Fintech Revealed: Deep Dive on Vertical Fintech with Increase and Tekion

This episode is part of our occasional Fintech Revealed series, where we do an extended deep dive into one topic with two industry experts. The topic today is vertical fintech, and I am joined by Matt Hennessy, the Business Lead at Increase, the modern banking infrastructure company, and Jamie Fox, the General Manager of Fintech at Tekion, the AI-native cloud platform that runs the entire business for auto dealerships across the US, Canada, and the UK.

Tekion built its embedded banking on Increase, so the two of them give us both sides of the same story: the platform that lives inside the dealership and the infrastructure that connects it to the banking system. We get into the surprisingly large money flows inside a single dealership, why paper checks still beat instant rails for many operators, how compliance and trust get engineered into the product, and just how big this embedded banking opportunity gets.

What We Covered

• What vertical fintech is and why it matters now
• The money flows hiding inside a single car dealership
• Why outbound dealer spend is roughly 2x inbound
• Operating account vs. ledgering account adoption paths
• Dealer-to-dealer payments as a ledger change with zero rail fees
• Instant rails: RTP, FedNow, and Request for Payment
• The persistence of paper checks and the cost to operationalize them
• Direct Fed access vs. layers of middleware
• Compliance as code, codified into the product
• Building trust in building blocks
• Where agentic payments and “know your agent” fit in
• How large the embedded banking opportunity ultimately gets

Key Takeaways

• Owning the financial system of record inside core operating software is the defensible position in an age when light “systems of engagement” can be replicated with AI.
• Outbound payments, not inbound, are the bigger prize: US auto dealerships pushed out roughly $1.3 trillion in 2024, about 2x what they took in.
• The barrier to instant rails is education, not technology. Many dealers do not know RTP or FedNow exists, or that they can pay a vendor any day of the week.
• Trust cannot be launched all at once. Holding a dealer’s operating cash is a different level of trust than processing a payment they can fall back on, and it is earned in building blocks.

For the founding story and more about Increase, check out my conversation with CEO and Founder Darragh Buckley from last year.

Cleaned Transcript

Peter (00:10): This is the Fintech One-on-One Podcast, the show for fintech enthusiasts looking to better understand the leaders shaping fintech and banking today. My name is Peter Renton, and since 2013, I’ve been conducting in-depth interviews with fintech founders and banking executives. Today’s episode is a little different. It is part of our occasional series called FinTech Reveal, where I do an extended deep dive into one topic with a couple of industry experts.

The topic we are covering today is vertical fintech. And I am delighted to be joined by Matt Hennessy, the Business Lead at Increase, the modern banking infrastructure company, and Jamie Fox, the General Manager of Fintech at Tekion, the AI-native end-to-end cloud platform that runs the entire business for auto dealerships across the US, Canada, and the UK.

In our conversation, we get into what vertical fintech really is and why it matters now, the surprisingly large money flows inside a car dealership, why instant rails like RTP and FedNow still have to compete with paper checks, how compliance and trust get built into the product, where agentic payments fit in, and just how big this embedded banking opportunity gets. Now let’s get on with the show.

Peter (01:36): Welcome to the podcast, Matt and Jamie.

Matt (01:39): Great to be here.

Jamie (01:39): Thank you, Peter.

Peter (01:36): Okay. So let’s kick it off with just a little bit of background, describe what you’ve done in your career, just some of the high points briefly, and then what you do today. So, Matt, we’ll start with you.

Matt (01:55): Cool. I’m the Business Lead at Increase. Before coming to Increase, I spent six years at Stripe leading go-to-market functions focused on vertical software platforms who are embedding payments and financial services to their customers. I came to Increase about a year ago. Increase exists because we think that the way that most companies access the banking system is needlessly convoluted, by no fault of their own.

We have seen that the norm in the industry has been, with more functionality comes the introduction of layers, a fintech program on top of middleware, on top of a sponsor bank, on top of a 50-year-old core that finally talks to the Federal Reserve. Our founder and CEO, Darragh Buckley, was the first employee at Stripe and saw some of the problematic nature of those layers, data and capabilities that he saw within those networks. Things like trace IDs in the NACHA file and the ability to send wires programmatically via an API were things not available to him when building Stripe. And that was purely a function of the software not contemplating technology-powered use cases. And so Increase provides modern banking infrastructure that lets technology companies like Tekion access banking rails to build the best possible financial products for their customers. We believe fintechs, software companies, and traditional enterprises should be able to open real bank accounts, spin up cards, and move money across the financial ecosystem all via an API.

Peter (03:24): Actually, I did interview Darragh on the podcast last year. And I’ll make sure I link to that in the show notes as well. So, Jamie, over to you.

Jamie (03:33): I’m Jamie. I’m the GM of Fintech at Tekion. I’ve been at Tekion six and a half years. My first day at Tekion was actually our first dealership customer on site. I went from a high-rise Brooklyn tech office to a dealership in the middle of Connecticut. So a very different change. But yeah, so I started my career at Tesla. I was an intern and ended up being there seven and a half years, managing infrastructure and global expansion across European and Asia Pacific regions for Tesla. And I met Jay and Guru, who founded Tekion at Tesla. So Jay was a Chief Information Officer and Guru was the VP of IT. And I always knew I would end up at Tekion. It was just a matter of time before they said that they were ready to have me, because we had worked together for so long.

But I think it’s important to talk about, you know, Tekion is a vertical SaaS platform in franchise automotive. So we are an AI-native end-to-end cloud platform for dealerships across the US, Canada, and UK to operate their entire business. And when we talk about that, we mean service, sales, parts, and accounting. And just as Matt was saying, Increase exists because of the old-world nature of the banking industry and access for people to build in a tech-forward environment, Tekion’s very similar. The automotive dealership software market is incredibly antiquated. My first couple of dealers, I was pulling physical servers out of IT rooms that ran their entire platform. It’s this concept of, it worked all day, I’m going to close the doors, turn the lights off, and I kind of hope it works again the next morning. And we totally changed the game on that. But we don’t consider ourselves a dealership management system. That’s one part of Tekion’s retail platform. So we consolidate dealers from about 30 to 40 disparate products to somewhere around 10 to 15, centralizing the majority of their employee interaction and operation, both internally with each other and with customers, into one single platform.

Peter (05:32): Gotcha. Okay, that’s very helpful. So let’s get right into it. We’re talking about vertical fintech today. And let’s just maybe define exactly what we mean, and why does it matter today more than it did a few years ago, Matt?

Matt (05:48): Yeah, we think vertical fintech is financial software built specifically for one industry rather than for everyone. The reason vertical fintech works is that every industry has unique financial workflows that oftentimes generic tools handle poorly. A law firm needs to handle trust accounts with strict compliance rules. A property management company needs to handle receivables like ACH transfers and checks, but also payables for home repair. An automotive dealership like the ones Tekion powers needs software that manages dealership inventory and flows through their accounting system. They need to pay auto-part suppliers, other dealerships, rebates for customers. It’s the financial operating system for money in, money at rest, money out. It matters now as much as ever because it’s so critical to be the financial system of record in an age where light systems of engagement can be so easily replicated with AI. Matt Brown, a fintech investor, famously says you can’t vibe-code a bank. And so embedding those deep financial workflows within core operating software, we feel like, is a really critical place to be for vertical software platforms.

Peter (06:57): And so Jamie, is that sort of how you’re looking at the world as well?

Jamie (07:02): Yeah, I think when we look at our platform specifically, or auto specifically, dealers operate on essentially one core platform, which is a dealership management software product. They integrate with a bunch of other tools, they use a lot of things in their daily life, but everything from the time you take your car in for service, or you buy a car, all the way to the financial statement being transmitted to the manufacturer, that is all done inside of their DMS, or their dealership management software platform.

And for us, dealers today are looking for efficiency. They’re looking for ways to not only reduce cost, but also reduce context switching and have actually reconciled amounts inside their system. For years and years, it’s just been a system of record that you’re recording things in. Dealers are looking for that penny-perfect actual reconciliation to make their lives easier and know that their books are going to close correctly and faster. That’s why vertical fintech for us is a huge, huge opportunity.

Peter (08:05): Right. So every industry can use a general-purpose bank, it can use general-purpose tools. You worked at Stripe, I’m sure there’s a bunch of industries that are just relying on those general-purpose tools. What is it about industry-specific money workflows that the traditional tools, or the more general-purpose tools, do so badly?

Matt (08:28): It’s oftentimes the case, I think, that these tools in and of themselves are not bad. It’s just that they’re disparate. The moment a dealership runs its business out of its operating account, it’s dealing with inbound payments, expense management, you’ve got accounting to deal with, payroll, your core banking software. If those things aren’t stitched together, you have to build people and process around them to do that. And it has been our experience in the market that those folks that can provide the software experience around those workflows, because naturally the dealership is reaching for Tekion, where they operate their business day in and day out. If you can provide those workflows within the software experience that feels most naturally adjacent to the business, that’s where value is created, stickiness is created, and that’s where we feel like there’s opportunity for vertical fintechs.

Peter (09:22): And so Jamie, for you, obviously you are working with Increase and you have a specific focus on one industry. What is it about the money workflows that led you to Increase rather than some of the more general-purpose tools?

Jamie (09:39): This isn’t confidential information, but we’re a Stripe platform on the inbound payment side. And so, of course, naturally, our first thing was to look at a traditional banking-as-a-service model to bring treasury, treasury operating accounts, and card issuing to our dealers. And ultimately we see incredibly higher value in having additional visibility and direct access to things like banks and the API provider on the layer on top that connects direct to the Fed. Because we’re not talking about small gig-economy accounts where they have a few hundred dollars coming in, or they’re transacting a few thousand dollars out. We’re talking about billions and billions of dollars every single month that comes in and comes out of these dealers’ businesses on their P&L. And we have to have a better solution in order to get them to the point of wanting to use a product like this. Because they expect immediate support and immediate visibility and action, which you can’t necessarily get when you layer on a BaaS provider on top of a bank that you can’t even contact.

Matt (10:47): They want a real bank account. If you’re joining with billions of dollars, those customers inherently want something that looks and feels and acts like a real bank account. It’s FDIC insured, it’s got access to all the rails, just with the added benefit of it existing within an existing software workflow.

Peter (11:06): So then Jamie, do your clients move their banking? Because I imagine there’s a bit of friction there if you’re trying to get them into a new account. How do you position that?

Jamie (11:17): Yeah. So when we look at it, we own publicly 90% of our dealers’ money-in product on the acquiring side. So we’re already dealing with their AR and their inbound receivables. Now, we built our product in a way that allows for flexibility. So if you ask me, I’m more biased to using this as your operating account, but it makes a lot more sense if you’re doing that, because every inbound ACH payment, RTP, wire payment, anything coming inbound, automatically is reconciled against your journal entries and your general ledger. If you’re not doing it that way though, Peter, it’s a good question. You can use it as a ledgering account, where we’ll actually have bill-pay requests that need to go out, bulk money that comes in, and then programmatic payment to vendors on the other side, along with accounting treatment. So the best way to do it is to use it as your operating account, and you can sweep in and sweep out money as you choose, but we did build it in a way to accommodate dealers that may not have interest in that, being able to still use it for its purpose of automated reconciliation and visibility and efficiency, but keep their money outside if they want. So keeping your money as your operating account on Tekion has its advantages just from an overall company visibility perspective to the penny, but we don’t require our dealers to do that.

Peter (12:34): So it sounds like it’s better from a productivity point, for being efficient, to do it with the new operating account. How are your dealers viewing this? Do a lot of them just try Tekion and then say, this is great, or do they jump in feet first and start with a new account?

Jamie (12:53): It’s a good question. It totally varies and it’s all based on factors of the business. When I think about how we interact with dealers, there are some dealers that they’re all in on Tekion. They said, this is the future, I’m ready, I’ve been on my provider for 30 years, I need something modern. They are generally the ones that are quickest to adopt everything across our vertical software stack. But then you have dealers that need to build to it. So you need to build and gain their trust over time, and they’ll start entertaining more and more pieces of their business decentralized. Because as a business owner, I can relate to this, putting all your eggs in one basket is very scary. And when you start doing things like payroll and banking and outbound payments, inbound payments is less risky, in my opinion, than the actual bill pay to operate and continue to maintain your business. But you have to have the trust in order for them to do it. Some give it to you a lot sooner than others, and you have to prove it every single day.

Matt (13:48): One of the things we’ve seen is that platforms who are providing the ability to move money out programmatically, “I need to pay suppliers and I want to give them the option to receive money by same or next-day ACH, by real-time payment, by FedNow, by wire, by push to card, by check, you name it,” platforms who do that well, we see oftentimes earn the right to keep money at rest. We saw a version of this with Ramp, who built their bill-pay product on Increase and then also launched their Treasury product, building out and earning trust by launching software experiences that enable the platform’s customer to say, yes, I can actually do something with this account. I might as well keep my money here because it sits with a trusted entity and it enables me to do things with it that help me run my business. And that’s a common playbook that we’ve seen play out.

Jamie (14:39): I think the biggest thing that we see from an adoption perspective is everything looks and feels the same inside of Tekion. And so if you’re already well versed in other pieces of Tekion’s business, everything from your searching to your filtering and titling and buttons, they all look the same across all the different products inside of the platform. And adoption of that is huge compared to saying, well, we integrate with a third party, and here’s how it works, and here’s what you get, and here’s what you don’t.

Peter (15:08): Maybe we could just dive in a little bit deeper on that. I’d like to talk about the actual money flows that Tekion set out to handle. You’ve touched on a couple of them, but maybe you could tell us a little bit about the money flows.

Jamie (15:21): Yeah, sure. So dealers today, if you think about how much money they take inbound, dealer outbound spend is 2x their inbound. And so from a money flow perspective that we’re going after, Tekion Spend, which is the product that we’re discussing here, is primarily targeted right now at AP. So you have every vendor that you’re paying monthly. You’re paying OEMs, you’re receiving money from OEMs on a daily, weekly basis for things like incentives, vehicle floor-plan payoffs to banks. Every invoice that comes into Tekion can be paid with Tekion Spend. But what this does is create a foundation for us to be the true operating layer for the entire dealership.

And some examples of that will be business to consumer. So, Peter, for example, if you took your car in, you wanted to buy a new car and they’re going to buy your vehicle from you, and you want to take some of that money out as an equity takeout and you want to take it home as cash. Right now, a dealership is going to cut you a check, a physical check in the back office. They’re going to run to find someone to sign it. And then they’re going to hand it to you and say, here, Peter, thank you for your business, here’s your paper check. And what we want to do is use this infrastructure that we’re building on the back-office bill-pay side to then facilitate consumer payouts, where we can actually pay things like direct to debit, send them a payout link and let them decide, do they want to get it standard, do they want to get it immediate through RTP, do they want a push-to-debit payment? We’re building infrastructure where we can accommodate and facilitate any workflow in any department. And it is truly something that we can do, anything that involves money movement, including dealer to dealer. Because as our network grows, if you have two dealers that are both using Tekion Spend and they want to pay each other, that transfer is a ledger change and it’s immediate at the bank that’s underlying running the money. So that money is immediate, there’s zero rail fees, it is just a payment to each other, and the ledger changes and both dealers go on their way.

Peter (17:28): Both dealers have to be on the same bank, right, if that’s the case?

Jamie (17:31): Yeah. And so we see that as our network effect grows, as we continue to grow at the pace that we’re growing, inevitably dealers are customers of each other. And so we see this highly efficient way of transacting across the dealer ecosystem. And I’ll actually make it a little bit more closer roadmap: if you have a dealer group that has three or four locations in the same area, some of these sales reps can actually, in our system, sell a car out of another dealership to a customer that’s sitting in front of them at some other building. And then in the back end, that car is sold wholesale to each other, and money has to go out to a bank, transact all the way over, but it’s the same company. And so in that case, the workflow is highly efficient because the money just moves in between the two dealers that are in the same group on Tekion. Accounting is 100% done, and the deals are both booked. So time to revenue is faster, time to reconciliation is shorter. The human-capital increase in efficiency is truly remarkable when you take out manual accounting and manual payments.

Peter (18:36): Sure. So Matt, what’s happening under the hood there? Just some of the use cases that Jamie just covered, they’re somewhat complex. Take us through what you’re doing under the hood to enable all of this.

Matt (18:48): Yeah, there’s two things really. One is enabling the instant provisioning of an account, and it’s a real bank account. You make an API call to Increase, and as the bank’s core, as their software, their ledger of record, we’re opening a true, fully featured account that is FDIC pass-through-insurance eligible and connecting to all of the US depository rails. So the first thing is the core underlying account. And the second thing is actually moving money directly to the networks. And so Tekion makes an API call to Increase to issue an ACH credit out to a supplier. Increase goes and communicates directly to the Federal Reserve. There is a very thin layer between Tekion, who is initiating the payment, and the underlying networks. The dependencies are Increase and the Federal Reserve. There’s no other middleware in between, which enables things like speed. We hit all the same- and next-day ACH windows. We support instant payout rails via Real-Time Payments and FedNow. It is useful for things like transparency. Where is my money? Wouldn’t it be nice if you got real-time tracking updates from FedEx about where the check is in the mail? Wouldn’t it be nice to know that the ACH file has been sent to the Federal Reserve, so you can tell the supplier, rest assured your money will land at the next ACH window at 8:30 in the morning? And then the last thing is just reliability, knowing that there are fewer layers introduced within the system. So you make an API call, you expect that money to move as you think it should.

Jamie (20:27): Peter, one thing I’ll add to this is that for us as a system of record, 10 to 20 different functions of their business in one place, if we have an issue with anything under the hood, especially dealing with money, it is Tekion’s problem. Dealers are not going to Increase or to Stripe or any of these places and saying, hey, I have this problem. Tekion has to be able to not only understand where the money is, they have to provide that support as if we were moving the money ourselves, which is why our selection of underlying partners has to play into that. Because dealers come to support at tekion.com or 1-833-TEKION-C. They don’t want to hear, sorry, that’s a bank problem, or that’s an Increase problem, or that’s this. It has to be done and addressed and fixed by Tekion.

Peter (21:12): One of the things I think that’s really interesting about Increase is that you do have that direct access to the Fed. Most companies of your size and your age don’t have that capability. So when Darragh explained that to me, I was quite impressed that that was the case. Because that means that you don’t have to rely now on a third party to move.

Jamie (21:32): Yeah, we don’t have a layer to a layer to the Fed. I always put it that way. I go to a banking-as-a-service provider, there is a bank under the hood that generally you don’t talk to, and then they have their infrastructure, as Matt said, going to the Fed rails. That creates a huge blind spot when you’re operating business with businesses like ours that see millions of dollars transact through their account every single week, almost every single day for some of our customers.

Peter (21:56): For sure. So one thing that Dara shared, a story that he shared on my podcast, which I want to reiterate here because I think it’s really interesting and it points to a lot of what we’re talking about: he talked about someone on a weekend inside a car dealership and wanting to buy the car and sending the money via your rails to the car dealer, and having it appear on a Saturday afternoon. And the car dealer was like, that doesn’t happen, you can’t do that. And they said, I don’t…

Jamie (22:24): “I don’t believe you.” They go, “I don’t believe you,” right?

Peter (22:26): Go check your bank account. And sure enough, it was there. So Matt, let’s talk about instant rails like RTP and FedNow that make this possible. Do they matter a lot? And if so, why today?

Jamie (22:39): Yeah, I’ll start with this one. I think when I look at, and I’m super bullish on digital payments and instant rails and all these things, so my conversations with dealers, we still see a ton of physical checks. I’m talking thousands. I have dealer groups, large dealer groups, that potentially send 50 to 100,000 checks a year as a group. And so when you do the math on one, first question, why are they doing that? Second question, the cost to operationalize that. Because these people are printing these checks, these dealers are printing them physically, they’re signing them, they’re licking the envelope, they’re stamping it, and they’re putting it in the mail. And then it’s gone. They don’t know when it’s going to show up, they don’t have any visibility. It’s very hard to operationalize that. We think that RTP and instant rails are huge. And as Matt knows, I am also very bullish on RfP, which is the inverse of RTP, so you can request a payment when it comes to consumers. And these are things, it’s really funny when I talk to my engineering team, when I talk to people in other countries, they’re like, why are you making such a big deal about it? We have always been historically very behind in the way that you transact money in a B2C, B2B.

Peter (23:52): Well, since the 1970s. In the 1970s we were cutting edge.

Jamie (23:56): Yeah. And then we decided to stop. And when you look into the code and you see it on the bank rails, that’s probably still some legacy code from there. So when I talk to my team that’s in Bangalore, they’re like, we’ve had this for years, like New Payments Protocol in Australia and all these things. We’re finally getting to that point. We just need consumer adoption to do that. And we are super bullish on instant rails because right now, in Darragh’s story, where they didn’t want to take a check from him, he RTP’d the money, his friend RTP’d the money to buy the car. Dealers take checks and wires and ACHs on Saturdays and Sundays, and there’s this kind of concept of, I want to do everything I can to book the car. If something happens on the Monday after I deposit it, or the Tuesday, I’m going to sort that out with the customer. But what’s so critical about automotive is that in a car-deal scenario or a services scenario, that service is already performed, or that asset is leaving that day, and you have a check that may or may not be good in your hand. And so when we talk about instant confirmation, irrevocability, and instant settlement, that gives a back office at a dealership the ability to say, yeah, you can sell a car for whatever amount of money you want, you can take any amount of money on a weekend, and it’s going to increase their ability to actually sell to consumers six, seven days a week instead of really honing in on potentially five. So it’s mitigating risk for dealers that still accept it, and allowing dealers that have put a kibosh on it for the weekends to open it back up and have a whole other selling day every single week.

Matt (25:25): Jamie’s obviously touching on the inbound payment experience. And we think it’s important that every automotive dealership should be able to accept a real-time payment or a FedNow from a customer. But there’s also the money-out concept, which we see quite a bit, which is an automotive dealership wants to pay a supplier. The supplier gets a notification, hey, your money is ready to be sent, how would you like to receive that money? Some folks, to Jamie’s point, do want a check, and you should be able to. You should be able to hit an API and put a check in the mail and get real-time tracking events and get the image from the Federal Reserve about the inbound check deposit. But also, if the supplier says, yes, I want my money now, you should be able to do that with a normal account and routing number, and the platform should have the opportunity to monetize.

Jamie (26:08): Yeah, agreed. And I think on the outbound side too, the big thing that instant rails will fight is the ability to pay outbound at any time of the day, seven days a week, 365 days a year. When you look at the construct of automotive, cash flow is king. Margins are slim when you look at overall dealership groups. They make good gross profit, but vehicle deals and those kinds of margins are very low. And when you think about being able to sweep money, hold money and float it for as long as possible, and then issue that real-time payment at a price point that is magnitudes lower than a wire with no cutoff windows, that is a highly valuable operational payment method for any accounts payable or accounting teams or back-office team.

Peter (26:53): Yeah. So for those of us who are close to fintech, I’m sometimes astounded that not everyone is using it, because of the benefits you just described, Jamie. But the reality is, as you said, there’s checks still flowing through the system, and check fraud is increasing dramatically right now because it’s seen as relatively low-hanging fruit for a lot of these fraudsters, and the technology is so good now to wash checks and to do all sorts of things to them. But I’m just curious, the car dealerships that you’re talking to, I imagine there’s a lot of them that probably don’t even know that RTP or FedNow exists, or they certainly don’t even know that you can do an instant payment. How much is instant payments part of their thinking?

Jamie (27:34): I was just on a call with a dealer earlier today, launching them on our Spend product in, let’s call it, early beta. And they said, well, we only have two vendors that we send wires to. And I said, well, why do you send a wire? Why don’t you send them an RTP payment? And to your point, they said, what is that? It’s not, if you go into a bank portal, they’re not offering you an alternative when you go select a wire, if it’s the way that you’ve paid that vendor. They’re not going to say, hey, would you like a less expensive option, or a faster option, or one that is more convenient for your business, you should use RTP. And so for us, it is a very educational piece, because one, traditionally our customers and most B2B customers are not constantly looking at their vendors and saying, what else will you accept, or let me change my process. When you’re paying out so much money, you’re not constantly looking to improve the process. You’re basically just trying to survive through your check run, or through your payment run. And that is a mindset that we kind of set out to change with our dealers. We’re doing all these things to make them paperless and more efficient. Now we have them still cutting actual checks on a MICR printer in the back office and sending them out. Something is wrong here.

Peter (28:49): Okay. So Matt, I want to turn to you and talk about the various payment rails, because there’s obviously, I don’t know how many there are, there’s a lot. We’ve got ACH same-day, ACH next-day, we’ve got RTP, FedNow, push to card is another one. What do you think are the most important? I presume you support all rails, right? Which are earning their keep, shall we say?

Matt (29:11): Our answer is, we don’t choose favorites, not to play the parent card of, I can’t choose my favorite child. But no, that’s sort of fundamental to what Increase believes. We think if the US depository system has a method of moving money, it should be available and it should be good. The Federal Reserve protocols are very good. The challenge oftentimes is the software that folks interact with most frequently doesn’t expose the full data and capabilities. But to answer your question, yeah, we think the flexibility of option set is what’s most important from the customers that we talk to. You want to send a slow check and monetize money at rest, you should be able to do that. And the check should be good, you should get the real-time tracking events, you should be able to customize the check, you should be able to append invoice data to the skirt of the check. Those things should be available to you via an API. But yes, if you want to send money instantly, you should be able to send a real-time payment or FedNow. And of course, there’s lots of gradations in between. The same- and next-day ACH protocols themselves are pretty good. And so we think it’s important to offer that flexibility and optionality for platforms to decide the ways in which they want to offer the rails to their customers, and the ways in which they want to monetize those rails as adoption grows across various schemes.

Jamie (30:32): I’ll say two things about this one. I think that what Matt’s saying without saying it is that the more layers you add, the worse the rails seem, but in reality they are actually good, because you’ll have multiple cutoff windows and processing times and all these things. But to ground it in auto, when we build a product, at least from my seat, I want to build it to be as flexible as possible. I don’t want any “yes, we can do this,” and then you get down further in the conversation and there’s some gap. Even if it is antiquated, even if it is not what you want your customer to do, you’re going to lose adoption as a vertical SaaS platform if you are missing something that they do today, they want to keep doing, and consider changing. You need to provide as much optionality as a platform instead of forcing them to immediately change everything that they do. And so, like Matt’s point, I mean, I’ll say it, I don’t like checks. I’ve done some work at my house recently and I had to dig for checks that I think were issued in 1992 when I opened my first bank account. And they’re still good, which is equally scary, the fact that that’s still a good check. But we have to offer all the options. We will try to persuade our customers to use more forward-thinking digital payments, but if they don’t want to, we still want to cater to them. And that’s why we need all the optionality that companies like Increase can provide.

Peter (31:50): Are you educating them about the benefits? I imagine you are, but they’re, I guess, people running car dealers in their 50s, 60s, “we’ve been doing checks since I was a boy, we’re going to continue to do checks.” But there’s real costs. You talked about it a little earlier, real costs for a dealer to do checks, not to mention the chances of fraud. Are you educating them into these other areas actively?

Jamie (32:17): Yeah, I’ll give you another example of a dealer that I spoke to recently, big group, 26 stores roughly, big business. Their response to me was very similar to what you said: we want to physically touch and see what that check is and sign it, so we know that we are good to send this money out. And so we’re educating them that you can accomplish the same level of visibility, same level of approvals, approval tiers and structure in our platform, and then digitally transact the money. And so it’s definitely an education. And we see this potential, as this program scales, to be massive, as we truly believe it is, just based on the performance of our other fintech products. We see the ability to help them in a more programmatic way with data and AI. We know that there are 30 to 40 percent common vendors across automotive. When you think about paint suppliers or parts suppliers, all of our customers are using one flavor of a single vendor for all of their businesses. No matter the OEM, they’re all ordering shop supplies, they’re all ordering paint. We can alert them programmatically that, hey, why are you sending a check to this company? They will take an RTP payment, they will take an ACH payment, save money, time, and transact digitally with them. But the only reason that they’re exposed to that is not because the vendor is saying we don’t want to take your check. It’s because we now know from network-effect data that we can enact that change as the system of record. But to your point on cost, we have polled our dealers on the cost to operationalize a physical check. It is somewhere between, I’ve gotten answers from four dollars all the way up to ten dollars, based on the complexity of the business, in time, postage, printing, signing, approvals, discussions, all of it. And our goal is to centralize that. And if they still want to send a check, they still can, but we don’t really want them to.

Matt (34:05): If they’re going to do it, you should make it good.

Jamie (34:07): If you’re going to do it, do everything you can digitally and don’t ever print something, right?

Matt (34:11): And it’s funny, we see some people who, they’re trying to drive conversion to digital payment methods. Increase is also an issuer and processor. And so you can spin up one-time or multi-use cards, and folks will send a check and then educate the customer, here’s a payment link to actually go pay by card if you prefer next time.

Peter (34:30): Okay, so I want to switch gears a little bit and talk about compliance for a second. Because when I was talking to Darragh, he was talking about how at Increase you sort of built that into the product from day one. Now, that’s not the way everybody operates, because it’s expensive and annoying. So from your perspective, Matt, when you’re looking at your products and what you offer, the fact that you’ve built compliance in, what does that actually mean in practice? And why is it different than building the software and then adding the compliance layer later on?

Matt (35:04): It’s such a good question. We believe your banking application code should also help you manage your regulatory, your compliance, your legal requirements. It should be your source of truth for all the obligations that you have. If there are compliance obligations to access these networks, they should be codified within the product. Dara likes to talk about this travel rule for wires above a certain price, I think it’s $3,000. And if you don’t have this codified in this product, then, gosh, I hope when you go talk to the regulators, you’ve collected the requisite data for wires of this size. We believe that should be a part of the product experience. The transaction should fail if you haven’t done the thing that the regulator says that you ought to do. And so having compliance as a core part of the engineering experience for accessing the banking networks is really important for us.

Jamie (35:59): Yeah. And I think from our perspective, a lot of times…Matt and I were at an event recently and they said, yeah, fintech can be your number one money maker as a product. And for some platforms, that’s true. It’s not the case for us because of the breadth of our product. But what that really translates to, Peter, is we don’t want to have to, as a vertical SaaS platform where our fintech products are incredible and high-penetrating, we don’t want to take on the load of the regulatory nature. And so when we go and partner with someone, we want as little internal compliance effort that we have to take on, monitor, and mandate ourselves. We want to delegate that. And it allows us to operate faster, focus on our core, and let our partners that do it arguably better, and as their daily job, do it for us in as many ways as they can.

Peter (36:54): A lot of what you’re doing here is built on trust. A car transaction for a consumer is usually the second largest thing they’ll ever buy outside of their home. It’s a big deal for a lot of people, literally. And there’s a lot of trust that’s needed throughout this transaction. And I’m looking at a dealer, the amount of trust that they’re going to have to have in a company like Tekion, because they’re basically handing over the keys to the kingdom in some ways to you, where you can store money, you can move money, that requires a lot of trust. How do you go about building that?

Jamie (37:29): Yeah, it’s a building-block scenario. The first step, and I think it’s a little different for us, and I’ve had this conversation with the team at Increase too, and the bank that we’re working with: the first step is they have to trust you with their core business. And that’s what they do when they come onto Tekion, is they are saying, I am trusting Tekion to run service, parts, retail, make sure my rates, residuals, calculations, everything is right. So I think we have a little bit of a leg up because we’re already entrusted to run their business, transmit their financial statement to their OEM for them on their behalf, integrate with every piece of the product and manufacturers that they want. But financial technology has an entirely different layer of trust. And so you have to create it systematically. I can tell you that if I rolled out spend management and my payments product at the same time, I don’t think we would see equal levels of penetration. Because payments is a really easy thing to say, okay, I’ll try it out, if it doesn’t work, I just fall back. If you start storing money in an account that happens to be embedded inside of Tekion, that is an entirely different layer of trust, because that’s where people are doing things like paying their employees, paying their bills, keeping the lights on. That is a different layer, and it comes with time and performance and consistency, which is why this product is primed to launch now, after we’ve hit some pretty good scale and have a lot of people that have entrusted us with other pieces of their financial business, whether it’s inbound payments, payroll, any of the things that we do on that side. The trust has to be built in building blocks. They’re not going to give it to you all at one go. And so we had to be very intelligent about when we launch things and how to make sure that people actually use this product and give us a shot at it.

Peter (39:13): Matt, with you, you must be thinking about trust because you are the money mover and the money enabler. So the trust has to be built in from day one, right?

Matt (39:24): Absolutely. And for us, that trust comes both in terms of money at rest and money in flight at the networks. On the former, it’s really important that when Tekion goes to its customers and says, we’re going to custody your funds and we’ve got this, that the system that they are interacting with is the same one that the bank would point to as the ledger of record, or where money reconciles, or whose money is whose. If your technology provider is telling you where money sits is different than what the bank would say, that is in and of itself trust-eroding. And then on the money-movement side, we’ve talked about this a bit, but the visibility of where money sits in the ecosystem is a really big part of engendering that confidence. Certainly within Tekion’s use case, but you think about payroll, the payroll engineer who runs the job to execute payroll transactions on Thursday nights. Wouldn’t it be great if they knew that the file was actually sent to the Federal Reserve, that the Federal Reserve has acknowledged the payment that the technology company is issuing back to the payroll provider, that we expect funds to settle at 8:30 in the morning, and that they actually did so? And so that level of visibility also builds that trust in money moving through the ecosystem.

Jamie (40:40): Yeah, Matt, having also overseen payroll, I understand from tickets and things we’ve received from people, that having the visibility, which we do thanks to Increase by way of others, allows us to support them. And we know everything from ACH trace IDs to landing times from our side.

Peter (40:56): Okay, so I want to talk about agentic commerce and AI for a moment. We haven’t discussed this yet. Buying a car, I don’t think we’re going to be at a point where a consumer is going to type into ChatGPT, “buy me a car.” They might say “find me a car,” but not “buy me a car” without sort of a human interaction. But there’s a lot of things happening in payments, like payments moving around, that may end up having an agentic component to it, when you’re typically moving money between suppliers and car dealers. So question for you first, Jamie: how are you thinking about agentic payments, when you’ve got agents, not people, that are actually moving money?

Jamie (41:38): Yeah, this is very top, very high on the list of topics of discussion internally. We have the data that we have as a business to help our customers, it’s remarkable. When we convert a dealer to Tekion, we migrate between seven and ten years of historical data. Our conversion process is essentially one step down from an electronic health record system. And so we have, when you think about it, we just get the luxury of them closing from 7 p.m. to 7 a.m., whereas a hospital doesn’t close. So we have a little bit of airtime to do our conversion. But I think for us, we see, especially I think first between our dealers, it’s going to be big where they’re trading cars and interacting with each other. And when they’re on the same level of ability to commercially interact with each other, we can build really, really unique tooling, so dealers can have agentic workflows between each other.

I think for us, a lot of our agentic commerce will likely come on the consumer side first, mostly things like interactions, paying for one-sided agentic to start, where you’re interacting with an agent, they’re billing something for you, they’re sending you invoices, billing cards on file. That’s a very easy, like, I’m just talking to a system. But then you layer on the agentic piece where you can search for consolidated services, you can schedule and pay for services via MCP servers and having access to that and interacting with these dealers with shared payment tokens. We think there is a very big future for that. I just think we’re a little bit far, a little bit of a ways off. And I think, to your point, what you said earlier, being their second biggest purchase in their life, there is a level of comfortability. There’s a difference between buying a necklace on Etsy through ChatGPT or agentic commerce, than buying a vehicle. And I think we can build the tools and the mechanics. I think adoption will be driven by the market’s willingness to accept agentic commerce. So I think a lot of people are like, a little bit of cart before the horse right now, talking about it, but it is coming. It will be something that we see grow on our platform because we handle all parts of auto business. But we think it’s a little early right now from a comfortability level.

Peter (43:51): Right. So Matt, how are you preparing for agentic commerce, for the money-movement side of agentic commerce inside Increase?

Matt (43:59): Well, we think the robots should have the same clean set of APIs that humans should in order to move money. We’re seeing quite a bit of experimenting on the Increase platform at the moment. We think Increase is a great solution for this type of testing. You can easily give an agent its own account and separate credentials and budget, and if something goes wrong, you can freeze one account and nothing else is affected. We think in the future your software will dictate to the agent the desired outcome, and the agent will do the research on where to buy things, provision itself a card, check its available balance, make reasonable judgments based on human preferences around speed, price, reviews, whatever is the thing that the individual is looking to optimize for. But certainly you can imagine a world where for supplier payouts, it’s “cool, here’s my budget, here are the types of auto-part supplies I need, go forth, execute the payment, make sure these things land by this date,” and you take it away.

Jamie (44:57): So Matt, are you saying that we’ll have KYA at some point? Like, know your agent, maybe?

Peter (45:01): There already is stuff like that. There’s already things like that out there. And I actually think we’re going to start with small purchases and small things, but B2B is going to be more motivated if it saves time.

Jamie (45:14): Yeah. And Peter, I will say, the one other thing I’ll add is on the spend-management, outbound-payment side, we will start with things like very basic agentic approval flows. The agent recognizes that you’re paying the same bill for the same amount every single month. If you want, give the authority to approve that within a buffer, 5% up, down, whatever you want, where that flow becomes entirely automated and approved and obviously transacted through a more agentic workflow, saving you any sort of human time doing it, unless there is a problem in the bounds of what you’ve given them for rules, where you actually need to get involved. We think we should automate as much of this back-office work as we can to free up people’s time, because right now it’s just a huge slog as far as manual work.

Peter (46:02): So Matt, we’ve obviously talked a lot about one vertical here, the auto-dealer vertical. There’s obviously lots of other industries with vertical SaaS players. I’m curious about what you think about this embedded banking layer that you guys have built. How large does this ultimately get?

Matt (46:21): We think it’s fairly unbounded. We’ve talked about automotive. We see a lot of banking activity in the healthcare space, where they’re talking about compliance as code. There’s plenty of compliance restrictions around different accounts. Think about health reimbursement arrangements, flexible savings accounts, health savings accounts, different card types tied to those compliance regimes. We think about real estate and property management, where you’re collecting money en masse, but also need to put that money to work to go pay the lawn-care company or get your deck repaired. We are seeing a lot of uptake in construction, where you think about the money flows. You’ve got the owner who’s sending money to a general contractor, who’s sending money to a subcontractor, who’s sending money to a supplier. Wouldn’t it be great if you could ledger money movement up and down that ecosystem, monetizing up and down that value chain? We think there’s tons of opportunity to power accounts-receivable flows, true operating bank account, expense management, tying cards to an account, and of course the supplier payout, which, as Jamie’s talked about, is huge as compared to the inbound payment.

Jamie (47:33): I agree with you that it’s relatively unbounded. The TAM, basically, the total addressable market for you, Matt, is anyone that’s sending money, theoretically, within reason. I mean, that’s a very, like, “I’m going to go get my Series A, the TAM is the world” kind of vibe. But in automotive specifically, the total outbound flows, including everything, in 2024 was $1.3 trillion in the US alone. It’s 2x the inbound. So $650 billion inbound and $1.3 trillion in just auto, which makes up roughly 1.3 percent of the GDP.

Peter (48:09): I want to close with a prediction, if you would. So if you could both give me some sense of where this vertical fintech is going. What’s a prediction you would make for the next three years? Jamie, I’ll start with you.

Jamie (48:23): The prediction I’ll make is that I think we’ll continue to see more and more business customers of these vertical fintechs and vertical SaaS platforms that embed fintech wanting more embedded financial tooling. Because once you get into the flow and one part of your business that deals with money becomes optimized and efficient, and you’re just running smoothly, you’re going to want to add more and more and more. And so to me, any industry that has vertical SaaS and has money movement, which the definition is, every industry has money movement and software now, will continue to yearn for more of this. And if you do it right, adoption will be super high, because it will beat out any disconnected system, even if there is a price difference. The difference in cost, if it’s more to come to vertical fintech, it will prove that it is more efficient and worth the money. And I think it also changes the way people look at their business decision-making, because it becomes less about where can I get the cheapest price, more of what is making my business as efficient as possible, and does that justify the cost? That’s a much different way to look at it, rather than getting the best possible deal. You want the most efficient tooling. And I think people will be willing to pay for it.

Matt (49:35): That sounds absolutely right. We certainly see a world in which just the proliferation of financial services, bank accounts, cards, lending, insurance, grows pretty dramatically. I think the thing that we feel is that this concept of offloading user interfaces, or UIs, to sort of no- or low-code solutions will fade into obscurity. The proliferation of AI and capable engineering teams, certainly we saw this in the integration, the speed with which the Tekion team got up and running. Vertical fintechs are becoming more sophisticated themselves, and with AI tooling, where they can build a layer deeper. Integrations shouldn’t take months, they should take a matter of short weeks. The product should feel like your own, it shouldn’t be a third-party experience. Jamie was talking about some of these third-party experiences that folks run into, it’s like, yes, go and talk to your bank dashboard. That’s not the experience that the software company is looking to create. And so too, the actual user interface, the actual interaction should feel inherent to the platform. And though some of these low-, no-code options, I think in today’s world do help some folks get started early, I’m bearish on those types of tools lasting deep into the market.

Jamie (50:50): And Matt knows, the first thing I ever asked Matt when I first talked to him was, this all sounds great, but do my customers ever have to work in your product? And he’s like, no. And I said, all right, we can schedule another call, basically. Like, okay, I want to know the skeletons now before I get any further deep into this, you know?

Peter (51:09): Okay, well, gentlemen, we’ll have to leave it there. Really, really interesting conversation. I know I learned a lot, and I’m sure the audience did as well. So Matt, thank you for your time. Jamie, it was great to meet you, and best of luck to you guys. Thanks.

Jamie (51:21): Thank you.

Matt (51:22): Thanks so much for having us.

Peter (51:29): What I kept thinking about throughout this conversation was the topic of trust. Jamie made the point that you cannot launch payments and money storage at the same time and expect equal adoption, because asking a dealer to hold their operating cash inside your software is a completely different level of trust than processing a payment they can fall back on. You earn it in building blocks, and you prove it every single day. And Matt said that the ledger the customer sees has to match the ledger the bank would point to, or trust erodes instantly. In a business built on moving other people’s money, that consistency is the whole product. And this trust is the key to growing the vertical fintech space. Anyway, that’s it for today’s show. If you enjoy these episodes, please go ahead and subscribe, tell a friend, or leave a review. And thanks so much for listening.