They left Stanford to face Amex and now their Brex company is worth $12.3 billion (2023)

Henrique Dubugras and Pedro Franceschi, now co-founders and co-CEOs of corporate cards and expense management start-up Brex, got an early start. While they were in high school, Henrique began programming to create clones of games that his parents didn't want him to buy, and Pedro was the first to jailbreak the iPhone. But when the patent infringement and cease and desist letters came in, they had to find new things to work on.

When Henrique and Pedro were in high school, they met on Twitter and then decided to work together in a company. That company was Pagar.me, a payment processing company that serves customers in its home country of Brazil. They employed more than 100 team members, processed more than $1.5 billion in transaction volume, and were acquired before ever setting foot in a college classroom.

After selling Pagar.me, Henrique and Pedro decided to enroll at Stanford, where they spent less than a year before dropping out to join the YC Winter 2017 group. There they hoped to work at a virtual reality startup, but after a few weeks they realized realized that this was not the right domain for them. As the day of the YC demo approached, they switched to Brex after realizing that most of their batchmates couldn't get company cards despite being backed by venture capital.

Five years later, Brex has raised $1.5bn from investors including Greenoaks Capital, Tiger Global, YC Continuity and others; served on top50,000 customersthrough its expense management and corporate card offerings; and employs a team of more than 1,100 people.

I recently had the chance to sit down with Henrique to get a deeper insight into how Brex was built from the ground up. In our conversation, we examined the granular details of Brex's growth: areas like Brex raised its first round of funding without a product or launch pad, what the team focused on in the 15 months it took to get to a public launch, reflecting the co-founders' philosophy on hiring engineers and other roles in the early days compared to today, and Brex's recent decision (and justification) to focus on serving larger accounts. We also had time to step back a bit to discuss the tailwinds and headwinds that Covid-19 has brought to the Brex business; the "new world we live in" and the role Brex hopes to play in it; and the main priorities of the company.

Steven Li: You and Pedro founded Pagar.me together before you started Brex. What have you learned about proper segregation of duties and how have you applied those learnings to Brex? And how has this model evolved over time?

Henrique Dubugras:At the beginning of our first company, we were both programmers. But Pedro was a much better programmer than I was, so he basically told me to do the other stuff, and the "other stuff" at the time was sales. At first, I didn't think sales was something anyone could do because I always thought it was a role for really extroverted people who could play high school sports and really talk, and I was an introverted programmer. But I've read this book called The Sales Bible and it turns out that if you just do what the book says, it works. That's how it was at first, but over time it evolved.

We discovered that I was a better seller than the leading salespeople. And not just sales in the sense of selling customers, but also discussions with investors, press and other external communication.

So we evolved into this kind of insider-outsider model, one that we also use at Brex today, in the sense that I do a lot of external relations and he does a lot of internal things. Brex was a repeat offender in the sense that when we started over, Pedro focused on the technology and I did a lot of other things like navigate the regulatory landscape, build partnerships, start the company and raise funds.

Today we are working on a co-CEO model where I run the external stuff and Pedro handles a lot of the internal stuff. Now that Brex has developed more companies, my role has also been particularly relevant because with a lot of the larger client meetings and things like that, it helps divide those responsibilities between the two of us.

Brex started because none of the companies in the W17 stack could get a corporate card, but Brex's vision was much more than that. How did Brex approach customer awareness and what did you learn about what are the critical features by bringing in the first one? Do customers have to be present?

We knew from our time at YC that every business needs a corporate credit card, so we didn't have to do much convincing; We just had to make sure that ours is better than the solutions on the market.

And then the first version of that was, okay, let's go to the customers who can't really even get one. And that's why anything we do is better than nothing. So we didn't compete with anything. And so we set priorities. We thought about how we could create the minimal working version where a card is swiped and a transaction is processed.

Then we went feature by feature. Once we got a card working, we needed to create a dashboard where customers could see their transaction history. But we've had customers use Brex even before they could see those transactions because we weren't competing with anything.

Brex graduated from YC in March 2017 and was able to raise a $6.5 million Series A the following month. As far as I know, you did this without code (since Demo Day was too soon after walking away from another VR idea you and Pedro were exploring) and without a deck. Can you explain to me how you did it?

I think a couple of things happened and the first was that we had previously sold our payment company (Pagar.me) and when Mickey (Malka) and others called the company that won us and our clients over by checking references, these they came out positively. . Being a second-time entrepreneur helped us build confidence that we would build what we said we could build.

The second thing was that fintech was all the rage back then; We were in this phase where Stripe and Square were doing really well. In general, since we were early, I think it was a bet on both equipment and space.

I remember your fireside chat with Anu (Hariharan) at Techcrunch Disrupt that you and Pedro were keen to bring Max Levchin on board as an inventor and the plan to open the door was to do an interview on Affirm and then ask for an interview to ask Max before accepting an offer. Can you share how it helped Brex in the early days?

In the early days Max was very good at helping us figure out how the landscape works. He told us about Marqeta, the first vendor we started issuing cards with, and how he got the first debt deal from him.

We knew quite a bit about the fintech ecosystem and payments space from our previous company, but Max understood a lot about the US environment and what was going on here. He also helped us figure out how we would launch and build Brex in the long term. As Brex grew he continued to help us with mentoring to build the business.

How did Pedro and you define PMF in the early days and what were the first signs that Brex had it?

After our public launch, we started to grow very quickly. Customers signed up and started spending, and we got revenue. We went from an ARR to around 10 million within five or six months of launch.

To us, that seemed to be what people would describe as product-marketplace in the sense that people sign up, use our product, and like it. We are also guided by revenue metrics and payment volume.

It took 15 months for Brex to come to market after the seed/Series A (combined round) raise, which you mentioned because you're doing a lot of things right and really wanted to bring a differentiated product to market. And in that time, Brex has gained over 100 customers and tested and reproduced the product. What did you learn during that time and what do you think needed to happen for you to feel comfortable with a public release?

We thought we really wanted to capitalize on the public release. We wanted signup to feel very fast, so we tried to allow 70-80% of customers to sign up right away, without manual verification. We also wanted full features compared to what's already on the market so that as we grow our customer base, they don't feel like it's a terrible product. Although looking back our feature set was very incomplete, we felt it was complete enough to release at the time.

When Brex didn't launch publicly, I'm assuming the first team was deeply involved in each customer onboarding. But as Brex became public self-service, I'm assuming the customer acquisition strategy had to change a bit to keep up with the scale.

yo I'd say it's pretty similar now vs then. Overall, performance marketing hasn't worked as well for us. Lots of startups had Google and Facebook as their primary channels and they just doubled down on them, but to be honest, they never worked that well for us.

We've always been more focused on community-focused things, like channel partnerships, referrals, brands, working to build more general awareness, or bottom-of-funnel stuff. But as we move more into the midmarket and enterprise, there are a lot of things that we should be doing that we're not doing, and we're working on them right now.

Brex increased the number of employees very quickly: from 12 to 100 to 400 to 600 to 1100 (numbers are year over year since inception). In the early days, he mentioned that his approach to hiring non-engineers was to pick the right people to hire as soon as possible and grow into those roles if they weren't already a great fit. And for the engineers, he mentioned that he had a technical challenge creating an authorizer to distinguish between young and old people. Can you explain the reasons for this recruitment strategy? He is curious as to what he was looking for technically and what qualities he was looking for in these crucial early candidates.

At first we didn't hire a lot of junior staff; Our initial strategy was to hire fewer engineers who better understood the architecture and could build things right. As for the authorization issue, we wanted to get an idea of ​​how well someone could do something they could actually work on in Brex compared to things like how many tennis balls can fit on a bus.

A big differentiator on the technical side was the test. Testing things is important because scaling a payments company is different than scaling Facebook, for example. One scenario is if a million people click the "Like" button on Facebook at some point, but it turns out you're not counting some of those likes, it's not the end of the world. Many of the strategies you should use to implement and grow Facebook revolve around having many users access a resource at the same time, but this process may not always be reliable. The question is how to create algorithms and systems that can solve such problems?

While every transaction is very important in Brex, it is vital to make sure we have the correct balances and correct fund movements. In that sense, our authorization question was quite relevant and really relevant to how engineers would think about architecture, databases, and maintaining consistency and reliability.

Listening to some other interviews, I understand that in the early days you always wanted to hire great engineers the way you saw them. And then on the non-technical side, you've often gone for people who didn't necessarily have the experience in the role they were applying for; They didn't want to miss the opportunity to hire talented people, even if they weren't exactly right for the part. Would you say that this philosophy has been maintained as the company has grown?

I would say with most startups you need more generalists when you're smaller and when you grow you want more specialists. Today we try to hire people who have similar and highly relevant experiences with what we do, because every little bit is worth tweaking.

We used to need someone who could learn and do multiple things. Just like Brex today, there are one or two people whose job it is to manage social media. There are other examples, but in general, we would probably want to hire someone who has experience in a specific role, unlike before where we might have had a marketing rep managing social media, email marketing, and other related roles.

Intuitively, it's painful for companies to transition their spend management from something like Concur to a new service, because even if those services suck, they "work" and companies (especially those unfamiliar with the software) can resist. the change. How does Brex feel about this challenge? Can you walk me through the steps in a typical sales cycle that are being taken to mitigate it?

For a long time, Brex was very focused on attracting startups that would grow. So our whole model was, how do we sign some small companies and then grow with them? For example, Scale AI was our first customer and now there are almost 1,000 people still using Brex Empower.

Now that we're targeting bigger customers, those who may already be big, we need to present a good enough value proposition to be worth the cost of switching. There are a number of value propositions that can be compelling reasons for a company to change.

For example, one is the shift to working from home, with many employees now being paid a salary instead of eating lunch at the office. You need a better way to manage this process because when you have 1,000 employees and everyone is submitting a grant expense every month, it's agony for the finance team and the entire company.

Another example is that many business trips have focused on sales reps visiting clients or entrepreneurs visiting partners. Now, with remote work, all companies are off-site all the time and traveling to see each other as teams. So many of the built-in workflows in Concur and others failed when you look at these distributed businesses and their travel patterns.

Then there is the idea of ​​going global. Many companies are starting to hire people everywhere, and to do it right, you need a system to compensate team members in different countries. This paradigm shift is forcing organizations to look for new solutions and re-evaluate their current systems, and we are definitely thinking a lot about it.

How has the Covid-19 pandemic changed the way Brex works? One thing that comes to mind is that when startups started saving cash, they probably cut back on spending, which lowers the POS and therefore exchange rates. At the same time, businesses need their credit cards more than ever to work. If these assumptions are correct, how has Brex adapted operationally in this environment?

If you look at payment processing in general, it's actually gone up. If you look at Stripe and Adyen, for example, their numbers have actually increased significantly due to the "Covid-driven" nature at the request of companies like Doordash and Instacart.

For corporate cards, we've actually seen a drop in processing volume because people haven't traveled. And I will say that during Covid there weren't a lot of companies that wanted new software because everyone is focused on surviving.

Now, in the macroeconomic context of the recession, we see spending fall. But now there's a tremendous focus on doing more with less and on software that makes your business more efficient. So if you haven't helped increase sales over the past year, you've often missed the CFO's time of day. But now all CFOs are looking for new ways to help with cost management, so we've focused on making that happen.

You mentioned that your fundraising philosophy is to build investor relationships over time, raise money only when you need it, and see those fundraisers as an opportunity to partner with great investors and change the price. Brex has raised two rounds of funding in 2021, one in April and one in October (Series D and Series E). Why were these salary increases critical at the time?

We definitely only raise money when there's a lineup where we find the right investor at the right price and in the right market. And we always like to be very overcapitalized. There are some companies that like to optimize to the maximum so that they always get the minimum dilution and only get the capital they really need at any one time. We like to have money left over for those moments. Luckily up to this point we have a lot of money left over because now we have a lot of track.

But we also don't like to raise money from people we don't like to spend time with or who we don't think are good long-term shareholders. We want people who can hold our shares for a long time, because as a public company, the price goes down when a majority shareholder sells a bunch of shares. We are optimizing for long-term holders who are not only not selling, but would consider buying more as we grow. That's why we like to spend our time getting to know people and understanding them, and allowing them to understand the vision. When we discover that an investor really understands our strategy and "gets" what we're doing, he gives us the confidence to execute it for the long term.

Timing wise, I think 2021 was a special year for fundraising. There was so much money in the market that not raising funds last year was almost stupid. The cost of capital is really low and could just go up, and we raised over a billion dollars last year. With relatively undiluted reviews, you could be paid incredible amounts of money. We found the partners we wanted to do it with and we did it.

Tiger led first and Green Oaks led second and I think if you ask them they don't regret making those deals. They are very excited about Brex and happy that we have been able to work together.

At the beginning of June this year, Brex made the decision to stop serving small businesses (defined as companies that have not raised venture capital). And in the context of the Brex Empower launch, that direction definitely seems more business-focused. But double-clicking on Pedro's thread about not being able to provide "white glove service" to these smaller businesses, when did that first become an issue?

We started out serving startups and had three expansion packs: eCommerce, Large Business, and Small Business. For the latter, we thought of serving traditional neighborhood stores and expanding horizontally. We tried to edit and expand on all three segments at the same time.

What we didn't expect was that small businesses were consuming all of our resources; We went from taking 500-700 clients per month to taking 5000-7000 clients per month. Then we mainly focused on improving support for those customers, initiatives like building more automation and anti-fraud tools. At this scale there have been many challenges and we have allocated many resources to serve this customer segment and have neglected the other two.

When we look at the business, we think, okay, it turns out we can't do all these things at the same time because all these problems are so complicated. We have to choose some that are not so complicated or choose some that we think will be the best for the company to solve. For us, that meantI had to choosebetween larger clients or small businesses and we went with the larger clients because we really had no choice.

Our major customers, like Scale AI, have told us that if they don't build something for us, we'll have to think of something else. So do that or you'll lose a lot of our biggest customers. So we decided to go offline and focus on those customers, the startups that are growing, and also the larger customers.

How did the team come to the conclusion that venture capital-backed companies fail? Why not something like income?

It's hard to define, but basically we think that if we look at our product roadmap and the things that we build, we're building things for a company that has reached a certain scale. And by size we mean a large number of employees.

Because if you look at expense management, for example, and you have a lot of employees, it's often very valuable to make this process very efficient. If you don't have many employees, there aren't enough expenses to process that you don't have to do anything particularly efficient.

Also, a startup going big happens all the time, but a retail store turning into a Walmart, for example, is a bit more rare.

What do you see as the top three initiatives at Brex over the next year or so with the launch of Empower and the refocusing of efforts towards enterprise?

Number one ensures that we support global business. As businesses go global, they need to figure out how to issue cards anywhere, refund people anywhere, and pay taxes anywhere across multiple entities.

Number two is building expense management at scale – all the functionality needed by businesses on the scale of DoorDash and Coinbase, which we've already signed up for. We want to be able to say that as you are a VC-backed company and you grow, you can be sure that Brex will allow you to scale indefinitely.

And then number three is building our go-to market for this new world that we live in. Previously, all of our marketing and go-to-market was designed around corporate cards – how we would serve startups and other segments. Now that we're building for larger companies, many of our marketing, sales, and other functions must evolve. That's why, among other things, we hired a new CRO.

Since the corporate cards and expense management categories are so large (even in the US), what was the reason for allocating resources internationally now or later?

We believe that all companies in our category will eventually become a global company, and this is especially true in this macroeconomic context. The cost of hiring an engineer in Brazil is much lower than in the US and honestly they are just as good.

I think we interviewed for a HR position and in Toronto the talent is literally half the price compared to San Francisco; and the person we ended up hiring in Toronto was better than the SF candidates. I think every company, especially technology companies that are efficient and profitable, are already a global company or will hire people from all over the world. The talent market will be the global market.

How would you describe "this new world we live in" and the role Brex hopes to play in it?

If we look at what has happened in the last few decades, organizations are becoming more and more distributed. And by "distributed" I don't necessarily mean remote work, just that companies now have multiple locations.

If you look at Walmart in the '60s, they opened one store and it was really hard to open a second store, but they opened it and figured out how to do a lot of business. Back then, Sam Walton bought a plane to fly back and forth between the stores just so he could handle them.

But then phones went mainstream. And then we had the internet, then email, and then Zoom. With all these new tools, companies could open more offices to hire employees at a lower rate from a previously inaccessible talent pool or have team members closer to their clients. We continue to move in this direction with Google Docs, Figma, and other tools that allow teams to work better without being in one place.

We create a lot of tools, but no one has done it for expense management, but money is a type of resource allocation process of every business. And when we talk to a lot of companies and ask them if they want to save money, they say, "Yes, we want to save money, but more importantly, we want to save money to put it where it's needed most." “Businesses want to make sure money goes to the right places in the business, not wasted in unimportant places. And I believe that with Brex Empower, our expense management solutions, we're enabling this next generation of distributed business to empower every employee to make better financial decisions.

If we can empower people (the 99% who do the right thing) to break free of red tape, we can help companies move much faster, no matter where their teams are located. And that is especially true on a global scale.