Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Podcast Episode 4: History of Platform Payments

Podcast Episode 4: History of Platform Payments

Expand for full transcript

Reggie Young: Welcome back to another episode of Fintech Layer Cake. We’re super excited to have Jareau Wadé as a guest today.

Matt Janiga: Jareau is the chief growth officer at Finix and has really unique insights in the history of platform payments. We’re very excited to chat with him.

Reggie Young: Jareau, let’s start by running through a quick tour of your career in payments.

Jareau Wadé: Thanks for having me. I’m excited to do this. I like talking about payments and nerd out about compliance and stuff. So a quick tour, my background in payments is I’m currently with a company called Finix, which is a famous technology company for platforms. I actually started my career in payments at a similar company called Balanced that I co-founded in 2010. And in between, I’ve done a number of other things, including lead, product partnerships for e-commerce and visual search at Pinterest. So those are the highlights. There’s a few other things I could get into, but those are the main things I’ve done around e-commerce or payments for most of my career.

Reggie Young: We will be digging into some of those in more depth. First, we would love to chat generally about platform payments. I think today companies like Stripe and Finix can help customers with platform payments. But it might surprise a lot of listeners to hear that even 10 years ago, platform payments weren’t really a thing and companies in the space, companies like eBay, Amazon and Uber, they had to build everything in-house.

Matt Janiga: Now you’ve got a front-row seat due to your time at Balanced. I also had some nice front-row seats due to my time at my Square and Stripe. And we’d love to hear your perspective on how platform payments got started. How does that sound?

Jareau Wadé: Sounds good. Let’s do it.

Reggie Young: Let’s kick off with your time at Balanced, which was a pioneer in the platform payments. How did Balanced get started?

Jareau Wadé: So we started the company Balanced around 2010, 2011, so as myself and my co-founders, Matim and Mahmoud. And it started off actually as a Venmo-like competitor. So it was supposed to be a peer-to-peer app where you could send money instantly from your bank account to another. And one cool thing, we ended up changing the business to focus more on building an API for platform payments. But in between starting the company, which was first called PoundPay, and then pivoting to Balanced, we actually created a lot of the technology in-house that started in building on our own. So it’s something like a login with Chase.

So we would actually allow someone on the mobile phone to log into their bank account, and that’s one of the things that allowed us to move money instantly for free. But we saw this opportunity behind the scenes for building infrastructure for a company like ourselves with our peer-to-peer app and it just didn’t exist. So this was 2010. Everyone told us to use PayPal. Stripe was barely, barely around. So we ended up just building it ourselves. So we ended up doing processing, escrow and payouts for marketplaces and other platforms and I think raised around $6 million from Y Combinator, Andreessen Horowitz, SV Angel and a couple of prominent angels, including Brian Chesky, the CEO of Airbnb.

Matt Janiga: Now one thing quickly, I’d love to know what your first office was like. Before my time, Stripes’ first office was a house in Palo Alto and maybe just like a walk around a normal neighborhood type of thing. What was your first office like?

Jareau Wadé: It was a house in Palo Alto. I remember we had these sliding doors that would open up onto the patio or in the backyard, and that’s where we would whiteboard essentially all of the ideas. So it was just a house on Williams Street in Palo Alto. There’s three of us living there with a giant dog, and that’s where the first idea came up.

Matt Janiga: And was your dog’s name Balanced Payments? No, probably not.

Jareau Wadé: No, definitely not, But I did work at a company Milo.com, where we had the name for the company and a dog mascot and then the CEO bought a dog that looked like a mascot and named it Milo.

Matt Janiga: Terrible reason to buy a dog, people. Please don’t follow that example.

Jareau Wadé: He ended up treating the dog very well and had him for years, but it was a funny reversal. Everyone thought the dog came first.

Matt Janiga: That is really funny. All right, Reggie. Sorry, I know I hopped in there. I’m happy to turn it back over to you because you’ve got some great questions lined up.

Reggie Young: Oh, no, important facts we want to record. So I think everyone would think starting a payments platform company is a no-brainer today given Plaid and Stripe’s valuations. But what was it like pitching to investors with this idea before those companies were a sure thing?

Jareau Wadé: There’s a few things that are definitely different now. So one is everyone just assumed PayPal could be used for everything, and I don’t know if you have looked at PayPal’s API. Certainly in 2010, they were not great. But the idea of a developer-friendly experience for payments was not that well-known at the time. And this seems silly now because that’s what everyone is moving into. Obviously Stripe’s successfully popularized this idea. But at the time, we were one of the first to do that.

One of our big things is we wanted to move money quickly. So I think our backgrounds being low income or my co-founders being immigrants, we all had experiences in our personal lives which have been really, really useful to get money that day from something useful than Craigslist. So when we were doing the peer-to-peer app, one of our big value props was instant payments, instant fund transfer for free. And I remember Matim, my CEO, was showing this off to a few investors and they were like, cool. And we were like, no, we just moved money from my bank account to your bank account, live demo. You have 20 bucks in your account. It was free. It’s instant. They were like, well, I could just wait for the ACH in the two days. I have a million dollars in my bank account. Why do I need 20?

And so we started to realize that there was also not necessarily a lens on exactly how transformative financial services and digital payment tools would be to the majority of humans on the planet. So that’s one reason that we wanted to build more infrastructure so that other experiences could be built on top of that. But at the time, I don’t think the VCs really got that idea.

And then the last one I will mention is I think a lot of folks were used to seeing PayPal or Venmo, which had a lot of winner-take-all market dynamics. And when you’re building infrastructure, that is not the case. And so as we were scaling, we would get that question a lot like, whoa. At first it was PayPal, then it was Stripe. How could there possibly be space in the market? And you then look across to Europe and you see a company like Adyen who has built a $50 billion company at the same time as Stripe. And clearly, there’s room for more.

So those are some of the misconceptions that I don’t think are around as much anymore. People clearly understand the bottoms-up in terms of giving access to more financial tools to people. They understand payment infrastructure is a huge market, and most people are moving in the direction of instant money transfer.

Matt Janiga: No, I think it’s really good to do that. I’m not a VC myself but I think your education in the market and obviously other folks working to educate the VC market, that they were missing the opportunity in real-time payments really led to some breakthroughs. It’s primarily because none of them missed the 15-minute food delivery start-up. I’ve seen a lot of those companies out there. So I think we owe that to you as well.

Jareau Wadé: They ported their understanding from payments where they missed the boat over to the grocery delivery, absolutely.

Matt Janiga: Instant delivery. They’re like I missed Stripe. I’ve got to get in the 15-minute food delivery window. All right, more seriously. So this is great to hear this kind of background, especially for folks who are starting a company now. They know it’s a long slog but they may have peaked around these corners. Tell us more. So you’ve raised funds. Balanced obviously has a great team. We’d love to know more about your journey to scale. So today my view is that companies getting started, it’s a lot easier. You can plug into various APIs. You don’t have to do as much thinking about funds flows and money transmissions or things like that because you have companies.

In the past, it would have been Balanced. Right now, it’s Stripe and others than can take care of those things for you. But take us back, take listeners back to that time period. So I think it really is magical in payments innovation. What was it like for you guys to build Balanced back then, back before you had all these APIs and other service providers?

Jareau Wadé: It was a lot of fun. It was hard. It's funny because you ask these questions and I’m like, oh, yeah, I’m not that old. But we were starting Balanced before it was an obvious foregone conclusion that you would use web services. So I remember definitely through YC where I think we got some of those free credits to use AWS. But my co-founders were both computer science majors in college, and we’re just on some of the cutting edge of web services that were coming out. But this was 2010 and there were a lot of businesses, especially Wall Street, where my CTO, Mahmoud, he was working at Wachovia Securities, where they were doing high-frequency trading.

They were racking their own servers. So he saw that, and it was very clear we weren’t going to rack servers in our living room in Palo Alto. So we went all-in on our web services in the first sem in the beginning. On the payment side, it was very different. There weren’t really APIs or, if they said they were, they were not up to even modern standards at that point. But we used Stripe. I think we were part of the first 100 customers of Stripe. They were called dev/payments. Didn’t have what we needed. I remember getting 80-page PDF specs from some of the bigger processors and banks and working through those. But it was slim pickings. It was not clear how to build it. So as a result, we built a lot, not only the integrations, but some of the internal services ourselves so that we could build on top of them to create whatever experience we want with our API.

Matt Janiga: It’s just so funny that you mentioned racking your own services. I remember that’s what I experienced when I moved out here to work at Square in 2014. And so I got to Stripe, and I never asked and I didn’t read enough. I read all about the open email culture and everything but I didn’t read enough about their servers set up. So I didn’t realize they were all just cloud-based. I think I went three months without realizing everything was on Amazon. And when I did, I had this moment where I was like, oh my God, is that safe? Did I make a big mistake leaving Square to come to Stripe?

And then you realize, no, this is the future and you get over it. But it was funny. Every lawyer that joined after me I think had the same month lag, didn’t realize there were no rack servers. And they were like, well, where’s our data center? We’re like with Amazon. And I thought this was really funny.

Jareau Wadé: The funny thing now is it’s clear that you could pay Amazon to provide a lot more and better security for your data than you could provide this individual company. Unless you’re a massive company like Facebook, maybe you have a chance. It has only been a decade really, and now obviously you’re going to use AWS or whatever web services you want.

Matt Janiga: Yeah, absolutely.

Reggie Young: We’d love to hear more about finding a processor and that whole experience. It feels like the biggest chicken and egg thing for payments companies is like, at the time, you need to work with a bank like Wells Fargo and JPMorgan. But they probably won’t talk to you until you’re big and have your own team, controls and everything set up. So how did Balanced deal with that cold-start problem?

Jareau Wadé: It was difficult because we were getting market signals and growth. So we were starting to have real customers and real volume. But we were too small to be underwritten by a large payment processor. So we had to do the resourceful, scrappy start-up thing, which is just find a way to make it go forward and deal with the next problem as you get there. So specifically we set up a series of merchant accounts through various ISOs, and we processed payments on those. So it was like buying T-shirts. So say we had a T-shirt company, that’s what Balanced was, a hypothetical story. And we were screen printing them, but we were getting our inventory from Costco.

So yes, you’re getting some discount but we’re really just marking up something that has already been marked up. We needed to go wholesale. We needed to go straight to the processors and the banks. And so specifically, we would prototype things on early-day Stripe or PayPal or one of these ISO merchant accounts we got. And then as we saw that we’re getting volume or usage on it, we would move it on to another ISO merchant account. And from there, we actually set up a fake website for selling pound cakes, and this allowed us to process payments. The company’s name was PoundPay at the time. So we were selling pound cakes and that was the cover.

But it allowed us to scale through various merchant accounts until we had enough volume to be signed by Chase Paymentech, and I believe that was our first processor. It was either them or Wells Fargo. But I’m pretty sure it was Chase. And at the time, the PayFac, payment facilitator model, it existed but it wasn’t in wide deployment. So for every aggregator, as they were called, that Chase had on their payment stacks, so for example Square, Chase would have to go out and do a site visit to do risk assessment.

So I remember having like the chief risk officer of payment tech come to the Balanced office, which was in SoMa. Our floor that we were on across the hallway was a medical cannabis shop, and it didn’t scream buttoned-up payments company. But we had to do it. And I actually remember actually putting on the access control doors with the key fobs and everything so that we could just say we had an entry system for part of our PCI underwriting. There was a lot of stuff like that. I’m just kind of surprised we made it through.

But to simply answer your question, we did whatever we needed to to be scrappy and eventually we got everything legit, processed with eventually Chase Paymentech, Wells Fargo and then finally on Vantiv, which is now part of FIS. Then on ACH, we use Wells Fargo, Fifth Third Bank and Push2Card. And some other things, we did Forte, which is a processor that’s still around now.

Matt Janiga: Really awesome, and it’s funny because I think people forget about the large role Chase played in payments, especially over the last decade where they were powering Square, PayPal, folks like you guys.

Jareau Wadé: Yeah, Intuit.

Matt Janiga: Yeah, that’s right. Back before we could go plug into Stripe or Braintree or something like that, if you wanted to plug payments and you were a start-up, you’d go work with Chase. They also have a big and still do have a fairly big venture arm. So if you’re a start-up and you’re looking around, Chase can be a great start-up partner. Vantiv as well and obviously Wells. So we’d love to ask you. One of my favorite things about Stripe, I joined early enough, where we still got to see the early customers. Some of them had reached huge scale, like Lyft and Shopify, by the time I got there.

But I remember one of the most fun things was working with these early customers. So we’d love to hear what was it like for you at Balanced with your early customers. And what shape did they take? And what were some of the challenges that you faced with them?

Jareau Wadé: So we started off focusing on marketplaces. So if you go back to 2010, 2011, this was when Uber, Airbnb, TaskRabbit, Etsy were first taking off. There was this huge explosion in every type of marketplace model possible, gig economy, collabora con, whatever you call it. So we focus mainly on marketplaces. We ended up having a number of crowdfunding services and then one thing, which actually led to some of the insights at Finix, is vertical SaaS companies. Our favorite example is this company that was based in Napa that did SaaS e-commerce inventory and website building for vineyards. And then they added payments on top. So you can actually transact through the website or consumers could transact through the website.

But some of the more notable companies, we had Reddit. They had a gift marketplace that they did every year, and we powered payments for them. Square’s cash app actually uses our ACH API. Thumbtack, one of their early experiments with payments they did with us in the early, early days. Vinted, which is a European-based secondhand resell marketplace, was with us, Tradesy, Zaarly.

Do you guys remember The Fancy at all? I think they’re still around. They had a partnership with Apple at one point. It got a fair amount of notoriety. But they were basically a marketplace where almost anything that you could see on a retailer’s website, they had scraped/crawled and made available on their marketplace. So they were one of our customers. But yeah, it was fun watching customers who had, as you mentioned, tiny scale. And then six months, two years later, you look back and you’re like, wow, this is a very large company now. So we’re seeing some of that at Finix. I saw that at Balanced. It’s actually a very cool experience.

Reggie Young: I love it. It sounds like a fun time. I know something we see now is customers don’t know how to evaluate various providers in the card issuance base. So you might go talk with Stripe, Marqeta and Lithic and find that the Stripes of the world seem to be playing in the best space, whereas Lithic and others seem to be focused on modular card issuing. I imagine you saw something similar, needing to help customers understand, help with their various payment models they could take. When you were at Balanced, what kind of customer education did you have to do?

Jareau Wadé: So the biggest thing that we found is that customers would come to us for their hair-on-fire problem. We’ll start at marketplace. We need card acceptance. They might go get a solution then. It might be a traditional merchant account. It might be Stripe at the time, Braintree, whatever it might be. And then they would be making manual payouts through PayPal or writing checks or something like that. Over time, they started realizing, hey, we can’t scale, either for compliance reasons or for just manual operational reasons. We can’t scale the payouts like that. So we actually started positioning ourselves.

We had three things. We had processing, escrow, payouts. But we started positioning the payouts piece first because we realized that was the best way to meet the customers where they are. And then eventually we would sell them on the full stack, the full solution. Similarly we would often find people would get cards in their mind first, and then we would sell them on bank payments and then cross-sell them into cards. So part of our model was educating folks on the different payment types, the different payment models of a platform is not just accepting cards, it’s also disbursing.

And then just working with them over time to expand and further penetrate into their full payments volume. That I think was the biggest misunderstanding. And then one thing we did also was we started actually adding changes to our API schema that would better allow platforms to link payments in and payouts. And so that’s something they didn’t even know that they wanted or needed because they’re doing it manually. But at scale, obviously it’s a lot better to do with an API. So we provided some affordances for that, like a meta tag on every API resource, things like that.

Reggie Young: I think that’s something a lot of fintech struggle with, the customer education and hand-holding piece. And kind of related, I think some fintechs also have to deal with educating regulators and I know you were building Balanced during an interesting time period when the fintech industry was more advanced than regulators and the banks were at the time, like what we’re seeing with crypto now. My understanding is regulators had spates where they would play catch-up and banks would also have these spates where they suddenly derisk a bit.

The main example that comes to mind is Operation Choke Point, when the DOJ investigated banks and payment processors that did business with certain types of companies that the DOJ thought posed certain fraud or money laundering risks. And ultimately the government functionally pressured the financial industry to cut off access to certain financial services for these types of businesses they were worried about. So I’d be curious to hear about your experience interacting with regulators as you were growing Balanced.

Jareau Wadé: We can talk more broadly about Choke Point. I actually think you, Reggie, have more insight into the specifics. I saw some of this come down through the acquirers. So a lot of the acquirers like the Chase, the Wells Fargo, etc., Vantiv, they got spooked by what was going on in the treasury department with Choke Point and often there were questions. Is there payday lending? Not aggregation per se was in the sites but there was a lack of visibility and granularity into some of the aggregated payments volume. And so when the processors would come up empty-handed about who was the actual underlying recipient, I think that’s one of the things that spurred on the development of the payment facilitator process.

There was some collateral damage though. As far as I heard, Chase actually went in and just kicked off a bunch of aggregators except for the big ones like a Square or an Intuit. We were caught up in that, which is I even know about Choke Point. I don’t think the processors and the inquirers were in scope of Choke Point per se but there’s definitely some rattled cages where they decided to be more risk-averse. So as you said, Reggie, they overcompensate or they might fall behind. This is definitely an area where they think they went a little bit too much. But I’m curious what you saw in private practice.

Reggie Young: So it was interesting. I remember we had a partner who was representing online lenders, basically payday lenders. And it was a big issue for them because one of the things Choke Point seems to be heavily impacting is payday loans and regardless what you think of if these products were lawful, if you’re offered with the right disclosures and consents and things like that and the right operational backend. So I remember there was a big scramble around, well, is the government trying to use enforcement against the payment processors, which I think is where all the news articles came out, to choke out this conduct that from a policy perspective they don’t like but is lawful?

And it was interesting because as you dug into it, it wasn’t that clear-cut and dry, that's what was happening. It was more about the processors that were there. If people keep an eye on the FTC’s enforcement actions, if our listeners do that, especially if you’ve been doing that for a couple of years, you will see every now and then they come out with an action against some tiny payment processor that has facilitated a really big fraud. They’re looking for things like somebody took all of grandma’s money and now she can’t make her tax bill or her mortgage payment or her rent or something else like that. Those are the things they look for. And unfortunately, there’s a lot of fake businesses that are out there.

Jareau, you saw this. I saw this at Square and Stripe. The payment facilitation side now, there are all these rules and structures and diligence that the banks help the networks do, and the whole industry operates in concert to make sure you’re not enabling the ripoff schemes. They’re going to rip off grandma or the single parent or something else like that. But tiny processors, especially ACH processors, they don’t make enough money. They never commercialize those types of controls on those things. And so what was happening was less around, hey, let’s go get the payday lenders that are operating online. It was hey, here’s this really seedy ACH processor that will take everybody, including the payday lenders because they were in the high-risk space.

And it was something else happening by and large with these processors that really tipped the government off. And so the government squeezed the processor. And then you mentioned earlier collateral damage. And we saw that with the payment platforms with Chase tightening, smaller payment platforms that didn’t make the cut, collateral damage. Same thing on the ACH side. The payday lenders themselves were not expressly targets of this in most cases. It was really just this collateral damage of the tightening on the ACH processor.

And so that’s a really, really good springboard to say it’s important to have good lawyers, important to engage with your trade groups, especially if you’re at scale which is important to your business. You’re past product market fit. You have the revenue to spend on this type of effort, just to keep an ear to the ground and see what’s happening. And the regulators, by the way, FDIC, OCC, FinCEN, etc., are really great about engaging with the industry. So we have a lot of great proactive channels that folks can take, listeners can take if they’re interested. And the best lawyers and all of that, so go hire a good outside counsel because we are not gold on this podcast.

Jareau Wadé: One last note. For as much as there was collateral damage, I do think that helps spur on some of the innovation in platform payment technology. So Vantiv in particular definitely wasn’t the first but they go ahead of the curve, Fifth Third Bank with them on the core benefit of accounts, which we can talk about later, dynamic payouts, a lot of innovations around changing the soft descriptors so what shows up on a credit card statement for our consumer’s bank account or a bank statement. So there are things like that, that actually helped both gain more granularity but also in some ways create a better consumer experience or some merchant experience, which we now all take for granted today and use.

Matt Janiga: No, absolutely right. Obviously, Reggie and I are nerds in this space around regulations. So let’s talk a little bit more because I think you guys had to deal with this. What happens when the regulators come knocking, especially if you’re operating in payments? Because I think you have some really unique insights that listeners will benefit from here.

Jareau Wadé: So we had state inquiry letters from states like Washington, Illinois, things like that. And one funny thing now looking back, so I was just talking about a For-Benefit-Of or FBO accounts is we would have to explain to these regulators what a For-Benefit-Of account was because they weren’t just invented for payment facilitation or aggregation, but they were deployed in a new way. And then strap technology onto it, oftentimes, regulators would get worried about things. But yeah, it’s just funny in hindsight. We would have to send these detailed flow fund diagrams who our banking partner was. They would have to thumbs up this and who was getting the money and how. And going back to them and saying like, hey, you’re worried about us doing money transmission or some other money service business activity.

That’s not the case. We don’t touch the money. We have no custodianship over it. Our bank partner does. But in this case, it’s in the For-Benefit-Of. Everything is cool here. But I think it was not well-known that you could do that. It wasn’t just the assumption that you were doing that. And we ourselves were probably a little naïve about how to interact with the regulators. But that was the initial thing was just we’re moving money for platforms who have buyers and sellers, and we’re doing relatively novel funds flows to make it happen. And then, of course, strapping the API on top of it, putting developer docs on our website instead of sending PDFs out and having this bottoms-up approach. And so some regulators became curious, let’s say.

Matt Janiga: No, that makes a ton of sense. And I think very similarly, I think what I saw both in private practice and then over at Square and Stripe and also Lithic because we did a lot of money transmission exams with those companies. It’s really fascinating and it’s a good reminder of why it’s important to shop around or at least talk to a couple of different outside counsel if you’re doing something novel in fintech. Lawyers only know what they’re fed.

And especially if it’s a law firm lawyer, somebody who’s homegrown, been at the same firm for 10, 15, 20 years, they only really know the work they’ve done or if they’ve gone to industry conferences and have these types of conversations. Not on podcasts but informally over dinners and things, that they get these insights. So it was interesting going into Square because there was a lot of stuff we knew about payments on the legal team but also a lot of stuff that we were encountering sometimes for the first time. The mandate there was the same at Stripe, solve problems. And same at Lithic by the way, that’s how we operate as well. And we know you do it as well at Finix. You guys are out to help customers solve hard problems on things.

But it was absolutely fascinating. And I think we experience the same thing, where regulators sometimes didn’t know what the FBO was. So if you got really good counsel, especially really good outside counsel, if your in-house counsel didn’t have that knowledge, and I did not at first around FBOs, that you could do it. Otherwise, it was interesting. There were some business executives that were hard charging. They’d be like just get the money transmission licenses, not thinking about [crosstalk] the team.

Jareau Wadé: Yeah, exactly, which has its place, you can do both. But oftentimes, it’s just more straightforward and faster to do the FBO account and work with your choir. The interesting thing is this problem doesn’t go away. It just changes. So at Finix now, I’ve had conversations with Push2Card, for example. I won’t name names, so with different banks, so not a counsel but risk or compliance officers explaining to them that the push of money to a debit card is coming from a good funds account. So the Visa/Mastercard operating procedures that define transactions settling to a merchant, in this case, going to a card for a consumer are completely reversed.

So you’re not worried about some chargeback scenario where it’s going to come back. These are good funds from someone who is putting money into a pot to send out to end users. And it just breaks the existing written codified law or rules in this case. So it takes some conversation to work that out and eventually people get on board, but it still exists today.

Matt Janiga: A really good tip there. For our listeners, one thing to keep an eye out for is the CFPB has been reported to be working on some guidance around friendly fraud. They actually just released some chargeback rule updates, if people are interested. If you work in that space, go take a look. And so it would be interesting to see what the CFPB does around it. When you tick the 30,000 policy level, it makes sense. You don’t want somebody ripped off. But anybody who has worked in payments long enough and, Jareau, we know you’ve definitely seen your fair share of these stories. We will get to them at the end of the podcast.

There’s a lot of friendly fraud, When people present that they’ve been ripped off but they’re actually in cahoots, which is my favorite legal term.

Jareau Wadé: Is that a legal term?

Matt Janiga: I’m pretty sure. I’ll pull my Black’s Law Dictionary after the podcast. Not legal advice but my favorite legal term, I’ll go find the page for it.

Reggie Young: Pretty sure, it was on the bar.

Matt Janiga: That’s right.

Jareau Wadé: You’ve defined cahoots. I love it.

Matt Janiga: I think Reggie you’re up next.

Reggie Young: Looking back, is there anything you think you’d do differently if you were to start a platform payments company now?

Jareau Wadé: So I mentioned before, we were naïve. We were all engineers. We had stumbled into payments as a product that we were passionate about. But I think as a result, we didn’t know a lot of what you should or should not do or who to talk to first. We ended up obviously becoming confident because we had to. We were able to drive our chargeback rate down. We were able to handle inquiries. We were able to sign up with three major banks. Eventually, we got it fixed. But I think in the beginning. We were a bit intimidated/unawares of how you deal with regulator inquiries or a compliance officer asking certain questions. It’s not a skill that you learn in college. My first job was writing web scrapers at an e-commerce company. I didn’t understand that.

So just in general, being a bit more open to diving into the industry earlier versus trying to just completely ignore it. That’s the main thing. The one thing that I think we did do well, which I would recommend to others, is we ended up being forced to build really modular and extensible infrastructure and then eventually partnerships, so that we could move between different providers and partners who would say yes to a certain funds flow or to a certain MCC or whatever it might be. That allowed us to be very, very agile in terms of not stepping over any lines because we work with a partner or an underwriter or a bank or a processor who said yes to a certain model. So there’s some complexity in that, both operationally and technically, but it’s manageable. And I think because we were engineers, we thought of that first. And then eventually, that actually turned very beneficial for things like PCI compliance or moving processors or things like that.

Matt Janiga: It’s really great to hear those insights, especially for folks who are early either pre-product market fit or just down product market fit because it’s really important advice. So we’ve touched a little bit on this. We talked about Operation Choke Hold. You mentioned a couple of the early-days things, finding the dog, very important early on. But we know payments veterans, especially you because you’ve done a couple of tours of duties here through the financial system. We have some really great stories. And so we would love to hear about some of those. Tell us more about the headaches you ran into helping folks move money. What was breaking and scaling as that was going on?

Jareau Wadé: Some of this stuff is boring. It’s like, oh, we missed something in reconciliation or a ledger, and we create a new schema or version. But some of the fun ones were around fraud. So if you have a $25,000 credit card transaction run through your system, it’s fraud. There’s no way someone is going to buy something for $25,000 on a card. So we work with the marketplace where we saw this come through. We at some point refund the transaction. It triggers our fraud detectors and then we get a panic call from the head of ops at our customer, who’s like, hey man, I just got a call from this woman who is in the royal family. I think it was the UAE, and she wants to know why her card was declined. We just had to realize that was actually a valid transaction. We kept getting these. I think this woman in particular spent hundreds of thousands of dollars in this way. I’ve never spent $25,000 on a single thing except for maybe a house or a car. But that was fun because it just obviously, obviously seemed like fraud, and we were wrong.

On the flipside, we would do a micro deposit validation. So again, before Plaid was really at scale, where you can verify your bank account using your bank login, we were doing micro deposits where you put a few cents into account and you tell us what those cents are. This is something we just didn’t even think to put any thought into detecting fraud. But we ended up having someone essentially siphon I think it was $400 over the course of three months. I think they probably wrote a script repeatedly creating bank accounts, registering them for micro deposits. And then we even talked about this internally. We’re not going to return the ACH for a few cents. It costs more to put the money there. So it will stay there. So they would just harvest three cents, eight cents, two cents, seven cents for months, and eventually we detected it. We’re like, oh, this account is doing something really weird. So again, it was something where from then on, I never underestimate the incentives someone has to defraud you if there’s any amount of money available.

I’m trying to think of other ones. Crypto is booming at one point during Balanced, and we kicked off all the crypto exchanges in the platform. So there’s one that was cleverly disguised as a contracts marketplace. I didn’t know what it was. We let it go for a bit. And I remember having to call this guy and explaining to him that he lost $40,000 because an ACH transfer had come in. He was effectively a node on the ripple network, and so there’s an IOU token essentially that goes somewhere else. Someone was able to exchange through another node, get their cash, and then they file an affidavit with their bank saying I didn’t register that deposit. So now the money is out both ways, and this customer was responsible for it. So that was a tough call.

Then the last fun one, maybe Matt, your former employer is involved in this a bit.

Matt Janiga: BlueVine?

Jareau Wadé: No, Stripe. We used to run Balanced as an open company. So what this means is we did product development and a lot of open conversations with our customers on GitHub issues. And one of the things that would come up was Push2Card. So this is why we created this because customers asked for it. We did our whole this is the spec, this is the design of the dashboard, this is the pricing, this is our launch date, etc. One of the downsides of an open company is that competitors are watching this. And so on the day that we’re supposed to launch, I remember Stripe launched the same thing with what appeared to be a somewhat hacked together version using unreferenced refunds. And so it was just a bummer. We got into a hacker news spat with them, but it was all in good fun. So those are the war stories that come to mind that are interesting and colorful.

Matt Janiga: That last one is really fascinating because who knows what was malicious and what wasn’t. You will never get the straight story out of Stripe, what was happening inside. I don’t think I was there at the time.

Jareau Wadé: It wasn’t like they did anything illegal. They just looked at our GitHub and launched a product.

Matt Janiga: Same features. Yeah, highly coincidental. The interesting thing will be to see because Moov is operating in that way now. I’m in. I don’t have much time to read it anymore. I really appreciate what the folks there are doing. Anyway, all the folks at Moov are really fantastic. We’re excited with Lithic speaking at Fintech DevCon. We have some great folks from our revenue team going to talk about how we build products and work with consumers, and that’s a really fantastic thing if folks don’t know about it yet. But the interesting thing is they do so much in the open. And if you’re in that Slack channel, you can go peek into what people are up to, what are the hard problems that they’re solving that are there.

And I know on the Lithic side, we’re not malicious about it. We don’t have anybody in there that, oh, what’s Moov up to, kind of thing. We’re cheering for them from the side. We think what they’re building is really fantastic. But it will be interesting to see can they sustain that long term if someone does take a malicious tactic to it or they close ranks at some point with the channel, which should be unfortunate because it is fun to have the opportunity to [crosstalk].

Jareau Wadé: It’s good to have that collaboration and openness in the community. Do we have time for more?

Reggie Young: I think we got a few more questions. First up, if you had to articulate Balanced legacy, what would it be?

Jareau Wadé: I feel like someone who’s thinking back fondly on his high school football days. It’s impossible to prove if we did any of these things first, but I know we were among the first to do them. So the biggest one is the concept of doing a white-label merchant onboarding as a PayFac just didn’t exist before. So even Stripe Connect, which they released to compete with Balanced, I think, 2012, 2013, the first version of it was they just sent your merchant to a Stripe dashboard to sign up for an account. So it wasn’t likely but it wasn’t fully embedded. So that’s I think definitely something that is now at scale. There are various providers and companies doing that experience. So we were one of the first to figure out how to do it via API, not as the actual platform, but as the provider for the platform.

A few other things. This is very specific and technical, but we added an item potency parameter to our API calls, specifically for transaction processes.

Matt Janiga: Very important, very important now for API providers to have this.

Jareau Wadé: So funny enough, we actually pushed that to production at the request of Square’s cash app team who wanted to make sure they weren’t multiple pinging our same endpoint to send ACH transactions. And now every payment provider does it, Braintree, probably BlueVine, Stripe, Square, everyone in it to the point where it’s going to be adopted as I think a web standard. So that was cool.

Matt Janiga: I can see that. Yeah, DDM, Internet consortium, all that. It is like WC3, I think, who had a big hand in the running because they have some folks for some time now.

Jareau Wadé: We did a partnership with GroupMe back in the day where we were doing in-message payments, which obviously now is like WhatsApp, Facebook, Messenger. All of that stuff has that, so that was cool. And again, these are all things that we just did early. Who knows if they were going to happen eventually or not? And then the other thing, PCI architecture. Actually a shoutout to my CTO and co-founder, Mahmoud. He actually came up with this very novel PCI architecture when we narrowed the scope to just 1:1 subset of our architecture, that actually had raw card numbers. Everything else uses tokens.

So he actually started a company to raise money for Andreessen Horowitz and everyone for that business. So a bunch of stuff like that, that is still in the wild. So it’s cool to see now platforms obviously are a huge engine for the economy. That’s what we focus on at Finix, and we think it’s going to be one of the big drivers over the next couple of decades.

Reggie Young: Those are all great legacies to have. Even if you’re not the first, it’s still something to say. You were one of the folks at the helm figuring stuff out. So we’d love to fast forward to today a bit. Tell us more about Finix. What does a typical customer and use case look like? And what’s generally going on with the company nowadays?

Jareau Wadé: So I mentioned Finix, like Balanced, is focused on platforms. So we moved away from marketplaces, which was a very specific thing to 2010, 2011. But platforms so it could be vertical SaaS. It could be marketplaces. It could be an e-commerce platform. Anyone who’s moving money between one party and another. That's what Finix provides payments technology for. So we are working with start-ups that are doing zero dollars per transaction volume all the way to publicly traded companies doing over a billion. And we provide an API dashboard to move money.

So you said what’s going on with this now, updates. We’re doing billions of dollars in volume. I think we recently shared in May that we actually doubled our transaction volume from 2020 to 2021, which is huge because I think we quadrupled from 2019 to 2020. 2020 was obviously a massive year for e-commerce and that growth continues. And another fun stat because I think this is some of the impact we have. Like I said, the economic engine of the future, through the platforms we support, we’re now settling funds to over 12,000 active sub-merchants, so it could be small businesses, restaurants, gyms, places of worship, on and on and on. So I think it’s a really cool stat that I like to personally track.

Matt Janiga: We are so glad that you guys are doing well. There’s so much space in the payment space. If you take a look at the numbers, still I think less than 20% of commerce is happening online. And so if you think about all the things that a company like Finix can help, and they’re still so much to go. I don’t carry cash anymore. And so far, it has worked out okay. The only time I get hung up is at the softball snack stand where my daughter is having games. The season just ended last night. So I don’t have to worry about cash for about another year.

Jareau Wadé: So you’re good now?

Matt Janiga: I’m good for a year. Yeah, it’s really funny. But it’s just fascinating to see all that. And so it’s great to have so many great providers, and Finix is certainly one of them out there.

Jareau Wadé: And similarly, we are fans of what you guys are doing on the issuing side. It’s not something that we play in but it’s nice to see a developer-first platform for issuing. It’s great.

Matt Janiga: Thank you. If you’re out there and you’re looking for these solutions, understanding what you need, what are your jobs to be done and which providers can best help fit those jobs is really critical because a lot of companies look the same from their website’s landing page. But what they offer and how they can help you get those jobs done is really different. Lithic, for example, we can be highly modular, which some people really value. Obviously, Finix has a lot of great attributes that set it apart from some of the other providers in the space. And so we encourage folks to check it out if you’re looking for services from Finix.

We have some Internet questions. Do you guys want to turn to those? I actually have that Twitter thread holed up where people were pinging. Do you guys mind if I just ping a couple of them?

Jareau Wadé: I put them at the end of the doc too.

Matt Janiga: Oh, okay, maybe we will do that. So we have some good ones here. Do you guys just want to run through the bullets? Or do you want to hop around? What do you think?

Reggie Young: I’m down to run through the bullets.

Jareau Wadé: Yeah, let’s do it.

Matt Janiga: What are the mysteries of PCI compliance? This is actually really fantastic because I think we’ve each seen pieces of that now. Jareau, would you want to kick us off on this?

Jareau Wadé: I’ll go into some of these questions that came up for the legacy of Balanced. One of the first things we realized at Balanced is that the default assumption is that you have to have your entire system locked down for PCI, which makes it really difficult to do things like give access to anyone in the company because there’s only a certain group of people who should have access to PCI-sensitive environments. And so I would say one of the mysteries or misunderstandings is you actually don’t need the card number for that much. So there are use cases where you might want part of the card number, so the last four or the first six, for customer support or to look up what the processing rates might be for a card.

You can actually do that in a PCI-compliant way. The use cases where you actually need the full 16-digit PIN are usually to send it out and transact with a network or a banking partner. So the biggest tweak I would put on people’s understanding of PCI is to interrogate where you actually need that full card number and then limit the scope as much as possible to some environment that can store the full 16 and then use either the limited parts, the last four or first six, or token everywhere else.

So again, I’m talking of our own book because this is what we did at Balanced. But I haven’t seen a better PCI architecture or model since then in the last 10 years because it complies with all the use cases for customer experience, complies with the PCI data security standard. And we have the technology to do this. It’s called a reverse proxy. It’s called a secure vault. You have all of that stuff. I’m curious about your guys’ takes.

Matt Janiga: I’ll probably approach it from the builder's side. So I think if you’re a founder, you’re listening, or you’re an early team and you’re listening and you have to worry about it today, I would say definitely go explore some of the solutions like very good security that out there, talk to you payments provider, the ones that you’re actually working and running things through.

Jareau Wadé: Basis Theory is a new one too.

Matt Janiga: That’s right. Because some of the companies will offer you tokenization and vaulting solutions, which is what Jareau was just referencing. It’s super helpful. Otherwise, because I’ve seen now how PCI compliance is run at a couple of different companies, you have to effectively staff a small team to get you up and running. And that team goes away and that they are fully absorbed by your PCI audit once a year and it’s usually a couple of months’ worth of time. And so you don’t have to spend those resources when you’re small and you can leverage a very good security or someone else like that, I’d say explore it. I have not used them but I have helped teams through PCI audits, usually helping prepare paperwork where they have questions on what’s in scope or out of scope and what they need to disclose to the auditors. So I’ve seen enough of this to know what it is.

The other thing that may surprise people that are new to the space, PCI is pass/fail but not in the way I got through law school. It’s pass/fail in that you get all right or you’re wrong. And so one screw-up on PCI will cause you to flunk your PCI audit or you will have to go in and fix it. And we all know because we’ve all worked in technology. And our listeners certainly know, it’s not like there’s some magic switch you flip to do it. Somebody has to likely go touch code or you’re going to re-architect the way systems interact with each other, which takes time.

So if you fall out of PCI compliance for something like that, it can be a really big headache. And that’s why I think for folks who can, go explore a provider. I think what AWS and Google Cloud and Azure can do today for folks, these providers can potentially do for you in the PCI space. Any payment provider that generally is helping you touch card rails can also help you tokenize or provide solutions to help you be compliant, understand how those fit into your PCI compliance needs as well.

Jareau Wadé: Matt, maybe think of one more. One thing that people don’t understand about PCI audit is that often that results in a human being sitting in a room in your office for days to weeks. So you mentioned they absorb the attention of your whole team. That doesn’t happen for most technology companies, that you have a third-party auditor who just sits in usually a windowless conference room and looks at logs and things like that. So it’s a strange thing.

Matt Janiga: No, that’s absolutely right. We got another good question from one of our favorite Twitter lawyers. How should a payments company diligence its clients? I know we’re happy to share a couple of high-level things we do on the Lithic side. Many of our customers have come through the process of experiences. We hope it hasn’t been too much of a hassle for you all. Jareau, we are also curious, how do you all think about this at Finix? Is the client safe enough to work with?

Jareau Wadé: I think we had another question, it was like, how do you balance the innovation and the risk? And I think they’re all merged together because it’s not just innovation but it’s also growth. The biggest kind of failure mode that I’ve seen is that folks will be too binary. They’ll conflate a lot of the different risk factors. So if you are just looking at it by, does this present any risk, like binary yes/no? Then you’re going to be overly conservative to the point where you won’t, one, build the innovative products or tools that you need to support them. And then two, you actually won’t experience the growth that customer might bring to you.

So maybe a controversial thing but I think when you’re diligencing a customer on a portfolio basis, if the answer is that you’re going to have a zero risk, I personally spending most of my time marketing growth would be very concerned about that because there is not enough value in that customer in terms of the upside to justify reducing risk to zero. So I think that’s my first tweak. It has to be, one, a portfolio approach where you’re assessing risk over a group of people or cohort. And then two, it has to be balanced both on risk and revenue or growth opportunity or whatever it might be on the plus side. And too often, I see folks say, well, this is something that could get us in trouble with compliance or could get us in trouble with risk or financial loss, whatever it is. And all of a sudden, we are going to shut it down. I don’t think that’s an appropriate answer.

The second one is there are different types of risk, so if it’s financial risk. And this is why I think they are linked. You might say because they’re going to bring us X revenue on again a portfolio basis, we’ll say yes to that financial risk. Then there is compliance risk which could be binary and say, if we step afoul of this rule, we get shut down or our business is gone. That might not be a good risk to take. And so disentangling those is actually probably more useful than having a binary approach to things.

Matt Janiga: I’ll add another dimension too, which I think is interesting or for folks to chew on. Because if you just take what Jareau said now, it’s really good insights. But if you apply it to the wrong business model, you might find that your business partners are going to get rightfully upset with you. I think the other thing is velocity to using the product. And it’s funny. We work with this great guy, Gil Rosenthal, who Reggie and I know from BlueVine. Now Gil is out there consulting. So if you’re a smaller start-up and need help in the risk space, we can help you connect with Gil. He is absolutely fantastic. Gil runs a risk session and he talks about, are you a carnival model where you pay per ride, ride per ticket? Or are you a theme park like Disney where there’s a really expensive entry point when you come in?

I think something like Lithic, you can’t just launch a card product instantly. There are too many laws and rules around it. We’re working at creating some safe spaces for that and we’ll continue to chip at that. But by and large, if you’re going to do a consumer card product, let’s say in crypto, you’re touching a couple of very, very heavily regulated areas and areas that also partners want to see some thoughtfulness around. So that’s an interesting thing where we can do things like we do background checks. We will look for negative news right around stuff like that. And we have the time to do that because you’re not rushing to get live tomorrow on your product.

I think the flip of that, I think about Square where you can walk into Target or Staples and I think still buy a Square reader. I’m old so I used to be able to do it. And that’s how my mom actually got her first Square reader. She used to sell things at the craft fairs and take Square reader with her. You just go and buy it. And you don’t care if my mom, she doesn’t, if she’s been arrested for something or anything else like that. Because at the end of the day, if she can pass KYC after the bank standards and she is not a known fraudster according to fraud risk models, you can give her the reader. And you can do some other things on payment processing like you can slow down when you pay her if anything looks funny on that end to help protect yourself and protect the network as a whole.

So that’s an interesting model where like you wouldn’t clobber my mom selling tooth fairy pillows at the craft fair. But you might dig a little deeper on somebody standing out platform-based or risk issues. So that’s another thing I’ll toss out to folks. And then in terms of the other details, we look at the normal stuff. We look at what’s your financial health. We get a sense of what your business plan is or what you’re looking for and things like that. And that’s how we look to control risk. And each way, we are looking to see if there’s something funny or fishy around it or is there a problem with the founding team. Again, negative news, background checks for arrests or other things like that, bankruptcies that may come up. So that way, we can understand what’s going on.

One tip, if you have a common name and you know you have a name twin out there that is bankrupt and constantly getting arrested or other things like that, come prep with your provider. For example, somebody showed up at Lithic and we had a name twin issue recently and something like this. If you know about it in advance and you tell us upfront, this is somebody else, this is why I’m different, it helps us spot that out of the gate. And then we or any other provider can move faster. So if you find yourself getting blocked on stuff and you’re not sure why, that might be something that’s out there.

And then there is the reputational risk when we talk to our bank. I’m sure you guys talk to your outside partners as well fairly regularly.

Jareau Wadé: I thought you were going to recommend folks to change their name if they run into that issue.

Matt Janiga: I’m pretty sure that’s why Meatloaf changed his name. I think the old one, they wouldn’t let them play in venues. Rest in peace, Meatloaf. He would do anything for love, just not that.

Jareau Wadé: I got to go tour a party facility for my daughter’s fourth birthday soon, so I got to wrap up in a bit. What do we want to hit? Anymore of those?

Matt Janiga: Do you want to do the questions for us?

Jareau Wadé: Yeah, sure. That would be fun. Yeah, we could do that. So I have a question for you guys. I know you’re supposed to be asking me the questions. But actually, I’m curious to get your take on this given your building a payment business on the other side of the stack. So I’m curious if we can compare notes. So if you are building a payments company today, and you don’t have to speak to Lithic specifically, but just generally hypothetically. What strategy would you use to approach that?

Reggie Young: Can you say more and give a little more color on what you mean by strategy?

Jareau Wadé: So let me break this down. As I look on the acquiring side, there are a couple of different flavors of how processors are essentially approaching the market so you have someone like Adyen who is almost religious about not acquiring any company, building everything in-house, soup-to-nuts. On the other end of the spectrum, you have the legacy providers, the Pfizer, the FIS, who are essentially a collection of acquisitions over the last few decades. And then somewhere in the middle, you’d have a Stripe who seems to do this partner-first to enter a market or new functionality, and then they will essentially build it themselves and go as far as the metal middle as they can.

An example would be to enter Europe, Rapyd/Valitor, which is an acquirer from Iceland. And then now, they are registered as an acquirer in the EU. So generally, what is your take on the tradeoffs of how you would approach building something maybe on the issuing side or maybe not in the space?

Matt Janiga: That’s a really interesting question. I think again it’s great to think multidimensional, and so you teased out a couple of good strategies. I’d say it depends on scale. If you are a founder and you don’t have a product yet, you’re just trying to figure out, “I want to do something in payments. I want to bring my domain knowledge to it or I think it’s a hard set of problems. I want to try and solve a piece of it.” I’d say focus on things people aren’t doing well or aren’t serving the market today.

I think that was the genesis of Lithic. We went out. We were not happy with our issuer processor for Privacy.com. We kicked the tires on a lot of different solutions, companies we now compete with, and they could not meet the very unique needs we had for Privacy.com. And somebody internally said, hey, if we go talk to these companies, we could do this ourselves if we line up the right set of partners and build the right backend systems. And that’s how we got started on that path. And I think that’s one of the things that focuses us.

We keep looking at, how is the market not served? How can we bring our knowledge, stability? Our technology is pretty good but it gets better every day. You look at companies like Alloy, Hummingbird, Unit21 and others, they are doing that in the identity space, a really critical need to get started in anything in payments or fintech. And they are helping take that modular piece and do it really well. Now let’s say that scale, so now you’re really big. And I got to operate in this space from Square and Stripes. It sounds like Finix, you guys are there as well, which is really fun. Now you can say, hey, we’ve solved a really hard problem once. Can you do it again? And either you could be pulled by customers, you can hunt the market and say here is the market that isn’t working well. And a bunch of providers, they’re overcharging customers.

The economics are flipped. We can do it cheaper, more efficient, more modern and capture the same margins by doing it this way. And I think you can start to do this and expand. We were joking earlier about Stripe launching a similar product to you guys on the same day. That’s where I think some of the consternation comes in because you get so big and you’re placing a lot of bets that eventually you have a hand in every pot. Yeah, it looks like you’re competing with everybody. But really, what you’re doing is you’re serving customers. They want that supermarket. It’s like the super Walmart. I could buy a toy. I could buy a swimsuit, and I could buy milk and cookies.

Reggie Young: Yeah, anything you can possibly need.

Matt Janiga: That’s right. Whereas if I want really good cookies, I go to my local bakery. That’s all they sell.

Jareau Wadé: We call that the SUV effect, where it’s like I’m going to buy this car because someday I’m going to go off-roading and so I need all the suspension, 4-wheel drive. And 90% of the time, you’re driving it to Whole Foods and then back. That’s it. But you could use that if you want.

Matt Janiga: Reggie, how about you? Any other thoughts on this? Because I know you wrote a really great piece for some of these newsletters on this, who is going to build the next payment network?

Reggie Young: I think you hit the key thoughts I had, Matt. I think these questions are fundamental questions around making sure you’re checking all the boxes of who you need to actually support. I think the big framework that I have in my head is if you want to build a payments company, like a network type company, you should think about the consumers. You should think about businesses. You should think about in-person payments and online. You have to think of four quadrants like that, a two-by-two grid and each quadrant has its own unique considerations that depend on what your company’s strengths are at the time that you could use as the best wedge. So it varies by company use case.

Jareau Wadé: I appreciate you guys being good sports about this. I think about payment model strategy and also some of the nitty-gritty, and so it’s always fun to talk to folks who are in the same space but from a different perspective. So I appreciate the thoughts.

Matt Janiga: Absolutely. Absolutely.

Jareau Wadé: That’s all I got for you. Any last ones for me?

Matt Janiga: What's your favorite kind of cake?

Jareau Wadé: Pie.

Reggie Young: Any favorite type of pie?

Jareau Wadé: Apple. I have fond memories of my mom making me apple pie as a kid. A quick funny story, as an adult, I didn’t realize you could just go and buy pie from the store. It was just a foreign concept to me. I remember I was working at a company and my co-worker was like, if you’re jonesing for some pie, you just go to the store and buy some. And it dawned on me like a revelation that I could spend 12 bucks on a pie and eat it at night. Because that childhood memory is imprinted so strongly, so that’s why I say pie.

Matt Janiga: That’s funny. That’s really good.

Reggie Young: Love it.

Jareau Wadé: Well, thank you guys so much. This was a lot of fun.

Matt Janiga: Thanks for joining us today. This was great. We’d love to have you back, so let us know when you want to come back in the podcast.

Jareau Wadé: I’d love to be a repeat guest. That would be great.

Matt Janiga: Thanks, Jareau.

Jareau Wadé: Thanks all. Bye.

This episode of the Fintech Layer Cake podcast features a deep dive on the payment processing space with Jareau Wadé, Chief Growth Officer at Finix and co-founder of Balanced, who’s had a front-row seat right from the start.

Fintech Layer Cake is presented by Lithic and hosted by Matt Janiga, our general counsel, and Reggie Young, our product counsel and author of the Fintech Law TL;DR newsletter.

Highlights

What was it like starting a payments company 10 years ago?

Reggie: I think everyone would think starting a payments platform company is a no-brainer today given Plaid and Stripe’s valuations. But what was it like pitching to investors with this idea before those companies were a sure thing?

Jareau: Everyone just assumed PayPal could be used for everything, and I don’t know if you have looked at PayPal’s API. Certainly in 2010, they were not great. But the idea of a developer-friendly experience for payments was not that well-known at the time. And this seems silly now because that’s what everyone is moving into. Obviously Stripe’s successfully popularized this idea. But at the time, we were one of the first to do that. We just wanted to move money quickly.

We started to realize that there was also not necessarily a lens on exactly how transformative financial services and digital payment tools would be to the majority of humans on the planet. So that’s one reason that we wanted to build more infrastructure so that other experiences could be built on top of that. But at the time, I don’t think the VCs really got that idea.

What was it like before all these APIs and other service providers?

Jareau: On the payment side, it was very different. There weren’t really APIs or, if they said they were, they were not up to even modern standards at that point. But we used Stripe. I think we were part of the first 100 customers of Stripe. They were called dev/payments. Didn’t have what we needed. I remember getting 80-page PDF specs from some of the bigger processors and banks and working through those. But it was slim pickings. It was not clear how to build it. So as a result, we built a lot, not only the integrations, but some of the internal services ourselves so that we could build on top of them to create whatever experience we want with our API.

How did Balanced initially find a processor?

Reggie: We’d love to hear more about finding a processor and that whole experience. It feels like the biggest chicken and egg thing for payments companies is like, at the time, you need to work with a bank like Wells Fargo and JPMorgan. But they probably won’t talk to you until you’re big and have your own team, controls and everything set up. So how did Balanced deal with that cold-start problem?

Jareau: It was difficult because we were getting market signals and growth. So we were starting to have real customers and real volume. But we were too small to be underwritten by a large payment processor. So we had to do the resourceful, scrappy start-up thing, which is just find a way to make it go forward and deal with the next problem as you get there. So specifically we set up a series of merchant accounts through various ISOs, and we processed payments on those. So it was like buying T-shirts. So say we had a T-shirt company, that’s what Balanced was, a hypothetical story. And we were screen printing them, but we were getting our inventory from Costco.

When you were at Balanced, what kind of customer education did you have to do?

Jareau: So the biggest thing that we found is that customers would come to us for their hair-on-fire problem. We’ll start at marketplace. We need card acceptance. They might go get a solution then. It might be a traditional merchant account. It might be Stripe at the time, Braintree, whatever it might be. And then they would be making manual payouts through PayPal or writing checks or something like that. Over time, they started realizing, hey, we can’t scale, either for compliance reasons or for just manual operational reasons. We can’t scale the payouts like that. So we actually started positioning ourselves.

We had three things. We had processing, escrow, payouts. But we started positioning the payouts piece first because we realized that was the best way to meet the customers where they are. And then eventually we would sell them on the full stack, the full solution. Similarly we would often find people would get cards in their mind first, and then we would sell them on bank payments and then cross-sell them into cards. So part of our model was educating folks on the different payment types, the different payment models of a platform is not just accepting cards, it’s also disbursing.

What was your experience interacting with regulators as you were growing Balanced?

Jareau: A lot of the acquirers like the Chase, the Wells Fargo, etc., Vantiv, they got spooked by what was going on in the treasury department with Choke Point and often there were questions. Is there payday lending? Not aggregation per se was in the sites but there was a lack of visibility and granularity into some of the aggregated payments volume. And so when the processors would come up empty-handed about who was the actual underlying recipient, I think that’s one of the things that spurred on the development of the payment facilitator process.

There was some collateral damage though. As far as I heard, Chase actually went in and just kicked off a bunch of aggregators except for the big ones like a Square or an Intuit.

What did Balanced do exceptionally well in the early days?

Jareau: So just in general, being a bit more open to diving into the industry earlier versus trying to just completely ignore it. That’s the main thing. The one thing that I think we did do well, which I would recommend to others, is we ended up being forced to build really modular and extensible infrastructure and then eventually partnerships, so that we could move between different providers and partners who would say yes to a certain funds flow or to a certain MCC or whatever it might be.

That allowed us to be very, very agile in terms of not stepping over any lines because we work with a partner or an underwriter or a bank or a processor who said yes to a certain model. So there’s some complexity in that, both operationally and technically, but it’s manageable. And I think because we were engineers, we thought of that first. And then eventually, that actually turned very beneficial for things like PCI compliance or moving processors or things like that.

If you liked this episode, subscribe to the podcast on your favorite podcast app and give us a review on iTunes.

About Fintech Layer Cake

Fintech compliance. It can be complicated and overwhelming — even if you've been in the industry for a while. But what if there was a podcast that made learning about it a piece of cake? That's what Fintech Layer Cake is about.

It's hosted by two popular fintech lawyers, Matt Janiga and Reggie Young. In each episode, they use their experience from working at companies like Lithic, Stripe, Square, and BlueVine to break down some of the toughest topics in fintech.

Listen on iTunes, Spotify, or your favorite podcast app.