Metacorn: How to Build a Platform that Fuels an Ecosystem with Plaid Co-Founder/CEO Zach Perret

Please welcome Zach Perret, Plaid co-founder and CEO, and Ari Levy, CNBC senior technology reporter.

Go for it.

All right, happy post-lunch, post-lunch interview. Um, I'm I'm Ari from CNBC. This is Zach from Plaid. And I guess, well, it's interesting because you're not a SaaS company. So we're at Saastr, but we'll talk about what it's like to not be a SaaS company at Saastr. Saastr-ish. Saastr-ish. But first, just for people who don't have a good idea of what Plaid does, maybe give a kind of brief overview.

Sure. So Plaid, we build the infrastructure that lets consumers interact with their bank accounts. On the web. The core thesis behind it is banking was built for a world that didn't envision the internet, and as such, most things that we do in our financial lives are frustrating, difficult. They require you to walk into a bank branch, talk to a banker, send in a fax, print out 50 pages of bank statements in order to apply for a loan. All those things are kind of frustrating and annoying, and so what we build is the infrastructure that lets you programmatically interact with your bank account. Think of us taking your bank account, wrapping a set of APIs around it, and then enabling you to interact with that bank account through any application that you want to out there on the web. Easy way to think of it is Venmo was one of our first customers. We do the connectivity between Venmo and the bank account, allowing the two to talk to each other and allowing you as a consumer to come to Venmo and say, hey, link my bank account, move money in and out. So infrastructure for fintech is a shorter way to say it.

Give a sense of— so I as a consumer don't know Plaid necessarily. I never know that— not yet, but I never know that I'm interfacing with it. What's the difference in sort of a Plaid world versus a pre-Plaid world and what my experience is as a consumer?

Yeah, so as a consumer, the way that you interacted with your finances on the web pre-Plaid was doing a variety of manual processes. So If you've ever had to start a new job and walk in with a voided check, and then they take that voided check and they figure out how to process your payroll, that's pre-plaid. Or if you've ever gotten a 2-cent and a 4-cent deposit into your account as you're trying to set up direct deposit or payments transfers, that's a pre-plaid world. Or frankly, using your debit card to— or credit card to do all sorts of payments that you really should be just doing directly from your bank account. Applying for a loan. 20-odd pages of documentation. If you've ever applied for a mortgage, there's 50 to 70 pages of documentation per mortgage. And so we build all the infrastructure to digitize that.

And so instead of getting that 2 to 4 cent deposit in my bank and having to acknowledge that I got them and then waiting a few more days, what happens now?

It's instant. You link a bank account, and then you can immediately do what you want to do.

And you work with how many financial institutions?

So in our backend, we've integrated with about 10,000 financial institutions in the US and Canada, which is substantially all. Our lawyers say that substantially all means almost everyone. And so about 10,000 in the US and Canada, and then about 3,000 to 4,000 different fintech applications that have meaningful size, and about 70,000 more that have almost no size. So lots of fintech and then lots of banks in the backend.

[Speaker:JASON M. LEMKIN] And so you've had a lot going on in the last couple months. In December, you raised $250 million at a reported $2.7 billion valuation. And in January, the following month, you spent a reported $200 million on Quovo. So raise $250, reportedly spend $200. Are those related?

Reporters report things, so I can't confirm any of those numbers. So yes, yeah, of course. So we were— I'll talk a little bit about the story. We were in the midst of doing our Series C fundraise, fortunate to have a really great set of investors. The process was going quite well. Around the same time, kind of this conversation that we'd been having for a long time with the Quovo team kind of reached a point where it made sense to think about joining forces. So, at Plaid, we build integrations into bank accounts focused on letting you interact with your money in your bank account in whatever way you want to on the web. Quovo had focused on investment accounts and allowing you to programmatically interact with your investment accounts through third-party applications. And there's actually a really nice synergy between those two, to use the dreaded term. And so as we were having this conversation, we did of course raise a little bit more money in that round in order to enable the acquisition. We couldn't be more excited with how it's gone so far. People ask if we had an acquisition strategy. We definitely don't. We have a product strategy and a company strategy, and it happens that this acquisition was a fantastic fit there. But we are now in the midst of figuring out how to integrate a new 100 people or so into our company.

Which brings the total to what?

About 260. So 90 new people coming on and figuring out how to do that integration.

No wonder it's so hard to reach you. So Mary Meeker led the round, the financing round, right? I think we all probably know. About Mary Meeker. Can you talk about how that came together and how you met her?

Yeah, so we feel so incredibly fortunate to get to work with Mary. She is one of the, of course, most well-known investors in Silicon Valley, longest-standing names. But the really wonderful thing that we've been searching for is we wanted a mix of finance and technology. And in our past rounds, we brought in a lot of great financial investors. They kind of knew tech quite well. But we didn't have anyone that came from a deeper-seated finance background, coming kind of off of Wall Street, having those relationships with banks. And in this round, we'd actually been talking to both a number of investors but also a number of independent board members trying to find that ideal mix. And when we met Mary, we were just lucky that it was a great mix of having those deep kind of finance roots and then thinking about it in a truly innovative and kind of forward-leaning way. And so it's been wonderful to work with her so far. And we're excited for the years to come.

So when I imagine, again, a reported valuation of $2.7 billion, and you've been around for 5 years, basically?

Our product's been live for 5 years. We've been working on the company for a bit longer.

So the image I have as a vapid tech reporter is sort of everything is up and to the right for that period of time in order to reach that. Is that true?

No, of course not.

So talk about just, you know, fintech, particularly dealing with banks, is brutal business. Talk about like a particularly difficult period in your early days.

There's been a bunch. It's funny, I'm sure many of you have heard of this concept of peacetime companies and wartime companies. Plaid has always been a wartime company, and we're trying to continue to scale that out. We've had lots of difficulties. Earliest days, trying to integrate with the banks when the banks didn't like us so much. We were enabling an industry that was competing with them. Lending Club was an early customer of ours, and Lending Club was directly competitive with the banks, and the banks didn't love the fact that we were enabling Lending Club to make loans, enabling Lending Club to steal their profit.

[Speaker:JASON M. LEMKIN] So they weren't worried about what you were doing, they were worried about what you were enabling other people to do.

Sure, yeah. I think there was, there's always this tension between does a consumer own their financial data and thus if a consumer wants to use their financial data to apply for a loan, does the bank have any way to stop that? I think that has become quite clear through regulation over the past few years, but yeah, there's definitely hesitation amongst the banks kind of pushing us to frankly do less. We've gone through that. Another interesting story, when we were an 8-person company, we actually got sued for patent infringement on a handful of patents that under today's patent standards are not patents. So, for example, passing financial data via an API. In that case, all of Silicon Valley infringes on that patent. So when we were 8 people, we fought a patent lawsuit for about 2.5 years, raised a Series A and a Series B into the patent lawsuit, ended up doing just fine in that lawsuit. It didn't end up affecting our company long term. But that does kind of cement that— call it wartime mentality, that mentality of needing to grind to get through the challenges.

So you effectively raised money so that you could fight a lawsuit?

Sure, yeah. That is certainly one way to put it. We raised money to build the company, and it happened that we had to spend some money on the lawsuit.

How do you explain to investors that you have a— $7 million, or however much—

$7 million in lawsuits.

Millions of dollars in lawsuits as an 8-person company, and you need them to give you an amount of money that's roughly equal to that.

Yeah, so that was actually when we were going into our Series A. It was about that time. We raised, I think it was a $12 million Series A, and knew that we had to spend at least a few million on this lawsuit. I think we were fortunate to find a great investor that really— a set of great investors. They really believed in the long-term potential of fintech. It's interesting, in our first round and our second round, we went to raise and we said, hey, fintech is going to be a really important market, and we're going to build some of the foundational infrastructure, and we think we're going to have 1% to do with the fact that fintech is going to grow. And I still think probably less even than 1%, just right place at the right time. But some investors were willing to go on that journey with us, and of course, per the lawsuit, I spent a lot of time on legal analysis, We spent way too much money on lawyers getting VCs comfortable with it. But it is, it's wonderful to find a great investor that truly does believe in the vision, hard as it might be to identify them. But we were lucky to find that.

[Speaker:JASON M. LEMKIN] And so what is your relationship with the banking system today? I mean, do they, I guess, do they view you as a partner? Is there some level of concern that you're sort of disrupting, disaggregating, just something them?

Fair question. So we have, we've worked a lot with the banks in finding ways to help them build their businesses and make their businesses better. Historically, banks were hesitant because we were bringing change, and with anyone that has deeply entrenched long histories of making profit in a certain way, creating change is scary. and we were coming to them and saying, hey, you can do these things in a digital way, in a different way. There's actually an interesting example that I like to cite. If you think about the heyday, the early days of Wealthfront and Betterment, they came out and people started investing with them, and they were really excited to grow their assets under management. And over time, they worked really hard over kind of 3 years maybe, they got to a combined $10 billion in assets under management. About that time, Schwab decided that they needed to react. And so they spent 9 months internally building this product called Intelligent Portfolios. Some of you might use it. And this Schwab product, they launched it, and a month later they had $10 billion in assets under management. And so what we've actually seen is a sea change in the way that the banks think about fintech and technology in general. They're starting to say, well, fintech is becoming my innovation lab. and I'm able to look out there and see what's new, see what's different, and then react to it. And so we've seen the banks not only launch their own products, like intelligent portfolios, but do partnerships with a lot of lenders, a lot of other companies to distribute their products to their customers, ultimately driving that combined customer experience using the fact that a bank maybe has 50 million customers to distribute a smaller product and kind of sharing the profit on the backend. We've seen a sea change of banks saying, you know, fintech is bad for me, to now saying, well, fintech might be bad on the margins for me, but I can actually profit from this, so let's embrace it.

So you're distinctly not a SaaS company. So quickly, how do you make money?

Well, I guess in the truest sense, we do sell software as a service, but we tend to differentiate the way that we think about ourselves from SaaS in a variety of ways. More specifically, we're a platform, and we treat platform as a distinct or perhaps odd type of SaaS, and as such, we do a lot of things very differently. So when you start to look at the metrics of a platform, there's incredibly high engagement and incredibly low churn. If engineers are doing days or weeks or even sometimes months of work to install something, it's incredibly unlikely that they're going to quickly rip it out and change it. If we are kind of a core vendor to enable the user experience you want, or more specifically enable consumers to actually move money in and out of your platform, it's quite unlikely that you're going to turn that off. And so we grow very much as a flywheel. So little by little doing work, getting more and more installs that then pay off years down the line. And that allows us to have a very different go-to-market strategy. It allows us to think more about, where we enter the organization, how we interact with the organization, and allows us to spend a lot of time on account management and figuring out how best to kind of create champions out of those customers we have.

And so you get paid every time someone does what?

The best way to think of it is the number of monthly active users that an application has that are interacting with their money on that application. So we get paid on a usage basis for the amount of data or amount of utility that a consumer gets out of an application.

Organization. And who are you selling to? Who's adopting the product within an organization?

So it's a, generally it's a developer that will introduce it to a product manager, and the product manager will do the sales within the organization for us. We have a very strong kind of developer focus. We acquire a lot of our customers, or acquire a lot of our interest, from forums, Hacker News and Reddit actually being quite a good channel for us. From referrals from one developer to another. Honestly, we have a lot of direct Google search saying, how do I connect bank account, or how do I see my bank balance, or how do I process an ACH payment?

Developers doing those things.

Developers doing that, yeah. Stack Overflow being another good source. And so there's a lot of initial developer interest.

And so they'll do that within their small group.

Sure.

Kind of a classic bottom-up, one of these consumer software in the enterprise.

Yeah, we call it selling through the basement. So we're trying to identify the lowest level engineer, build a fantastic experience for them. So all of our user acquisition is focused on someone hits our docs. Our docs are our marketing page. Someone's able to quickly build something in our sandbox. They're able to quickly launch an application. And we love applications for fun. So we have a few thousand meaningfully paid applications, but we have tens of thousands of fun applications. So people building— some of my favorite things are people built a bot that texts you every time you get paid. It just texts you a bunch of emojis. Or someone built their own accounting system or their own TurboTax just right within the Plaid platform. They do it for free. And then the great thing is, over time, when their boss asks a question or someone says, hey, how do I do that thing over there? Then the developer's saying, well, try this. Have you used Plaid?

So is that how you got installed at, say, Venmo? I mean, was it a few developers that started using it and then it sort of spread throughout the company?

Yeah, so Venmo was quite an early version of that. Actually, a friend that was head of engineering at Venmo came to us and said, hey, can I license this thing? And we said, well, we don't know how to license anything. We just built a product. This is before we had any real revenue.

Yes, you can use it.

Yeah, but please use it, yes. And if you'd like to pay us money, we'd love to collect your money. But yeah, so that story kind of growing out of— this is at the time the company was located in New York. And there was a fintech developer community. And it was just kind of growing out of that. Of people using it. And then actually Venmo became a fantastic lead source for us because an engineer that used Venmo to do a linkage working on a project at home, they say, hey, they email someone at Venmo or even actually ping Venmo support sometimes to say, hey, how did you do that? I'd like to do that too. So it really is this kind of network-driven acquisition model that starts with activation of developers, moves very quickly through kind of the phases of a traditional sales funnel, into an evaluation phase. And in the evaluation phase, which is something that developers will do on their own if you give them the right set of tools, that's when you really get hooks. And over time, they'll start to introduce to their product managers. And as we built our sales team, initially it was basically all self-service with a couple kind of human interactions. And now as we built our sales team, it's all about helping the product manager sell up through the organization. So actually, not truly doing that much sales ourselves, but equipping a product manager to go get their boss to sign the check or ship the product.

If you're dealing with a mega bank like a Chase or Wells Fargo, are you still getting adoption group by group, product by product, or is it possible to get sort of an enterprise-wide deployment?

Yeah, well, I guess when I'm talking about the way that we do customer acquisition, that's mostly in non-banks. For banks, we do have a more direct sales team. And we do some enterprise-wide adoption, but generally trying to sell to Chase as an enterprise is nearly impossible. But drilling into the specific use case is much easier. There's of course sub-business units and sub-product leads there. And so for example, with Chase, we've sold directly into their wealth management division to help them build wealth analysis tools for their advisors. And in that case, yeah, we'll sign actually a narrow MSA. And over time iterate that MSA to be many other use cases. But we'll start quite narrowly focused again on that product person. But we do have a different team that does the bank navigation.

So I remember when we first met, I think early last year, early to mid last year, you had just started doing PR. You had just hired like a PR agency to start getting your name out. Now I knew of Plaid because I was covering fintech and finance. The name would come up. But had never been reached out to by anyone at the company. And so we were having breakfast, and I asked you why you got a PR person. And I don't think you exactly knew why at the time. But it did make me think, you're not a consumer-facing product. Consumers never need to know that they're using you. Even business, it's not like a SaaS product where even people within the organization necessarily need to know they're using you. So what—

but you are—

you did hire a PR person. You are doing some marketing. So I guess what do you see as the objective of being better known, and how are you spending your dollars to do that?

That's a great question. I think you're correct. We don't need every consumer to know who Plaid is, frankly. What we need is for every consumer to know what Robinhood is and then use Robinhood, because that's wonderful for us. They don't need your publicity.

By the way, they're good.

Yeah, I think they've done well on their own. No, so I guess we couldn't hope to build the consumer brand that our customers have built, nor would we want to. We truly are an enabler behind the scenes. I think there's a couple elements where we do think about wanting to be a bit more public. One is acquiring customers. So as we start to think about new geographies, how do we get initial recognition in those geographies? So we launched Canada midway through last year. How do we get initial recognition in Canada? And that was important. It's important for recruiting, but mostly our communications are focused on financial institutions and helping them understand what the new world of fintech looks like. You could say in some sense banks are our supply side. Truly it's the consumer that's our supplier because it's the consumer's own data that's being passed through the platform, but banks are a key constituency there, and as we think about building a message and kind of sharing a message externally. It's about getting the financial community comfortable with the fact that fintech is here to stay and that this new normal is actually quite good for them. They just have to understand it correctly. But yeah, I think a lot of the brand building really does focus on how do we get people to come join our company. We need to hire a lot of people. And how do we get the banks to understand what's going on?

So what are the marketing channels that you use?

I talked a lot about kind of developer-focused communications. Our blog is a fantastic marketing channel, getting that distributed through Reddit, Hacker News, all of those typical places. And then we do a bit of direct marketing. We've actually been experimenting with this. We started with a $2 million marketing budget last year and spent less than a tenth of that in trying to run some experiments. So figuring out how to scale that up. But truly, developers don't buy through ads. They buy through content. They buy through deep engagement and being truly authentic to the community.

So I know you're traveling a lot. Maybe you could talk about who you're meeting with when you go to organizations, who you're interfacing with. And I would imagine even if you are a bottoms-up product, and again, when you're selling infrastructure to big banks, you probably have to get people at the very senior levels comfortable with that, I would imagine. So maybe you could just talk about where you're spending your time, how you're spending your time, and who you're meeting with.

So I do spend a lot of time with banks, but luckily we have a team that is not me that spends a lot of time with banks as well. I have long hair, I wear sneakers, I generally wear jeans when I'm meeting with banks. There's a certain value to being a bit different than the mainstream of everyone else that's trying to sell something to them. Are you allowed?

Do they let you in the bank, at least?

Oftentimes, there's been a lot of jokes, but I get a lot of questions, I guess, on the way in.

I'm not the first person to ask.

No, you're not. Security teams on the way in give me the evil eye. But no, so we have a wonderful team that does a lot of that relationship building. There's a certain value to having a dichotomy, to truly being a tech company and talking to a bank as a tech company. But you asked how I spend my time. A lot of my time is focused on internal infrastructure building, making sure that the company is functioning as it needs to function. These days I do a lot less product work than I once did, much to my own chagrin, and now the company has become my own product. So you tweak the levers, the company becomes your own product, and you focus on building the right services and interfaces there. I spend a lot of my time on the road with our customers, just understanding exactly what they need and how we can help them better, and then partners on the backend, making sure that the banks are happy and doing what we need to do, both in the U.S. and increasingly abroad.

So speaking of partners, how did you meet your co-founder, William?

So William and I, both developers by background, both had built a lot of software, and both ended up at this consulting company, Bain, somehow. And in the first week of him starting, I was told to teach him how to use Excel, which is the last piece piece of technology I've ever taught him. So I was told to teach him how to use Excel, and then we ended up going down this rabbit hole—

So you could get a job at Bain without knowing how to use Excel?

Well, they teach you. This is the thing they do. The first 2 weeks, they teach you how to use Excel, and then they said, now you have to be good at it. And of course, in the initial meeting with him, we decided that Python would be much more effective in doing the analysis we needed to do. And so we became fast friends, and actually rock climbing buddies because we both rock climbed a lot.

And this is in New York?

This was— no, we lived in Atlanta at the time, which is interesting for a variety of reasons. And the first time we went rock climbing, I dropped him from about 20 feet. And that was— after that, he kind of forgave me, which was fortunate. He didn't break anything. And then we became kind of friends for life and have since worked on everything, both building the product in the initial days and then After about 18 months of just heads down building it, once we finally launched the product, someone had to go sell something, and I grudgingly admitted that he's a dramatically better engineer than me. It was not hard to admit that. It just took me too long. And so now he leads all of our kind of product and engineering side, and I lead the go-to-market side.

So the seeds of the company were from you dropping him while rock climbing?

Apparently, yeah. Yeah. And the forgiveness that came with that. We talked about wartime companies before. You have to have a little bit of— a little bit of— of craziness in the beginning.

What does financial inclusion mean and how do you help foster that?

So one of the core principles behind Plaid, or one of the uniting principles, is to make money easier for everyone. And money for most people is the most stressful thing in their lives. It's the thing that that acts as kind of called the limiting reagent to the things that you want to do. It determines if we can go to XYZ or if we can do XYZ or buy XYZ or have XYZ. And one of our, again, uniting principles is to make money easier for everyone. Most people don't, at least in the past, when we started the company, they didn't trust the financial system. So 2008 was 10 years ago, but consumer frustration with banks is still quite high. Most people treat, treat banks as, or think of banks as a tax on their lives. You put your money into a bank and it's just a matter of time before they charge you enough fees that it's gone. And so we have this deep-seated focus on making money easier for people, and we believe that by creating a very dynamic ecosystem of financial technology products that are trying to do good for consumer, and these products themselves competing so that ultimately a consumer makes a choice based on the simplicity, the ease, the communication, the quality, the price, So that kind of deep-seated view of making money easier for people is one thing that unites us. What we found is when we built the company initially, we focused on just the biggest banks because that was the most volume, that was the most people, and that's the way to build the product the fastest. But we left out the people at the edges. We left out people that had smaller banks, community banks, generally the lower-income folks that weren't able to have a big bank account or apply for a loan with a big bank. The past 2 years we've actually done a lot of work to focus on making sure that everyone at every bank can use Plaid, and thus the applications powered on the backend. We'd like to do more work on allowing people— I think it's something like 10% of the US that is underbanked or not able to get a bank account at all. So allowing people to open bank accounts for the first time themselves. And we've done a bit of work on the edges there. But financial inclusion, making sure that everyone's able to live the financial life that they want to live, is core. Forward to kind of the principles behind the company.

So the title of this session was something like Lessons Learned in Building a Platform. There we go, 6 Lessons on How to Build a Platform. So maybe just like 1 or 2 quick lessons on building a platform.

Lessons on building a platform. So this session is titled that. However, I wasn't aware that it was titled that until recently. [Speaker:JASON M. LEMKIN] So a couple lessons in building a platform. I think the first one is truly go slow at the beginning and get it right, because it's very hard to modify the way that a platform is built over time. Launching an API, versioning an API, think of versioning before you have to version. We actually didn't the first time. We shipped an API with no versioning infrastructure in place, versioned once. Luckily, that was a year in and we had 30 customers. And as we versioned that first time, we built in the procedures and policies to make it quite easy to version in the future. So think carefully before you launch something, because people get deeply entrenched in it, and then it gets really big, and it's hard to fix. So that's one. Second, I would say, is be truly authentic to the community of users that you have. They will determine the outcomes that you have as a company. They will be your best evangelists. And platforms don't necessarily have a great deal of users on them. Oftentimes you can build a platform that is quite successful with a small community of highly engaged users, because you are the backend. You are the infrastructure that then scales rapidly on the other side. So be truly authentic to them, and don't ever— people said this before— don't ever eff the customer, because they will determine your success. And then I think the last one is this concept of a flywheel that keeps coming up over and over again. Realize that the things you do now will pay off in 2 to 3 years. And it's interesting, as we were fundraising, we put together this slide that basically just said, we are not SaaS, for a variety of reasons. And when you put together SaaS multiples on a platform company, they look ridiculous. And so we put a couple multiples up. We said, this looks ridiculous. Please don't consider us SaaS, please consider us a platform. And by the way, we'd like to go make these platform decisions. So we're gonna go optimize for market share over the next 2 years, and we're not gonna optimize for revenue because in years 3 through 5, it's gonna be much larger. And this is what we said early on in the stage of the company. And so taking a patient view to understanding that kind of the flywheel will pay off on a decades-long time horizon is crucial. And frankly, finding investors and partners and employees that think the same way is important.

20 seconds. What is the single most interesting thing in fintech that maybe isn't directly related to Plaid, but that you're watching?

So there's a bunch, but there's been a ton of volatility in fintech broadly, or more specifically in the markets. And I think one of the most interesting things is we're seeing these high-yield savings accounts come up. Over and over again. I suspect that that is going to be the hook in the same way that Bitcoin was the hook at the beginning of last year. The hook for this year, like the acquisition strategy of fintech, is going to be around high yield. So hopefully a lot of options for consumers.

Great. Thanks, Zach.

Thank you.

Thank you, guys.