Matt Harris of Bain Capital Ventures on Fintech’s Future

Matt Harris of Bain Capital Ventures on Fintech’s Future

Matt Harris is often asked about the history of fintech, and rightly so — as one of the earliest investors in the space, he witnessed its creation more than 20 years ago. He was drawn, some might think paradoxically, to its complexity. Fintech is a specialty. For the person on the street, fintech can be difficult to understand, explain, or predict. This struck Harris as a good place to start.

One of the many things fintech has brought greater recognition to is data, Harris said. Before companies like OnDeck, a now public small business lender in which Harris was the first investor, loans to small businesses were made almost entirely on the basis of the owner’s FICO score. But the digitization of business means there is excellent data on many other areas of operations that may be better indicators of the risk of the loan. This reliance on more streams of data is now routine, but that has taken time.

One of the subtler effects fintech has had on banking is culture — the way people dress in the office and at conferences. But those are surface issues — the underlying attitude toward ideas coming from frontline employees and other issues affecting operations may take years to change. Banks can be enthusiastic about fintech and innovation but unless there is a true cultural shift, it will be a wasted effort.

Harris is now focused, naturally, on where fintech is going. In a recent two-part article in Forbes he outlined fintech’s future as part of the tech stack with which new companies will be built. This is what some have called embedded fintech — every company can be a fintech company with enablers like Finix and Plaid.

Hear more from Matt Harris in the podcast below, and if you like it, be sure to subscribe on Apple Podcasts or your favorite podcast platform.

Read the full transcript here:

Phillip Ryan [00:00:02] Hello and welcome to Bank FinTech Fusion. This is Phillip Ryan, Director of Communications with CCG Catalyst. And I’m joined today by Matt Harris, who, as I’m sure you know, is a partner at Bain Capital Ventures. Welcome Matt.


Matt Harris [00:00:17] Happy to be here.


[00:00:19] Thanks for being here. So you were one of the pioneers or trailblazers in the space. What drew you to get involved in investing in financial services? Was there a particular problem that appealed to you?


[00:00:33] Well, but my formative experience was in the 90s, I worked at Bain Capital in the nineteen nineties and we bought a company in nineteen ninety six then called TRW Credit for any veterans out there. And one of the things we did was change the name of that company to Experian, which is as it’s now called. And it was a great return for us, but even more for me. I spent a year of my life in Orange County working with those executives, was understanding the power of data and technology in their case and in the lending and insurance businesses. And so that was, as I said, a great result. And that got my attention, but also some really compelling themes intellectually. And I was frankly drawn to the complexity of it as well. I think there’s a lot of things in the venture business where the man on the street or the woman on the street can readily understand what Ellerton is or or what Facebook is. And so to me, that seems harder to derive real competitive advantage as an investor in things that are easy to understand. So I was also drawn just to the fact that these things are hard. They’re hard to understand, hard to explain that hard to predict. And to me, that meant that if I spent, as I have now, two decades of my life thinking about it, that perhaps I could gin up some competitive advantage.


[00:02:03] What is it look like now, seeing all the money pouring into FinTech and all these all these Johnny-Come-Lately coming into the space where you are, where you were so early?


[00:02:14] Well, first, it was relatively disorienting.


[00:02:17] I would say if I go back 10 years when really fintech started to be a category of interest and those early payments companies like Square and Braintree and of course, PayPal have been around for a while, Stripe getting pounded.


[00:02:35] That to me was, as I said, disorienting. It was unusual and a good thing on the face of it that that so many people were starting to think about and care about payments, whereas for the prior decade it had felt relatively lonely and unloved.


[00:02:57] And then then the alternative lending fad, which started maybe eight years ago with Lending Club and Pontiac and Kabbage and Prosper, etc., then I started to feel confused because it’s one thing to like to have more of a technology investor to like payments. I can understand that payments is fundamentally a technology business, even though there are elements of risk and and certainly complexity. But lending is not fundamentally a technology. So that was that was a very strange patch when lending was so much in favor. But now I’ve kind of gotten used to it.


[00:03:34] I think it has become the conventional wisdom that financial services, which have historically been manufactured and distributed by these regulated entities that are typically kind of relatively risk averse and slow moving or those those same financial services, be they payments or lending or insurance or investing, are being manufactured and distributed by entrepreneurially led technology oriented startups.


[00:04:07] That’s the new state of the world. And and therefore, I’ve got a lot more competition from investors who also recognize that, but also a lot more targets to invest.


[00:04:18] Right. What’s the greatest lesson that you’ve learned in your career?


[00:04:24] Well, I think to me, I’ve learned that I have to just kind of stick by my guns and actually lending provides us a good example here. I was the first investor in a company called On Deck, which I mentioned is a now public company, small business lender.


[00:04:42] And we had a view that at the time, small business loans were primarily made by banks based on the owner’s credit score, their personal score.


[00:04:55] And the idea was that there were way better sources of data with which to underwrite the health of a small business and on work worked really hard alongside Kabbage in funding circle a few others to sort of pioneer the use of those data sources. But what I learned, which didn’t come as a real surprise, was that it’s incredibly hard to build the lending company just so much harder than a technology company or even a payments company because of the new and perilous types of risk involved credit risk, regulatory risk, liquidity risk. And so I more or less stayed out of lending since then. And so it was challenging for me as an investor to go through a phase where the lending universe was so popular and so highly valued, going back five or six years, companies like so five, four or five billion dollars Lending Club at its peak, well north of 10 billion dollars, you could go on and on. But there was in one given year, it was twenty thirteen. There was five billion dollars of venture capital invested in lending companies.


[00:06:08] And so I raised a billion dollars at one time.


[00:06:11] I remember that was said on one thread. And by the way, I like the supply story. They do good things for their for their customers. It’s not to be critical. It’s just that it is true and has been true for a century or more that lending companies trade in some proportion to their book value because of the risks that I mentioned and the constraints on their growth. And so during a period of time where they’re trading at revenue multiples, it’s pretty easy to get sort of lost and lose what had been a pretty well thought out and well developed conviction that that’s not how lending companies should be valued. And so to me now, having seen the crash of alternative lending, I’m glad that I stuck by my guns there and didn’t kind of get caught up in that. And I feel the same way about insurance companies now. You know, insurance carriers take a huge amount of risk. They require enormous amounts of capital. And you often don’t know how well they’re performing for many years. And so it strikes me as similarly is not a great area for venture capital. And yet at many of my competitors disagree with me. So to answer your question for me, I need to stick by my guns. I need to develop conviction. I need to constantly reassess and be open to new ideas, but not follow the crowd.


[00:07:46] What’s keeping you up at night now, whether about what you’re doing at Bain Capital Ventures or just the industry in general?


[00:07:56] What really worries you?


[00:07:58] Well, I do think.


[00:08:01] The valuation rubrics that are in use are confusing to me, and I think I think worrisome in that it’s not good for companies to be valued so highly. It’s surely not good for their investors to value too highly. And then further, it’s not good for anyone if there are too many companies getting started in any one lane. And I feel that way. For instance, we have all these Neil Banks now, all the ones in Europe are coming to the US. There are two of them here in the US, a couple of dozen. And then any consumer facing technology company of any scale is thinking about whether they should add banking services we just saw over this week becoming much more of a full fledged bank for its drivers. And so I think it’s scary for banks should newly wake them up to the fact that there are going to be many, many diverse options for their retail banking customers. But it’s also troubling, I think, for the industry and just the magnitude of funding and the quantity of companies being created that generally doesn’t end well, creates confusion for customers and will clearly create carnage at the end of the day. So I worry about that. I, I worry about the consolidation that’s happened in the payments industry. I think it’s just the beginning of twenty eighteen. We had six plenty large enough companies could be great acquirers of other payments companies and those six and now three through that was a crazy wave. Yeah. And and honestly, for the for the field of entrepreneurial companies, it’s better to have more for potential acquirers and fewer obviously. So I do worry that we may be coming to the end of an era in terms of the value placed on payments companies.


[00:10:16] Ultimately, most industries do consolidate and and and die down somewhat in their in their frenzy. And it does appear to me the payments may be coming coming to the end, at least in terms of sort of plain vanilla merchant payments. Those are the things that are on my mind right now. Yeah.


[00:10:36] What what would you say is key to banks and fintech working together? We were I was just talking to Ron Shevlin, who doesn’t believe that partnerships exist really between banks and fintech vendor relationship. But so that’s one point of view. But what what to you allows banks to work with fintech and vice versa?


[00:11:01] Well, what I’ve generally seen is that banks have a hard time developing and maintaining beautiful and functional software. There are some exceptions to that. Some banks have more capability in this area than others, and some banks are more inclination in this area to build products, software products versus others. But we stipulate that the general case is that banks are not the best in class of building software. Then the most fruitful partnerships are with companies that do build great software. And so the question is what types of relationships with software companies lend themselves more to? I want to hire a vendor to help solve a problem for me versus I want to partner with somebody with whom on an ongoing basis we’re going to collaborate.


[00:12:02] And in general, I think when it comes to either a sort of go to market partnership or a deeply embedded relationship, I do think with respect to Iran and now I’m afraid to make this point because I find it very compelling and is telling quite savage.


[00:12:27] But I will say we have examples of companies where it is a fruitful relationship and I will talk about somewhere. It is a fruitful. So we think about it. There’s a trend right now where in the merchant payments world, a lot of the activity is moving to software companies. So if you’re a yoga studio and you use mind body software to run your yoga studio, you almost certainly now also use mind body to take payments for you. And if you’re a retailer, you use the lightspeed point of sale system. Increasingly, Lightspeed Speed is also helping you take payments your restaurant. You use Tost as your software vendor. They’re also taking payments for you. So this is an inexorable trend. I would say that small and medium sized businesses will purchase payments acceptance through their software vendors. Now, banks, and particularly the processing and merchant acquiring banks, they have a hard time interfacing with software. They just don’t have the tools. And so we have a company, Phenix, that is that bridge that helps the software companies take on the capabilities to become payment facilitators and helps their bank partners plug in in ways they’re used to to these software companies. So I think that partnership, the banks and now almost all the banks work with Phenix to make that process easier. That seems like a durable relationship so that no bank is going to have the scale to invest in this tier of fact enablement software. And it’s not a good use of their time to do so because Phenix can really power the whole industry or similarly, the different use case is in need to be payments. We’re seeing that really what’s going to solve the problem of paper checks in B2B payments is software on the accounts payable side and the accounts receivable side. And so banks are never going to get cut out of that picture. But they need to partner deeply with accounts payable software companies and accounts receivable software companies in order for their issuing side and their merchant acquiring side to be present where the transaction is. Banks just walk me into a treasurer and saying, hey, we’d like to issue you virtual cards to make your payables happen. That’s not working in an embedded software world. And so there are good and healthy partnerships between HP software company and the commercial card issuing divisions of banks. And without those partnerships, there’s no clear winning go to market strategy for banks in commercial payments in the long term. So I think in those scenarios where there’s highly distinctive software. Enabling acts as one example, power accounts payable as another example, highly distinctive software, often deeply vertical eyes, there’s no chance the banks are going to build it themselves and managing a suite of vendors to do it also unreasonable. So I think those partnerships can work. Where it hasn’t really worked is actually on the lending side, to be honest. And and, you know, we’ve seen I’ll never forget and I won’t name a vendor here, but I was meeting with an executive at a large bank who has, as they thought about their lending, they had partnered with a fintech lender to, in effect, take the capability of the fintech lender developed and use it as their in-house loan origination system. And it was it was about 18 months into this. And the guy came into my office to complain and he said, look, the next time I choose a vendor to help me with the core function, I’m going to choose somebody who thinks of themselves as a vendor. And I think that was real wisdom in that I think know companies that build in-house systems to solve problems do it and inevitably highly bespoke in particular ways as compared to companies that build software from day one that’s designed to delight back as customers. I think it’s a very hard transition to make and in particular, as it comes to lending where banks for them payments is often strategic, it’s increasingly strategic. But but let’s look at the history of banks involvement in payments. The majority of banks that are involved in payments are doing so through a joint venture with first date. They’ve acknowledged for decades that they don’t know how to do it and that it’s a scale business if you leave other than US bank and chase, basically the rest of the banks have said we’re going to do this through partnerships because it’s not our expertize. No bank said that. Aboutaleb There is not a single bank in the country that doesn’t think they are good at lending. It’s a deeply core function. And so the idea that they would go to Ron’s point on an ongoing basis, more or less outsource that. To a partner is way less comfortable than keeping it in-house and just buying technology from vendors to supplement their in-house.


[00:18:17] Right, and yet there are lenders are lending software providers whose business is built on that is working with banks in some capacity, but I guess there’s a lot of banks here that will never build it in House.


[00:18:32] Some banks will most most maybe will not. Yes, but I think, though.


[00:18:38] That when it comes time to think about your mortgage lending operations and on one hand, you’ve got the lender labs where we’re not unfortunately not an investor, but a terrific company, and Glen or Ruth Stiffy will help you modernize your point of sale in the mortgage space. Or alternatively, you could strike a deal maybe with a better mortgage or rock and mortgage, who’s actually kind of a competitor of yours, but might let that you use their software so they can have kind of a to be revenue item. I think you’re much better off going with black or rigidified and saying, like, let’s let’s let’s work with a company that’s set up and has DNA around serving and delighting banks as customers instead of the one that was born to be a lender. And it’s kind of helping us on the side. So I don’t think it’s a question of banks doing it all themselves. But I think it gets to this point that Ron was making that in some lines of business lending being one, the natural relationship between the bank and the fintech company is the vendor, not partner.


[00:19:54] Right.


[00:19:56] We’re seeing a couple of companies you alluded to this briefly before, Robin Hood Harbage expand way outside their original value propositions to include deposit accounts or whatever.


[00:20:10] Does this mean the end state of every fintech company is ultimately being a bank and providing all these these banking services, let’s be honest at scale, that is an interim phase. To me, I think the.


[00:20:25] What that is, what that is evidence of. Is the dematerialization of these financial services, the fact that Uber can cobble together what it needs to provide a complete banking solution tells you that these elements, lending elements, payments, elements, custodial elements, broker dealer and investing elements are now ingredients instead of businesses themselves. So I think what you will see is just what you said, which is that everybody who has deep customer relationships, either with consumers, small businesses, medium sized businesses, will is already contemplating about how they might layer in some kind of banking product to both further lock in those customers and gain more economics. That is what’s happening right now. But to me, the logical end state of this is forget about a quote unquote bank. That same logic will just go further into, OK, I have these customers. What financial services can I offer them? Can I offer them payments, acceptance services? Can I offer them the ability to pay? And based on what we would think of as issuing services to help them make payments more conveniently, can I, through my use of data and my understanding of their business or they themselves as a consumer, am I in a position to make loans to them on an integrated and embedded fashion such that they don’t have to go out and get a loan? I’m right here offering one like Square does with square capital. Am I well positioned to make them an offer of auto insurance or of a business owners policy insurance or workers comp based on what I know about their employee base? And so, you know, to me, I think this idea of fintech companies is going to go away entirely. I mean, just like 20 years ago, we used to talk about Internet companies and now we don’t talk about that anymore because every company worth its salt uses the Internet as an ingredient. Ten years ago, we talked about mobile companies, but we don’t talk about that anymore because, again, if you have a company that’s not leveraging mobile, they’re not in business. So I think it may take five or 10 years, but I’m pretty sure fintech will be more like an ingredient. And a bunch of different technology businesses, then it will be a business model unto itself.


[00:23:09] It’s interesting, so it’s really about you have this group of customers, you know them so well, you know what they need, and now the technology exists that you can provide that cost effectively.


[00:23:21] Yes, much of financial services is about distribution, right? As I look out my window, I can see three branches still it’s been about brand and it’s been about finding customers, whether it’s innovations like MBNA with their affiliate marketing or Capital One with direct mail, financial services, most of it is really about sales and marketing. And so the embedded models totally flipped that on its head. You’re starting with the customer has a, as we say, zero zero customer acquisition costs. It’s Shopify payments is a great example, shopping the size of a shopping card company. But they realized that they were incredibly well positioned to offer merchant payments alongside the shopping cart in an e-commerce context for zero additional customer acquisition. So a big part of the logic, embedded payments, embedded lending, embedded insurance is the zero customer acquisition costs. But it’s also this data. That not only can you offer a loan and not have to pay the direct mail or other marketing cost, but you can actually offer some version of a pre-approved loan. Because if you’re in the flow of that person’s life or that business’s finances, you know a lot about them and then the extent to which you stay in that person’s life through software that business’s finances, you can actually monitor that loan. Much better than any sort of over the top lender ever could. And then finally, you can create really novel functionality. You can tie your loan to what you know about their receivables. You can have it be a revolving or floating loan instead of a term loan, because, again, you’ve got this ability to understand their life. And all these analogies apply in every line of financial services you can offer as you go auto insurance, if you’re embedded in the car in a way that, yes, you could theoretically sell somebody that and have them do something complicated to track their auto mileage. But why would you ever do that if you could offer it directly through the vehicle, etc.? So write for those three reasons. The zero customer acquisition cost, the data risk relationship and the ability to develop novel functionality. I think those either banks or fintech companies that are metaphorically in the business of calling up their customer and saying, hey, do you want a loan? Hey, do you want some payments today? Would you like to buy some insurance? That model is going to die. It’s all going to be about this embedded experience within technology, bringing these benefits of data, richness and novel functionality.


[00:26:19] Right. Last question, what currently developing technology, what new technology do you think has the greatest potential for financial services? Did you read anything about that Google quantum computing thing? That stuff is crazy.


[00:26:34] But yes, it is crazy craziness I quadrillions of times power versus a traditional computer.


[00:26:43] Yeah, insane.


[00:26:46] Yes, I find it these horizontal technology is exciting because I’m a geek, but I tend not to think about them in the abstract and their potential implications. So long chain, of course, and a small quantum computing and wearables. And there are many people in my field who we think about advances like that and then kind of quest about for problems that they can solve. And to me, this is sort of backwards. I much more live in the problem space talking to customers and executives and entrepreneurs and what are the main points and then and then rummage around the toolkit for what might solve the problem. And that’s served me pretty well. That said, I mean, if I had to pick one.


[00:27:47] I do think that the.


[00:27:49] And I hate this term, but I do think, like advanced analytic techniques are creating this ability to really anticipate customer situations and customer need. So I’ll give you an example like I think. The fact is that banks.


[00:28:09] To this day, make a lot of money off their customers laziness, right, and I think about it most acutely in terms of deposit pricing. This is a lot of lazy money sitting in an extremely low yielding savings accounts. And that is responsible for a lot of net interest margin at banks.


[00:28:33] And it’s actually a solvable problem. There’s a company called Flourish that I work with just personally, who, you know, it’s a separate high yield savings account and they have connectivity to my bank. And through their analytics, they just sweep money out every day, multiple times a day and sweep money back in. Yeah, and I’m just feeling extremely optimized.


[00:29:01] Observe the yield on capturing. And so, yeah, I’m definitely kind of an early adopter on these things. But, you know, that would have been really tricky with kind of a pretty gross and rough analytic techniques that existed five, six, seven years ago, but that the ability for that to be just never wrong, always right, ever more precise is it’s happening really quickly. And I think it would be great for consumers. It’ll be good for businesses like florists, and it will not be pretty bad for banks, actually.


[00:29:37] Yeah, yeah. I love the solution at that dynamically moving money around without you having to say, oh, let me move this block of money over here and having it done for you is is genius.


[00:29:50] Yeah, I agree.


[00:29:51] Matt, thank you so much. Really appreciate the time today. Great to be with you and best of luck with everything.


[00:29:58] Thanks very much.


Subscribe to CCG Insights.