The OpenAI Memo: Why Invest? Is it too Late to Catch OpenAI? Are OpenAI's Models Truly…

Thank you so much for having me, Harry. I’m really excited to be here. I’m a big fan of the show.

Started my career like many folks. I studied business and accounting in undergrad. I went to a big bank out of school, which was Goldman. And in a lot of ways, going to these big banks is like doing an MBA. It’s a two year program for the most part. And if you go there, you wanna on these really complicated big companies. And most of what I worked on were that was that. It was things like AT and T and Verizon, or actually Dell announced it was gonna merge with EMC at the time, and so they shipped us down to Austin.

And we worked on carving out a bunch of software companies to go finance that big transaction. And after doing that for about eighteen months, I ended up wanting to do something different, and I got staffed on Flipkart, which is this e commerce business in India. And lo and behold, the largest shareholder was Tiger. And so I worked pretty closely with the folks at Tiger on Flipkart for about three or four months. And as I was gonna leave Goldman, the stars kind of aligned, and I ended up joining Tiger to go work really on private company investing for the Sky Leaf Fixel.

As luck would have it, the first company he handed to me was a software company. And so I spent the first three years of my career with him looking at lots of software companies and trying to really find what were the next generational big companies there. Lee ended up leaving Tiger, and I was also chasing Airtable at the time. And the folks at Thrive, Josh and Miles, led around at the company. And what I’ve learned from Thrive is always be recruiting. One coffee chat led to another coffee chat, and about four years ago, I ended up joining the team at Thrive.

First of I feel really lucky to have started my career working with Lee’s. I credit a lot of where I got started to working closely with him. Tiger also is in the news a lot more recently, but the firm’s been around for more than twenty years. And if you look at how it got off the ground, it is very much in these kind of hedge fund roots. Chase Coleman, started the firm, he was in his mid twenties. He was really young. He had this hedge fund mentality mindset coming out of Tiger Management post the dot com bubble.

And the way we thought about investing in companies was very financial. Look at the p and l and understand deeply how do the numbers tell the story of what the business does, and how does that ultimately make a good business a great investment. I learned a lot just from the financial rigor of doing that. But what’s interesting is when you contrast that to what we do at Thrive and how Josh has built the firm, we started from the roots of where Josh was, which was a founder.

He was the same age as Chase when he started Thrive, but ultimately, he was trying to build a startup. And so the mindset was very much how do you empathize with the entrepreneur? And we focus much more on the product and the customer, and ultimately, how does that manifest itself into a business. At the of the day, we’re looking for the same thing. I was looking for a Tiger, which is iconic market leading companies that are gonna generate great returns. I think every investor wants that stuff.

But to make it tangible, at Tiger, I would have asked the question if you were the founder and I’m trying to figure out who your customer is, I would say, who’s your ICP? Or, like, who’s your core customer? That’s a very investor lens. At Thrive, the way we’d ask that same question is, if I’m a SDR on your sales team and I’m trying to qualify a prospect, what am I looking for? At the end of the day, it’s really the same question. You could even argue that the way we do it at Thrive is less efficient, but I think it shows a much deeper level of empathy with the founder, and it’s a very different mindset.

Everyone’s motivators are to actually unpack and understand. We talk about this a lot in founder assessment. And I just like you’re saying, I believe that we’re all a byproduct of these accumulated experiences. And I’m sure have had a lot. I grew up in Michigan. I’ve got three brothers. I’ve got two parents. They’re divorced. And so there’s lots of things in that I’m sure manifests into who I am today. But I’m sure, like, a lot of people that are listening or listening to your show, I’m trying to figure out who I am, and I’m trying to figure out how that impacts the decisions I make and how I react to what I’m seeing as I go through the world.

And I think for me, personally, the motivator is just maintaining this mindset. How do you maintain the steepest slope possible that you’re learning on? One of the things I’ve grown on a lot over the last eight years of my career is I think when you start, you take a very solo mentality to doing things. Because it’s the ship on your shoulder. You want to accomplish things. And so you do that really on your own. But over time, I’ve realized your friends, your spouse, your coworkers, the people you’re around and so forth, all of those folks, you can compound your learning curves together.

And so for me, I think a big part of where I’ve come from and where I’m going is trying to make sure I utilize all of these people or ecosystem to really compound how I learn in that slope. And, ultimately, that’s been, I think, the most powerful thing for me over the last ten years of my short career and really my entire life.

We talk about this as a team from our culture. A lot of when you join the Thrive team, the focus is how do you build trust with the organization. And trust is one of these things where in great organizations, it can be given by default. In many organizations, you have to earn trust. I think we have a culture in Thrive that definitely people get a lot of trust by default because of how small our team is and how autonomous our model is. For me personally, I actually had really great investors that you guys would know told me something that was interesting, which is trust with founders is actually just being very predictable.

People want to not feel like they’re getting surprised by how you’re thinking. They want to understand how you think, how you’re gonna react, and feel like they understand you. And that ultimately kind of breaks down these boundaries between people and allow you to have some mutual trust and empathy with each other. And so I think with founders in particular, when we talk about building trust, it’s like any partnership. You’ve got to increase the reps, get in the water, in the trenches with them, and they have to understand how you think.

And ultimately, I think you’ve got to telegraph how you’re going to make decisions with them and make sure that they get there alongside of you. They don’t feel like you’re superimposing things top down on them, and they don’t feel like they’re getting surprised at a left field because if they feel that mutual level of partnership with you, I think it’s really actually pretty easy to have trust with founders.

I think part of it is you’re a byproduct of your environment. I think for me, I’ve gone through a bunch of different waves in my career, and that’s helped me understand what volatility looks like, feels like. And so I don’t think you just become levelhead as a person. I think you kinda build into that psyche over time. Someone said to me once, which is a line that I’ve been saying to the team a bunch internally, is things are never as good as they seemed, and they’re never as bad as they appeared.

And I think just keeping in mind that the rest of the environment around you does react to these peaks and troughs of your emotion in the market and the volatility. And ultimately, in good times, people overextrapolate. In bad times, people underextrapolate. If you maintain this kind of more balanced approach, I do think it helps you hold more clearly what are we ultimately looking for in solving around for a given investment or person or situation? I’ve just found that it’s not productive to necessarily get caught up in the emotions.

You gotta try to think clearly. And if you can remove that noise from the volatility of the emotion, it allows you to focus on the core a lot more easily. But that said, think you do need to trust your gut, and you do need to be emotional to react that way. And so I wouldn’t say it’s all about being just this kind of robotic level headed person. You need to figure out what are the types to do that, and what are the types to lean into your instinct.

This is the question I think every investor is probably asking themselves right now or certainly the ones I talk to. And it’s the quintessential question of, like, do you sit on the sideline and be patient and stay disciplined, or do you jump in on the gold rush? Because if we don’t get in now, we’re gonna miss all the seed in Series A companies that create tens of billions of dollars of value, which understandable. I think there’s a reason why we’re in the business of being optimist in venture capital.

And I think for us, and really Thrive in general, we try to get into the psyche of why are people so excited and less about chasing the next deal and more about what’s the core value to the customer. What’s the product? What’s happening? Crypto obviously had an amazing run and I think has kind of pulled back a decent amount. We’ve been comparing and contrasting how much is AI potentially the thing people are latching on to, like, crypto maybe for the marginal investor a couple years ago versus how much is it real.

I think just to use crypto as an analogy, it was very ideologically driven. Bitcoin came out after a financial crisis, and it was all about take this centralized financial system and make it decentralized. Allow people to take their money anywhere. Don’t let the government insert themselves in the financial ecosystem. And, really, like, you had to believe in that as a concept to take all of the trade offs of using crypto. As a consumer, was a pretty bad user experience. And so when we think about other hype cycles, I mean, you go back to even the .com bubble, when that ran up, Microsoft, Intel, Cisco were the top three technology companies in the peak of the dot com bubble.

They were all infrastructure related in some way, and they were 50% of 3,000,000,000,000 in market cap in the hype cycle. And so people really latch on to these companies, and they can run for a long time when people believe. And speculation fuels more speculation. And so at the end of the day, you know, if you invested in Microsoft in 2000, it would have taken fifteen or twenty years to break even on your investment. And so it’s hard to really time and predict hype cycles. And so we ask ourselves, what’s the core primitives of AI?

We don’t know them all today, but we can try to figure out what some of them are, and we can really try to understand products. And there’s obvious benefits to those things today. Obviously, all the search that’s coming with ChatGPT and the LLMs, but also just, like, companies that allow you to trigger actions without having to do 20 clicks. Or, I mean, for your show, I’m sure you’re using AI in some way to edit or create content or something like that. The marginal cost of content production has come down a lot with these tools.

I think when these are the foundational questions you’re asking because of technology shift, it forces every company to think about what could happen to their business over the next five years that would be really disruptive, or how should we be thinking about re disrupting ourselves to ultimately take advantage of the platform. And all the boardrooms I’m in and many of the founders I work with are thinking about that question.

Yeah. We really first started focusing on the company maybe eighteen months ago because there were a number of startups that were really rising with software products with these kind of tools that you could use to do marketing copy, or you could create blog posts with them. I think many of these companies, we spent a bunch of time with them, and we kept coming back to the fact that they were thin user experiences on top of this cool thing, which was this model, which really no one talked about, at least not mainstream, eighteen months ago.

And so that triggered us to go spend time with the company. And ultimately, maybe in the classic Sam kind of way, the way we kicked off the round was he did almost like a closed demo with lots of investors on a couple of calls of the technology they were working on and ultimately in GPT-four. And we were on that call like many other investors. And I think I remember getting off the call, and Josh was also there. And we looked at each other. We’re like, wow.

Like, the way companies are gonna get changed is gonna be incredible. Software is gonna look so much different. And we were reflecting on it, And that kind of transformation is just so rare in our job, company after company, and so many things seem marginal. But when you see these things that are discontinuous or seem so different, we trigger this inner instinct to pause, focus, reflect, and ultimately, we have to spend more time learning about the financials and the products they’re releasing and the customers, but that’s what encouraged us to really lean in.

I think there’s a lot of people that said no. As much as I like to think we want a really competitive deal, it was not obvious to everybody. And even now, I don’t think it’s still obvious to everybody, even folks that use the product.

I think a lot of investors get tripped up on trying to be so precise on TAM and market and defensibility and the moats around businesses and trying to map that all to price. And those are so important in the investment decision making process. But when it’s so early in a technology cycle like this, it almost is a little bit more of a venture mindset where there are gonna be 50 reasons you can say no, and we’re not gonna have answers to every question. But we need to really think about the things that can go right.

We’re not talking about building the next unicorn or decacorn. We’re talking about disrupting search or Google. I mean, that’s a trillion dollar opportunity. Sure. I can’t put a TAM around ChatGPT, but I can tell you, like, we’re not talking about a small prize at the end of the day. And so I think when we say it’s not obvious, people aren’t thinking about it in this lens. They’re thinking about it in the box of, what does this chat interface do, and oh, those use cases aren’t that valuable.

And I think it does take a higher level of creativity or imagination to ultimately think about the world that way.

Yeah. As these new technologies scale, it’s so hard to be precise about a TAM. And so more of what we got into the psyche of thinking about is if you looked at other big technology movements, how did they scale? The iPhone went from a million phones a year kind of post launch to a 100,000,000 in five years. AWS took six or seven years to get to a $100,000,000,000 of run rate, but they went from a 100,000,000 to 10,000,000,000 in six or seven years. Google went from nothing to 10,000,000,000 in the first six years of monetizing.

Obviously, this is rarefied air we’re talking. I mean, these are three of the most transformational companies on the planet. But the technology, if you really think about it and the zeitgeist it’s captured, it is of that elk of the transformation it can have of the world. And so for us, it’s less about thinking about the exact TAM, and it’s more about if we think about what the world’s gonna look like in five or ten years from now, how are we gonna look back and say, wow.

I can’t imagine the world without this kind of thing. When you have experiences like that, that’s what gets us out of bed. These are the kinds of things that create new categories, new companies. That was what a lot of the discussion was. And maybe that sounds less precise to people, and they think it’s a little bit two finger in the air. But I think for us, a lot of this is instinct mapped with why it’s a really great business.

I mean, obviously, these are my opinions, not the company’s, just to make that clear to everybody. I don’t think I would characterize OpenAI as one model to rule them all, and obviously, there’s all kinds of talk right now about it being closed source, and the model should be open source for the value of the community and all of those factors. But I’d say, actually, if you give the company credit for this stuff, which they’ve released, they have done a lot in the open source and contributed back to the community.

CLIP was a model that they put out there that helped a lot of the image generation open source models. Whisper was for audio. They open sourced the inference framework called Triton on top of GPUs and CUDA to ultimately drive more efficacy and scalability of inference. And so they have done a bunch of that side. They obviously have kept the core GPT model behind an API and paywall. I think thinking about it as this entirely closed ecosystem, in my mind, doesn’t really give credit to what the company has done on the open side.

Founders we at least talked to are choosing their model has quickly changed from who’s got the biggest model and who has the lowest cost model to these new dimensions, which is if you’re gonna build a company today, Harry, you don’t wanna have to think about the scalability of it or the reliability of the infrastructure or supporting all of that stuff. If you use an open source model, you’ve gotta go figure out all of these infrastructure problems. And so now if you talk to OpenAI and you think about what they’re spending time on, it’s how do you support this onslaught of scale that has come into their system, and that’s infrastructure engineering problems.

It’s the same thing we talked about in cloud ten years ago. I’m sure there are banks out there saying, you know, we’re not gonna move to public cloud because we can run all our infrastructure ourselves and do it really well. You would never have told a startup ten years ago to go run your own infrastructure if Amazon was there doing it for you. I think we’re kinda getting into this world now where, sure, there are lots of open source models and they can be used for great things, and some companies will choose that path.

But by and large, for the vast majority of companies, particularly ones that have scarce resources, which are many of the ones I work with, I don’t want you spending time building infrastructure and scaling open source models. I want you just getting your product out there with lots of value to your customers. And I think OpenAI is quickly figuring out the cost curve, which is probably the biggest competing interest against that. They put out 3.5 turbo, which is, I think, at least one, if not two orders of magnitude less expensive than before.

And in my opinion, I’d bet on these ecosystems to scale the infrastructure, and ultimately, that’s becoming more important than decision factors.

is a question we talked about a lot. I think part of it we’re already seeing the evidence of, in my opinion, where commoditization of the core model output is not really what people will make decisions on over time. You know, I’m sure there are lots of companies that give similar search results to Google, but the reason these companies get built up to scale is because the ecosystem around them grows. And you’re even seeing this. They had launched plug ins about a month ago, and ChatGPT has taken off and taken the world by storm.

I think it’s the fastest company in a 100,000,000 users ever. And so even if the raw model does get competed against, there are maybe companies like Google or Facebook or Microsoft or some of these startups that can compete on that. I do think, again, the dimension by which people are gonna think about this in five years from now looking back isn’t gonna be about the commoditization of the model and the raw output. It’s gonna be, oh, the ecosystem around the model has become much more robust such that you can do a lot more with the model than just get text outputs.

Or the convenience of putting these things together such that it’s not just text. It’s now text and image or video and audio all through one interface. That’s complicated. And, again, it’s not to be dismissive that the model doesn’t matter in terms of accuracy or size, but it’s to say that people are gonna optimize for companies, builders are gonna optimize for the inputs that allow it to be much easier for them to build and much easier for them to capture this new channel of hundreds of millions of users that are flooding here.

And that is not about commoditization model. That’s about ecosystem and the kind of classic things that make great businesses.

Elephant in the room is big tech companies. This is not something where they’re sitting flat footed. Facebook, Google, even Microsoft. I know they’re a great partner in opening that deal.

I think it’s hard to be dismissive of these companies. If you think about where the best talent in AI is right now, I think it’s OpenAI, and then I think everyone would tell you it is Google and Facebook and Microsoft and Amazon. And maybe there are folks that have kind of dripped their way into the startup ecosystem, but by and large, the talent is so clustered in these big tech companies. And so I get it, that they’re tripping over themselves trying to figure out how to navigate these giant organizations they’ve created.

But let’s not also kid ourselves. We’ve talked about this maybe outside of the context of OpenAI to up level it to AI. We work with lots of startups that compete on things like presentations or content creation. What’s scary to me right now is if you’re a startup, large companies are shipping product. The kind of canonical examples were that, oh, the big incumbent can’t move, they’re slow footed, and you got years to execute before they do something. Microsoft, 200,000 person company, they’ve shipped AI in Bing, AI in PowerPoint.

They’re rolling out the other products. Adobe’s got the content creation in their product already. Even the big like Notion. Those folks, I’ve been so impressed with how much they’ve shipped in their product so quickly. And so if you’re a startup, I get that you have the advantage of speed, but you need a multiyear execution window to be able to take advantage of all that product you can ship quickly. If these big companies are able to ship this quickly, they’ve got so much distribution, so much talent.

I just think it’s gonna be very hard to compete with them.

I think it’s too early to call. Maybe that’s a cop out answer. But to put it in context, if you go look back in time at these big technology cycles, they are great moments to create new categories. And so maybe that’s a better lens to look at it through. Where will new categories get created where being an incumbent or being a startup won’t matter? And if you look back at the.com, Google, PayPal came out of it. Google about finding information. PayPal about bank online. Social came after.

You know, Facebook was connecting all these people online, and Twitter was the town square for broadcasting information. Know, mobile, the answer wasn’t Salesforce was gonna build mobile CRM, and that’s where all the value was. The answer was you put a computer in everyone’s pocket. DoorDash and Uber came out of that because there’s geolocation, the ability to have connectivity. In my view, what we’re looking for, if I was to answer that question, yes. If you’re going toe to toe right now with an incumbent on their home turf and their shipping, I think it’s a hard bet to take the opposite side of, you know, in our business.

But if you’re creating something that’s totally a different user experience that no incumbent has today, I’d bet on startup 10 times out of 10 in that case.

I have to believe it’s gonna accrue mostly to the application layer. I think if you think about infrastructure companies as platforms, you know, if you think about the software market as the relative value of AWS and Microsoft and Google Cloud versus the software companies built on top or all of the internal software tools built on top, I think it’s probably an order of magnitude to one. You know? So I just have to believe where value gets created to end customers is where it will get captured.

And the nice part about the infrastructure business is that there are effectively toll roads on that whole ecosystem, and applications might compete more aggressively with each other. And OpenAI or Amazon or the infrastructure provider might be able to clip a coupon on it that’s really valuable. We might capitalize it at a high multiple, but I have to believe that there’s gonna be a lot more value in the new experiences and applications than the infrastructure.

It’s hard. And I think at the early stage, you know, we think about investing in founders first and foremost. You want extraordinary people in really great sandboxes, and those folks will be able to figure out how to pivot and adjust and iterate to take advantage of opportunity. In this environment, if you told me, I rather be in the sandbox with the great founder experimenting and iterating or sitting on the sidelines? I’d rather be iterating even if things changed a bunch from underneath you at the early stage.

But that said, I think you can’t make every investment without knowing what’s gonna happen. So some things, there is an obvious evolution that makes your business less interesting than it was six months ago. If you think about where we were on infrastructure companies nine months ago, there were all kinds of companies branding themselves as MLOps or these end to end solutions for AI. Now in this environment we’re in now, there’s specialized companies that are doing individual components of the infrastructure stack, whether it’s laying chain on connecting all of these tools and stitching them together or these vector databases that are allowing you to store information.

And so things specialize. You know, our reaction, it’s not that hard to say, I have a friend and mentor of mine that I spend a lot of time with who says there’s only two questions that matter for a company: Who is the customer, and what is the product? And you’d be shocked at how many folks can’t really answer that question with clarity. And if you can answer that question with clarity, everything else about the business stems from there. And so if you’re really answering that question, you’re struggling because you think in two weeks from now, there’s a chance that the customer doesn’t care or the product won’t have any value, that’s hard to build a company around.

And so in any of those environments, I think you’d rather just wait for that there’d be more clarity than try to keep firing investments into that market.

I don’t wanna speak for the company on anything because I know this is a topic that’s out there in the mainstream. I think regulation will come to AI, and it will be necessary. I don’t doubt that at all in my mind. I think it has to be done in partnership with the companies that are building because, you know, regulation for the sake of regulation is not gonna be what solves the problems people are concerned about. You’ve gotta go understand the nuances. You have to understand the technology.

And so I would hope that other companies building an AI are also working with regulators to understand what is the technology and help educate them to get to the right decision. I do think there’s kind of some negative stigma going around OpenAI on regulation, security, and safety. And in my view, I think they’re being pretty proactive. As an example, on GPT-four, we saw the demo in the fall. They didn’t just release the product then. They took, I think, four or five months to test and learn about safety and the edges of the model and then ultimately released it to the world.

And so they’re not gonna get it perfect every time, but I do think trying to let people build it and understand it is important, and poorly designed regulation on a technology this early is not gonna be effective for anybody, and we’ve gotta find the middle ground.

This is a solvable problem. There’s information out there. There’s experts out there. Regulators, I’ve been impressed with how up to speed they’ve gotten on crypto, and they’ve really taken time to learn about it. And sure, they’re gonna make mistakes, but ultimately, that gap is solvable, and it’s gotta start the dialogue between the people working closest on AI and these regulators. And it might take years to get to a solution, so people gotta be patient. But this isn’t an unknown. This is something that you guys can work together on.

We certainly had a discussion, a heated discussion around it. We have a very team oriented firm, and so we disagree and commit once we make decisions. And we make decisions to this team, we live by that. No one person makes a decision. And the price here was high. Just on an absolute basis, it doesn’t matter what company you’re investing in. At the prices that this round was done, you’re talking about very upper echelon type outcomes to justify good returns. But I kinda come back to the intangible.

It’s really hard to understand the pace of adoption of these major technologies, and any numbers I would put on paper for you would look insane. You would look at them in a spreadsheet and tell me there’s just no evidence to support this. But then you look at these iconic technologies, and the great ones all follow that insane curve. And so the balance we try to hold, I think this what makes Thrive a really special firm, is we’re able to kind of separate ourselves from the kind of quantitative rigor that we rely on for a lot of investments we do and hold the attention of what could go right, think creatively, understand how this could look like the most transformative things.

And if that happens, again, you know, we’re talking about search potentially getting disrupted or something that enables workflow automation for hundreds of billions of dollars spent on, you know, different jobs and categories out there. And even now, and we didn’t know this at the time of the investment, but with ChatGPT Pro and the traction that’s seeing, maybe that won’t be the durable revenue engine for the company over time. But the velocity with which that’s ramping you know, Karim, my partner Karim, has a way of saying this.

The scalability properties of what they have built are so good that we’d rather bet on the compounding upside of scalability than sit on the sidelines because once it’s clear, it’ll get priced up very quickly. And so we ultimately decided to lean into that.

throwing it out the window, and it’s more about putting it in context. A financial model is a tool to help you make a decision. You need many tools to make a decision. It’s less about is that the only thing we use and we throw it out the window, and it’s more about what’s the weight by which we put on the financial rigor and model as the tool to make the decision. And I think for us, there aren’t that many iconic companies that could create it. And so no iconic company, at least that we’ve been a part of, was it clear and obvious in the early days of that technology.

They always get priced well ahead, and they look cheap in hindsight because they defy the gravity of the model. And so it’s kind of the quintessential humans think linear and the best things happen exponential. We have to understand, and going back to being level headed, when do our psyche or heuristics break down in an investment decision making process? And when they do, we need to compensate them with a different tool in our toolkit.

think we had the same level of excitement that we have in most investments. We don’t have a culture where there’s this kind of sharp elbowed mentality or people feel on edge for making investments because prices are high or checks are really large. We have a culture where we support each other, and we have a growth mindset and learn. And so if this ends up not working out, I don’t think we’re gonna look back on this and put all the onus on one person or two people that made a decision.

I think we’re gonna look back on this and say, we got there as a team. It was the right thing at the time, but here’s the learnings from it, and here’s how we’re gonna adapt our lens and course correct in the future. It is intimidating, and it’s nerve racking sometimes to write large checks and investments. But in the right team and culture, this is how we enable ourselves to make these kinds of transformational investments. And frankly, I think there are a lot of firms that could not have done this because their organizations are not set up to make these kinds of big decisions.

I think the thing that enables us to make this decision is we have this single small team that thinks extremely creative about how the world works, and we are ultimately trying to concentrate our investments in the best products that are out there in the world. And so when you simplify it down to that, and we allow this autonomy and focus to happen on a single team across the entire investment cycle from early stage all the way to growth, we are enabled to go find these kind of iconic companies.

And we’re young in our career, obviously, but we also we can appreciate when these kind of transformational technologies come around. And certainly, was the most transformational thing I’ve seen in my career. And being able to lean into that and not be afraid that if it goes wrong, we’re gonna get fired. Or if it goes wrong, we’re gonna be pushed down in the firm. I think having that psychological safety, that enables a firm to do it, and I think we have that at Thrive.

It’s actually very clear. The thing I’ve developed the most on in my investing mindset is this deep empathy for the customer, trying to really think deeply about not just what’s the product and trying to write that down on paper, but really to understand how the business is gonna get built and mapping those nuances to who the person is, what the product does, how does that manifest itself in the business. Do you have a sales intensive product that’s gonna require people to be kinda constantly out there and on edge and with their customers?

Do you have a product that’s more middleware, so you need somebody who’s gonna be willing to grind it out and not be in the limelight? Do you have a product that requires a lot of creativity? That means you need to set up your org to be creative. There are certain org structures that promote that more. And when you think about where you can develop as an investor, think where I have developed the most is continuing to understand the connectivity between this rigorous financial lens, very much of where I started at Tiger, to how do you build a company.

And at that intersection, it’s really hard to get to clarity on, but when you do, it is clear. And I think the best companies have very simple explanations. It trickles down in the company. The CEO can articulate in one sentence, but the manager, four rungs down, also can articulate it. And that means that manager knows how to go left when they should go right, or they know how to make that decision and communicate it to their team. And so having appreciation for that, I think, is where I’ve grown the most on these investment decisions.

And, again, it’s hard to quantify, but I think it results in just, like, a deeper empathy for what a great company looks like.

lots of mistakes. The one that I think stands out the most to me, or at least comes back to me a bunch in my psyche, is when I was at Tiger, I flew out to Sydney and spent a bunch of time with the Canva team in person. And obviously now people know about it. It’s a remarkable company. At the time, it was much smaller than it is now. And at the same time, we were so focused on investing in enterprise software as an emerging category at Tiger.

And so once we were spending time with Canva, I very much and this is early in my career, but I very much let the pattern matching and the DNA of what I was thinking about of a great enterprise software company creep into us looking at Canva. And the learning, I think, just to distill it down to something, is every company is unique. Even within the bounds of enterprise software companies, they’re unique. And we took this lens of what a great enterprise software company was. We’ve retrofitted to Canva, and we said, okay.

Well, the churn looks a lot higher than what grade looks like, and the engagement looks very whimsical relative to deep workflows and integrations. And so we shouldn’t value this like a highly recurring software business. We should value it like a consumer subscription business. And there’s merit to that. But ultimately, because we were so much in this one dimensional mindset of enterprise software, we missed what was so special about the company. The learning for me is just you can’t walk in biased about what you’re looking for out of the metrics of a company or what the product should result in metrics.

You should walk in and try to have a very open mind and say, do these metrics explain the qualitative of the business that I’m so excited about and reinforce why it’s special? And if you think those things are true, I think then you should lean into it a lot more and try to really get to the guts of could it be a great investment? And I see this a lot with investors. When you get too pattern matchy on why companies should be the way they are, I think it leads to a lot of mistakes.

You’re only as good as your next investment, Harry, and I’ve still got a lot to prove. So it’s hard to say we’ve won. But I do think an important part of winning and just understanding how to embrace kind of this culture of making great investments is you’ve got to figure out how to be authentic to yourself and map that to how you go win. You know, what a founder might be attracted to for me might be different than you, Harry. It might be different than the next investor that’s listening to this podcast.

And so you can’t just apply someone else’s style and say it’s gonna work for you, and you’re gonna win. You gotta try to figure out how to map what’s really authentic to you and evoke that emotion in the entrepreneur and get them bought into why that could be a really fruitful partnership for them. And it’s hard to do that. You can’t do a lot of those things at once. You can treat it transactionally. You’ve gotta invest the time. There’s no substitute for the time. The things I look back on that were the most exciting and rewarding for us to be a part of, we’ve been able to establish that kind of ground with whoever is most important in that process or the team.

And when you do that, I think those things become obvious. And the decision is less about all these other variables in the process, and it’s much more emotional because they wanna work with you.

The short answer is just reading anything. I think for me, I like to try to subscribe to lots of different angles of reading. So sometimes it’s podcasts on topics. Sometimes it’s newsletters. Sometimes you go find the blog focused on the developer to go learn about the thing. Most importantly, like, learning about the history of things is really important. All this AI stuff is happening right now, and some of where I’ve been reading is trying to trace the key figures from all the way back to, like, nineteen eighties in AI to today.

And so for me, it’s really just about variety. Simulate the personas, simulate different people that are in the ecosystem, and go learn about them. And wherever you can find content on that, I’m willing to read it.

Harry, I’m only investing in Thrive. There’s no other answer to that.

We’re concentrated. We only invest in ourselves. I can’t give you one, Harry.

firm, can you

I can. I don’t know him personally, but Nat Friedman, who’s now focused a lot on AI and I think has his own fund, I really respected some of the stuff that I’ve seen him go do, and I would be certainly interested in putting money in his fund.

This is not for the rapid fire, but one thing that’s become clear as we’ve kinda gone into this more difficult environment to operate in is in the good times, we probably overattribute to teams and products how good they are. And in the bad times, you can’t just blame everything on macro, but I think it makes you reflect on the overattribution you probably did in the good times. And so one thing I’ve changed my mind on is you gotta be more balanced about how much credit you give to the momentum of a company from the market environment versus the actual execution they’re doing and know that there’s the balance there.

And good execution doesn’t always mean that it leads to great momentum. Sometimes great momentum is also influenced by these market environments or variables that are harder to quantify.

Crypto stands out. We talked about peak market cap in .com. I think crypto’s peak market cap for tokens was, like, 3,000,000,000,000, somewhat similar. The most iconic company that people talked about was fraud. A I think, like, we’re gonna look back on that, particularly from an investor lens and say, there’s these hype cycles that lead to massive speculation, and sometimes even the things people think are most real are just not.

The person that stands out that I’ve learned so much from is Eric Fisher. He’s on the Benchling board. He’s on the board of this company I work with called Airplane. Eric is, I think, just an incredible blend of has the operational instinct rigor, but also is fun to be around, and he’s able to land his messages in a really effective way with entrepreneurs. And he also just has a really great balanced perspective on being commercial and understanding how all of that works and replaces the strategy of the company, and I found that perspective to be something that I’m continually learning from as I’m listening to him.

And when you find those kinds of people, I think you just want to surround yourself with them.

There’s so many lessons. It’s hard. Josh is an amazing person. This is not about investing, but I think with Josh, biggest lesson is really that, like, to be successful, you don’t have to compromise on all of the things that are important that are not your career. Your family, your friends. Josh, one of the most amazing parts of being around him is his warmth and empathy and his priority of his family over everything else is so obvious when you spend time with him. I’ve respected that so much about him, and I think it’s even changed the way that I prioritize how I run my life.

Hopefully, we’ll be talking about some amazing AI companies we invested in that created lots of value for us both. But I think, you know, we have a lot of ambition at Thrive as a firm, and I hope to be a big part of us building it and ultimately going and backing some of the next transformational companies, but also building our team and maintaining this amazing culture that I think we have and attracting some of the most talented people that wanna go invest in these kinds of companies to come work with us.

And so I hope if we talk again in five years, we’re talking about those companies, we’re talking the people on our team, and ultimately, we’re really excited about all that stuff.

Thank you, Harry. It’s really fun to chat.