Dave Girouard - Disrupting Credit Underwriting
This episode is brought to you by Stream by Mosaic, a product that is integral to any fundamental research process. Stream has developed an extensive library of expert interviews that cover a variety of industries. https://streamrg.co/BB features over 300 expert interviews per week. 70% of Stream's experts are found exclusively on https://streamrg.co/BB. Visit https://streamrg.co/BB today for a 14 day trial.
Album art photo taken by Mike Ando.
Thank you to Mathew Passy for the podcast production. You can find Mathew at @MathewPassy on Twitter or at thepodcastconsultant.com
Please leave us a rating in your favorite app store!
+ Transcript
Bill: Thrilled to be joined by Dave Girouard, today. Dave, we met because you listened to the episode with David Gardner and commented on it.
Dave: Yeah, I thought it was a great episode. I hadn't heard either of you before. So, I happen to just hear an episode where you interview David and I thought it was fantastic. Anyway, it was inspiration for me to reach out to you.
Bill: He's got an infectious personality. He's so positive.
Dave: Yeah. Amazingly, I haven't had a lot of interaction with Motley Fool for their long history for whatever reason until they started to have things that we're writing about Upstart. Of course, I started pay a little more attention to it. He and his brother, pretty amazing guys.
Bill: Yeah, I believe it was on Patrick O'Shaughnessy's podcast, but maybe not, you had mentioned that in public markets, the shareholders can choose to be your partner whenever they want and they can come and go. I think that the foolish investors that listen to this are not those type of people. They're the buy and hold type. So, that's something that I really respect about how David and Tom have taught 'The Art of Investing.' Still trying to learn it myself.
Dave: Yeah. Hard to do. It's very emotional game, but those guys clearly have figured something out and made a great business from it.
Bill: Yeah. Do you want to give people a brief background of where you came from, and then the Upstart story, and how it got started? We don't need to go through everything, everything, but I do think hitting on you were at Google, and then how you came up with the idea, and I don't know if you still own liquidgrad.com, but mention that.
Dave: [laughs] Yeah, very briefly, I'm an East Coast native. I grew up outside Boston, but came out to California in the 1993. Pre-Netscape as I like to say. I got into the technology world really first with Apple and really a long time ago, frankly, before Steve Jobs even came back to the company, if you want to talk about long time ago. Then, found my way to another startup for a bunch of years and then I did end up at Google starting in 2004. Quite a while ago now. Any case, I, for eight years was at Google. I built what's now their Cloud business. Without belaboring that built into a fairly large business, but under the domain of a mega giant business. So, it was always what I would call other.
When we would be reviewed at Google, the Cloud app business, what we call Google Apps was always in that other category. Some sense it still is. Even Cloud is still modest compared to the crazy ad business. In any case, about a decade ago, I had been there eight years. It was great, amazing experience, but I got that edge, I was in my mid 40s at the time and said, "Hey, I've never actually started a company." I definitely had this thing in my head, where I was suddenly open to the idea of, "Hey, maybe I could start a company. Maybe I didn't need to be Google. I could actually do something outside of Google." A bunch of period of time where I was just thinking through ideas and got very interested in access to credit or access to capital and how the system worked.
For really, just honestly, conversations I happen to have some people that just struck me as it was a really broken system, one that was really limited, hadn't really been affected in any way by the Internet, meaning, it existed as it did in 2011, when I was thinking about this and pretty much looked it did in 1992. Just felt like an opportunity and decided eight years in to Google, I was going to leave and found this company, and my wife was supportive. We've done pretty well at Google. We weren't putting our life style at risk, if you will, I made enough money. I wasn't that kind of bet where I was betting the farm, if you will. So made the leap after thinking about it for a few months, talking to a bunch of people about whether they might join me. Most of them said no. Then, fortunately found a couple of founders right at the time I was really pulling the ripcord and leaving. That's how Upstart came to be. It was the beginning of 2012 now.
Bill: When you talk about your upbringing, you mentioned that your family, I don't want to put words in your mouth, I don't know that struggled is exactly the right word, but worked really hard and didn't have a ton. Did you perceive like your parents to be credit restricted? Is that some of what the seed germinated or was it mostly? I've also heard you talk about when you were looking at hiring grads at Google and you were like, "This is absurd. These people are coming from great schools and they can't get credit. This doesn't make sense. I just didn't know if it was deep in you that you'd been working on for a while or not.
Dave: Well, I grew up in a family that never had any net worth at all. My parents were both like the first in their families to go to college. They grew up in North Cambridge and in Somerville, respectively, and then, they moved to the suburb and bought a house, which to them was the greatest thing in the world. They were that generation of like, get a job, buy a home in the suburbs, wow. It's amazing, have a flock of kids. We had a little house, a lot of kids. We never felt poor, trust me. My parents probably always felt that, "Wow, we're so fortunate to be where we are living in this beautiful suburb of Boston," etc. That's just what I grew up in. There was no entrepreneur or anything. But when I think about my history beyond that, I went to Dartmouth College, I went to grad school, University of Michigan, I moved to California, bought my first car, bought my first house. When I think about my own trajectory, it really was actually access to credit that unlocked all these things. We just never had money to pay for anything. So, it was all done on credit.
I'm a product of the system when it works well and it just later, begin to dive the thought of like, "Well, what would have happened if I couldn't have gotten this loan or that loan, or if I was still paying too much for X, Y, or Z." It really struck me that in some sense, the price of credit is the price of mobility or opportunity, and there's an opportunity to really overhaul that, and make it work much, much better for many, many more people. I wouldn't say if anything, that was the genesis of it. It is a little bit of the history of my own person. Again, I never remotely felt I was growing up poor. It was just life. We were perfectly fine. We were never hungry or anything like that. But clearly, getting on the trajectory on I got on, it was certainly unlocked by access to credit, which is just part of the American system, I think.
Bill: Yeah. My takeaway from studying Upstart is what you're talking about. The fact that you're saying, you've repeated now a couple of times like I didn't grow up poor. The point that I think that I have adopted from reading you and listening to you is that, there's so many underbanked out there that it's not justifiable why they're underbanked. They are on a trajectory towards prime and traditional FICO scores simply forget about them or don't even rank them, and they're left out of the access to credit, and it seems to me that that was one of the problems that you were looking to solve. Is that a fair characterization?
Dave: Yeah, I think as the idea got peeled apart and thought about a lot more. It became obvious that the credit system works really well for a thin slice of America. In some sense, credit scores themselves are inherited. One of the things everybody tends to do is, you get a credit card for your teenager and your teenager suddenly starts to build their credit really based on yours, which is a form of inheritance. That's all well and good. Nobody should be blamed for that. But there's a whole set of people who are left out of that. Their parents don't have good credit and they certainly don't have the ability to bequeath that on their kids.
In some sense, you're outside the system, you're locked outside of it, and everything just gets much harder. It's not to say you can't possibly break your way in and do the right things, but it's just really hard. Our notion became, there's this perimeter and the fortunate people live inside this perimeter, where banks work with them, offer them services, offer them credit cards, it might give them a mortgage. Then there's this whole part of the population, which is frankly 10s of millions of people, who live outside the perimeter. Banks just don't serve them. Our view generally is, we can move that perimeter dramatically. We can't solve all of it. There's some structural inequities and problems that we can't solve all of it, but we can play a big role in helping to identify more people that should be worthy of credit, reducing the price of borrowing.
It's just many of those people who at a moment in time, they're either going to head down this bad path in life. It's going to be very difficult financially as long as they'll live or maybe they'll just get on the other side of it. They'll suddenly live in a world where they don't worry about money every minute, every day. We really feel we can affect that fork in the road for a lot of people and get them into a better state.
Bill: Yeah. One thing that I enjoy listening to you speak about is the idea of driving down the aggregate cost of credit. If you can bring a more accurate pricing decision, there are people that would be quoted. I don't know exactly what it is and what you can save, but long story short, I believe I was listening to a speech like I think he gave it four years ago, and I think you said, "FICO scores capture like 43% or something like that of a borrower's probable behavior." Your models at that point that you decided gotten to north of 80%. So, that's significant delta.
Dave: What that really was was, there's many ways to show the world that there's this enormous inefficiency in the system. You could illustrate it many different ways, but one of the most stark ways we found that people can just scrap quickly is, the numbers do change a little bit. I don't know the exact number as this moment. But something like 83-ish percent, 80% to 83% of Americans who have gotten some loan or credit have never defaulted on it, okay? But less than half of Americans, maybe it's 48% actually have a credit score good enough that a bank would give them a loan. You just have this enormous gulf between 48 and 83 or wherever those exact numbers are, that just says like, "Where are those people? What is going on here?" It's a very simplistic way to highlight the enormous inefficiency in the system and how much it's underserving many millions of Americans.
Bill: One thing that as I've studied your business, I'm curious not just within Upstart, but within the economy, and I think you have a really good idea of the answer to this question. Machine learning, how much is this going to change in our lives and how much is the edge that data provides in reality? It's something that not dealing in the field. I don't have it, I just read about it, and it's like, "Well, data is the new oil and machine learning is great," but it's hard for me to conceptualize. So, I was curious to hear you talk about it a little if you're willing.
Dave: Yeah, sure. I would say, sometimes referred to as AI, sometimes as machine learning and in some sense, they can be interchangeable but the domain really in some sense is the next level of computation. What's possible to do with computers is that humans either couldn't do really well or limited in some fashion. It's built on some building blocks of things that have come to maturity itself. The algorithms that are famous, and known, and shared in AI, that cutting edge of technology has really been in the works for a few decades. Obviously, the notion of machine learning--
But the idea generally is, it's software that learns on its own. Meaning, as it experiences the world, as it interacts with the world, it actually gets better. It's better at predicting what will happen next. That's the heart of machine learning is, the prior world would be the software engineer needs to tell the thing exactly what to do. If this, then that, and enormous complexity in a set of rules to find in software, that's what the software does. Almost our entire history of knowing what software is in the business world or consumer world, anywhere is exactly that.
But suddenly you had this software where the programmer is not telling you exactly what to do and it's actually learning from experience. That's why it's notionally described as artificial intelligence is it's learning the way humans learn, which is through interaction with the real world. That's how a baby learns to talk, and to do everything else, and walk. That notion is interesting because computers theoretically, the microprocessors get faster, storage gets cheaper, Cloud computing has really just made enormous amounts of computation available at radically lower cost than it would have been 20 years ago. You put that all together and suddenly, it's possible to do things in software or in computers, generally, that would have been unheard of a while back. What we're doing this is one particular domain of it. The domain of credit, and lending, and it happens to be a good fit, because there's a huge amount of risk and unknowns.
You can think about as every time as a bank, or credit union, or some type of lender that you make a loan, you're making a prediction about the future. You're expressing your opinion about the likelihood that this money is going to come back to me and when, etc. It's just shouldn't be at all surprising that computers can do that better than people. Now, with AI and modern Cloud computing, etc., the difference is increasingly dramatic, how much better a computer can be than a person at making a loan. That's the heart of it. There're so many other domains. You can think of medical science, and autonomous vehicles, and learning languages or translating languages. There're just so many domains that AI can have enormous impact over time on. I think it was really gated by Cloud computing. Meaning, some of these ideas have been around a long time, but you needed so much power of computation, which is so expensive that it didn't really come into the mainstream until Cloud computing began to unlock the availability of that much compute.
Bill: When you were building out what's now known as Google Cloud, at what point did maybe you realize that--? I guess, did you go into building that out thinking like, "Boy, we can build out a toolset that smaller businesses can use and look at what it can do for everybody or were you more building it for internal use and then figured out how to generate revenue out of a cost center? Does that make sense? Is it even the right question?
Dave: It's a great question. I left Google 10 years ago. If you listen to Google today, Sundar, who was my last boss of Google, he's the CEO of Google, he will talk about AI every time he can. He is clearly very central to how Google sees or Alphabet, I guess, as they're called, sees the future and their play in the world, if you will. If you went back 10 years when I left, my whole time, there was just focused on what I would just call a prerequisite to that, which was just Cloud computing itself. The idea that Google could in various ways deliver value to a company or to a consumer through this massive computational capability it had. Most of my time, there was really focused on just the applications. If you use Gmail, or if you use Google Docs, or Google Calendar in your business or a school, etc., that was largely the focus in the time I was there. Now, it's become much more general purpose, this compute and competition with Amazon.
But as I said, I think of that as a prerequisite to AI and it's a no brainer now. When we started pitching to people in let's just say, 2006, 2007, "Hey, you can get rid of your email servers. You can shut down your exchange servers and just use Gmail on the Cloud." It was like, "Are you out of your freaking mind?"
Bill: [laughs]
Dave: We would never do that. I remember this woman, who was a CIO at a large university, and we already had maybe 10 universities that had signed up to use Gmail for all their students, which by the way, every school in the country does now. I don't remember what school she was from. It might have been, I won't even say. I don't want to implicate her. But she said to us, "I just want to tell you, all these CIOs that you've convinced to move their students to Gmail, they're all going to get fired." I was like-
Bill: [laughs]
Dave: -that would be really bad. Anyway, a few years later, literally-- [crosstalk]
Bill: [laughs] I thought I was doing a good thing.
Dave: Yeah, I thought this was good. Literally, a few years later, virtually, every university in the country had moved to Gmail with their own look and feel, etc. This is the fun of this stuff is, is it's cutting edge stuff, it's scary to people, and it's a long journey to get the world to accept it exactly what we're doing today, different type of problem into a different industry, but it has the same challenges.
Bill: I think your experience in building Cloud is interesting to me, because when you left, here we are 10 years later and it's easy to say, "Oh, well, of course Upstart worked." It was inevitable. They're public, and they're winning, and sure[?]. But do you think that some of the ethos at Google and the willingness to lay the groundwork and wait, do you think that that's what helped you maybe start Upstart in the way that you did? Because as I understand it, it was quite a push uphill. You had to get banks on board, you had to get the algorithms to learn. That's a lot of patients that maybe today looking back we don't appreciate.
Dave: Yeah, I will say, first of all truth be told, we barely made it. There are several times, where our ongoing existence was highly in doubt to be frank. The first version of Upstart frankly didn't really work. We went through a significant pivot a year or so into the company's existence. Suddenly, got to something that worked better, but we either felt late to a market that was already had taken off or then the market itself meeting online lending crashed in 2016 and suddenly we were doing okay, but it was a market that went out of favor. [crosstalk]
Bill: You're talking about the lending-- Not Lending Tree, what was it?
Dave: Yeah, Lending Club had some issues back then and then it just sort of reset the whole industry. Then, suddenly, we were doing well, but the industry was so out of favor that it didn't even matter. People were like, "Oh, that thing. Oh, God that--." We had no shortage of challenges. I would just say, the thing that made it work for us is we had a team that really worked well together, but we never felt like we were down to last our dollar, or we're ready to quit. None of us remotely felt like we were ready to throw in the towel, none of us wanted to go get a job, or go back to Google, or anything like that. There was just a conviction that the long-term view of what we wanted to do was sound that we had a great team and we could navigate all these terrible things that were happening, investors not wanting to give us money, or the market turning upside down, or whatever it would be, there're just lots of them.
But that's the startup story. You just have to stick with it. Not all will work out. But in our case, we always just had conviction that we have a great team, and we're going to get to the right answer, and hopefully, we'll get to the right answer quickly enough, and repeat the process enough that we'll build something sustainable. But probably, it wasn't until quite a few years later, 2017, 2018 that suddenly, we really had our feet and we felt confident.
Bill: Huh. It was that long before you really felt you had traction and that you-- Okay. Let's rewind real quick. Something that you said, it was great to get your ex-employer as a seed investor, but then when they didn't re-up in A round it created a bit of a problem perceptually. As you're struggling, and the business is pivoting, and you're just trying to get one person to give you a check, how do you get the grit to stay with it? I know that that's maybe a hard question to answer, but is it innate, is it something that the team had together? That takes a lot of character to keep pushing.
Dave: That was the hardest part for the company. We hadn't actually yet pivoted, but we needed to raise our series A after having done a series seed, which was trivially easy when I was leaving Google. Yeah, essentially, we made what I would call a rookie founder mistake, which is, we took seed money from three institutional investors including my prior employer. When we needed to raise our A round, in short, all three of them said, "Ah, we'll wait a while and see how you guys do," but we needed to raise money. Suddenly, you're out raising money and everybody's going, "Well, what about them, them, and them? Aren't they going to leave this round?" Well, that's a signaling problem and I was warned about it and said, "Oh, we won't have that problem." We had that problem. But in case, we've probably got 30, 40 nos in that round itself. For that reason, essentially, there must be something wrong.
Then frankly, one of our early investors, who was a much smaller one, First Round Capital is the one that stepped up and said, "Screw it. We think this company's going to work and we will lead this round.' They didn't lead as big of around, because they're seed funded. They don't generally lead later rounds and it was a smaller amount of money. But it got us to the next step and that's effectively what we needed. Which is why, we will always tip my hat to First Round Capital is the company that stuck with us and frankly saved us. We wouldn't be here today and they've been great partners all through the journey.
Bill: That's awesome. Good for them, good outcome for everybody. That's how it's supposed to work.
Dave: That's right.
Bill: At that point, you had young children, right? But I guess, you said your lifestyle wasn't at risk. But I would think, there's something-- I ran a small flooring company nothing like Upstart. But man, it sucked. I was working so hard and the sales just weren't coming. It was 2009, and I'm out there, and I'm busting my ass. Nothing was working. I remember sleeping on the office floor and just being like, "Why can't I push this rock uphill?" It's very, very cool that you've pivoted and crack the nut so to speak. That was the beginning of exit velocity, which occurred what six years after that or something, right?
Dave: Yeah, not that many, but it took a while. The first pivot was definitely the glove save for the company, if you will. But we had plenty of just mix analogies turbulence [laughs] after that-
Bill: Yeah.
Dave: -before we really were on solid footing,
Bill: What was the pivot just so that people understand exactly how you were thinking, and what you saw, and how you saved it?
Dave: Yeah, the original notion for Upstart was something that became known as an income share agreement. The idea generally was, instead of getting a loan at a particular interest rate, you would actually get some money, but you would pay it back as a fraction of what you earn over a period of years. Something almost out of the University of Chicago School of Economics and so it has a lot of history to it. But we were the first really to try to create a company in a marketplace around sharing your income. Man, we got so much press, because it was the craziest idea. People would just-- New York Times, everybody would cover us and be like, "Oh, my God, is this indentured servitude, or is it ultimate freedom, or what is the thing?"
Bill: [laughs]
Dave: We didn't really know we were stepping into this thing with all kinds of political overtones. It's libertarians dream, but of course-- It was interesting. We had our zealots. We suddenly got a couple hundred people who sold pieces of their income, we had investors who put the money in. It was like Kickstarter. You would, [crosstalk] in buying a fraction of someone's future income. But ultimately, it wasn't going to scale. You think, "Oh, well, we thought we would have a thousand people funded by now and turns out we have 45. What's going on?" If you're an entrepreneur, you keep dialing the knobs and trying to figure out if you can make it work. We did that for quite a few months. I always like to say, people don't know, like, there's this metaphor of when is it time to quit? Well, quitting never sounds like a good thing, but of course, there's times to pivot and realize that what you're doing isn't going to work and move to something else.
Ultimately, there was a time in late 2013, I told my founders like, "We need to talk." We went to this coffee shop in Palo Alto that we frequented. I said like, "It's time to pivot. We got about maybe six months of cash if that left and this just does not appear to be scaling." I'm just afraid the next turn of the knob is not going to solve that problem. So, it was a forced pivot and we moved really urgently to do it over just a couple months.
Again, it saved the company. I think we're fortunate. Most pivots don't really work if I had to guess and most people end up and eventually just given up but ours worked pretty quickly. That was the good thing. When we launched a standard loan product, which is we moved from this income shared notion to really just a standard, unsecured amortized loan, our business overnight started to take off. Not to say, we didn't have problems, but at least relative to the prior pain, it felt like Nirvana.
Bill: Yeah. Did you have a bank partner that was extending credit at that time? Who held the credit risk on the loans?
Dave: Yeah, that's a good question. First of all, in the prior income share world, we had no bank partner at all. We were doing the stuff ourselves, there was a lot of regulatory ambiguity. Once we were getting into this loan concept, we clearly needed a bank partner. That was one of the urgent tasks. It was literally, the week before Christmas in 2013 and I'm smiling and dialing to 20 different small banks, who might be willing to work with us. [crosstalk]
Bill: Praying somebody who's not on vacation. [laughs]
Dave: Yeah. Literally, finally one did, Cross River bank decided that it would make sense to work with us. The rest of them of said, "Ah, no. No, thank you. We needed bank partners. The loans at that time were a mix. We funded some ourselves, probably 20%, if I had to remember back in those days. Then mostly hedge funds and all would buy them, and they will take enormous bite out of us for doing that. The first hedge funds that funded our loans, guts got big chunks of equity in the company. It went from being what amounts to really expensive capital and not really scalable, but it gets you through the first proof point if you will of what you're doing. Then over time, of course, we move to more efficient capital. That's why I say it was the trying phase. We configure how many loans we could fund in a month based on the deals we had. I learned my lesson about New York hedge funds. I'll say that as well, which is a little different world out there and had some experiences that I really spun my head around in terms of how that works out there.
Bill: Yeah, well, it sounds like, they were willing to take the paper, but the terms weren't exactly the nicest.
Dave: That's right.
Bill: Wow, that's really interesting. Were you doing your own syndications at that time and calling up the hedge funds to find partners or did you have a third party that was placing, getting the meetings for you to place it?
Dave: We were mostly hustling. These were done what amounts to what are typically called forward flow agreements. The bank originates the loan, and holds it a few days, and then it sells it on to a deal that we structure to a hedge fund. The hedge fund might say, "Hey, I can fund $10, $20 million a month or whatever it is. You just have this roster of funds that are-- We weren't paving that road. There were others that had done this. There was a known set of people you go out to and-- Yeah, it was constructing these very complicated agreements.
We had our very first institutional agreement to buy these loans was with a firm that's notoriously difficult. The term sheet for this thing, before it even got to an agreement was 36 pages. It was like, "Dave's children will handed over if loss rates exceed X, Y, or Z." We fortunately never ended up signing that agreement. I think we'd still be working to sign it today.
Bill: [laughs]
Dave: We had someone else to come in, because the market was pretty hot and just gave us a simple two-page version, where my kids did not need to be handed over in the event of excess defaults. That was just the time where we were learning a lot about stuff we didn't know that much about. We knew just enough to get through it and survive. Fast forward, we were sponsoring securitization starting in 2017. We have banks now funding a lot of the loans, which is much more efficient funding. There're still credit funds and such that buy lots of the credit or buy bonds and such. So, now, it's grown into a very diverse ecosystem of sources of capital that make it all work. But that was the very modest beginning to it.
Bill: Yeah. So, Cross River Bank, you still have a good-sized relationship with, right?
Dave: That's right. They were our only bank partner for the first five years. Then just a few years ago, we started to say, "Look, we're going to expand out. We're going to become a provider of technology to banks." Suddenly, that was a business model decision that was made several years into this thing. We started to add additional banks. But Cross River Bank is still our first bank partners, still a very important partner to us. Most of the loans, not all of them that end up in the capital markets actually go through Cross River and one other bank. They provide a role that's like a conduit. Meaning, the certain amount of risk that banks are willing to take on their balance sheet, but there's a much broader world out there. So, there's a whole bunch of credit that makes sense for us to facilitate that ends up in the capital markets or in the broader market, institutional markets, where there's a much broader appetite for risk and it all works out best for us.
Bill: Yeah. When I started to research, I saw the concentration to Cross River, and I have seen it cited is at least a yellow flag, and I would highly encourage people to look into the risks themselves. But then as I dug deeper, I realized, "Well, a lot of this paper is getting sold out to the securitized market and it's not as if--."
Dave: Well, the way to think of it is, banks always have cheaper funding, because they fund loans with deposits. Think of it as this 40-ish banks and credit unions in the platform and anytime someone applies for a loan, one of those banks or credit unions, if they make an offer to them, if it fits in their risk profile, etc., that's going to be the best offer and that's what the consumers going to see. If it doesn't fit them, because they might say, "Look, we just don't want anything with expected loss rates higher than 3%, 4% a year," whatever their risk appetite is.
Suddenly, there's this overflow, where it can be funded at probably a little bit higher rate and will be sold for it into capital markets. That's what somebody like Cross River bank does on our behalf is that latter part. They actually do keep a bunch of the loans as well. That's why I call it a conduit. It is interpreted or read when you see us as a customer concentration risk, but probably, not exactly that we don't necessarily think of them as a customer. They're pretty much a business partner, who plays a very important and specific role for us.
Bill: Yeah, as you think about that strategically, not that there would be a reason to have another partner, but it's not. I don't perceive them to be the only potential partner to do this for you. Is that a fair characterization? If you needed a second partner, it would be possible to get one. Is that fair?
Dave: Oh, we actually have a second that doesn't get named because it's not as much in volume and we expect to have a third. It's almost like any business, you don't want a single point of failure, no matter what it is, if you're buying silicon or certainly, the supply chain woes most physical businesses are experiencing today would highlight that. Any good business, you don't want a single point of failure. We do have alternatives to Cross River, if for some reason they were unable to originate processed loans. Yeah, it's a little different than how I think the world interprets it.
Bill: Yeah. Something that I find really interesting about your platform and you talked about it on the earnings call, the third quarter earnings call is, I believe you're moving into auto. Is that right that it's the first tangential market that you're going to move into?
Dave: That's right. We started in unsecured personal lending and then decided a couple years ago now that the next sector that made sense is auto lending. So, we're starting to rapidly ramp into that now.
Bill: There's a large TAM for lack of a better term. If you're going after all the credit, it's going to be really interesting to watch. The thing that really clicked for me over the past couple of days is like "Boy, the savings that consumers could have and the benefit that the smaller banks that can hold some of the paper that otherwise would not be able to have a chance to bid on some of this stuff, this could really be a win-win solution." If I'm remotely correct, that's got to be a heck of a thing to create. I'd be very proud if I were you.
Dave: Yeah, well, we're on the way. Better risk models in credit significantly better not rounding error better, but way better. There're no losers honestly. It's better for the consumer, it's better for the lender. The only extent, there's a loser is that some banks will move more quickly into this era of technology and others won't. That's the kind of technology like, where our business success in our view is totally aligned with our mission. Our mission being improving access to credit for consumers in our business objective, of course, is to grow across flavors of credit and become really close partner to as many financial institutions as we can.
The good news is, they're not in conflict. The more we can cover more flavors of credit, the more we can get banks to adopt the technology. We're effectively delivering a better product to consumers or businesses, where we are going to get into business lending as well. So, it's a win-win. It's a long, hard journey. But as you say, the TAM is unimaginably large and we just have to decide, you always have this question about how much if you spread yourself too thin or stay focused more, that's a judgment call we're making every day.
Bill: Yeah, that makes sense to me. One thing, I come from the value investing school or I don't know, some value investors may be mad at me for even saying that at this point, because I've transgressed in some people's minds. But one of the things that's tough about watching and analyzing a growth company is figuring out how much of SG&A, for example, is growth spend versus maintenance spend. I don't want to put you in a weird spot by asking you the question, but when you have something that's working, and you have such a big TAM, and you want to win, as an operator, how much are you thinking of generating profit today versus like, "Let's just go win and then we'll worry about the accounting profits down the road?"
Dave: Yeah, it's a good question. Part of that is an internal debate and discussion, but part of it is a bit of a market reality as well. Particularly, if you are public company you are operating in an environment, where you have to be thoughtful about what the market is valuing these days and how it's weighing growth versus profits, and this and that, and also like the subcategory that you exist in. We're in this category where, to be honest, there's just always has been a lot of skepticism. Lending as itself is this almost necessarily evil thing that the world needs and-- A lot of companies tried to do it better, and most of them have failed, and run aground eventually, when a recession hit or whatever. There's so much skepticism built in that it doesn't get the--
When I think about Rivian, somebody like that coming out, more or less zero revenue, and they're going to be a $50, $100 [chuckles] billion-dollar company, where they we're like, "That's one where the market is ready to really lean forward and take chances." We're not in that kind of market. We're a business that's growing super quickly, triple digit growth rates, and we're profitable. To some extent, we do wrestle with, maybe, we're too profitable. Maybe, we should be looking to grow market share this and that. But importantly, the thing we can really invest in more than anything is the technical talent, the machine learning engineers, the software engineers, product managers. That part can only grow so fast before irrespective of how much money you are willing to throw at it. It just won't work that well. We're doubling or more that growth in that technical talent, which really limits how fast we can go.
We're thinking hard about that being a profitable business in a heavily competed industry such as lending, I think it's a great indicator to the world that there's something special going on here. But there's this inherent skepticism whether you talk to technology investors, financial services investors. For different reasons, they both end up with a fair degree of skepticism in the category, which is okay. It means it's harder for other people to get funded in it, it just means there is this what we call a belief adoption curve that is elongated. That means, those who can get comfortable and understand the questions to ask and where to dig in early have, as investors, enormous potential. I still believe that's where we are. There is still, despite our success, we've been public for a little over a year now, done incredibly well in terms of delivering the numbers, growth, etc.
But I just think there's inherent skepticism. It's like this AI thing, either it's completely impossible or everybody's doing it. I'm not sure which is true, but one of those two things must be true. I can't really invest in this. I think that's what we're up against in many ways that people can't believe we're doing something so distinct and separate that everybody else wouldn't already be doing it or maybe it's all just smoke in mirrors anyway. So, that's the foundation of the belief adoption curve and that's okay. Because we have time, we're generating cash, we're growing quickly. If it takes a while for the world to come around and really appreciate what we're doing, we're okay with that.
Bill: I think you're actually talking to myself just a couple years ago, maybe, but that has been one of my-- One of the things that I think happens when you read investment transcripts and whatnot is, the buzzwords come around. AI and machine learning is one of those buzzwords right now and I've wondered, what's the unique special sauce that can't be replicated? But without putting words in your mouth, I think as I've realized how much more accurate your models have gotten over time, there's got to be some advantage to the head start, I would think.
Dave: Yeah, it starts with the team and having people that are capable of building these types of models. It's really hard to bootstrap that. Because if you don't have real deep machine learning model building expertise in your company, that first person you hire, you don't even know whether that person's good or not. How can someone who knows nothing about this hire a great machine learning model developer? You have this bootstrapping problem, which makes this a hard problem for anybody to solve. Not unsolvable, but we did it, but mostly because my co-founder was that person. [chuckles] We had that advantage. But it starts with that like, you have to have the right team and then you need critical mass of that team. Once you have a lot of good machine learning engineers developing and doing cool things it becomes a magnet for talent. It's just much easier to hire the 20th from those first, or second, or third.
Then those teams have to really be working on a mission within a business that works. It has to be funded; the business model has to make sense. You can't just build great AI. A lot of things have to come together. But over time, you have the software that you're building is getting more sophisticated. As we described earlier, you need a lot of training data. That means your business has to grow and provide that training data. You do have this growth trajectory that means the amount of data, the amount of variables, the talent that's working on it, the sophistication of the algorithms that you can implement, they all play off each other. You're building what amounts to an enormous moat.
Somebody couldn't build maybe a cheaper, cheerful version. There's no doubt people will say, "We're using machine learning for this or that." I don't necessarily say, they aren't, per se, but Venus Williams and I both play tennis, too, but like--
Bill: [laughs] I like how you said that, yeah. Yeah, Tiger Woods and I play golf just very differently. [laughs]
Dave: But the subtleties of what it takes to really do it well and do it in a meaningful way that can move a market and put a company into an elevated position, it's hard at first glance to understand. You have to honestly really dig in to get there. Ultimately, I will say that, "Look, it has to show itself in the company's financials or who really cares anyway." That's where you get into this thing. Our numbers look really awesome, but why is that, is it sustainable, and will tomorrow, some enormous bank like Wells Fargo or somebody else replicate what you're doing? We feel pretty good about our competitive advantage and where it sits in the market.
Bill: Well, that's where I used to-- that what you're saying used to be what I would say and I'd be like, "Well, look at how much JPMorgan's spending on technology and aren't they going to be able to do the same thing?" I think what I've started to appreciate a little bit better is, why does the engineer pick JPMorgan over Upstart, or JPMorgan over Google. It's a scarce pool of talent that you're all competing for, so, there's got to be a reason to go somewhere.
Dave: Yeah, and I have no doubt. JP Morgan, as well as some other of the largest banks, they have amazing technical talent. I would never say that they can't build the types of things we're doing. We would hope to win them over to say, "Look, you can work with us and with much more assurance without the investment. This is just an area that's better left to a specialist, a technology company like us. Now, that hasn't totally played out. It's unclear whether a large bank or I think it's pretty clear, a mid-sized bank is not going to do that. But the largest of banks, they may try and take shots in various ways. But I think, it will playout whether this is something banks feel they can build it on themselves. To us it seems very obvious that it's just not going to be in the sweet spot for banks to build models we've built over time and that's why we're optimistic. We think over time maybe they'll try, maybe they'll fail, maybe some we'll see success. It's such a large disruption in such a large opportunity that'd be shocking if there weren't several choices in the market. One of those choices might just be a homegrown solution by a very large bank.
Bill: Well, one of the nice things that you provide at least as I perceive it is the small bank of fighting chance, because for a while I was like, "These small banks they just don't have the scale to invest in the technology to be able to compete." If they can outsource something this credit decision, I'm sure they have their own parameters. Like you said, they don't want certain-- They've got their own capital restrictions that they don't want to incur or whatever. But to the extent that they want to take the paper, it's nice for them to be able to have good loans to match against the deposits that they're growing, because I was worried for a while that small banks would just be in a tough spot.
Dave: Yeah, community banks, even before the onslaught of FinTech have just been challenged by the largest banks who have much more money to spend to just move their operations, their technology forward, so then you start to layer in FinTech and the changing needs of consumers, and it really is hard to be a community bank or maybe a credit union these days and keep up with the world, if you will. But having said that, small banks are so important to the communities they serve. It's one of the real caution points when bank branches close in underserved communities. That's just bad news. It's just as bad as when the grocery stores close.
Keeping banks, keeping their business models working, keeping them up to date on technology is one of the great things. Virtually, all the banks on our platform today, they're smaller, regional, or community. But I think by partnering with us, we can make their business models viable, we can keep their branches open, begin to integrate a fully digital experience with an in-branch experience in ways that, yeah, they'll be here in 20, 30, 40, 50 years. I think that's obviously great for them, but it's great for the communities they serve. I just think consolidation into a handful of giant banks isn't really where the US wants to go. We want a more vibrant market than that.
Bill: Yeah, definitely. I think the nice thing about community banks is, they know the community better, and they're willing to take risks on community businesses that maybe a bigger bank wouldn't, and to the extent that they can have a chance to survive, which I know a lot of community banks have been thriving. But I'm glad that this is another route that they have, because I think that it's an important part of the economy, like you said. Banks are great for headlines and punching bags, but banks do a lot of really good stuff in the economy when you get to the nuts and bolts of banking, when you get to the investment banking stuff. Maybe, we can debate some of that.
Dave: Yeah, that's for sure.
Bill: It's like plain vanilla commercial banking is extremely important to vibrant economy.
Dave: No question.
Bill: One thing that you had said in one of your interviews, I think you've said four or five Upstarters have not come into the office at that time. I don't know if it's still true. Is that true that you hired so many people remote or am I making something up?
Dave: Yeah, well, first of all because of COVID we've been more or less entirely remote for coming up on two years now. I don't know. Again, I'm not sure the exact numbers, but we're probably 1,600 plus people now. When COVID started, we were maybe 300 or 400.
Bill: Wow.
Dave: That means the vast majority of people at Upstart first of all were hired during COVID, which means, with a few exceptions, they've probably never even been to an office at Upstart. For a few months, it was an oddity. Now, it's just reality. Most people have never been in one of our offices and they only know the version of Upstart, where they're generally working from home or somewhere else. We've quickly adapted our plans over time. It took us a while to recognize like, "This isn't temporary. Even when COVID goes away, we're not going back the way we were." It took me a while. I've been around the block a while. For me, work is get up, and get in your car, and drive to work. It was that way for many decades and just kind of realizing, "Wow, that is not coming back." I used to laugh when I read some of the first people that said, "Oh, that's never coming back." I was like, "Oh, please."
Bill: [laughs]
Dave: Few months go by I'm like, "I think he's right. This is not coming back." So, we're still adapting to it this day.
Bill: How do you instill a sense of culture remotely when you hired that many people over the pandemic, that's got to be difficult to, I don't know, to keep the culture together? It's fascinating to me.
Dave: Yeah. What we've done now is probably a little different from what we'll do going forward when theoretically, the pandemic is less of a thing. But for now, we do certain things. For example, every once a month, three founders get together on a Zoom call for an hour with all new employees. We just talk about ourselves, and where the company came from, and it's a bare bones introduction to who we are as people, what we built, what happened in the last X years to get us here, and what our mission is. We just do that. It's every fourth Friday morning at 9 o'clock I'm doing this. It can get tiresome, honestly, but people are floored like, "Wow, I was at this other company for eight years, never met the CEO once, and never met--."
The idea that they're going to spend an hour in a relatively small group, we are hiring a lot of people in a month, so it's not that small. We do things like that. We have a TGIF every Friday, literally, a 45-hour long thing where updates on the company, open mic Q&A.
Bill: Oh, that's cool.
Dave: Yeah, that's an old Google thing that we replicated. We've been doing since the beginning of the company. Then during COVID, I was doing a lot of sending out updates and email. You just try to do what you can do digitally and it's worked well. We do anticipate as we're hopefully getting beyond COVID, we're not giving up our offices entirely. Our notion generally is, we're going to have teams come in about once a month to either our California office or Columbus, Ohio office, and spend two or three days together in what we call on sites, and those are semi-structured, semi-unstructured. But the idea of being is, look, you can live anywhere in the US you want, but once a month, not for everybody, not necessarily for the individual engineer, but for most people, you expect to come into one of these offices spend a couple of days with your team. We have this core belief that all those important things that we all love about working together, being in person building relationships, can actually be done--
The less volume meeting only a couple days a month, but high, high-quality. That might really end up being the best of both worlds, where you can recruit across the country. There's some disruption. People needing to travel to get together once a month. But this is the model we're aiming toward now. We call it 'Digital First.' It's just our name for it. But we'll see. We haven't really proven it yet, but that's where we stand now. That's our plan going forward from here.
Bill: That's cool. I like that concept. Why Ohio?
Dave: Well, most of our first years, we were only here in the Bay Area. Pretty much, I was always in the like, we love being on one floor of one room, just so much happened, because we were a small team, just figuring things out quickly. I was as resistant as anybody to having multiple offices. It's became clear a few years back that we just would never be able to hire enough people that we wanted to. We were always behind on hiring and just decided, "Look, we can't hire all the technical talent in the Bay Area, particularly, with the competition from the big tech companies here, who were just raising the salaries, their offers, this and that. It was just getting crazy.
We looked around. Our in our qualifications were fairly simple, which is a lot of talent, number one, technical talent, predominantly. Number two, no competition at least hopefully for a while from big tech. We didn't want to drop into say, Seattle or Austin, where there's just the big tech companies already [crosstalk] very aggressively scaling there. In Columbus, Ohio, and we happen to have one of our very first employees, Grant Schneider, who's our first and leads our machine learning team. He's a three degreed Ohio State guy. That was the closer. Columbus is a big city for almost a million people. It's the second largest city in the Midwest after Chicago, which is stunning.
Bill: Yeah, that is.
Dave: Yeah. It's growing and it's high quality of light now. I'm the University of Michigan graduate. Let me tell this didn't--
Bill: Yeah, that had to hurt.
Dave: This did not come easily to me.
Bill: [laughs]
Dave: When I first told friends, they were convinced. It was a practical joke that we were opening an office in Columbus, Ohio. But in any case, we didn't know it's funny. When we first announced the company, we leased some office space and thought, "Man, this office space could just be empty. Maybe we're just not going to find the people. It's not going to work." By the time that office space opened, we had already overfilled it. Within a couple of years, it was the biggest part of our company.
Now, since that's almost in a strange way been obsoleted, meaning, we're now hiring everywhere. But we still consider Columbus, Ohio and in here, San Mateo, California as our co-headquarters.
Bill: Interesting.
Dave: People will spend time there. There'll be a lot here. But I think in the last quarter, I think the numbers are two thirds or three quarters of the people hired outside of our Ohio, California footprint. So, COVID changed everything.
Bill: Do you get a lot of applications for University of Illinois or from University of Illinois? I only ask because I'm an ex-Chicago guy, and Illinois people used to tell me it has a great computer program. I just wonder from the Midwest, whether or not you get some good Big Ten talent?
Dave: Yeah, for sure. University of Illinois is definitely renowned for computer science. I think it's where Marc Andreessen went right, the founder of Netscape. It's definitely a famous school. There's great talent throughout the Midwest and the Big Ten For sure. We've historically hired people with some years of experience technically. When you really want to grow fast, you need talent. You can't always go find the people that already have the skill. I think we're going to start moving closer to doing just University grad programs like hiring people right out of university, which we haven't done that much of. But honestly, when you're gating item to your growth and your success is technical talent, you have to really constantly rethink your model, because it's a battle out there. Especially, this whole crypto explosion, suddenly. there's a huge vac-- Magnet for talent into crypto. We aren't today, really a crypto company. Maybe someday, but not today. So, that's just another wave we have to deal with.
Bill: Yeah. Well, if I were in your shoes, I'd say, "We've got a big opportunity here. Let's win this opportunity. We'll worry about crypto down the road." So, I can understand why that would be a secondary priority.
Dave: Yeah, that makes sense.
Bill: It's got to be tough to compete with talent just given-- for talent rather. Along those lines, do you mind-- Tell me a little bit about your philosophy of share-based compensation. This is something that comes up a lot in the value investing school, where a lot of us look at free cash flow and they say, "Well, and I've done it myself." It's like, "Well, it's all share-based compensation." Where I have got to in that is I like financing with some employee skin in the game as well and giving everybody upside, but it's something that took me a long time to get my head around. So, I'd be interested to hear you talk about it.
Dave: Yeah. We have from the beginning, I guess, in true Silicon Valley style always had equity ownership of some form or another part of employment. Every employee has some equity ownership in the company. That's pretty fundamental. We don't just accept that. We have definitely had some discussions and looked at the fact that particularly when you're early in the equity, it's really unclear if it's going to be worth anything. Cartier says, "Well, let's just give them cash if they want cash." Companies like Netflix are famous for that. I think Netflix, I don't know, if it's true today. But for most of the history, they really just said, "We're just going to give people great salaries and we're not going to bother with equity." They just took a distinctly different path in most of the technology world.
We didn't go there, though, we've debated it in such. I would say, first of all, I think one of the issues there is, it just has the risk of selecting for the wrong people, because you need people that have an optimism, and a belief, in an appetite for risk. Meaning, this equity might be worth nothing. But it could be worth a lot someday. That's one of the more fundamental reasons we did not go to pure cash as a way to compensate people is, we just fear we'll select for the wrong people and we'll end up with a company of risk averse people, [laughs] which in our world is not really going to work.
Bill: Yeah. That makes sense. It would also then make sense if you were to say, "Well, the senior people that have taken the most risk get the most future equity and those that--." Now, the equity is de-risk. Your company is at escape velocity, at least certainly appears to be me. So, that's interesting. That's an interesting way to think about it.
Dave: Well, certainly once you're public, the equity is a little more cash like. Meaning, there might be a cliff or vest period and there's only trading windows in that. It's not as pure as cash. But I think done right. It's the right system. It shares in the upside of the company. One of the best things about going public and then the success we had since we went public is really just how many people who is inside the company whose lives have changed. Of course, the stock's been up and down and all around and that's part of the journey that you can't go without. But to see a meaningful success and really changed the lives of people, who might not even have expected it, that is one of the most gratifying things of the journey so far.
Bill: Yeah, that's cool. Speaking of the stock volatility, how do you keep people focused when the stock is flying up and then--? When I say your stock, it's not just your stock, the whole space had, I don't want to call it a blow off top, but that's what's in my head. Certainly, some speculative valuations, now they've come closer down to earth. How do you keep people focused on the job at hand rather than looking at the stock price during the day?
Dave: Yeah, it's always a bit of a challenge. I think it's just a constant reinforcement of some basic ideas. Relative to some companies, we had some advantages. Meaning, when we went public, we went public for $20 a share less than a $2 billion valuation. Then we had enormous run up from there, but we at least have that base that we started at a modest place. There was a lot of companies really went public at crazy valuations and they've just suffered from there. Little bit of a less severe problem than others might have in that way.
I think, in general, of course, the story is that I will always say like, we don't control the stock. We can only control the business, and build the business. It might sound like a platitude, but the market will recognize the business over time for what it is. We have neither power and hopefully we're not getting too distracted by the stock of we're just saying like, "Look, eyes on the horizon. We have this place we want to be in three years or five years and swimming, we can get there. There's no doubt the market's going to appreciate that and it's going to work out." I think that's the only way to say it is--
The reality is, I'm sitting here seeing our stock go up and down as much as any random investor on our stock is and in so many degrees of it are outside our control. Over the long period, of course, it's not outside our control and that's what's important is. As long as we're doing the right things, and we feel like we're on that trajectory, and we're fixing the problems, we're addressing whatever issues we find, stocks going to do what it's going to do. But I can confidently say like, "Look, there's this place we want to get to. I think we can see it, it's not perfectly clear, but we can see it, and we know it's a matter of execution to get there, and you shall feel pretty damn good about what the company's likely worth at that timeframe if we're able to achieve these things."
Bill: Yeah, that makes sense. Focus on the long-term and execution. I just think somebody that looks at markets a lot, it's hard enough to train myself not to look. I can only imagine if I worked. I worked at Bank of Montreal. Our stock was not very volatile. But even then, people knew and it was just hard to prevent the conversation in the office. I would imagine with a stock as volatile as yours was last year, it would be a topic of conversation among some.
Dave: I'm sure it is. We're not in the office, of course. So, it's a little different [laughs]
Bill: Yeah, well, on Slack or whatever. I don't know.
Dave: Yeah, you can't really tell people not to watch the stock. I think it just under a certain context. Of course, there's some entertainment value to the whole thing. You get a little numb to it, whether it's going up or down. But I don't tell people ignore the stock. I think that's unfair. It's, of course, so meaningful to their compensation and all they've done. But I just tried to provide the context. If you want to spend a lot of time looking at the stock, I'm not going to say one way or the other way that, that's right. But I would just say, "Look, of course, you have to think about this over a long period of time," because that's clearly what we're trying to do is not about next week or next month.
Bill: Yeah. How do you think about whether or not you're going to increase the amount of the loans you hold or whether or not you want to sell all the loans? Is there a philosophical ethos? I like how you avoid a lot of the credit risk, but sharing in some of the credit risk also, you arguably have aligned incentives, right? So, just curious as you grow, how you think through that?
Dave: Well, I guess, our position is really-- We're a technology provider to lenders, but we're not a lender ourselves as a business. Just like many others, who would provide technology to do various activities to a bank or any type of company, that's our position is, is we're building better models, but we provide the toolset to banks and other lenders to use it in a way that makes sense for their business, but it just hasn't been. Having said that, because banks are regulated, because they're conservative as they should be, we do fund a certain small fraction a few percentage of our loans to test and experiment in new areas.
When I say, we're not a lender, we're never a lender, but we do buy back some of the loans when we're trying to experiment and push the boundaries. As a technology company, that's really powerful. I view that as effectively, it's a part of our balance sheet that's going into R&D. Because if we had to say like, "Okay, every time we're going to upgrade a certain thing or launch a new product, we suddenly need a whole bunch of banks to appreciate that and want to buy into that on day one. It would be really difficult for us as a business to move quickly. One of the things, I would say, puzzles you're always trying to solve from our seat is, "How do you stay innovative moving quickly in a very regulated and very conservative industry? We do that by taking some credit risk at the margins to experiment and do things where if it doesn't go well, it's only our money that's at risk. That's worked well for us.
But generally speaking, a lot of FinTechs have moved into a place where they're becoming banks. They are holding the credit. In my mind, that's actually a completely different business with its own set of rules, and guidelines, and limitations, and advantages. But it's a completely different business and that's not one we're particularly interested in.
Bill: Yeah, that makes sense to me. I was trying to figure out, I knew that you held some small portion of credit risk, but doing it to test new products. It makes perfect sense to me and it's not really fair to tell your bank partners, "Hey, you guys take this credit risk. Well, we test out a new product."
Dave: Right.
Bill: It's nice to see you eating your own cooking on that. For people that don't know, you get a referral fee, platform fees, and then servicing fees. That's how you generate revenue, right?
Dave: Yeah, that's right. Those are all fees generally paid to us by banks or whomever ends up holding the loans that they're sold forward from the banks. Yeah, it's a fee-based business. It's transactional and volume based. In many ways, I mean, it's reflective of I don't know, Twilio, companies like that, who really are volume-based, fee-based businesses and it works really well for us.
Bill: Yeah. Have any of the changes in Facebook's targeting, has that impacted your marketing spend at all or is that not typically the type of channel that you're using? I'm just curious how if whether or not it's impacted your business. I'm not asking you to disclose anything material.
Dave: Facebook for related, but different reasons made it harder to target for anything related to a financial product years ago. Facebook advertising is not an enormous part of our mix. It is part of that we do some advertising on Facebook and Instagram. But I would just say it's very modest part, it's always been. It just for variety of reasons not nearly as productive as other channels. So, no, frankly, the recent changes with Apple and those haven't been material to us.
Bill: Yeah, it's interesting to watch. I just figured I'd ask. I know you're a busy guy and I appreciate the time that you've given. I hate this question, but is there something that I should have asked you that you want to say, I want to make sure that you have the ability to tell people what you want to tell them. I did think actually a very interesting factoid is that your medium borrower is 28 or 29.
Dave: Yeah, that's right. Listen, it wasn't an intention on our part. It just happened to be the first enormous demographic that was underserved by the legacy system, if you will, the incumbent system was young people, who couldn't get loans, couldn't get access to credit for all the reasons we said they don't have a lot of credit history. Yeah, that's just carried forward to this day. It still tends to be a younger demographic. It's a fully digital product. They appreciate the convenience and all that. Of course, most 20-year-old dare I say, it's probably very rarely if ever been to a bank branch. So, it's well suited for them.
Bill: Yeah. Well, there are worse demographics to have traction with.
Dave: That's right.
Bill: So, congrats on the success and I'm just going to kick it back to you. Is there anything else that you think that we should cover? Do you think we covered some base as well?
Dave: No, this is great. It's been fun conversation. I really appreciate it.
[music]