trying to push the premium tier at all costs. And so my question for you is, is that an immediate response to the backlash regarding this change in the energy bar? Or did you see an impact in the number of daily active users basically after this change?

29:27Shawn O’Malley:Not really. I mean, Duolingo actually saw better retention numbers, as at least if you believe the CEO on the earnings call. I would speculate that this change benefited less serious learners since, you know, they would make more mistakes with the old system and thus they would lose hearts more quickly. Whereas more serious users, who are not paid subscribers though, dislike the change because they could complete more lessons using the heart system. And obviously, the more engaged users will also be the first to start rebelling against those changes. And I think generally Duolingo is very data-driven.

29:58Shawn O’Malley:Otherwise, you wouldn't conduct 750 A-B tests a quarter. So they conduct all these tests and then they check not only for what is better at driving conversion or user growth, but also for figuring out which smaller metrics have the best predictive values in terms of driving these numbers like daily actives or paid conversion. And one metric that Louise emphasized recently was the average number of words spoken per max subscriber, which is kind of a long way to describe it. And basically, it's a metric that, you know, they were able to double in 2025, in part because of the new AI tools, like talking to Lily, for example.

30:33Shawn O’Malley:And he likes to point out that Duolingo generally, as a company, is very good at optimizing metrics once they decide to focus on them. I'm not doubting that for a second, honestly. But I do think as investors, we also need to be careful not to lose sight of the bigger picture, because there are a lot of metrics that technically make a lot of sense to focus on. But the question is, do they really drive the bigger picture? And, you know, I'm just pointing this out because I've seen so many cases in the past where management talked about a lot of secondary metrics and how great they were doing every single time.

31:06Shawn O’Malley:Overall, though, the fundamentals, they didn't do that great. And again, this is not what you're seeing with Duolingo right now. But after looking at over 60 companies on this show, I'm getting increasingly sensitive to this. And I could imagine the same is true for you.

31:20Daniel Mahncke:it's something that famously happened a lot during the dot-com bubble you know like pets.com would say well hey we don't have any revenue but look our web-based traffic increased 30 percent and that's a great sign and eventually that does wear out right but for a time these techie growth companies have definitely been able to keep investors excited about the future with these secondary you know non-traditional non-monetary kind of metrics and you also just can't hide reality forever though and actually match group has had the opposite problem in a way their financials look pretty good or at least like the earnings per share has been growing but their number of paid subscribers from year to year just keeps declining and declining of course you can only sustain financial growth with higher prices for a smaller and smaller user base for so long which is really effectively what they're doing and just pivoting a little bit here away from match group and back to Duolingo, something else we talked about briefly in one of our last meetings is whether people in different countries learn languages differently.

32:24Daniel Mahncke:It's kind of a funny point, but I do think it's remarkable how, for example, social media mostly works the same across the world despite cultural differences, right? I mean, there's some local trends that occur, but if Instagram and TikTok and Netflix has shown us anything, it's that the same type of content, more or less does work around the world. And if anything, the difference is usually along generational lines rather than, you know, between countries and cultures of what people like and are interested in and how they behave on these platforms. And so I think you mentioned that the CEO, Luis, pointed out that it's similar for learning languages online.

33:04Daniel Mahncke:And, you know, this idea of the gamification and a streak that you want to maintain, that works in every country, no matter where you do it, right? It's getting to very fundamental human impulses that supersede culture. And so you might use some different sentences as examples, but otherwise there's just really no material differences in how this is implemented.

33:26Shawn O’Malley:That's exactly what the CEO says. I mean, the only difference, again, is between generations. You simply get slower when you get older, but this is something that the Duolingo course more or less adjusts for anyway. And there was another interesting data point, though, that just came to my mind now as we've spoken about metrics and their impact. As you mentioned at the beginning of the episode, Duolingo is famous for posting quite unhinged content on social media. And those ads are almost always about Duolingo's owl mascot or some other of the more popular characters, for example, like Lily. The strategy is generally centered on turning this normally friendly looking owl into a passive aggressive chaotic character that threatens you when you don't do your lesson.

34:08Shawn O’Malley:And in early 2025, Duolingo launched a dramatic multi-week campaign announcing the death of their mascot, including holding fake funerals, investigating the cause of death and basically changing app icons to show the owl with dead eyes. And in Q3, Duolingo then decided to stop posting this sort of unhinged content in an attempt to make, you know, the brand a bit more grown up, a bit more serious. And according to CEO Luis Van An, the original idea came from the Duolingo community itself. However, the content shift toward, you know, talking more about Duolingo's mission resulted in significantly fewer views, which was expected, but it also resulted in an immediately visible slowdown in user growth.

34:51Shawn O’Malley:Honestly, I don't know what I should think about that. Sure, if you have highly efficient marketing strategies, that's fantastic. But if the impact of posting fewer unhinged ads is so severe that you stop the plans of kind of rebranding the company, then I don't shake the feeling that this might point to a weakness of the business. And I'm thinking about this in terms of, you know, the target audience. There are two different ways to see this, obviously. And they basically decide whether you like the investment case of Duolingo or not. These unhinged ads are generally funny. but I think they primarily target people who sit at home on a Tuesday night and think to themselves you know what I should learn Spanish.

35:29Shawn O’Malley:I would also not be surprised at all if you know the gamification aspect works even better with exactly that audience too and when you think about the monetization potential though you generally want people seriously interested in learning.

35:43Daniel Mahncke:So I think you're saying that if Duolingo's marketing engine only works when they post funny memes and unhinged content, but not when they post things about their actual mission and use case. It probably indicates that people don't take the app very seriously, which I think is fair, and that those same people were likely never really interested in paying for a service like this in the first place, or at least they're probably not all that loyal to the brand, and then they're more likely to switch over, which is, of course, going to be a big problem for Duolingo.

36:13Shawn O’Malley:That's the one side or the one way you can think about it. However, I think there's another side to it as well. When, you know, Luis came up with the idea for Duolingo, he and his co-founder were both in the process of learning a language. And they basically had a challenge where they both had a daily 30-minute lesson. And after some time, they figured out, well, they often skip those. And I get that. When I was still in college, I set a challenge to spend a month or two on something new. And I wanted to learn that. Essentially, I chose a topic, I set a goal, created my own study plan, and then I started.

36:43Shawn O’Malley:Then after four to eight weeks, when I hopefully reached my goal, I would switch to another topic. Actually, chess was one of those projects with limited success. And I had to create those little habits and routines to actually stick with it because even spending, let's say, half an hour a day concentrated on something that's not work or your primary goal can be extremely difficult. I mean, it's basically your free time and you can spend it, you know, with something that's fun. And the point is, just because Duolingo seems like more of a game than actual studying doesn't necessarily mean it's not attracting people who actually want to learn something or, you know, a language.

37:18Shawn O’Malley:And it's not like people who are serious about learning a language don't want to be entertained while they do it. So generally, I think that gamifying is just an incredibly powerful tool and Duolingo's success using it kind of speaks for itself. And just because it's a bit more fun than other apps doesn't mean that people who seriously want to learn a language don't go to Duolingo and pay for it.

37:40Daniel Mahncke:Well, how about we get to the elephant in the room, which is a massive 80 % drawdown in the stock that it's gone through in the last 10 months. And I think that's one of the more severe drawdowns we've seen here on this show. And the first company that comes to mind that we've talked about already is being in a similar situation as Match Group. And that was fueled, though, by actual declines in their user base, as I was mentioning. So what happened for Duolingo? Has growth just been slower than Wall Street expected? Is it AI or just valuation repricing of SaaS companies everywhere? When you think about what ChatGBT did to Chegg, I'm sure that made many Duolingo investors nervous.

38:23Daniel Mahncke:So, I mean, how much does that factor in? And I mean, yeah, what is driving this 80 % sell-off, Daniel?

38:30Shawn O’Malley:I would say it's all of the above, except for a slowdown in growth, which we have not seen. But, you know, honestly, I would also not have looked at Duolingo if we didn't see a sell-off of 80%. I mean, when the stock was at an all-time high in May last year, it traded at a price-to-sales ratio of 30 times. Now it's trading at a price-to-sales ratio below 6. So even on an EV to free cash flow level, the company came down from ridiculous levels of over 80 times to a moderate 12 times. So that's an incredibly low multiple for a company growing 30 to 40 % a year. However, if you adjust free cash flow for stock-based comp, which is quite high for Duolingo, you get a more realistic EV to adjusted free cash flow of 20 times.

39:11Shawn O’Malley:Still though, for a company growing at these rates, that's very cheap. And the big question is whether Duolingo can sustainably grow at these rates and then also keep churn low. If that's the case, I think you have a pretty good deal here.

39:24Daniel Mahncke:So the stock used to be massively overvalued, but now we're looking at what is a fast-growing company, definitely trading at a more reasonable price. And then honestly, you might even make the case that it's cheaply priced compared to its recent growth if you just even extrapolate a fraction of that growth going forward. And so I assume the market then is really just pricing the risk that AI will disrupt the business, kind of like with Chegg, which seems to be so often the case with the companies we look at, especially Adobe comes to mind, where it just feels like the market is completely soured on some businesses we love because of the uncertainty of what the implications from AI are going to be.

40:03Daniel Mahncke:And I remember last time I traveled internationally, it was to Portugal with you, actually. And I didn't feel like I needed to try and even learn Portuguese. And I think my wife was all excited about it and she downloaded Duolingo. But Google Translate is just so freaking good now. It feels like I could lean on that if I ever had to, right? Most people are probably going to speak English anyways, at least in Western Europe. And then if you actually are in a situation where somebody really can't speak the language, you can lean on Google Translate. And so personally, that's probably made me less and less inclined on the margins to get really excited about learning a language like German.

40:41Daniel Mahncke:This is my excuse. I'm rationalizing. Where Germany is a country where a very large percentage of the population already speaks English. So I don't know. I find that hard for me to get motivated. But maybe there are some obscure languages on Duolingo. Maybe I need to learn Navajo. Maybe that'll get me more excited.

40:59Shawn O’Malley:Well, I don't know about that, but you're right. I mean, the market believes that LLMs or just AI native apps will be able to either do what Duolingo is doing today and perhaps much better and much cheaper. And also there's the argument that you say, well, learning languages has just become less valuable of a skill in today's world.

41:17Daniel Mahncke:Well, so how do you think about that? I mean, I think the answer is going to depend on whether you believe the business has any kind of moat, but still.

41:24Shawn O’Malley:Well, I would say there are three modes, or maybe not yet modes, but more so competitive advantages that are worth mentioning if we talk about Duolingo. And the first would be Brent. I mean, at this point, Duolingo is the first thing that pops into your head when you think about language learning. And that's the case in most countries on earth, whether you are in the US, in any country in Europe, or even other parts of the world. And I remember back when I lived in my student home in Frankfurt, how many people there used Duolingo to learn and improve their German. And those were people from two dozen different countries and pretty much all of them use Duolingo.

41:57Shawn O’Malley:And a big part of that brand building process has been Duolingo's phenomenal social media work, as we already alluded to. But the fact that so many people stay is simply that the app works well, at least when you look at the daily active users and the trend that it shows. I totally get that many people on the app never actually end up learning a language. but I would like to point out as we have done today is you know that most people I know and that most people that you know are probably not the best examples of serious language learners and we do it as a hobby I mean if you want to improve your German again so that at some point we can have a talk in German so I still hope that you use Duolingo it's a hobby for you and most people that I know me included use it for a third or fourth language not to learn English because we needed for a job or something else that's really important.

42:45Shawn O’Malley:And the point there is that whether you really learn a language using Duolingo or not in the end, it still comes down to your motivation. The app is highly effective in making people log in every single day and keeping them engaged, even if it's just a couple of minutes per day. And that's already 80 % of what's required to actually learn language, if you ask me. And beyond that, there is simply the factor of whether you're truly motivated to learn something and if not even the best language app in the world won't

43:13Daniel Mahncke:make you succeed i don't think i know anyone who's become fluent with the language primarily using duolingo but like you said i'm sure it has a lot to do with my social circle um i don't know many people whose professional lives really depend on them learning a language and actually the valedictorian of my high school works at duolingo so i'm sure she can vouch for it and and gosh i I mean, she was someone who could have gone anywhere to work. So the fact that she went to Duolingo has actually kind of stuck in the back of my head for years now. And I think she still works there. And it's like one of those stupid anecdotes that I'm not proudly saying biases me, but it does make me think subconsciously, you know, hey, this must kind of be a special company or at least a special company to work for to be able to attract that kind of top talent.

44:04Daniel Mahncke:And clearly other people see real value in the mission that they're working on.

44:09Shawn O’Malley:It's actually interesting you say that because Luis von Aan got quite a big shitstorm when he once put out a memo that Duolingo is a quote unquote AI first company. And in part, that's because both users and employees see Duolingo more as a mission driven company than one driven by profitability and growth. And Luis even once said that he had to change the culture a bit within the company because it seemed like too many employees didn't fully understand that a company also needs to make money. And perhaps, you know, the valedictorian of your school joined because she believed in the mission too.

44:44Shawn O’Malley:But getting back to the second potential mode for Duolingo, that's data. Duolingo is the most used language app in the world. Again, 50 million daily active users generate a ton of data for Duolingo, which can then be used to feed its AI tools. And every exercise produces labels like correct or incorrect, time taken to answer, hints used, numbers of attempts, and also the type of error that you did. And all of those data points are why Duolingo can train its AI models so efficiently. And the weakness is that Duolingo has no obvious network effect. Of course, you could, you know, start a challenge with a friend where you both start to study a language, but it's just way less direct than, you know, on apps like TikTok or Instagram, where you directly benefit from other people being on the platform, either by sending, you know, memes to your friends or also just by other people posting the content.

45:34Shawn O’Malley:I talked to one of our members in the intrinsic value community who is quite bullish on Duolingo, and he actually compared Duolingo to Spotify in terms of how far ahead it is compared to its competitors. And I partly agree. You could argue that Duolingo's network effect is the data, for example. The more people learn, the more Duolingo can use the data to then improve the learning experience for you. Think about the advantages of the international learner base that you have. Duolingo can not only personalize lessons for you, but it also knows that, let's say, German speakers struggle with a certain word or structured English.

46:09Shawn O’Malley:And it can then include that learning, basically, in the lesson for all German Duolingo users. And in that aspect, Duolingo is similar to what Spotify can do with its data. However, Spotify has a flywheel that spins since users and artists reinforce each other. So the more users Spotify has, the more artists want to join the platform. And the more artists Spotify has, the more users want to join the platform. And that's a dynamic that you simply don't have with Duolingo.

46:39Daniel Mahncke:I was just thinking about this the other day with Spotify, because I think I turned it off, or maybe just Spotify got rid of it. But I used to discover music almost exclusively by being able to see what my friends were listening to in real time. I thought that was like the coolest thing ever, maybe a decade ago. and that's not the strongest network effect ever and Spotify has a much more legitimate network effect that you were just describing but it was pretty effective with me and even this I think is stronger than Duolingo's friends network that they try to promote on the app I mean it's just never really resonated with me like you could see that a friend was learning a language and I think you can like virtually high-five each other and support each other and it's all very nice since like, okay, you know, that was really only one time information though, right?

47:25Daniel Mahncke:Like, okay, Daniel's learning Spanish. Got it. I don't really need to see continuous updates the way I might be interested in kind of continuously saying, oh, look at what Daniel's listening to. You know, that's interesting. Let me listen to that song. Because they're probably going to stay learning the same language. So it just doesn't make sense in the same way. And I think what can be easily miss is the fact that Duolingo's main focus is on keeping people engaged and sort of changing their behavior. Uber and Airbnb change people's behaviors, for example. And so the market for ride sharing and well, house sharing used to be much, much smaller, but they convinced people to change their habits and that unlocked hundreds of billions of dollars in value.

48:07Daniel Mahncke:And that is what Duolingo is pretty good at. If you tried learning a language with the help of any LLM, you will know that it's a complete nightmare. And so there's no way that you're going to stick with learning a language through chat GBT. And it's similar for most AI native apps, I think too. The hard part is getting users to log in every day for a long time and matching the lessons to their current level of skill. And the fact that Duolingo does run so many AB tests and has so much user data and has been doing this for so many years, I mean, there is some sort of real advantage with it that they can optimize the experience, even despite my frustrations with it, in a really good way.

48:46Daniel Mahncke:And for the most part, it is a really good experience.

48:49Shawn O’Malley:I don't know what call it was, but some time ago on one of our community calls, you kind of said to yourself, don't overthink a dummy. Sometimes it is just that easy. And I think we tend to get into the weeds of the companies that we look at. And especially when we follow them for a while, because they are in our portfolio. And then you end up talking about all kinds of possible bear scenarios. But when you're on a great company, the best you can do is sit on it and not overthink every piece of news. I don't think Duolingo fits the category of a great company yet. So I don't want to make an argument for buying Duolingo by saying that.

49:25Shawn O’Malley:But when I think about the AI risk, you might do well if you don't overcomplicate things. Since ChatGPT was introduced, Duolingo has grown its daily active users from 15 million to 50 million. Revenue went from$250 million to almost a billion dollars. And operating margins went from negative 2.5 % to positive 11%. The point being, Duolingo has nothing but benefited from AI yet. And of course, this can change in the future. But that's true for 80 % of the companies we look at on this show.

49:56Daniel Mahncke:Well, I think that's right. And to your earlier point, I mean, I go back and forth on this idea of don't overthink things, dummy. You know, because you look around, you think, just stop thinking so much in a decade from now you're probably going to be glad that you bought alphabet or uber or netflix or whatever it is and then there is this other part of me that that feels like and we're going through such a truly extraordinary period of accelerated technological progress and that can humble any of the assumptions that you take for granted and i don't know before Before I get lost on an AI tangent, how about, you know, you talked about this problem with the monetization of users studying English outside of the US or Europe.

50:40Daniel Mahncke:And I think that's a recurring theme. You know, we see that with Netflix, we see that with Spotify, which is to some extent, users outside of their core developed markets tend to generate much less revenue per user. And so really how you square that is going to be interesting for Duolingo. So maybe you can speak more to that challenge for them.

51:02Shawn O’Malley:So in one of the latest earnings calls, Luis talked about how difficult it can be to monetize users outside of the US or Europe. So let's say South America, for example. He said something that reminded me a lot about the difference between the idea of paying for convenience that we also talked about when we covered MercadoLibre here on the show. One of its biggest competitors was Sharpie. a Southeast Asian company. And Shopee can compete with MercadoLibre despite longer wait times because South American users, they're willing to forego some convenience if that means they can pay less money. And for an American or an European, waiting a week or two for a package is by now unimaginable.

51:40Shawn O’Malley:If they can pay$4 extra to have it arrive the next day, most of them will do it. We all know that a subscription like Amazon Prime is even more effective than paying for shipping extra, but the point remains. people are generally more willing to pay for convenience while in many south american countries people would rather wait a week than pay extra dollars for a tour life early and it is the same thing for duolingo louis actually mentioned that as long as you can use the app for free people do that even the power users so they prefer watching ads and being limited and how much they can study to paying for super or max even and that's what he said if it would be at a price point of just 50 sense.

52:22Daniel Mahncke:To some extent, I mean, it just makes you wonder whether Duolingo is doomed to ultimately have most of its engaged users in countries where really people just refuse to pay at all because that's where the service is most valuable. And then that would mean that you primarily make money via ads or, you know, you just have to accept much lower ARPUs, which would actually suggest that the quality of the business could decline going forward as adoption scales. And to me, that's besides disruption from AI, I think that's really the big concern that I have with the business.

52:58Shawn O’Malley:I'm pretty sure we will still get into the AI risks in more detail, but I do think this right here is also one of the bigger potential bear cases. I mean, if Duolingo's popularity in the US or in Europe declines, it could still be a company that remains highly relevant for people who seriously need to learn English, but simply operating in a less attractive market. And actually, most users are already sitting outside the US. So while about 50 % of revenue comes from the US, only 20 % of daily active users are. So that basically means that, you know, 20 % of users are responsible for 50 % of revenue.

53:32Shawn O’Malley:And you might speculate that these customers are also less sticky, since, you know, most US users will be learning a second language as a hobby and not because they need it for their job. And again, it could very well be that Duolingo won't be successful monetizing users outside the US or Europe at all or at significantly lower up use.

53:54Daniel Mahncke:Maybe this is a good time to talk about the other big bear cases that are lurking here and I think are mostly AI related. And so one of the bear arguments I've repeatedly heard is that learning languages will become less important. And this is kind of what I was saying earlier about my trip to Portugal where I felt less inclined to feel like I needed to learn much of the language before I went as AI translation or real-time translation has become more widespread. And I think on that point, you could really imagine the international business world changing pretty significantly when you have real-time translation integrated into Zoom or Google.

54:31Daniel Mahncke:And so, yeah, how do you think about that?

54:34Shawn O’Malley:I think you mentioned some of the most important points, but I would only partly agree. I think in the end, the whole AI translation debate pretty much underestimates the power of human interaction. Also the complexity of communication. There are so many nuances, not just in language, but also in tone, pronunciation, facial expression. All of that is part of a conversation. And most of that is best communicated and transported when both speak the same language. I think I would even argue that communication works better when someone is not perfectly fluent in the language spoken than when they can't speak it at all and instead rely on translation services.

55:12Shawn O’Malley:I can't imagine how, for example, the two of us would sit in our almost daily meetings and I speak German and you speak English and then we have AI translate everything. Even just, you know, the couple of seconds it would take me to either read or listen to the translation, just the facial expressions you miss, the accentuation you would miss, it's just not the same at all. And beyond that, and that's, you know, a point maybe a native English speaker might not appreciate to the same extent. There is something special about, you know, someone talking to you in your language when you know it's not their native language.

55:46Shawn O’Malley:It shows some kind of respect, commitment, maybe even confidence. And perhaps I'm losing myself on a tangent here, but I think communication and language are just a totally underrated topic. I mean, The two of us had a meeting with Stick, TIP's CEO and co-founder, not too long ago, and he talked about how different people can see the world. So for example, if you're an engineer, you might have listened to my tangent on the importance of communication in the last minute or two. I thought it was ridiculous because as long as your translation service works, it's a much more efficient way than actually learning a new language.

56:20Shawn O’Malley:And you would be right. But, you know, with humans, not everything is 100 % logical. and how we communicate is probably one of those fields where logic takes the backseat from time to time. I mean, I know that, for example, when I was in Montana, you know, we had one or two members who actually made an effort to pronounce my name the correct way. And, you know, that just shows some form of respect. And maybe even, you know, if someone texts me on LinkedIn and they say, well, actually I can speak a bit of German and they text me in German, that's a pretty good way to make sure that I actually respond because it's just, you know, a form of appreciation that you otherwise wouldn't get.

56:57Daniel Mahncke:I'll have to start greeting you in German before we record, Daniel. But no, I'm kidding. I mean, there's definitely truth to that. And if you ask me whether I'm confident enough, though, to bet a portion of my portfolio on people continuing to invest hundreds of hours into learning languages themselves, we have nearly perfect real-time translation tools that become increasingly available. That's where I'm just, I'm not so sure. I think it's a really nice feel-good story, but a lot of work goes into, I don't have to tell you, learning a language, right? And I wonder if more and more people will find that the calculus is less worthwhile when it's not just Netflix and TikTok distracting you, but you also know that you have these apps that are really, really seamless.

57:45Daniel Mahncke:But yeah, in a similar vein, I guess I I want to ask you how you think about the risk that AI translation apps could create competitive apps that can basically quickly reach Duolingo's level. And one thing I might say before you answer is I think this gets to the heart of a question I've been thinking about a lot lately, and that's how LLMs are only trained on the outcomes of things. And so what do I mean by that? Well, they can see at a distance how Meta's algorithm works from maybe a user perspective, but the actual details behind the algos are closely guarded internal information that the LLMs haven't been trained on.

58:22Daniel Mahncke:So an LLM may give the illusion of being able to exactly understand certain things like how the meta algorithm works, but in reality, they're only seeing the output and not the input. And so for example, maybe to make things more tangible, you could connect your Instagram to ChatGPT and it can learn a lot about you by going through your account, looking from the outside in, but compare that with the amount of data that meta has about you right they've got all the metadata on your posts you know where you've been what time of day you post that what time of day you log in they can see everything you've ever liked on the platform all the posts you scrolled past how long you lingered on those posts everything your friends liked what your friends of friends are doing and who they are and i mean that's just like the bare minimum scratching the surface of what they know about you and again chat to bt doesn't have that i mean they can look at your output.

59:08Daniel Mahncke:They can see, you know, Sean and Daniel posted a picture together in Montana and they can get some info and context, but it's nothing like what the underlying platform has going on behind the scenes. And that's guarded, you know, gated information, right? That LLMs, right? They've been trained on the open internet, but it's not like they have been trained on the internal processes of all of these companies. And so that leads into this discussion of, you know, the barriers to producing software have fallen significantly with these advances in AI coding. And, you know, maybe AI can create a language app nearly as good as Duolingo.

59:47Daniel Mahncke:Yet, is it really possible to do that without access to the same underlying data that I'm sure Duolingo is not making available to the chat GBTs of the world or people building apps with chat GBT? The reality is that data is very valuable and Duolingo is, they're really the only ones with it. And, you know, what are the implications of that for the durability of their competitive mode?

1:00:12Shawn O’Malley:If you want to argue that Duolingo has a mode, most likely you would need to go into the data direction. And I guess this is one of the key questions regarding AI and, you know, how much difference will it make that AI can only be trained through outputs? That ultimately also answers the question whether all the A-B testing and the data that Duolingo has will turn into a mode or nothing else than, let's say, a bunch of data that is not the difference maker when it comes to actually building a competitive tool. And Duolingo is no matter. You know, the ecosystem that matter has produces a magnitude of valuable data more than Duolingo ever could.

1:00:48Shawn O’Malley:And if I had to bet, I would say Duolingo's data advantage wouldn't be enough to prevent it from, you know, a competitor building a similar app with AI. And switching costs and network effects are rather low too. But if anything, probably the scale that Duolingo has already reached, plus the brand, is why I would say it's likely to succeed. The data mode alone, probably not.

1:01:12Daniel Mahncke:I guess I'm just feeling a little bearish today. But before we look at the bull case, I just want to ask you if there's any other aspects of the bear case that you think we should consider that maybe we haven't talked about yet.

1:01:24Shawn O’Malley:I think we covered most of them. I mean, ultimately, it comes down to whether you believe the company has something that cannot be copied, or that makes it stand out among other language study or even gaming apps. I mean, maybe you remember how I mentioned in my meta episode, how much time I spent going through meta ads before my pitch. And besides watches, I also saw an ad for an app that is supposed to improve your accent. And that's not the same as a language learning app. But I got to say, not only is the app pretty good and mostly AI based, but it also showed me how competitive that space is and how quickly you can get hooked on another app.

1:02:02Shawn O’Malley:On that point, it would be interesting to get more data on how exactly people are using Duolingo, as in how much time do daily active users spend on the app. We all know how effective streaks are for getting people active every single day on the app. However, when I think back to my dual lingua use, and it looks like yours has been the same, at least the last three months of using it, I was just keeping my streak alive by quickly finishing a lesson. And that made me count into the daily active user statistic. However, I wasn't using the app for longer than two or three minutes. And I wouldn't be surprised if that's the case for a whole lot of people.

1:02:39Shawn O’Malley:And those people are unlikely to convert to a paid subscription. And if they're already on a paid subscription, it's likely they will churn eventually, except for this effect, which you basically talked about, that you could see it as somewhat of a gym membership. You pay for it because it feels good. But is that a business model I want to bet on? Likely not.

1:02:58Daniel Mahncke:I mean, we saw that with Snapchat, which faces a similar problem. And when I pitched that company here on the show, I realized how unprofitable, kind of counterintuitively, having a huge number of daily active users can be if they're not actually spending a lot of time on the app or it's very difficult to monetize them, right? Because the more people that use your service, if you're not actually profiting from them appropriately, then you're just outlaying more and more computing power to make the service usable for them. And basically you're in almost like a death spiral of declining margins and profitability.

1:03:34Daniel Mahncke:And yeah, we often talk about businesses where the subscription service feels like a no-brainer and perhaps Amazon Prime is my favorite example to bring up of that. and the value you get is so much more than you pay for that it just it makes sense to anyone signing up that it pays for itself and so oftentimes businesses that can create these no-brainer kind of subscriptions maybe think costco too they are benefiting from having some kind of larger ecosystem that they can tap into and so amazon for example they fund their subscription service to an extent with the marketplace and also with aws and so in turn the subscription service brings in more users and turns them into highly loyal customers.

1:04:14Daniel Mahncke:And if you lack that ecosystem or other profitable businesses that can subsidize your scaled economy's shared business, you need the subscription to be profitable. And it's more difficult than it's almost like a competitive disadvantage if you don't have these other profitable businesses that can subsidize some of the other places that you compete in. And it just makes it harder to bundle things in a way that it becomes a no-brainer for customers. And if you think about the streaming sector, even some of the biggest media companies weren't able to create a service that people would subscribe to or at least stay subscribed to.

1:04:48Daniel Mahncke:And clearly Netflix has been able to do that. And so I think it's often easy to forget just how hard capitalism is and how hard it is to do what the market leader has been able to do for a period of time.

1:05:06Shawn O’Malley:And generally, I mean, apps that have no physical footprint or assets, they just have a hard time becoming a successful business in the long run. I know there are some exceptions, but think about apps like Clubhouse, which had had a$4 billion valuation at one point, which is not too far off from where Duolingo is currently. And there are dozens more of these examples. And most successful app businesses, they have a mix of, you know, a digital and physical footprint or infrastructure. Uber, for example, has a huge network of drivers. DoorDash has signed up thousands of restaurants and the list goes on.

1:05:40Shawn O’Malley:Duolingo doesn't really have any of that. I mean, historically, that makes an app easier to disrupt and there's nothing to argue about that. And at the same time, Netflix was able to survive and thrive in a highly competitive industry. And while they have a lot of capex, which kind of can act as a mode or competitive advantage. I think it really wasn't one for Netflix because I mean, they had competitors like Disney, which should have been way better positioned to win the streaming wars. And to some extent, we just don't know why Netflix succeeded. I think Netflix is just the go-to service and everything else is on the margin or you just, you know, sign up Disney Plus to watch a specific show and perhaps Duolingo can be that for language learning and other subjects that you could learn on an app too.

1:06:25Daniel Mahncke:Well, speaking of future potential, I think we already have done a pretty good job talking about the possible upsides as well as definitely the risks or doubts that we have about the company. But I've seen so much talk about this stock as being a potential hundred bagger as some people frame it. And gosh, I would be very curious to hear what these ultra bull theses look like for dueling. How are people getting to numbers where they can look at you with a straight face and say that this company might be a hundred bagger?

1:06:55Shawn O’Malley:I think for some people out there, delusion is the only right answer. I don't know how Duolingo is supposed to be a trillion dollar company, honestly. If we talk about more realistic bull cases, it mostly comes down to the current market cap, which is again about$5 billion. And then Duolingo's competitive position and the fact that the education TAM, totally addressable market, is huge. Up until now, we mostly talked about Duolingo as an app to learn languages. But again, they are now also expanding to have courses in math, music, and chess. And chess is actually the fastest growing course ever.

1:07:29Shawn O’Malley:It's growing even faster than languages did back in the day. And one possible advantage of chess is the player versus player function. We talked about how difficult it is for Duolingo because they do not have an obvious network effect. Well, if you compete against each other, player versus player in chess, that would change that part. You would actually benefit from more users using and playing chess on Duolingo. So I guess the bold thesis is that Duolingo will not only be the market leader in language learning, but also in all sorts of other fields. And, you know, that's a trillion dollar market right there.

1:08:03Shawn O’Malley:And based on active users, Duolingo currently captures about 7 % of all language learners, but it only captures 1 % of all the money spent on language education. And, And as more of that moves into the online world and Duolingo improves its monetization, you could have a market leader with a double-digit market share in a trillion-dollar industry. And at least that's how the bull case works out in my eyes. If we take a step back, the company again is currently worth about$5 billion. That's after an 80 % drop the stock went through. So while it would take a lot to become a 100-bagger, it's not that difficult to paint a scenario in which the stock becomes a multi-bagger, which wouldn't be bad either, if you ask me.

1:08:47Daniel Mahncke:I think we would happily take that. But I heard that one step Duolingo is taking to increase profitability right now is diverting traffic to its own website so that when a user upgrades, they don't go through the App Store and then Duolingo can save up to 30 % on the App Store fees that Google and Apple like to charge. And if that works, I think it's another argument for the brand strength of Duolingo. I'm kind of skeptical of to what extent they can actually get people to do that. But whenever there's an extra layer of complexity, it needs trust and willingness from users to go through with that.

1:09:28Daniel Mahncke:And apparently that's the case with Duolingo where you have a percentage of users who are happy to take the extra step to go to the website to check out. and they probably don't know it's they're doing so to save Duolingo money but that is pretty much

1:09:44Shawn O’Malley:why it's one of the more immediate measures you know that Duolingo is taking to increase profitability and it's still in the test phase and you know due to an outstanding cult ruling we don't know if this will be something that they can do in the long run but apparently it works quite well I mean Louis said that they lose some people along the way but he made it sound like that's not a lot of people so churn is minimal and you know it's still way more profitable than going through the App Store if you lose a couple people here, then save 30 % on an annual subscription, which costs$100 upwards.

1:10:14Daniel Mahncke:I think we're getting pretty close to the valuation. But there's probably one more thing that we need to bring up for the sake of proper diligence here. And just recently it was that there was this announcement that the CFO is leaving. So I would be remiss to not ask you about that. Is there anything from that announcement that caught your attention that's you know, worth us looking more into? Or is this really just a transition that occurred without any drama?

1:10:42Shawn O’Malley:From what I could find, I think it was the latter. Matt Skarupa posted on LinkedIn about his decision and his gratitude for being part of the journey. And he will also stay on board to ensure that the transition goes smoothly, which I think it will, since his successor has served on the board since 2019 and has prior CFO experience. So I don't think it's anything I would worry about if I would be a shareholder. All right.

1:11:06Daniel Mahncke:Well, then it's time to talk valuation. And I got to say, I was surprised to hear that Duolingo, based on metrics like enterprise value to free cash flow, EV to FCF, just sort of like a different lens to think about. It's like an alternative to the price to earnings ratio. The company seems to be priced on the cheaper end of things. And when I look at the margin profile, it does look pretty promising. Return on invested capital has looked less promising. so I wanted to ask you if there's really any reason for that and how you think about that going forward

1:11:39Shawn O’Malley:yeah maybe first of all I mean for example why do you take EV instead of price here normally because this is a company worth 5 billion dollars and they have 1 billion dollars in cash on the balance sheet and no debt so that's why you would say okay well if that's the case and you have so much debt compared to the market cap it can make sense to use enterprise value which takes into account debt and cash and basically you would then end up with an enterprise value of only 4 billion instead of a market cap 5 billion so Yeah, first of all, Duolingo has not been profitable for that long. So when you don't earn a profit, you will naturally have a negative return on invested capital.

1:12:12Shawn O’Malley:In more recent times, though, what's raining on ROIC is the fact that Duolingo expenses the vast majority of its research and development costs immediately rather than capitalizing them as assets. So that means they're flowing through the income statement and that is decreasing profits. And since R &D is Duolingo's largest expense, ranging from 37 % to 44 % in recent quarters and years, this significantly depresses the return, which is defined as net operating profits after taxes used in the ROIC calculation. If you add to that, the Duolingo is still primarily focused on user growth rather than monetization.

1:12:48Shawn O’Malley:When that's the case, there's a metric you can look at for an app like Duolingo, and that's called marketing spend or sales efficiency. So you calculate sales efficiently, basically by taking the difference in revenue between this year and last year, and then dividing it by the marketing costs. And if you do that, you get a number, which is something like 2.4 for Duolingo. And that basically means that for every dollar they spend on marketing, Duolingo earns$2.40 in revenue. And that's a very strong metric. All right.

1:13:15Daniel Mahncke:So investments are certainly paying off, even if it isn't yet really showing up in the numbers, at least like the metrics we like to look at, ROIC, return on invested capital. So how about we actually get into your valuation model and try to get an idea of what is the fair value of the stock? And then really, what is the intrinsic value in the sense of at what price would we feel comfortable buying the stock given the risk return profile looking forward?

1:13:41Shawn O’Malley:Yeah, so first, something that I would like to say is we actually had a call with Adam Cecil in our community just the other day. And Duolingo was also a topic, only a short one, though. And one thing that he said is, well, for companies like this, and I couldn't agree more, you're a business analyst, you're not a financial analyst. So looking at the numbers now from Duolingo, you have no idea where they will actually be in five years from now. So what you should focus on is figuring out if this is a company that actually has a moat and a competitive advantage and cannot be disrupted by other competitors or AI, no matter how it actually evolves in the next couple of years.

1:14:14Shawn O’Malley:So I think that's what we did up until now. So that's probably the most important part, even more important than the evaluation right now. But having said all of that, yeah, so let me maybe quickly go into the model and explain a bit more about the accounting of Duolingo. Nothing complex, but something that's worth mentioning. Duolingo reports something that is called bookings. Essentially, that's cash received from customers, while revenue is basically just an accounting measure that can lag behind in cash. So what do I mean by that? If you sign up for an annual Duolingo membership for, let's say, $120, Duolingo receives the full amount upfront, and that's the number that booking shows.

1:14:51Shawn O’Malley:However, revenue is recognized gradually over the next 12 months. So basically, it's$10 a month. So if you sign up in June, Duolingo will only report$70 in revenue for that year, although the entire$120 are shown in the bookings number. So with that aside, in my model, I've broken down the revenue by first looking at bookings and then I put bookings into two buckets. So there are subscription bookings, just from the max tier, the super family tier and non-subscription bookings like ads or the standardized English test that Duolingo offers. Because ads and one-time purchases are recognized instantly, non-sub bookings are basically equal to the non-sub revenue.

1:15:30Shawn O’Malley:I won't go into detail about how I model subscription revenue. Just briefly, I assumed a mix of annual subscriptions and recognition rate that actually determines how much of the bookings will be recognized as revenue in any given year. So if, you know, five months are left, then it will be five months. If seven months are left, it will be seven months. It might sound complicated, but, you know, it's really not. And most importantly, the differences between bookings and revenues are also not that large either. So it doesn't have a material impact on the model. So long story short, low single digit dilution, something like 2%, low 20s, adjusted free cash flow margins, and an exit multiple of 21, which is about where we are today.

1:16:09Shawn O’Malley:I get a fair value of about$180 per share. But then given the right range of outcomes with this stock, I decided to go with a margin of safety of about 30%. And that leaves us with an intrinsic value of about$130 per share right now. Now, in a bull case, I can see free cash flow margins move even closer to EBITDA margins, basically reaching 30 % in two or three years from now. And if you combine that with the revenue CAGR in the low 30s, you get a$300 stock. So even with a 30 % margin of safety and a reasonable multiple of 25 times, this will be a very good investment based on today's price. And then you can just as easily, and that's pretty much the end of it, paint a bear case too.

1:16:54Shawn O’Malley:And, you know, probably both my bear and bull cases are generally too conservative. And still, I get$60 estimated fair value when I assume 20 % revenue growth. So still growing, you know, quite high, but only 11 % free cash flow, just simply due to lower margins. And in a bear case, it's also fair to assume that the share count will increase faster than in a bull or bear case. So I went with 4 % a year, which is quite a lot. But historically, Duolingo has had quite its fair share of dilution. So, you know, summarizing it all, the point is, I consider this stock's future to be highly uncertain.

1:17:31Shawn O’Malley:I could, you know, see this app become better, more popular, and also more profitable every year. But I also wouldn't be surprised if five years from now, the fundamentals have deteriorated significantly. And I know that's not necessarily a satisfactory answer, but you know, that's a game of investing. As value investors, we want to first limit losses. And, you know, that's hard to do in this stock. Again, I wouldn't be surprised if it's significantly lower in five years. And still, I got to say that to you, it does seem like kind of an asymmetric opportunity at today's prices below$120 for someone with the right risk tolerance.

1:18:07Shawn O’Malley:It's definitely not what we look for in the intrinsic value portfolio. So I'm not suggesting we should add it. But for transparency, I opened a very small, very small gaming position in my personal portfolio at an average price of$114. It's certainly money that I'm okay to lose.

1:18:25Daniel Mahncke:Well, it's tempting. And I think it's okay to have those gambling positions, right? I mean, you only live once. And as long as you're not being irresponsible with it, which I know you're not, Daniel, that's perfectly fine to do. And I've been guilty of doing the same thing at times. But yeah, I don't know. I can only stomach so much potential AI disruption in my portfolio. and I think Duolingo seems to fall more squarely into a box where I'm just not confident about how they'll be impacted or confident that they won't be significantly impacted to their detriment. And my very basic view is that deeply entrenched enterprise software is going to be very difficult to disrupt if that ever even happens.

1:19:04Daniel Mahncke:So that's the sales forces and Adobe's of the world where you have real professionals and decision makers relying on these systems with their careers depending on them. And so it's just very unlikely anytime soon people are going to trust these vibe coded AI apps to basically replace some of those operations and responsibilities. And so I think the barrier to writing software, though, is lower than ever. And so it is easier than ever to create an app. And so if I was a mobile app company right now, or B2C mobile apps, that just seems like an industry that is so much likelier to be disrupted than some of the enterprise SaaS behemoths that we've looked at on this show, even if the market has been panicking about SaaSmageddon.

1:19:51Daniel Mahncke:And I just, to me, there's such a huge difference between a mobile app on your phone that appeals to people that are kind of almost playing a game versus this massive enterprise system that is Salesforce and Adobe. And so anyways, that's kind of my spiel of in this world of AI disruption, not all software companies are positioned the same. And some are much more vulnerable and some are much less vulnerable. and Duolingo, in my opinion, skews much closer to the vulnerable side of things than some of the other software companies we own. And that's not me saying I'm certain that Duolingo is a bad bet.

1:20:32Daniel Mahncke:I'm not interested in shorting the company, but yeah, I don't know. It's just, it makes me a little too uneasy. And so I do hope things work out for Duolingo because come on, I mean, who could not love that little green owl, but it's a story I'm happy to watch from the sidelines. and if that very small bet pays off for you, Daniel, I'll be expecting you to buy me drinks or something next time we meet up. I will do that, but only if you ask me in German.

1:20:57Shawn O’Malley:Because I started on Duolingo and that would help my investment. Okay, we take that. But honestly, I do think going into this deep research evaluation, I did think that Duolingo would just be more expensive than it turned out to be, but I wasn't surprised to see that even in the end, I don't consider this to be a company with a mode. If you ask me what's the most likely to happen in five years, I do think this company will at least, if not being disrupted, have significantly more competition. And that's never a good thing for your investments. But all right, having said all of that, how about you tell us about next week's episode?

1:21:31Shawn O’Malley:What are the hints that you can give us?

1:21:33Daniel Mahncke:We just can't keep away from software because next week we're looking at another software business and one that has recently IPO'd and definitely skews more in the B2B platform enterprise side of things. like I was just describing than the B2C model that Duolingo follows. And I'm sure that's way too vague for anyone to know what I'm talking about. So I would just add that I decided to pitch this company after wanting to better understand Adobe's competition. I think I'll leave it there.

1:22:02Shawn O’Malley:All right. I think the last hint might've given me an idea of what company we look at. All right. So let me close it for today with a quote by Peter Lynch. He said, remember, things are never clear until it's too late. And, you know, I remembered that one and thought I would take the opportunity. And with that, I will close it for today. See you all next Sunday. 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.

1:22:42Shawn O’Malley: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. 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.

1:23:17Shawn O’Malley:Copyright by the Investors Podcast Network. All rights reserved.

1:23:43Thank you.