Product lessons from Cash App & Carbon Health - Ayo Omojola on going "unreasonably deep"

[00:00:00] Probably the one thing that I'll say that I've taken away like philosophically is to always look for those opportunities where it's easy to stop, where basically it gets tedious. And what that typically is a signal of is that somebody or a bunch of people just haven't gone deeper before. It's always a balance too.

Like you go deep on any number of things. And I think the real art is choosing where to go deep on that's worth spending time, because you know, you have a limited amount of time and you have to be thoughtful about where you spend it. 

[00:00:32] Welcome to in depth, a new show that surfaces tactical advice, bounders and startup leaders need to grow their teams, their companies, and themselves.

I'm Brett Berson, a partner at first round. And we're a venture capital firm that helps startups like notion, roadblocks, Uber, and square tackle company building firsts through over 400 interviews on the review. We've shared standout company, building advice. The kind that comes from those willing to skip the talking points and go deeper into not just what to do, but how to do it with our new podcast.

In-depth you can listen into these deeper conversations every single week, learn more and subscribe today@firstround.com.

For our second episode of in depth, we're excited to be chatting with IO. Majola, he's a product Wiz and incredibly sharp thinker, but what's most impressive is his depth and breadth. I recently joined carbon health as their VP of product where they're tackling healthcare, accessibility challenges like COVID-19 testing.

Before that he was the founding product manager on the banking team for cash app at square, where he co-created the popular cash app and help build out squares, technical banking infrastructure. He's also a former founder and active angel investor. As IO puts it. He's constantly thinking of new stuff to build from starting several companies to trying to open a restaurant to the running log of ideas.

He's kept for 15 years, it's this mindset, plus all of his impressive experiences that gives him a unique lens into finding and evaluating ideas. The topic we'll spend the most time on today, IO particularly excels at going unreasonably deep as he calls it on any given space. If he can talk your ear off about the Durbin amendment or the physical factory process of producing a credit card, thanks to his familiarity with the heavily regulated spaces of financial services and healthcare.

On today's episode, we chat about the early days of cash app and dive into how he untangles regulations and other market hurdles to find the opportunities where it's easy to stop. I will also tells us why he thinks a lot about problem selection. And makes the case for putting more effort into choosing what to work on.

It's a must listen to, for anyone who's thinking about starting a company someday, or for a product leader who hopes to help a new product take shape. But even if those aren't goals of yours, there's still tons to learn. iOS shares the leaders he learned the most from during his time at square and the frameworks he picked up from them such as how to get better at process setting context and optimizing for the outstanding last but not least.

We get into his management and hiring philosophy, including why he loves to hire former founders. I hope you enjoy this episode. And now my conversation with IO. So thanks so much for joining us 

[00:03:35] IO. Thank you for having me happy to be here. I 

[00:03:38] thought an interesting place to start would be the founding story of cash app.

And I think throughout our conversation, we'll try to sort of have equal parts business strategy and maybe product building perspectives and ideas. And one of the things that I've been fascinated with cash app is that it's one of the few examples of technology companies kind of in their first 10 years, launching an adjacent product that then goes on to have extraordinary success.

What you generally see is whatever. The one thing that gets a company going generally continues to be that one thing for a very, very, very long time. And many companies only ever do one thing and cash app is just an amazing example of a second tour. I'd probably third or fourth product. I think square is probably a good example of a company that has probably launched multiple products better than most.

And so I'm curious to start by spending some time talking about the origins of cash app and maybe why you think it worked, or what were the conditions or the environment that allowed it to work when it's so rare across technology companies. 

[00:04:42] So I'll start by saying I joined square when cash app was about a year and a bit old.

So I think the first line of code written for cash app was probably written around February, 2013. And I joined square in May, 2014. So there's certainly context that I'm not going to have. So I'll just start by saying that the original core idea behind cash app was about building. A, I mean, there were honestly quite a few things.

One of them was about building a debit focused payment network for peer peer-to-peer money movement. And then obviously once you achieved scale and had a network effect, there are many, many ways that you could leverage that. Thinking back, I'll go through a handful of the core ideas that today probably feel not sure what, the way we described it today are probably more commonplace that were not necessarily a given back in 2014.

So I think cash apps, first iteration, and what we were trying to move beyond when I arrived was I could send an email to Brett burson@gmail.com. I could CC. Send@cashappatitwassentatsquarecast.com or something. There was an alias that you would be able to append, and it would have an offline flow for handling.

I would say send $50 to Brett burson@gmail.com CC pay@squarecash.com or whatever the alias was. And it would check if the sender had an payment instrument linked, if not, it would request that a payment instrument got linked through like an offline email process. And once the sender linked their payment instrument, it would initiate the payment and move money to the other person.

And I always find like an interesting way to look at this stuff is a lot of launches over the last 10 years. Where I think prior to product hunt were launched on hacker news. So you can see, there's kind of like a history that you can see with all the early comments and like how people thought about it at the time.

And at the time the novelty was just this ease of avocado number linked to this thing. And we're trying to make sending money from one person another really, really easy and frictionless while still feeling pretty secure. And when I joined, we were starting to move from email based payments to the mobile app concept.

So the email-based payments does one core idea shifting from using ACH as initiation mechanism to debit was also another core idea. And you can just think of it as the behavior of most consumers. Like most individuals, humans generally have their wallet within their reach, but to get a random person to remember their account or routing number, it's still today hilariously to me quite complex.

The concept was just, if you shift the ease of the instrument to the thing, that's generally within reach, you're just taking out a bunch of stress and anxiety generally associated with payments. A third big idea was around the philosophy of most financial products in the incumbent world, like bank of America and E-Trade and things like this, the way those products generally tend to work.

You're applying for a payment instrument and then they approve you or deny you. And one of the core concepts with cash app was access first. So the default is that you get in and then we work really, really hard to make sure that people once in behave well. And so that's sort of removing friction from onboarding.

And then we had these concepts of what we call like a graduated onboarding, where you try to let a consumer do as little as possible to complete pleat the action that they want to complete. And then only ask them for more information as they need to do more things. And so I would say actually there's like really like a big part of focus on remote moving friction.

And then this is by no means a canonical list. Just the last big bucket that I would describe is we really did try to make our team go unreasonably deep on anything. So there were. Like all the product team for sort of their domain were insanely familiar with the regulatory regime of the product they were in.

So I would have discussions with our legal team about the Durbin amendment and the prepaid rule and things like this. And I think that kind of depth, it's not a thing that necessarily exists if you don't work in a regulated industry or certainly not at the same level of sort of hand-to-hand combat.

And having that level of understanding really did shape the product. And it wasn't just me like Brian, who was the founder of cash and still runs cash extremely familiar with the regulatory landscape and many others on the team. That's true for them as well. 

[00:09:29] That's obviously sort of like a set of ingredients that kind of came together to make this thing work.

Were there environmental factors at square that you think made this work? I could see them. A lot of other companies. It's a bit of Jason to the core. It's going to take many, many years to get to any scale. There's Venmo, PayPal, there's other similar types of things. And yet this was able to get invested in, and then obviously over a long period of time has grown to just astonishing scale.

And so do you think there's anything organizationally, culturally environmentally that made it? 

[00:10:02] So the real answer is, I don't know, the slightly more nuanced answer is before I joined square, did have a pretty good culture of experimentation. And even after there's lots of hack weeks and things like this, having not worked at other big Silicon valley companies, I have no idea how common or unusual that is.

So it's hard for me to tell like, Hey, it was different here. I think beyond square, there actually were real material macro factors that I should have mentioned in sort of the first preamble, but also were big factors. So I think things like the Durbin amendment and the Durbin amendment rot, I would say a bunch of changes that were like actually transformational for companies like square and Stripe.

So that's one bucket I would say. The rise of the app store and the sort of ability to do more and more stuff in a mobile context made both products like cash app possible. And it also created this distribution channel that was really orthogonal to traditional distribution channels and financial services.

I'd say that's another bucket. And then on square specifically, I think the couple of things that translated to cash that I can speak confidently to that I think were relatively unique is when square. And I think this is true for Stripe as well, but when square launched back in 2009, 2010, the company just had like reams and reams of people who were insanely focused on payments and card networks.

And that's a relatively sort of arcane domain, and there's a really large number of variables, the deeper you go. And so what that meant was the team coming out of square, working on cash app. Ended up being quite knowledgeable about things that were possible, that weren't really in the public domain, public domain is not the right way to describe it.

It's just for years, the core advantage cash app had was for a long time in the United States. If you want it to move money from bank a to bank B instantly cash app was literally the only way that you could do it. You could use wires for instance, and pay $25 and you could use ACH, which is free and you could move money slowly.

But if you weren't sort of in the bracket that one had to pay 25 bucks. To B of a, to wire money to chase. There were years for which cash app literally was your only option and that sort of technical stack the knowledge that it even existed in that it was possible honestly, was what took quite a bit of imagination and just came from in large part that team of people having spent years at square, banging their heads against the wall, working with card networks on things and realizing, Hey, there's this unexploited mechanism for instant money movement over the card rails.

And it happens to be the case that nobody's using it this way, because it's a slog to like pull it all together. So I'd say like that's a big part. And then square, I think for years has a really high technical bar and cash app early on was really staffed with very senior technical people for a really long time.

And I think it's now it's a bigger team. And so the seniority distribution is going to look more normalized. But I think early in its development, a lot of really senior people at square were working on that product for years. 

[00:13:07] And so does that sort of get at the idea that was as much as sort of a technical innovation as it was a product innovation.

And so recognizing that there was a large technical hurdle, a certain team had to be put together to sort of create this magical experience. 

[00:13:21] If I'm being honest, I don't think it's possible to name any one thing. I think it's under appreciated mark for this, but cash up does like a lot of things insanely well, relative to what's possible.

And so on the design culture, very strong, the engineering culture, very strong, really, really sharp at money movement, pretty thoughtful at risk, really, really thoughtful at marketing growth. And then just to focus on being original on the product front rather than following. There's lots of places that will have like one or two of those things.

There's just not that many teams that have, I would say that level of depth in so many areas. It's not without its weaknesses, but I think when I look at competitive products in market today and sort of given what I know and what I know is possible, and I see the things that they're not doing, it's clear that that team is executing well on us.

[00:14:12] You touched on some of this already, but what lessons did you take from working on different parts of cash app that you think applied to either what you're currently working on, or you'll just take with you 

[00:14:24] not being afraid to engage with the rags is a big one. Matters more in regulatory industries than not.

It's a useful way to learn what you don't know. And it's a useful way to get an understanding of like, if you think about like how sort of the law plays out, typically what happens is there are some event and laws change and then over a period of time, the agency's chartered with enforcing those legal changes, generates regulatory guidance and.

Candidly, it's relatively impenetrable. Like I'm doing some of this now reading some regulatory material at carbon health. I'm much less versed in it than I was in the financial services space. And I'm like, sort of realizing that, but I think it's just important to not be afraid to go to whatever depth necessary to get an answer that you really understand.

So that's one, and I think it's also incredibly tedious. So a lot of people for whom it's boring, they don't spend time doing it, things like that. And there's a lot of people for whom. And actually, I wouldn't even say there's a lot of people from whom. Even for me, for everyone, it is just easier to take the thing that you are told.

And in a lot of cases, the thing that you were told is correct, but there are enough cases where the thing that you were told is just a series of interpretations. Like the person telling you just got it from the person who told them. And it's been like a long time since anyone's gone close to the metal.

So that's why just not being afraid to engage with like the core subject matter, I think is really important. I think just being mindful of the macro environment that you're operating in. I think it's a, the analog I like to use is no matter how good you are, like, no matter how good you are or how hard you work or whatever your persuasion is, you are never going to be as powerful as forces of nature.

And so if like these larger forces in the economy and the environment play out, and the thing that you're doing is at odds with those, you're probably going to get steamrolled. And so just being mindful of what's happening in the world at large, and how does it affect the thing that I do would be another one.

And then I think there's also this, I understood we were doing a cash app, but we didn't, it wasn't really articulated this way, but I think it's a really nice way to articulate it. Is just focusing on what utility you can provide that is like, how can you make something sufficiently more useful than the state of the art that it like cuts through the clutter and the people who you're trying to make stuff for actually pay attention and are even willing to try you.

And then once they try you, they're thinking, oh, wow, this actually is solved my problem in a way that I didn't realize that I needed necessarily. And that Delta is reason enough to skip. So you can think of it. Like if you create a financial product, that's like 15% cheaper, is that gonna attract some people?

Sure. Probably. But if you create a financial product, this 90% cheaper, it's just inherently going to attract a ton more people because price elasticity will increase the larger, the differential in price. And that dynamic is true, even in areas of utility that you can't directly measure in dollars. Speed is another sort of advantage that you can play with cost is an advantage that you can play with simplicities and advantage you can play with, obviously the further away you move from like really quantifiable, measurable things.

The more judgment you have to exercise, but the crass way to say it is it is actually relatively important to be kind of original, but that's probably just me talking my own book. 

[00:17:55] In that case. I think some of the conversations we've had in the past, there's kind of this bucket of things of make something 10 times cheaper, which is obviously compelling.

But I think that you've noticed that sometimes just having something that's beautiful or cool or feels good actually makes a difference. And I think you've seen some of that in cash app. And you've seen some of that in challenger banks or other products that just feel or look cool or good. Do you think about that in terms of like the dimension of like quantitative it's 90% cheaper, super easy to grok and easy to understand why somebody would want it versus, oh, this is a cool card with my name on it.

[00:18:32] I don't think that characterization is exactly right. So I think, yes, I have a huge incentive to say this, but I do think the cash card is a really cool looking card, but I think that glosses over when we were first working on the cash card, The state of the art and market was a not really impressive in any real way.

And B was being executed far below the level of technical capability. So like, I'll say it this way with no offense to B of a, they're a massive company that is obviously very successful. The BBA cards, like the literal physical object hasn't really been played with in a really long time. Some technical features have changed like the emergence of tap payments, additional security features and so on, but roughly speaking the state of the art and the card industry, it wasn't really like looking at all the ingredients and figuring out like, how could we combine them?

So just that just left an opportunity. Like the opportunity really was that nobody cared and nobody thought that the aesthetic appeal would matter. So that's like one part. The second part is for the demographic that we were going after for a long time, which is the unbanked and the underbanked. The best in class products for them were not products.

Like they were products that had what I would describe as negative status signaling. So imagine you are anywhere with your friends and you pull out like a random prepaid card that you could buy at CVS. For instance, there is a feeling associated with that, that a person pulling out a chase Sapphire does not feel so.

One of our underlying objectives was to create a physical card product that people were proud to take out of their wallet. That was part of it. There was a part about making sure that the thing we created could take advantage of the tools of the day. So the cash card was one of the first card products that people like.

We're very, very excited to share on Twitter. And then the third part was we had to be like relatively disciplined about costs. So it couldn't like cost materially more than a BV card or anything like that. So that was sort of a big working objective. And then I'd say the last piece was. We want it to put the customer's brand first.

And so we wanted your pride to be much more representative of you than necessarily representative of us. And, you know, we experimented with that over time. But a big part of it was if you give people something that they're proud to take out of their wallet and you give them the ability to like really, really personalize it.

So they're now even taking it out of their wallet, for reasons that are unrelated to payment, like you take it out of your wallet to show, Hey, look at this cool thing that I made. Then what would happen is you would both have people wanting this physical object in a way that's divorced from the financial benefits associated with it.

So it's not like when the cash card launch, we didn't have like points or anything like that, but if we made it possible for people to want this object, for reasons divorced from the financial benefits, then what would happen is. It's an additional lever of customer acquisition. That's not available to, if you're just like straight up competing on like how much of interchange kind of giveaway in travel benefits for instance, which is kind of what the chases and Amex's of the world compete on.

And so it's like to us, like we're relatively privileged. So it looks like, oh, it's just like, it's a cool, fancy object to somebody whose alternative is a prepaid card off the shelf to which the card issuers slapped their brand all over it. The intent really was to change like the way they felt about their card and in 2016.

And I mean, we spent an insane amount of time and effort optimizing that like. I spent probably in combined hours in factories, probably a few weeks along with Robert who's the lead designer on the product. And for the first, probably six months of us working on it, there was just nothing that we felt actually I should give them credit.

There's nothing that Robert felt would like meet the bar. And it took quite a lot. And I think it ended up paying off. Like we launched it. Literally thousands of people share their cash card on Twitter. That's free. Those are impressions that you don't have to pay for. And it drove like viral grows of card acquisition.

[00:22:57] I want to spend more time talking about going deep, particularly in, you were talking about the regulatory space, but I think that's an incredibly underappreciated idea. And to your point that you kind of go on this voyage of discovery and you find out that the common wisdom. Is actually true. And then sometimes you go on this voyage of discovery and you find that the common wisdom is wrong and some incredible opportunities are created by that.

And so I was hoping maybe you could go a little bit deeper in that area and maybe tell a couple stories. And I'm also curious, do you have a framework now that increases the odds? When you go on this voyage of discovery and you're reading and learning and talking to people that increases the odds that you will land on conventional wisdom, that's proven wrong, or do you just start the voyage each time hoping that it lands there and you never really know.

[00:23:49] So the phrase I like to use, and this really was Brian who coined this was going unreasonably deep on a problem. And I think in, it's kind of weird, I think like if you're working on really hard technical problems, like if you're trying to solve climate change or something like that, you don't have a choice for.

Domains where the technical innovation is new to the world. In order to even be relevant, you have to be technically quite knowledgeable to create anything new. And so you see this, the types of problems that like aerospace folks are solving and natural resources people are solving. And so on. Like those are like hard science problems in domains, like financial services, where a lot of what you're doing is business model innovation.

And it's generally rare that your edge is going to be like a scientific breakthrough, the place where you can go pretty deep is in the regulatory domain. And so you can think of it, like what will happen a lot of the time is you run into a problem and you say, okay, in order to solve this problem, like this problem is, is some legal regulatory thing.

And you go to a lawyer and you say, Hey, I've run into this problem. Help me figure it out. And your counsel will, you'll do some original research. They'll talk some experts and they'll come back to you and they'll say, Hey, you should do this. And I think a thing early on that cash app did, was it started with, Hey, let's really put the customer first and make sure that what we're doing is customer friendly or customer oriented.

However you want to describe that and then really go dig in with like we had, I can think of a handful of discussions we had where there's a few of us with our regulatory team sitting in a room with a section of some regulatory, like guidance blown up on a projector. And we're literally just reading and just thinking through, Hey, a lot of this context was written like years ago before the technology that's available today became available.

And so. I think, I think what happens is an N of one in this, so don't quote me, but I think the thing that happens is the further out that you go from when regs are written. The more ambiguous directs seem because a lot of the time, the things that were available at the time have now like really, really changed.

And so what we spent a lot of time doing was thinking about, okay, this technology that is available today is like really, really different from the technology that was available when this was written. So how do we resolve it? So, like an example I would say is before the year 2000 all bank accounts are basically opened in a branch.

Maybe if you had a bank account at Edward Jones or something, you'd open it over the phone. And what would happen when you'd walk into a branch is you'd walk into a bank branch and you'd sit down and they would ask you much questions. You need to fill out some forms. And then they would ask you for identifying documents.

So they would say, Hey, you need to provide a driver's license along with two pieces of mail from either a telecoms or utility company that match the address on the driver's license. So that's like a way of, you have to go to a real Tory authority, like the DMV. To get the driver's license. And then you had to interact with, you have to have power at your house and like somebody has to pay for that power.

And so there's some reliance on the PGNE doing some level of verification that you at least own the place. And so if PGNE, if you're on a bill, And that bill is mailed to your address. And that address matches what's on the DMV. There's some other computations, like sometimes they last for a passport, things like that, but essentially the bank has done its due diligence to reasonably say that you are who you say you are.

And there's a question that we have to answer early on about like, how do you implement that operating model in a mobile app context? Because part of the verification process you can imagine is there's a person sitting at the branch and they're like looking at you in the face and they're looking at the image on your driver's license.

And they're like, Hey yeah, you Brett look like the Brett on this driver's license and so on and so forth. And in a mobile context, or even on a web context, like there is no one who can do that. And even if they could do that, there's things like defects and things like that, that that could create problems.

And so one of the things that we were trying to solve was literally looking through. I forget the exact name of the rag, but there is a, I think it's the FinCEN examiners manual that goes into real deep specifics about, Hey, you have to verify these pieces of information. Here's how you can verify them.

There's this concept of documentary sources where you enter your name, date of birth SSN, and the financial institution can compare that information with information available on a database. Well, so that opens a question, what databases are eligible to be compared with things like that. And then there's a section on non documentary sources of identity verification, which is it's roughly speaking to driver's license path, but then it begs like a couple of other questions, which are like, okay, Somebody shows up with the driver's license.

There's like a number of ways in which it could be fake. It could be straight up just forgery. It could be a real driver's license of somebody else. It could be expired, et cetera, et cetera. And so there's like all these questions you have to grapple with. We spent a lot of time trying to make sure that we did that thing.

Right. And the easy thing would have just been like, take a picture of a driver's license and then like get on video, chat with someone. But then what happens if somebody doesn't have a phone that's video chat cable, what happens if someone has a crappy connection? There's like so many sort of, of these trade-offs that you have to balance and to really understand what the parameters are.

A lot of the time you have to like really get to a combination of what does the guidance actually say? Does it address the cases that we're looking at? And if it doesn't, what's our process for going from something written on paper that was written some time ago to, Hey, we can implement this and know that we're doing the right thing.

[00:29:39] So sort of building on that. If I were to go to sign up for IO guide or course to going deep, are there other lessons or tactics that you would teach people to adopt this philosophy? Or maybe how does this show up in either angel investing you do or now at carbon health? How does it show up or where is it done?

Well, or do you often see it, but it's not done well. And then that creates its own set of challenges. 

[00:30:04] Yeah. Yeah. I've been at Carmen for six months and I would say there are areas where we are looking at and thinking about these, but it's just too early to tell, to be honest, I did see this kind of model play out a handful of times at cash app.

And so I'd be lying if I said I have a guide, I think like right now, the way that I think about it is when looking at problems types. So this is obviously not the same type of problem that exists. For instance, if you're building like a social network or a SAS software, but when you're looking at a problem of this type that has a heavy regulatory committee, I think you just have to be willing to continue to question anything that you don't think makes us.

And in order to do that, you have to be willing to read everything on the topic, like from a regulatory perspective. And then the reality is. Sort of the legal community has tools like really good tools for validating and perspective. And so what I try to do when looking at this stuff is get as close to the metal.

Our second episode is with Ayo Omojola, VP of Product at Carbon Health and the founding product manager for Cash App at Square, where he co-created the Cash Card. Tapping into his experience working in heavily-regulated spaces, we dive into how he finds startup ideas and goes "unreasonably deep" when building products. Ayo shares the frameworks he picked up from his time at Square, and details why he loves hiring former founders.

Product lessons from Cash App & Carbon Health - Ayo Omojola on going "unreasonably deep"

Read the article inspired by this episode

Finding Startup Ideas and Building in Heavily-Regulated Spaces