How Petal founder Andrew Endicott proved alternative credit works, and what he’s betting on next

If you’re interested in investing in and learning about emerging fintech startups, join our community of current and former banking and fintech executives. You’ll be part of an exclusive group of out-of-the-box builders and investors building a community to invest in the next wave of fintech startups around the globe.

Sign up to hear about the latest developments in fintech globally, and access exclusive investor materials and opportunities.

The credit industry is shifting how it evaluates borrowers. Traditional credit scoring has left over a billion people without access to financial services, but lenders are increasingly turning to alternative data—bank transactions, spending patterns, real-time financial behavior—to make more informed decisions about creditworthiness.

Andrew Endicott has been at the center of this shift. As co-founder of Petal, he pioneered what they called “cash flow underwriting”—using real-time bank data alongside traditional credit reports to approve people for credit cards who would otherwise be turned away. The approach worked: Petal raised nearly $1 billion, proving that alternative underwriting isn’t just better for consumers—it’s good business.

Watch the full episode

Listen to the full episode

Apple Podcasts I SoundCloud I Spotify I Google Podcasts

After starting his career during the 2008 financial crisis, Endicott went to Harvard Law School, worked as a corporate lawyer, then moved into investment banking focusing on specialty finance companies. Around 2015-2016, he spotted the opportunity that became Petal and left his banking job to build what would become one of the most successful alternative credit companies of the last decade.

Now as co-founding partner at Gilgamesh Ventures, Andrew is investing in early-stage fintech companies across the Americas. The fund typically invests at the seed stage, putting in about half a million dollars into rounds of several million, with a particular focus on companies in the United States, Brazil, and Mexico.

In our conversation, Andrew shares the biggest hurdles Petal faced in the early days—from lack of FinTech infrastructure in 2016 to sophisticated fraud attacks that required constant innovation. He explains how alternative credit has evolved from a novel concept to mainstream adoption, and why he believes we’re entering a new wave of FinTech innovation distinct from the post-financial crisis era.

We explore Gilgamesh Ventures’ investment thesis, including their focus on companies solving compliance challenges, AI-powered financial services, and B2B marketplaces. Andrew provides insight into why he’s bullish on the Americas as a region for FinTech investment and shares his thoughts on emerging trends like stablecoins.

The big points

• The catalyst for Petal: “I had a really good friend named Burke who actually also co-founded Petal with us, who’d come from Switzerland. He had all these amazing job offers and internships, and this guy was going to do financially well, it was kind of obvious. Yet him, in particular, he was this edge case for underwriting credit where you didn’t have any history.”
• The power of bank data: “The biggest and best source of really signal in the financial system that was available, and I think still available today, to make those decisions, really is the information in your bank account. That information is very fresh. It’s updated constantly.”
• Fraud as the biggest challenge: “The people that are committing fraud in the economy, particularly with financial products, are probably some of the smarter people out there… they never sleep. And you have to get it right every time. They have to get it right once.”
• The current FinTech moment: “Three or four years ago, the number of ideas that were financeable was beginning to narrow… today, the pie has been completely rebaked, and everything is up for grabs.”
• Investment approach: “We don’t want to be kind of the boring of kind of slow moving fund… We like to get in the trenches with people. We want to help in tangible ways.”
• Compliance innovation opportunity: “A lot of the frustration of using any financial product is caused by these other forces… You can really kind of stiff arm that problem in a lot of ways, and make financial products delightful to use.”

Read the full transcript (for TS Pro subscribers)

The journey from Harvard Law to fintech pioneer

Endicott: Yeah, absolutely. You know, I’ve kind of come to where I am from a couple different directions. So I kind of began my career, kind of in the midst of the financial crisis, which today is becoming less and less relevant to things, sort of how we mark time in finance.

Yeah, absolutely. So I began my career then, you know, kind of couldn’t really get a job out of college, so I decided to continue going to school. I went to Harvard Law School as a result. I spent my three years there, and then moved to New York, and I was a corporate lawyer for a while. Had to pay off my student loans, which were hefty. Really enjoyed that, but I didn’t really enjoy the fact that lawyers didn’t get to make decisions. And so I wanted to be a decision maker. That was important to me.

So I left the legal profession, dropped my bar license, spent several years as an investment banker, really, at the beginning of kind of what we call FinTech today. I worked mainly specialty finance companies and non bank lenders. At the time, we didn’t call it fintech back then, did we? Exactly. Yeah, it was, I think most people call it specialty finance or other things. That really wasn’t a standard, but did that for some time, built a network, started to see some interesting trends. It was a pretty interesting time, and then spotted the opportunity that eventually became Petal, and teamed up with a good friend from law school to build it.

Left my job a little bit before it was quite as common to jump out and start a company as today. You didn’t have quite as many stories, as much infrastructure, but around 2015-2016 started Petal, and then spent most of the next decade building it. As you said, in the lending and the alternative data, alternative credit space, learned a huge amount. Well, I’m sure we’ll talk about it some but, you know, had the experience of starting something from scratch and continuing, and we eventually exited the company several years ago, and ever since, I’ve been focused as an investor in the thing that I have come to learn is fintech.

With Gilgamesh Ventures and my partner Miguel, we invest in early stage FinTech companies at the seed stage primarily, but also things around that pre-seed and seed-ish, as I call it, and we do it globally. We invest a lot in the United States, but a lot outside the United States as well. Big emphasis on Brazil and Mexico in particular. And we’re really looking for opportunities to invest in people that are going to build the next kind of things that define this wave of innovation that we’re in today, which is, in my opinion, distinct from the wave of innovation that we were in kind of after the financial crisis. It’s become different. Things are a little more fresh now than they were probably five years ago. So it’s an exciting time to be where we are.

Finding the idea that became Petal

Endicott: Yeah, it’s tough to really pinpoint, you know, kind of where a company starts. I mean, you can go back and there’s a little narrative bias, I think. Sure, how you tell the story exactly.

Yeah, I’ve been fascinated by credit for a long time. It’s something that I just find interesting, but it’s a complicated topic. I think whether we need more credit or less credit sometimes is an object of debate. And in the United States, it’s not straightforward, but it is very important. And going back, you know, to when we were starting the company and before that, it was apparent, and it’s been apparent for a long time, that the way that we underwrite credit, traditionally in the United States, is not perfect.

It’s pretty good for people who have very established track records in using credit, so somebody that has paid their bills for a very long time, or somebody that has not paid their bills very often for a long time, for those types of folks, it’s pretty obvious kind of what, how using traditional tools of underwriting, how they should be treated, how you price it, whether you want to approve somebody, and so on. But there’s, coming out of school and beginning my career and meeting a lot of people who are my friends and others, I knew a lot of folks that had problems that didn’t really make a lot of sense.

Why alternative credit was necessary

Endicott: Yeah. So I had a really good friend named Burke who actually also co-founded Petal with us, who’d come from another country. He came from Switzerland. He had all these amazing job offers and internships, and this guy was going to do financially well, it was kind of obvious. Yet him, in particular, he was this edge case for underwriting credit where you didn’t have any history, because he came from Switzerland and hadn’t been in the U.S. Had he been born in the United States, it would have been a much easier decision to underwrite for credit and the FICO Bureau information, all this stuff is not going to fix that. You have to take an alternative approach.

And the biggest and best source of really signal in the financial system that was available, and I think still available today, to make those decisions, really is the information in your bank account. That information is very fresh. It’s updated constantly. You get to see tons of information about how you’re making money, spending money, saving money, all these things, and that data is populated much more quickly and more democratically than credit bureau data, which is slow and depends on taking very specific actions that not everybody takes. So that’s really what motivated the initial kind of run at using this very rich alternative data set to underwrite credit.

Now, when we started the business, we didn’t initially see that we were going to be underwriting credit cards. We actually wanted to sell the credit score to other people, which that today is its own business, and my co-founder runs it at Prism Data. Right? Selling cash flow scores to lenders in the exact same way that we used them at Petal. But we initially wanted to do the scoring, which is kind of what’s happening today.

Proving the model by taking risk

Endicott: That’s right. That’s exactly right. So we had to prove it out. And to do that, you had to build a lending company. And so we did that. We stood up and built a credit card company, which is no easy task, if you ask anybody who’s done it. But that also took off and then we spent, like I said, the next decade, building that and raising capital, building relationships, building a really amazing team to build that business. It achieved pretty significant size. You know, around the time we sold the company, we were doing over $100 million in revenue. So there’s something very clear that resonated with the market and with consumers.

The unexpected challenges of scaling

Endicott: Yeah, I think one that’s, it’ll be interesting to folks doing this today that really is different, is there wasn’t nearly as much infrastructure. When we started in 2016, you didn’t have all these kind of FinTech friendly processors and just various counterparties required to launch a credit card company. They kind of didn’t exist, or they weren’t as easy to access. That was one major obstacle.

And there are elements that today you have to – there’s a credibility gap that you have to jump to launch a company like Petal, which is very similar to a bank, you know, you have to convince a lot of people that you’re going to be around in six months. And that’s hard. That comes down to your team. It comes down to your capitalization, and kind of your other elements of just seeming stable. So all of that was a big obstacle.

I think, as we launched the product and we really started fighting in the trenches every day with go to market and customer acquisition and so on, the biggest thing that really sticks with me from then was really dealing with fraud. You know, fraud is such a complex topic, you can spend hours talking about it, and kind of only scratch the surface of what’s happening. It changes constantly as well. The people that are committing fraud in the economy, particularly with financial products, are probably some of the smarter people out there, and they never sleep. And you have to get it right every time. They have to get it right once.

And all this today is becoming so much more complex with artificial intelligence. But even before that, the sophistication of the kind of the bad actors that we had to deal with was significant, particularly once we started achieving some scale. The sophistication of the attacks that we would receive from pretty organized sources was really severe, and we lost money doing it. You had to change things that you didn’t want to change to be able to combat it.

Something that I really believe is that at the end of the day, most financial institutions, most financial providers, want to have a good product. If you lived in a perfect world, they would all dedicate a lot of attention to making the consumer happy, they would make things easy, so on and so forth. These products are hard to use and uncomfortable and kind of cumbersome, a lot of times because of other problems. They often are because of fraud, because of regulation, particularly anti-money laundering regulation, or just prudential regulation, and trying to really avoid risk. All these things cause financial actors to put in a lot of roadblocks, particularly from the consumer’s perspective, into their product, and to make things harder, which is really unfortunate.

Innovation despite regulatory constraints

Endicott: But that’s actually where the innovation is and your ability. So with Petal, we wanted to underwrite credit better. We wanted to assess risk better, using this cash flow underwriting. And that was something that no one had really done. It was when, like you said, when Plaid was kind of really coming out, we were able to leverage that new technology of bank aggregation tech, to see this financial information that had never really been used. And because of that, we were able to serve a lot of people for which there wasn’t this risk data.

But there’s all kinds of other examples too. On the anti-money laundering, there’s a lot of really fantastic companies that are doing great things to be able to comply with the regulations and also to make it easier for the consumer. Make it easier for the financial institution doing it. At Gilgamesh, we have some investments in that space as well.

So kind of going back to challenges with launching Petal, I think fraud was really one of them that we had to grapple with. Last one, I’d say, and this is not so much at the beginning, but it’s something that very much became the case as we grew: you have this challenge in FinTech, particularly in the lending space, of marrying innovation and people that are doing new and different things, marrying that with people that know how to avoid pitfalls and have industry experience and can keep you out of trouble. Those are two very different talent pools, two very different cultures, and getting them to work together is often hard, and I think great entrepreneurs do this really well. They kind of make it seamless.

You know, getting those cultures to be productive and to not waste a lot of time on just these cultural battles within a company is really difficult. And so that’s something that as an investor, we talk to our companies all the time, that you have to have both, but you have to find ways to get those people who are both right about how things should happen, get them to work together and kind of to get past these stumbling blocks as far as how the company ought to be run, is something that’s really, really significant.

Gilgamesh Ventures’ investment approach

Endicott: We’re investing typically at the seed stage, so that’s usually when companies are raising several million dollars, three tends to be a sweet spot. We’re typically putting in about half a million dollars into those rounds. So we’re usually the second largest investor in a seed round, that tends to be where we come in, and as a result, we have really friendly relationships with all the lead investors. And we, you know, they look to us for having some expertise in financial services, and that’s what we understand in FinTech.

And we love, conversely, that when we like a company where we really see, we have conviction, we see opportunity, we love to get in early, and then, you know, kind of pull investors to that company, the bigger investors. That’s something we really love to do, something it’s fun. It’s where investors and founders are amazingly, you know, kind of aligned and on the same team.

Active partnership with portfolio companies

Endicott: Exactly. Yeah. Unfortunately, when you’re an entrepreneur, and you’re raising money, it often does not feel like that. So anyway, we’re investing at seed stage, and we do follow on into companies as companies grow, as they raise a Series A and so on. We like to invest more in those companies. We reserve a chunk of our capital to do that, and we’re in it for the long haul with businesses. We want to help them to really achieve great successes.

We pride ourselves, Miguel and I, one of the things we really try to do is we don’t want to be kind of the boring or kind of slow moving fund that just wants reports and kind of talks to you every once in a while, and nothing happens. We like to get in the trenches with people. We want to help in tangible ways. We want to help build relationships with banks, with lenders, with processors, to help recruit, you know, FinTech specific talent, like chief risk officers, general counsel, CFOs, those are things we care about.

And then Miguel, he has tremendous relationships, you know, in the media space, which he does a podcast as well, the FinTech Leaders podcast, and kind of helping people to reach that ecosystem as well ends up being something really important to what we do as well. And so we get in early, we stay late. We want the companies to succeed, and we want to be really aligned with the founders, which is pretty easy when you’re that early investor, there’s not as many conflicts as there are for later stage investors.

Portfolio themes and investment thesis

Endicott: So I mentioned the idea of companies that are helping, or yeah, companies that are helping financial institutions and fintechs to manage these kind of challenges of being in the financial space, anti-money laundering, fraud and so on. That’s a place we’re really excited about. We have two major investments there, one being Base Layer that’s in the United States, focused on B2B KYC, so know your business, money laundering and fraud avoidance, that company’s been growing really fast and helping a lot of companies to really accelerate customer acquisition, to smooth onboarding, and just to make things easier for everybody.

We also have another investment globally named Niva, which is doing similar things today in Mexico and expanding into other markets, like Brazil, to help, kind of, in a very different setting, different jurisdictions, to manage very similar problems. So that’s a thematic bet of the company for the fund that we’re pursuing.

Another one, obviously, is artificial intelligence. You know, I think that this is something that cuts across the entire economy today, bit of an obvious one that we’re interested in, but we’ve made several investments in artificial intelligence in financial services. It’s kind of rejuvenated the way that financial services may be delivered in the future. And so that’s something that we’re really excited about, and it cuts across from neo-banking to lending to all kinds of basically SaaS products are being rethought with artificial intelligence. And so that’s something that spans everything we’re doing.

We still have expertise in some of these kind of – my background in lending is something that is, I think, a particularly deep area with us as a company. We have several large investments in that space, one being Nexu in Mexico. It’s an auto lending platform that radically accelerates the speed to purchase for auto purchase in Mexico, and helps dealerships to sell significantly more cars. It’s a platform for the dealerships, and then it connects kind of the actual sources of capital for the loans to buy those cars. So that company is growing magnificently and is profitable and doing really great.

And another one is Zeppelin, which is based in Chile, but has expanded in Mexico. It’s really kind of an office of the CFO, helping companies manage their working capital and payments in the B2B space. And so that company has been growing very quickly. And Nexu and Zeppelin illustrate the focus on the Americas that you mentioned earlier.

And the last one I’ll mention, we made some investments also in B2B marketplaces, which is, you’re not hearing about that as much as you were probably two years ago, but it’s a powerful business model, particularly in B2B in our opinion. And we have an investment in Brazil, Caena, which is in the restaurant supply chain space, so connecting suppliers to restaurants and managing the payments and the procurement and also the working capital of both sides of the marketplace there, that company is growing very quickly and just raising additional capital as well.

The rebaked fintech landscape

Endicott: There’s, I mentioned earlier, that right now, you know, it’s today compared to three years ago. There’s a lot more kind of just macro activity in FinTech. I can’t, I don’t know, I don’t really check whether that has appeared in the data, as far as the amount of investment in FinTech. But there is, probably three or four years ago, the number of ideas that were financeable was beginning to narrow, because you had these big – whether it was SoFi or Affirm or Bill.com, you name it, there were kind of pieces of the pie were being carved out and kind of were off limits. They’d been taken.

Today, the pie has been completely rebaked, and everything is up for grabs. And so it’s a really interesting time to be investing where we can. You can really always kind of think like a child and say, like, why can’t we do this better? There’s a good chance you can today. And that’s more the case than it was a couple years ago.

The last thing I mentioned on trends is around becoming, getting a lot of attention right now, stablecoins, we’re seeing a lot of activity there. And, you know, I think that’s a place that we have a belief that in 10 years, that’ll be a very important part of the economy, and we’re likely to make some investments in that space as well. So, but there’s a lot of things going on right now.

Why compliance technology needs reinvention

Endicott: I think they’re hard problems. They’re hard for a lot of reasons. There’s a lot of data problems. The regulatory requirements are quite difficult to comply with. If you look at Chase, or you look at Citigroup, and this statistic may be dated, but a truly obscene amount of their headcount is dedicated to compliance. And the vast majority of that is dedicated to anti-money laundering and it’s an incredibly labor intensive task.

It’s hard to do – I mean, if you look at, if somebody’s trying to launder money, and they’re trying really hard, there’s a lot of ways they can hide things. And the same thing is true of fraud. And so the sophistication of the problem that you’re trying to prevent is quite significant. As a result, the sophistication of the financial institution to address that problem and kind of indirectly, sophistication of the tools they’re using need to be great as well.

Artificial intelligence has enabled a lot better products with this. It’s stripping out the human in the loop for a lot of the easy cases, allowing you to focus on really the challenging places, or where there’s a lot of risk. You can see things like this. I think Palantir, in very different settings, does similar things, of really using powerful machine learning, LLMs, things like that, to allow you to focus on the big problems, and conversely, make quick decisions for 99% of those data points, means that everything’s a lot faster. Everything’s a lot easier.

It means you don’t have to have these – Chase is not going to need a third of their company doing this, which you know, if you go back 100 years ago, a third of their company wasn’t doing this either. It’s really a creature of the modern world that that’s happening. So I think the other thing I’d add is like, it’s not a structured data problem either, for the most part. It’s one of those problems where having these new capabilities through LLMs and other things, allows you to solve them in a way that traditional machine learning and kind of just other kind of post-great financial crisis FinTech tools really didn’t let you do. So that’s a couple different things.

Also, the last thing I say is the cost of getting this wrong is gargantuan. Right. You can just Google or otherwise search, you know, kind of money laundering penalties bank, and you will see a tremendous number – billion dollar fines and nine figure fines.

The massive stakes in compliance

Endicott: Yeah, I think getting it right is the more exciting part, because back to what I was saying earlier. A lot of the frustration of using any financial product is caused by these kind of other forces, one of them being regulation, it makes them suck to use. And you can really kind of stiff arm that problem in a lot of ways, and make financial products delightful to use, while the relevant financial party is still complying with the law and not taking risks that are going to come back to haunt them down the road.

They don’t have to make that choice of having a terrible product that’s compliant or a great product that’s completely non-compliant. You can have both things. So that’s the exciting thing. That’s what all the companies want, because it allows them to grow. It’s part of them being able to acquire customers and give them a delightful experience.