Title: $DUOL Stock Is Like Netflix 15 Years Ago - Here's Why It's Going To $1000
Show: Christian Darnton | Investing (YouTube)
Guest: Christian Darnton
Date: 2026-JUN-13
URL: https://youtu.be/MMgi6KeBfe8
Length: 22:06
Note: fillers (um/uh/you know/I mean as tics) removed and stutters/false-starts collapsed; wording otherwise verbatim.
=====

(00:00) Hey everyone, welcome back to the podcast. Today I wanted to discuss why I believe Duolingo stock is going to a thousand dollars a share long term and why the market is totally and utterly wrong about this business. What we can see, as you can see on the slide that I'll share on the screen, is that free cash flow per share is going up and up.
(00:21) The stock price is going down and down. And what do stocks do long term? They follow free cash flow per share. Now, it's very very rare actually in the stock market to get an instance whereby the stock has declined 78% from its peak, but free cash flow per share has continually been rising and is rising also at increasing paces.
(00:39) Often what you see and what I saw with my Palantir investment at $6.34 a share was that the stock came down, but free cash flow per share hadn't yet inflected. And the job of an investor as I've described many times is usually to anticipate the rise of future free cash flow per share, and when free cash flow per share inflects and increases, the stock price later follows.
(01:02) And so what we're going to see with Duolingo stock in my opinion is a violent re-rating. It turns out so far I have been correct on my thesis. I said many months ago now that there was precisely a 1% chance, a trivial chance — there's a greater chance that an asteroid hits you on the head than Duolingo being disrupted by AI. Why do I say that? People seem to believe that digital applications like Spotify, like Duolingo, like Netflix, like Instagram for instance have no moat, and that argument people apply it especially to Duolingo.
(01:33) But what they do not understand is that these digital applications cannot be vibe coded by a programmer in a basement, but instead actually consist of thousands if not millions of tiny iterations and improvements which stack up over time and makes the user experience much more delightful to use. We can see this actually in Duolingo's latest earnings transcript whereby they've cited what you can describe as the innovation stack.
(01:58) They highlighted several new features, tiny iterations and improvements that they've made to the platform, which from an outsider's perspective seem to be trivial and many people frankly disregard them. But when you are a user using the application, it is far more delightful to use said application with these tiny iterations and improvements applied to the platform compared to a product which does not have those characteristics.
(02:27) You notice this even in your own life when you're using, as I'm sure many of you have, when you've used applications, websites, or something that have not been specifically tuned to the end user. And thus there is friction, long delay times, and tiny features that stack up and are noticeable to the end user on the platform. This is a moat that Spotify utilized.
(02:51) If you don't believe that digital applications like Duolingo have a moat, then what is the moat of Spotify? And you have to answer the question as to how Spotify got over 60, 70% market share in the audio streaming industry despite being against larger incumbent players like Apple, like Amazon, like YouTube, and still dominated that industry despite having little less cash.
(03:14) And it stems down to the innovation stack. And the innovation stack can only arise when you have an organizational culture and structure that is obsessively focused on the end user. And in Duolingo's case, I would go as far to say that I don't think I've ever come across a company in my life that is qualitatively as strong as Duolingo.
(03:33) Now, of course, there are other businesses that have indisputable moats that are extraordinarily difficult to replicate, like Google. Yes, of course. But in the case of Duolingo, I'm stating that the qualitative moats, which are variables that I use to anticipate the rise of future free cash flow per share, are extraordinarily strong.
(03:53) You've seen in this organization an absolute obsession with the end user to a degree that is extraordinarily rare. You see in this organization a constant culture of iteration and experimentation, and we'll get back to that point later because it's incredibly important. But regardless, just for this introduction, free cash flow per share for Duolingo is going upwards and upwards and upwards.
(04:12) The stock price is going down. This is a recipe for a violent re-rating of the stock price because stock price long-term follows free cash flow per share. Okay, so that's the main part out the way. I do not believe Duolingo is simply a language app. I actually believe this is turning into a personalized AI tutor, a superhuman AI tutor that is being democratized to the world at the lowest cost possible.
(04:38) And it has the potential to disrupt what is a very static and slow and generic current traditional education system. I believe this will become as good as a one-to-one tutor, if not better, and will also be fun, available globally at the lowest cost possible. And so, whilst the market today, as we'll discuss — they were previously obsessed about the AI disruption risk, which has turned out to be totally trivial.
(05:07) So, on that mark, I win already. On that one argument, I've already won. There is no evidence at all, zilch evidence, that Duolingo is being disrupted by AI. Not an increment of evidence. That's just totally been made up. And it stems from a misunderstanding as to the moats of these digital businesses. That's the first point.
(05:25) The second point, which I am yet to be proven correct on, but I think I will be, is around slowing monthly active users. This is the top of funnel. So, Duolingo's funnel operates in a pretty simple way. You have a large V-shaped funnel at the start, and ideally you get a lot of people into that, and that then trickles down into daily active users, which are the users that are using the platform on a daily basis, and from that a portion of those daily active users are paying subscribers.
(05:51) Duolingo today only has about 12% penetration rate. Spotify, as the CEO cited in the latest transcript, has about 50% penetration rate of the users on the platform. Meaning that 50% of the people that use Spotify pay for it and pay a pretty nice price for it as well, which turns out into a very good business.
(06:08) Duolingo's management and CEO have said that they aspire to get to that level, but their approach is not to focus upon that this year. This year they've stated that this is an investment year. This is what I describe as the Bezos algorithm, the willingness to sacrifice short-term profit and monetization for longer-term dominance, and Duolingo is doing that absolutely a million percent.
(06:29) In fact, they're doing it in a way that is far more aggressive than I've ever seen before, and I love it. And so, whilst this may mean that there is a hit, and we're going through this hit currently, there is a hit in the short-term when it comes to monetization, long-term this means, if they're successful in their strategy, which I believe they will be, that this is an incredibly dominant and very difficult to disrupt business.
(06:50) So, they're strategizing looking long-term, 14, 20 years into the future, and they're deciding, "Okay, how can we make this a really, really big business?" And to do that, you're going to have to scale back some of your efforts on short-term monetization in order to basically get more users onto the funnel.
(07:08) And so, the mental model you can use here is that scale precedes monetization when it comes to digital applications, just like what happened with Spotify. You get billions of people onto the app, or millions, and then you monetize them later. So, that is the first point. Okay, so the real bear case we have with Duolingo, and I have some notes with me, is that the only serious bear case with Duolingo now is one thing, and that's execution.
(07:33) Can Duolingo keep bringing more users into the funnel? That is the only bear case. Now, that is basically the only bear case this company has. In theory, there are 3 billion people currently today that are learning math and English alone. 3 billion people in the world. That's not a small number. Duolingo has barely 50 million, just over 50 million daily active users on the platform.
(07:57) And so, they're not even at 10% penetration of that existing figure. 3 billion people just learning math and English globally. If Duolingo can penetrate a larger portion of that user base, then this is likely to be a much, much bigger business than what the market is pricing in today.
(08:14) They have a billion dollars cash on the balance sheet, no debt, free cash flow close to 400 million for the full year of '26. And it's a business that — one of the mental models that I use that you need to understand in the market is, can this organization overcome X difficult problem with increasing speed and efficacy? Now, why is that important? It's important because any business you invest in, whether it's Google, whether it's Costco, whether it's a cockroach shop, will encounter throughout their lifetime
(08:43) various problems that they must overcome in order to succeed as a business. I don't care if you're Google or Amazon or whatever it is. Google had this few months ago when everyone was saying that AI is going to kill their business. And so, the mental model that you use is, is there good reason probabilistically to believe that this organization with increasing speed and efficacy can overcome X difficult problem? In the case of Duolingo, there is outstanding reason to believe that this organization can overcome this
(09:10) specific problem, which is user slowdown top of funnel. There is outstanding reason to believe that they can overcome this problem. Why? Well, just glancing back across their track record, you notice that management has an outstanding ability to focus on one specific problematic area and very quickly solve that.
(09:30) In fact, the CEO cited that in the latest transcript. He said, we have an outstanding ability to focus on one specific vertical and solve the issues within that specific vertical. And so, that is one of various reasons that I cite to my Substack as to why Duolingo probabilistically will be able to overcome this specific problem.
(09:48) Another interesting thing to mention on user growth is that daily active user growth is still healthy. And so, management is expecting daily active user growth to be around 20% throughout 2026. Daily active users matter because that is indicative as to your retention and real engagement.
(10:10) And those are the users that are paying a damn bit of money to go on your platform. And so, the key quote from management is that we're seeing, they say, an increase in our daily active user to monthly active user ratio, which keeps increasing pretty much every quarter, and it increases again this quarter. And so, what that means is that out of the monthly active users, there is a high proportion of users that are opting in to go on to the platform on a daily basis.
(10:38) And their daily active user to monthly active user rate, I think, is on par or just under TikTok, to give you a sense as to how sticky and habitual this platform has become. That's not something trivial. Surely that in and of itself speaks towards the efficacy of teaching on the platform.
(10:56) There is so much I need to say, so my head has been in a scramble, but that's one point. So, this is important. So, their daily active user growth rate is actually improving and still growing. The monthly active user rate, which is the top of funnel, is where there are some problems. The monthly active user rate is the metric to watch.
(11:13) And management says, "The reality is the top of funnel, which is monthly active user rate, has been flat this quarter, and we would like to accelerate it." And that's exactly why they initiated this new transition last quarter because of that specific figure. Now, one thing to say here, by the way, is that all the growth for Duolingo so far, when it comes to monthly active users top of funnel, has come from word of mouth.
(11:37) The company has spent close to zero on performance marketing. And the CEO actually said in the transcript, "This is totally crazy for a company of our size. We probably should start investing in marketing." And that's exactly what they're doing. This is an app that has got over 50 million daily active users entirely through word of mouth, not spending a pound on performance marketing.
(12:03) Does that tell you anything about the business? As Jeff Bezos says very clearly, there's a quote that he uses about the importance of word of mouth and how word of mouth spread is indicative as to the utility and stickiness of a product. This is exactly what I saw, by the way, with Palantir at six.
(12:23) Palantir at six, they had about three people in their sales team. Meanwhile, you had Snowflake and Databricks, and basically about 70% of the people in the company are sales people. And so, word of mouth spread is indicative as to the true value of a product.
(12:42) And in Duolingo's case, you see outstanding word of mouth spread. Pretty amazing. So, that's one thing. So, as I say, this growth thus far has come entirely through word of mouth. Now, how does word of mouth spread occur and why does it occur? It occurs when the product is useful, when it translates basically into learning outcomes.
(13:01) And management said this very explicitly. They said, "With AI" — this is another point I'll go to later on, because there's so many misconceptions about this business — but management said that the product spreads naturally through word of mouth when we improve and iterate upon the teaching efficacy on the platform.
(13:19) And thanks to AI, which by the way the bears were saying was a massive bear case, but whatever — thanks to AI, the company now is better positioned to increasingly improve the teaching outcomes on the platform, and they cited various reasons as to how they do this. One of them was for instance utilizing voice mode instead of just manually clicking on a few buttons to choose the correct answer.
(13:42) And it turns out that implementing that across the entire application is much more conducive to retention and teaching outcomes and teaching efficacy in comparison to the more manual mode that was present before. And so if you want to learn a word in Italian, instead of you having a set of words in front of you, you actually have to verbally speak to get the correct answer basically in this gamified way.
(14:05) And so that's just one instance as to how they are continually iterating and improving upon the platform, which will translate into teaching outcomes and improve teaching efficacy, which then will later translate into word of mouth. They did say however that when they make changes to teaching efficacy on the platform — in other words when they improve the capabilities of the platform — whilst that may translate into better outcomes per student, there is a lag time between improving the teaching efficacy on the platform and then word of mouth spread.
(14:34) And the CEO even cited that because basically all of our growth has come from word of mouth, word of mouth is not something that you can click on and off as a switch. It comes when you improve the efficacy of the platform, and even then it's difficult to predict when exactly word of mouth spread will occur.
(14:55) So that's just a few things, but the point being there is really that the entire question boils down to: can Duolingo continually add more nodes to the network and monetize those nodes over time. I think they can, and that's why I think this is going to be a much much bigger business over the coming years.
(15:13) And one of the reasons I think they can is not solely because they have an exceptional track record in relation to solving specific problems of this nature in the past, but also it's because they're now focusing upon performance marketing. And performance marketing, especially in Asia and China, they're saying is a great profitable thing to do.
(15:31) And so, yeah, it's going to be pretty outstanding, really. Regardless, sorry, I'm getting a phone call. What else do we have? Asia, China is just getting penetration. It turns out that the Chinese and Asians absolutely love Duolingo, and they love learning. They're not like us in the West, where we just kind of watch TikTok all day.
(15:47) One important point as well that needs to be cited is around saturation. The consensus in the market today is that Duolingo has reached some sort of ceiling. And for whatever reason, everyone's done today, and nobody wants to learn anything. That's the bear case in one sentence.
(16:05) Management actually expressed that they explicitly denied this head-on in Q4 of '25, and they said specifically, "We do not believe we are anywhere close to market saturation." Luis von Ahn, the CEO, said that in the US, about 2% of internet users use Duolingo generally, not on a daily basis.
(16:29) 2% of internet users in the US use Duolingo on a given day. 3% in the UK, 4% in Germany. The key quote there is that he says, "Even if we assume that every country got to only 2%, which is the US penetration rate, we would more than double our daily active users." So, he then goes on to say that we don't think we're near saturation at all.
(16:48) Okay? And also, once again to mention as well, Duolingo has entirely grown in the US through word of mouth. They've done zilch marketing, not even a penny. That's something to think about. Another thing that the bears were totally wrong on this when they said that AI would disrupt Duolingo.
(17:04) We actually see some very interesting stats around how AI is making Duolingo stronger. They say that in Q1 of 2026 alone, Duolingo published over 20,000 course units, more than 10x than what they were shipping per quarter just 2 years ago. So, they've over 10x their output in terms of content in a singular quarter thanks to AI.
(17:27) >> [clears throat] >> The other thing that's interesting to say as well is that the CEO in the transcript said that these generic LLMs are not good enough. They need to be fine-tuned, which comes back to the moat of Duolingo, one of their various moats, which is proprietary data.
(17:44) Proprietary data is the most important moat ever. Maybe the most important moat today. And having that proprietary data allows you to fine-tune a specific application or procedure to a specific purpose. You can only do that with proprietary data. So, the idea that your grandma using Claude is just going to rack up a Duolingo competitor is totally insane.
(18:05) And so, yeah, just to wrap up because I've been rambling on now. Free cash flow is doing amazing. Absolutely tremendous stuff, really. Nearly 400 million free cash flow this year in 2026. Highly highly cash generative. They brought back, I think, 500 million quid in stock because they know that the market's been totally insane and it's full of people that are pretty dumb, I have to say.
(18:25) They're pretty dumb. They don't know nothing. That's just the truth. But whatever. Anyway, so yeah, you get the idea. The other point is that Duolingo literally is a monopoly on sale right now. They have zero competition in the education space. They say around 85% of daily active users across language learning apps globally, Duolingo has.
(18:48) Okay, that's quite interesting. So, the whole point of this video is to say that I think the market is wrong. And just like when I said that AMD was going to $500 a share by 2026 and everyone also mocked me then saying AMD is cyclical, it's run by this Asian woman who's a DEI hire — turned out I was actually right again.
(19:06) So, maybe I'll be wrong, and if I'm wrong, I'm going to own up to it, but I quite like being right actually. It's quite fun. Anyway, so yeah, the key question that investors need to pay attention to is whether management can re-accelerate top of funnel growth. And as I said, in theory, there is ample reason to suggest that they can.
(19:24) Not solely because the TAM is over 3 billion people globally and they've penetrated not even 10%, but also the various other reasons that I cited about the efficacy of management in this business. And so this is a free cash flow machine and it's a beautiful business. I love everything about this business qualitatively.
(19:45) Qualitatively, it's an outstanding organization, simple as that. And so the conclusion really is this: Duolingo is not a language app, it is building a personalized global superhuman AI tutor at the lowest cost, and God knows the true value of that in 10 years time. I don't know. Mark Andreessen seems to believe that it's going to be a huge business.
(20:05) I don't know. They're building a personalized AI tutor, superhuman capabilities for the world at the lowest cost. The bear case, the one bear case that is real, is around monthly active users top of funnel being flat. But that's something that I've addressed and I think there's reason to believe that that's not a death sentence for the company.
(20:27) Everything else I love. I love everything else. They have a moat, proprietary data; they have many moats in terms of the innovation stack, network effects, brand, etc. Organizational culture and structure, one of their greatest moats, just incredible. And yeah, everything is really now setting up for this stock to go parabolic in my opinion.
(20:50) I may be wrong, but let's see. I think the bears are just totally and utterly insane if you want the honest opinion. But anyway, yeah, so that's it on Duolingo. I hope you appreciate these videos, and as you can see over the past few months and years really, I'm very happy to publicize my thought process and my investments.
(21:10) And some of them may go down 20, 30, 40% in a year. I don't know. I don't know what stocks are going to do in the next 2 months. I never pretend to know that, but what I do know is that stocks that have rising rates of free cash flow per share increase over the long term. And this is a place for long-term investors to make a lot of money.
(21:29) And I know a lot of you made a lot of money on AMD. One guy messaged me today inside the private community saying that he made over 70K in about 4 months from AMD. And he thanked me for the videos and commentary on it. So, I think the same is going to be true of Duolingo. It's going to be fun to look back in 10 years' time about this.
(21:47) And management, by the way, put out a LinkedIn post, the CEO saying, "We're looking out for the next 20 years. We ain't going nowhere, dude." And so, take from that what you will. But anyway, those are my thoughts on Duolingo. If you're watching this in 2029, then you'll know if either I'm terrible at investing or good, and we can see from there. Anyway, goodbye.
(22:04) See you soon.