The Inner Workings of Banking-as-a-Service with Treasury Prime
Chris Dean is the co-founder & CEO of Treasury Prime, a leading US banking-as-a-service platform fresh off a $40m Series C raise.
Treasury Prime is a software company that connects sponsor banks to customer facing fintechs and by doing so powers bank accounts, cards and money movement. Unlike others in the banking-as-a-service space, Treasury Prime describes itself as building a network of interoperable sponsor banks and customer-facing fintechs in a ground-up approach to Open Banking in the US.
In this conversation, Chris and Will Beeson discuss the state of the banking-as-a-service industry, Treasury Prime’s approach compared to other software-based BaaS companies and tech-enabled sponsor banks themselves, the economics of banking-as-a-service, the evolving regulatory environment and more.
Thank you very much for joining us today. Please welcome, Chris Dean.
Full transcript:
Will Beeson:
Chris Dean, welcome to Rebank.
Chris Dean:
Thanks. I'm happy to be here. I'm happy to talk to you again, too.
Will Beeson:
Yeah, likewise, likewise. It's great to see you again, great to speak again. I think we first met, it's probably, I don't know, 2018 or 2019. I think you guys maybe were relatively fresh off an A round. Definitely had a bank partner or a couple, and some customers at that point. But over the past three plus years now you, you've really, I think hit your stride at Treasury Prime, and I think just this week at the time of recording you announced a 40 million dollar Series C. So, look forward to digging into all of that. But maybe to kick us off, we can get a quick intro from you and I guess a high level description of what you're working on at Treasury Prime.
Chris Dean:
Yeah, that's great. And thanks again for having me. I always enjoy these conversations. So, Treasury Prime is a software company and we make banking APIs to connect banks and enterprises. Typically, this is embedded banking applications that people do. We have a unique model in that we have a marketplace of banks and enterprises, fintechs if you'd like, who that we connect together. We're a software company. We don't hold risk, we don't handle compliance. What we do is we build tools so that the bank or the fintech can handle compliance and can handle risk. And generally we don't believe that there's any solution in the United States for an explosion of banking services that doesn't involve the banks. We just don't believe that. And there's a lot of banks in the U.S., 5,000-ish, 10,000 if you count credit unions, and I don't think that's going to change to a dozen banks anytime soon. I think we might have that at most, but generally Treasury Prime believes in banks and we believe that there's a change in the fintech world, and we're here to power that. That's what we do.
Will Beeson:
So, you meet someone on the street, you're based in downtown San Francisco so you can meet fintech people on the street, and they say, "Oh, so you're a bank as a service company." What do you say to that?
Chris Dean:
I usually agree, because it's better to agree, because that's mostly right. I would say that we have a word internally that most of the banking as a service companies are, fall into three buckets. One is people outside the U.S., and generally outside the United States it's much easier and I think a more straightforward problem. And maybe the right way is just to move to London and start a challenger bank. That's a great way to do that. And if you're in the U.S. though, there's us basically, who write software for fintechs and banks, and it's just a software company, and then everyone else who is what I would call a rent-a-charter system where they actually find one, maybe two, sometimes three, but not a lot of partner banks and act as a bank to the fintechs.
That is, they take on compliance risk, they generally pool people's funds together, they pool risks together, and they handle that. There's a fourth category, which I keep eagerly awaiting for, which I just never actually see, which is a U.S. bank actually doing banking as a service properly. Cross River and Green Dot, whoever else you want to talk to, usually they don't do things I would think of as banking as a service. But Treasury Prime is unique in that we have an API, and we have a lot of different banks you can use that API to connect to.
Will Beeson:
Maybe to make this tangible for the audience who isn't as deep in it as you and me, what's an example of a use case or a business model that you guys support?
Chris Dean:
Sure. Our client I've been talking about for a while that is very typical, is it's a SaaS software company. It's called Builder Trend. They make construction software to run your construction job site. You're building a bowling alley, you need permits, you need schedules, you need materials, you need people who work there, and they have software to manage all of that, and it's great. They've decided they wanted to add banking services to their app to help their users. Construction's a very cash flow sensitive business, and they wanted to be able to see the money movement directly, be able to do payments, everything from inside their app. So, they partnered with us, we found them a bank, and they use our software to connect to the bank but they have a direct relationship with the bank, and that is how it works.
So the end user, the person running their construction company, they can see the balance of their bank account, they can write checks on that, they can send ACHs and wires, they can see money coming in. They can set triggers coming in on all their inbound payments. They can have a debit card if they want. It's not super common, but they can have a debit card if they want. And so, from the end user it looks like a bank, but it's a bank that's very bespoke to that particular industry that's construction.
Will Beeson:
And what's the relationship like between that vertical SaaS construction company and the bank that they work with?
Chris Dean:
They have a direct relationship with each other. Treasury Prime is a marketplace, we have a lot of banks. The number we quote today is 16. It's a little bit more than that, but 16 banks. And when a new customer comes in, we have an intake process where we match them up with a bank. And once the bank is comfortable with this client, and not every one is, it's not always a good match, but we have enough banks we can always find a good match, that the bank and the fintech have a direct relationship with each other. They have regular meetings, it's their banker, they have their banker's phone number, bankers are great at this. They lean on Treasury Prime for things like technical support, but a lot of the day-to-day work is with the bank.
Will Beeson:
Okay. So, this space over the past handful of years, the banking as a service space, and you can kind of, I guess extend the terminology a little bit and call it embedded finance, it's been of great interest to many, many people. We've seen lots of activity in the space over the past number of years, and you guys have been building away day in and day out. What observations or learnings do you have, have you developed over the past, what, three, four or five years since you've been working on Treasury Prime that inform the way you structure your business? And indeed, looking around you, say, "Oh, it might have been interesting had we actually done this other thing instead." I'm sure you wouldn't admit to that, but to the extent you have those sorts of reflections.
Chris Dean:
No, I'm with you. So, our story here is that my co-founder, Jim and I, we had a company on some other folks called Standard Treasury very early in the U.S., where we were trying to make an open banking style solution for the larger banks in the U.S. Silicon Valley Bank was our first client, they bought that company and I ran the fintech group, Jim and I ran the fintech group there for two years. And we saw how the bank worked directly from the inside, we saw how the regulators interacted with the bank and how the fintechs interacted with, at a distance, with the regulators. I'm really an engineer, a startup guy. This is not my first rodeo. I've done multiple successful startups. It was eye-opening to see how the bank worked. And we developed our strategy then, and I think we were impressed by the overall great product market that we had when we were at SVB.
We were turning away clients, it was so great. I'm like, "Well, we should take advantage of this because it doesn't look like any of the banks is really going to go for it the way we want it to." We had some hopes for BBVA to do something big, but they never really got there, and they semi-recently just kind of left that whole space. But we said there's a way that you could handle this. There's a way that you can help accelerate the growth of fintechs in the United States. And everyone we looked at who was doing it seemed like they were doing it wrong. You can't rent a charter and pretend to be a bank, you have to be a bank. And if you look at it, you can't be a single bank. The deposits controlled now in the U.S. by fintechs, I would say are in the order of 100 billion in deposits, that kind of number.
That's small compared to the overall deposits in the U.S., but it's big for a single bank. It seems very likely that in three, four years the fintechs will control a trillion dollars in deposits, which is about 5% of the U.S. total deposits. That is definitely too big for any single bank. So, you're going to need a lot of banks participating in this to work. And yet we don't have an open banking style API, so what do we do? This whole viewpoint of being at SVB informed all this, our strategy of we have to have a lot of banks all using the same technology with the same API, so the fintechs can actually connect to whatever bank they want, wherever they can get the best deal. They can move banks if they want, they can have multiple banks if they want, but you have to do this from a technology first perspective. And that's really what we did then, and our opinion hasn't really changed in the past four years or so, that really the banks are the ones who have to handle the compliance.
The fintechs are the ones who have to handle the end user relations, and we are the ones who handle the technology that connects these two together. We can't do compliance for you. I'm sorry, just can't. I can't talk to your end customers for you. I'm sorry, I can't. But the bank is good at one, fintech's good at another, let them be them. That's our whole premise. I mean, if anything that's been proven to us. It's good to feel right. Right now I feel very right.
Will Beeson:
In terms of the API marketplace concept, and the idea of having many banks and fintechs being able to, the implication is via integration to one API or suite of APIs that you offer, basically interact with potentially and multiple banks on the backend, and or switch banks as necessary. Now, I'm sure there's a kind of contractual wrapper, there's a compliance wrapper, some operational considerations. It's more than just API calls. The way you describe that makes sense if you're talking about weather forecasts, or the results of sporting events where different providers are probably pushing you all the same data, maybe some are a little bit faster, maybe some are a little bit cheaper. But if you're talking about a checking account or debit card issuing, which has additional operational complexities and various other things, how in practice do you kind of work with both the banks and the fintechs to create that ideal situation in which multiple banks can be involved, fintechs can have flexibility? Basically the balance between what's purely software and what has an operational element.
Chris Dean:
Yeah, and it's a great question. Our biggest fintechs, that's our most active ones, one of them has say, I don't know what it is today, but like 700 million in deposits. They don't want them at one bank, they want them spread across at four separate banks. They interact with them as if it's one giant thing. For them, it's a virtual bank but yet they have contractual relationships with each one of those banks, they have their own bankers. Each one of those banks, they get different things from the banks, but for them they have to have these banks. And one, just in the deposits are too large. We have a 200 billion dollar bank that could hold them all, but the rest of the banks can't. The rest of the banks are just probably too little to take 700 million dollars. It's a lot. Even the 40 billion dollar banks, they're like, "I don't know. That's a lot."
How does it work day-to-day? Well, they use our API, we route, we do direct core integrations with every single bank we're at. They're all real time. So, we do reconciliation real time throughout the day. Most of the operational burden is taken over by that. It's way lower when you can do realtime reconciliation. And then the things that you can't do we have tools to do. Mostly the banks do them. Some of the fintechs do them, and we do some. We don't really like to do any operational things but we do some. Mostly it's the banks.
Will Beeson:
I mean, I guess it embedded in my question, in fairness, was kind of this implicit reference to complex banking products like checking accounts, like debit cards, which have maybe been the focus of a lot of consumer neobanks, and maybe even SMB focused neobanks over the past handful of years. But far from the only relevant banking products. And perhaps we're seeing, we've learned, not even the most relevant products to be building if you're a fintech. Maybe there's something in there which is worth exploring. I mean, I don't know. How much of the work you're doing with customers and bank partners is specifically around, I don't know, payments, or specifically around some sort of banking product? Which maybe is crucial and adjacent to the core business of and in fintech, but which is not trying to replicate a full service consumer or S&B banking offering.
Chris Dean:
Yeah, I would say that the category of neobanks, that's the minority of our usage. Mostly it's not neobanks. I mean, I use the builder trend example because it's interesting. Most of them are folks who are doing banking like services and they'll have one specific thing that they're helping their end user do. Sometimes it's payment, sometimes it's they'll spin up an SPV for a certain activity that will be there for a year or so, and that's a real bank account and their money moving in and out. They can have debit cards, virtual or physical, attached to those, which we have one customer doing a lot of that. But these are all deposit products. We look at the world as bank account centric, so that you create a bank account, you can create a retail, like a personal bank account, you can create a commercial, like a business bank account, create trust.
You can do any kind of regular bank account, and they're absolutely regular bank accounts, and you can do anything you want with that bank account that the bank allows. And the way our realtime integration works is that anything that bank can do, you can do with the bank. And because of all this, it's actually an easier problem than people think. Once you do the hard work upfront of these integrations, and setting up all the operational triggers and all the operational flows, which we spent the last four years doing, the whole thing becomes easy. We have this multi-bank solution we do where people can move money instantly between any of our banks. That works great, but it only works because we have done all the hard work. If you're hoping that you can pull everyone's money in a single FBO, and have a ledger, and have that work, that is an operational burden right there.
You just signed up to hire hundreds of people to do manual reconciliation all day, because it's hard. We don't do any of that, because at the end of the day we're a software company. We want to automate the whole world. The parts we can't automate, the banks do, because the banks are really good at this.
Will Beeson:
How much of your work over time, you talked about four years of integrating with bank cores, I mean, is it identifying one kind of core provider or subset of cores, and perfecting integrations with those and then trying to get in front of all banks that use that core? Or it's kind of individual relationships with banks that lead to an interest on their side, that lead to a request for an integration, and then you're going in and popping the hood?
Chris Dean:
That's a great question. We're the second, definitely. We don't approach it as we're expanding based upon technology we already can do, we've decided that we'll handle any hard technical problem that involves connecting to a bank. We have just a world-class engineering group. Mike Clark, who runs the engineering group here and I have been doing this problem for what, eight years now? And we know these cores really well. Even new ones we know well, because they're all kind of the same. And we do the realtime integration. It's not fast, it is not easy. But once it's done, it's done. And the way that we decide on what banks to choose are we ask if the bank as well run, because that's the hardest part. The technology's the technology, but if the bank is not going to be good at dealing with the fintechs, or the bank is not going to be good with dealing with the operational exception handling, which you were alluding to before, stuff happens. An ACH bounces, what are you going to do?
You have to do something. And if they're not up for that, they're not a good bank for us. But if they are, then they're a great bank for us, and the technology, we say we'll just make work. There's maybe one or two we've said, "That's too weird, we won't do it." Because some of them are like, "I have my own custom one, which I've been using for 40 years and someone who passed away wrote." I'm like, "How about we put you at the bottom of the list?" But the major vendors, the FIS's, is the Fiserv's, the Jack Henry's, we do those. We don't do all the FIS and Fiserv cores, but we do a good number of them. We do ones built in the '80s from ones, FIS's brand new one. We do all those. Once you've done, I don't know, maybe 10 of them, it becomes easy. And it becomes straightforward, let's say. It's never easy.
Will Beeson:
Maybe shifting gears a little bit, as I mentioned before, this week I think you announced a 40 million dollar Series C round. Do you want to tell us about that?
Chris Dean:
Sure.
Will Beeson:
Firstly, congratulations.
Chris Dean:
Thank you. I think it shows that we have such a strong business that even in times like this, we could raise our funds. It was a interesting process. If we had done this, I don't know, a year before it would've been seeming like an afternoon, but there was real diligence here. And everyone fights their last battle in some ways, and I'm still fighting mine. I had a bad experience a long time ago with someone on my cap table who was not a good fit. And so, we're picky about who we let on our cap table. And it's a relationship. I can't really divorce them. They're here until they decide to leave. So, at one point we were doing the rounds and I talked to this person I didn't know, his name was Norman Chen, and Norman is at BAM Elevate. He's a VC there.
I had never heard of him before. And we met and had coffee, and we talked, and I said, "Oh, this guy's smart. We should keep talking." And we started, and he went through some diligence and did the customer calls, and did all the things that a good investor will do. And he called me back at some point to, and he said the kind of thing that you were alluding to, where it was, "I don't think you're a BaaS company," is what he said. "I don't think you're a services company. I think you're a software company." And I'm like, "100% right." And he went on for the next 10 minutes to explain our business to us. And he was more right than almost anyone I've ever talked to about how our business worked. And that call was when we decided to take this money.
We didn't have a number, we didn't have any sort of terms or anything like that, but that's the point when we decided to take his money, because someone that smart, someone actually understands the business, that's who you want. Because it is clear to me that Treasury Prime is going to be a brand new banking standard in the U.S. We're just too big now to not be that. But it's not going to be fast because the banks aren't fast. It's going to take a while, and we need the best people around to do that. So, I don't know, I was really happy to find Norman. He led the round, he's now on our board, which is great. We had all our insiders participated too. They did [inaudible 00:21:12] plus we have new bank funds, which is a new person who also who put some money in, and we're really happy with the round. I am thinking about investors, and these are all A plus investors.
Will Beeson:
Excellent. Well very, very happy for you and the team. So, it sounds like you may or may not have had the calls with the skeptical VCs who are like, "Well, I see what Column is doing." They have a charter and they're building basically APIs on top of that. Plaid was super successful, so Column probably will be too. Or they look at a Synapse, or a Unit or whatever, and say, "These guys have brand names in the market. They have lots of ..." And so, but you're doing things differently. You say it's an API marketplace, and they basically draw comparisons. And so, why aren't you doing the thing with the license? Or why aren't you doing the thing that's more the bank in a box plus charter, I think you called it rent a charter? What are your top three responses to questions like that?
Chris Dean:
Right. We talked to all those people, 100%. We talked to all the big players and all the big names, everyone you've heard of, and like, oh yeah. And some of them were like, "Well, Synapse does this and that, and clearly unit and Solid are doing amazing. And what about increase?" I'm naming competitors, which you're never supposed to do. But I look at them and they all have the same solution. They're renting a charter. Let me explain to you why that will not work in the United States. Sooner or later, a regulator's going to come down and drop the hammer on them. And lo and behold, that's what happened. That's an easy explanation. It's like the regulators don't want compliance to be outsourced. Regulators want simple things. Outsourcing compliance is not one of them. So, this is not a model you can really do.
So, then next it'll be like, that's one. They're like, "Yeah, but what about Column, or Lead, or somebody like that?" And I'm like, "That's a great question. I would say Column is a great family business that's going to grow slowly and be an amazing asset for the children. However, banks are not designed to grow at the rates that a venture firm wants to grow." Even self funded and everything, the growth is constrained by the regulators. So, what do you need? You need multiple banks. If there's indeed a trillion dollars in deposits, who's going to hold those deposits? Is it going to be Column? Are they going to hold half a trillion dollars in deposits? That seems unlikely. So, the only thing that'll work is if you have a lot of banks, which leads to us. And the last bit they always ask, they ask us, "When are we going to get a charter?" And my answer is, "I mean, I never say never, but I hope never."
That's a hard problem running a bank. All of a sudden you're regulated a certain way and it's hard. But you would have to be really great executor to be able to run a bank and run a software business at the same time. And I think that was use our responses, and most people were like, "No, I don't believe it." Okay, great. I can tell you we're bigger than multiple people you've named on that list, but that's fine. We'll wait.
Will Beeson:
Tell me about the economics. It's something I've certainly thought a lot about in part having worked inside this space and built on top of some of the infrastructure you're talking about. And then also just having talked to lots of people that do this, both on the bank side, banks that have built their own in-house bank as a service offerings with other software companies, like yourselves, with what you would describe as rent a charter services type BaaS platforms with consumer fintechs. I have a view on a lot of this stuff, and specifically I have a view on the economics of the existing value chain, which to me feels immature. I see a lot of value leakage. I see value leakage to components of the value chain that, in my view, are not the ones that are creating most of the value. I would be interested in your gut reaction to what I just said.
Chris Dean:
I mean, yeah, you're right. The way most of the banking as a service providers work is they're pretending to be a bank, and they take a piece of what's going on. They charge you for different activities, and at scale they're taking a big piece. And I think that anyone worth their salt knows that they can go straight to a bank and get a better deal than they're getting with this bank as a service provider. Maybe the technology isn't as good, maybe there's some hiccups there, maybe it's hard to talk to a banker, but it's way cheaper. And at scale it's obvious that everyone's going to leave these rent a charter platforms and go somewhere else. We see the biggest players now, they're just leaving and going straight to the bank. Why? Because when you start to do 100 million revenue, you're going like, "Well, percents here and there."
And they start to matter. That really adds up. Treasury Prime's whole point is that we don't do that. You have to pay us for our services, for sure. But we're a software company. Our deal with you with the fintech is, hey, either you can build this or you can use our software. We are going to be cheaper because we do this for a living. This is our whole job. You're going to have to write bespoke custom software. You want to make a different economic deal with the bank, go right ahead. We have deals where we receive zero money from the fintechs, but the bank pays us for the rent service. Why? Because it's useful to the bank to have that fintech there. They've figured out some pricing scheme where they share some deposit interest, or they charge some fees. I mean, I actually usually know all these but I don't care as long as they pay us my wholesale price. We're a software company. That works.
Will Beeson:
Your wholesale price, I mean, it's like some sort of annual license fee? Or it's pay per API call? What's the-
Chris Dean:
Oh, what's the model? Model's always the same. It's just different numbers. The model is you pay us a monthly SaaS fee just to be on the platform. Banks pay that, fintechs pay that just to be on the platform. And that's just a flat number. And then it's different for different people, but it's a flat number. And then there's usage on top of that. Generally it's very account focused. So, you charge to create an account, you charge to manipulate an account. And for banks it stops there. And what that allows us to do is at scale, the price per account goes way, way down so that it's actually worth it to you. They're like, "Well, I might have a couple hundred thousand accounts, but they don't cost me very much so I'd rather pay Treasury Prime to use their software," which already does all the realtime integration, already hooks up into all these systems.
You got a team of expert engineers finding bugs all day long, rather than do it myself. And my analog here is always could you build your own CRM? Absolutely, you could. You could, for sure. Do you want to? Probably not. You got other stuff to do.
Will Beeson:
How important is Durbin exemption in your selection of partners? And then also, I guess that's ultimately driven by the demand on the fintech side, and how important do you expect that to be going forward?
Chris Dean:
For us, it's never been a big deal. I mean, most of our banks, all but what four, five? Five, are Durbin exempt. So, that's fine. It's not an issue. People want Durbin exempt so they can get an extra few bips of interchange. Most of our businesses are not interchange interested. They have a debit card, or not, based upon usage by the end user. That's not really where their fees or their growth comes from. They get money some other ways. We do have some retail neobanks who like the interchange, but it's easy enough to find them a Durbin exempt except bank if that's what they want. But it is the minority. It's not a significant portion of our revenue.
Will Beeson:
How about the asset side of the balance sheet, credit products? Is that something that you currently offer or something that's in the pipeline based on customer demand?
Chris Dean:
We don't offer a credit product or a loan product these days. So, we don't offer a credit card, or a charge card, or a loan product. We're really deposit focused still. Having said that, it is not unusual for our clients to have one of those things, to have a loan product. And the way they do it is they make a direct deal with the bank. And our deposit API is rich enough that it's easy to manipulate a line of credit and just draw down on it interactively. So, if you're worried about how you manage that [inaudible 00:29:57] manage a lot of that, and it just becomes you have to do the risk management. And that part is hard to outsource. It's cool, there's lots of great apps where you can, at the end of the checkout process go, "Yeah, I would like to finance that, please." That's fine. I like that. We help people do that, but we don't do any of that ourselves because I think the risk, that's something you got to figure out. Fintech, I don't know you with these clients, you do.
Will Beeson:
Before we break, maybe you can tell us what's next for Treasury Prime. You raised 40 million dollars for lots of good reasons, I imagine. What are you going to spend it on?
Chris Dean:
Yeah. Well, the thing that's really working right now, which we're doubling down on, is our bank network. Our bank network's exploding, more and more banks on our network. You got to think about it like any kind of network, whether it's eBay, or Airbnb, or whoever, the more people on it the more valuable the network is, and the more useful it is. So, we're doubling down on that. We have this unique kind of multi-bank model where clients can have multiple bank accounts at different banks. They can move money instantaneously between them. They can spread risk out in different ways. They can get services at one bank or another. We have some folks who have some cannabis deposits. They keep it up at one bank, some not cannabis deposits they keep it a different bank, and same for crypto and all that stuff. And that works well. So, we're going to keep doubling down on that. I think at this point we are excited by the number of banks we have. However, I would say that it is a small fraction of the banks we'll have in a couple years time.
Will Beeson:
Chris, great to connect with you. Congratulations again on the recent progress and the raise, and I wish you guys all the best.
Chris Dean:
Thanks, Will. Great talking to you.
Will Beeson:
Chris Dean, thank you very much for joining us today.
