The Exchanges

Q matter if the, if the economy is bad, if you can't get 10 customers, it's not going to work anyways. And so I thought it was a great little blog post talking about like kind of starting a startup during the downtime, which is where I think we're about to, or we just officially entered a recession again. So I was curious, Dharmesh, do you have any thoughts on that?

A Yeah. So my position is I, I lean more towards kind of Uh, Paul Graham and the spectrum, which is, I think it's either neutral, that it doesn't matter. It's one of the things you're suggesting, uh, towards positive. Um, and the positive elements of being in a, uh, kind of downturning economy or recession is that, uh, things become available that would not have been available to a startup before, like talent. Um, it's like, oh, well, there are people sitting at meta Facebook or sitting at Google right now that may be reevaluating their lives, or they may have just been let go from a venture-backed, uh, company that raised 200 plus million dollars, right? So now you've got this, um, Talent that's coming on the market that might not have come on the market before, uh, things that were super expensive, like buying Google AdWords or certain marketing channels because you have this glut of money and a glut of venture capital. Um, yeah, I think of venture capital as a very efficient machine of turning like money from LPs into Google AdWords revenue, right? That's, uh, they're just a conduit between the two. Um, and as in a downward economy, you're gonna get less of that kind of glut of money flowing in, which drives the cost overall down for certain things that I think are important to entrepreneurs. So I'm, I'm generally net positive that endowment economy is actually a relatively go…

AI assessment note: “a downward economy is actually a relatively good time to start a startup”

Q Wait, I thought you have like a billionaire chat group where like every billionaire just kind of says the back channel of what's going on. Do you not have like a billionaire WhatsApp?

A No, we do have that, but I don't have any insider knowledge from that particular chat group. I mean, it, my sense here is that, um, you know, building large language models as OpenAI is doing is, is like supremely capital intensive, which is rare for a Software company, which is what they are. And so it's expensive. They needed access to capital. Um, I think they structured it such that it does cap the profits. I think they've done like, if you had to do that kind of, well, we're going to have to spin off and have this for profit thing. They did it well. And, uh, and I could be wrong, but Sam Altman seems Like a reasonable, rational, non evil guy. I mean, he's, he's a capitalist. Fine. Uh, and I mean that in the most positive way possible, but, uh, I don't think he was out to mislead anyone. I think he's trying to.

AI assessment note: “No, we do have that, but I don't have any insider knowledge”

Q like when you're in year five, six, seven, eight, you're like, Whoa, wow. It's really created in 1011, 12. But, um, do you think that, um, what portion of HubSpot and your success has been, well, I'm just really, I'm just willing to shut up and focus on this for 20 years as opposed To, uh, you actually being amazing or good at product and marketing and things like that?

A Um, yeah, here's how I think about it. I think the outcome is a combination of Like dreaming really big and like iterating really small and having tight feedbacks, feedback loops. So the dream really big is you have to have something that can someday translate and it doesn't have to be as clear a vision as, as, you know, Sean has kind of laid out as far as, oh, it's going to be Pixar for the new generation. Um, but you have to kind of have start with something relatively big in mind, um, at least at some level, but then it really comes down to the, the iteration, right? It's, um, I think the, and others have demonstrated this through data is that the quality of the output is almost I won't say always, but frequently a function of the number of iterations, not the quality of the, you know, the people's. For instance, one of the common examples that's thrown out is, and I've heard it with a pottery class or a photography class, but it's like, oh, one set of students divided class in half. Let's say it's a photography class. One set of students is like, okay, every day you have to submit a photo for the next 30 days while this class is running. The other one is like, at the end of the 30 days, you're going to submit your best photo, and then your grade is going to be based on the quality of that best photo. The first class is like, as long as you submit, A photo every day, you're …

AI assessment note: “quality of the output is almost... frequently a function of the number of iterations”

Q And so was, was, was Brian, uh, was he successful like you were or was he?

A Like he didn't like need, he hadn't had like an exit because he was, uh, he hadn't done a startup before. This was his first one. Um, But he also was like, okay, this is my last swing at bat, right? This is not, I'm not, this is it. This is the one that's either going to do it or not going to do it. Um, and so that, that was the reason. It's like, we weren't worried about dilution. You know, we've been negotiate the fairest value. We could, we make sure the terms were meaningful, had good investors, but we were not worried about what our percentage stake in the company was and how that moved over time. We're like, okay, if this thing does what it needs to do, uh, the valuation and all that, and the equity dilution will not matter. Like on the margin, it just, it, It's, we want it to be a binary outcome. Either it's going to be this huge, massive thing, which, um, will change our lives, uh, and have a massive impact, or it goes down and crashes and burning flames, which is highly likely, but, um, we weren't looking to hold onto control. We weren't looking at this, like, we wanted to be We didn't want to be king. We want to be rich, I guess.

AI assessment note: “he hadn't had like an exit because he was, uh, he hadn't done a startup before”

Q we can move on. But when, and the reason why you were able to do those Node Direct Reports is I think that you were the first investor. And so like, was this, was HubSpot your idea? Or was it Brian's idea? But I think you invested like half a million dollars of your own money to like get it started. And aren't you the largest shareholder of the company?

A I am. Um, and so, So the idea of HubSpot is that you have to decouple the two things. Um, so HubSpot was not supposed to happen, and here's what I mean by that. When I sold my last company, you know, I had been running it for 10 hours, uh, 10 years, and working the proverbial entrepreneurial hours as one would expect, right? It was self-funded, uh, bootstrapped. And I promised my wife, um, that I would not do another startup. I was like, okay, well, you know, the one I'm doing, I'm in, I can't not be in. Uh, the analogy, um, you know, I use this, it's a little bit like being at, um, Like a gaming table in Las Vegas. Right. And it's like, and you don't know the rules of the game. This is how startup entrepreneurship works. If you're, if you're self-funded, it's like, you're at the table and the only real feedback you get is like, sometimes the chips go down. Sometimes they go up, you're playing, you're trying to figure out the rules as you go. You don't know, but there's one like cardinal rule is that if you leave the table, Whatever chips you have on the table, they're gone. Every now and then the house will come up to you and say, Hey, would you like to cash in your chips? They're worth X dollars. And this is the position you were in. So I was like, okay, would you like to sell your company? Here's what someone's willing to pay for it. And if you miss that opportunity, you hav…

AI assessment note: “I am. Um, and so, So the idea of HubSpot is that you have”

Q Amazing. Has that, uh, changed you? You've probably become a worse person. You seem like a pretty good person. I feel like if you listen to us for 60 straight days, there's gonna be a part of your brain that's good, but a part that gets, like, corrupted slightly.

A You know, it's, I think that's true. That's an astute observation. Um, it's, you know, we just have, uh, I won't say, like, completely divergent paths. I'm just a, yeah, it's been good, though. I'm, I'm, I'm a believer in, like, just these kind of orthogonal things. That's like, oh, this is not stuff I'm like, you know, looking at retail businesses, or like, you guys have a wide array of stuff. Um, but I think that kind of thing, like, builds a different muscle group, and I think it's been, it's been useful for me. Um, I'll, I'll put it to you that way. It's, I think we're not, uh, trying to separate at birth by any stretch. We have things that, you know, we have in common, but, um, there are things that I'm just, And it'll come out in the pods. I'm a weird dude.

AI assessment note: “I think that's true. That's an astute observation.”