Brex Business Model: The Enterprise OS Pivot

### The Insight (2017): Why Startups Couldn't Get Cards Henrique Dubugras and Pedro Franceschi were 22-year-old geniuses. They had already built and sold a payment processor in Brazil (Pagar.me) for tens of millions. When they got into Y Combinator, they noticed a bizarre problem: Startups with $5M in the bank (from VC funding) couldn't get a corporate credit card. Traditional banks like Chase or Amex looked at "Credit History." Startups had none. They looked at "Profitability." Startups had none. So they rejected them. Brex built a simple underwriting model: Cash Balance. If you have $1M in Silicon Valley Bank, we will give you a $50,000 limit. It was "dynamic underwriting" that updated daily. They launched "The Corporate Card for Startups." It exploded. Brex went from $0 to $1B valuation in under 2 years—one of the fastest in history. ### The "Growth at All Costs" Trap (2018-2021) Flush with cash, Brex expanded everywhere. They opened "Brex Café" in San Francisco. They put billboards on every bus stop in SoMa. They launched products for ecommerce, life sciences, and small businesses. They onboarded tens of thousands of "Mom and Pop" shops, bakeries, and consultants. But unit economics caught up. Small businesses are expensive to support (they call support often) and have low spend volume. Brex was burning cash to support unprofitable clients. ### The Great Pivot: "Firing" Customers (2022) In June 2022, Brex made headlines for the wrong reasons. They sent an email to tens of thousands of SMB customers telling them their accounts would be closed in 60 days. The message was clear: "We are focusing on funded startups and enterprise." The backlash was severe. X exploded with angry founders. Competitors like Ramp and Mercury pounced, offering streamlined "switch kits" to capture the churned users. But strategically, it was brilliant. Brex leadership realized they couldn't fight a two-front war. They couldn't be "QuickBooks for Bakeries" AND "SAP for DoorDash" at the same time. They chose the latter. They sacrificed revenue for focus. ### The Platform Era: Empower (2024-2025) Today, Brex is not a card company; it is a software company. Their flagship product, Empower, is a global spend optimization platform. It competes with Concur and NetSuite. * Global Reach: They issue local cards in 100+ countries, solving a massive pain point for multinationals. * Software Revenue: By charging SaaS fees for the platform (in addition to interchange), they essentially broke the ceiling of fintech valuations. SaaS revenue is valued at 10-20x multiples; Fintech interchange is valued at 2-5x. By late 2025, that bet started paying off in the numbers. Brex crossed roughly $500M in annualized revenue (up ~40% YoY), enterprise revenue grew 70-80% year over year, and in October 2025 the company turned operating-cash-flow positive for the first time. It now powers spend for 150+ public companies, including Anthropic, Robinhood and ServiceTitan, while positioning itself as an IPO-ready B2B software company rather than just a card lender.

### Fragmented Stacks Before modern spend management, a global company's finance stack was a Frankenstein monster: * US Team: Used American Express corporate cards. * UK Team: Used Barclays cards or paid expenses out of pocket. * Receipts: Uploaded to Expensify (manual scanning). * Approvals: Done via email chains. * Reconciliation: The CFO would spend the last week of every month manually matching Excel rows. This system was slow, prone to fraud, and gave zero real-time visibility. A CFO wouldn't know they were over budget until 30 days after the quarter ended. ### The Multi-National Headache Issuing a card to an employee in Brazil or India used to take weeks of paperwork with local banks. Managing multiple currencies, exchange rates, and tax jurisdictions was a manual disaster.

Brex Empower unifies the entire stack into one operating system. ### 1. Global Issuance (Local Rails) Brex allows a US HQ to issue a physical card to an employee in London, Tokyo, or Sao Paulo instantly. Crucially, these are Local Cards issued on local rails (Mastercard). This means: * High acceptance rates. * Lower FX fees. * Compliance with local regulations. ### 2. Embedded Policies (Standard vs Exception) Instead of chasing employees for receipts, the policy is embedded in the card. * Rule: "Employees can spend $50/day on meals." Action: If an employee tries to buy a $100 steak, the card declines at the terminal*. * Benefit: Zero need for awkward "please reimburse us" conversations. Compliance is proactive, not reactive. ### 3. AI Automation Brex AI automatically categorizes 90% of expenses based on MCC codes and context. "No-receipt" policies for small amounts (e.g., <$25) save thousands of employee hours per year. For a 5,000 person company, this time saving alone pays for the platform.

Brex runs a hybrid fintech-plus-SaaS model, and the whole strategic story is the shift in mix. The company crossed roughly $500M in annualized revenue in 2025 (up ~40% YoY), drawn from four streams.

Costs skew to rewards (~35%), R&D (~30%) and enterprise sales (~25%). The discipline — including a ~20% headcount cut in 2024 — pulled Brex to operating-cash-flow positive for the first time in October 2025.

### Startup to Giant Brex's secret weapon is its vintage cohorts. They acquired companies like Retool, Scale AI, and DoorDash when they were 10-person startups in a garage. As these companies grew to thousands of employees, Brex's volume grew 100x automatically. They "bet on the winners." ### Sales-Led Motion Unlike the early product-led growth days (billboards and self-serve), Brex 3.0 relies on a sophisticated Enterprise Sales team. They hunt "whales"—companies with >1,000 employees. The sales pitch is not "better cards," but "better financial control." They sell to the CFO, not the founder.

### The Numbers Behind the Turn The discipline showed up in 2025. Net revenue retention in the enterprise book ran above 130%, meaning existing customers spent more each year without Brex adding a single new logo. Enterprise revenue grew 70-80% YoY, the segment now spans 150+ public companies, and the 2024 cost cuts (a ~20% headcount reduction) pulled the company to operating-cash-flow positive by October 2025. Land small, expand with the customer, and let the winners compound—that is the entire growth thesis.

### 1. The "Switching Cost" Moat Once a company integrates Brex into its ERP (NetSuite/Workday) and issues physical cards to 5,000 employees, ripping it out is incredibly painful. This creates high retention. It becomes the financial nervous system of the company. ### 2. The Data Moat Brex sees real-time spend data across thousands of tech companies. They know burn rates, software adoption trends, and travel costs better than anyone. This data helps them underwrite risk better than banks who rely on 90-day-old quarterly statements. ### 3. The Global Infrastructure Setting up local BINs (Bank Identification Numbers) and licenses in 100+ countries takes years of legal work and partnerships. This global footprint is a massive barrier to entry for new competitors who only operate in the US.

### 1. Don't be afraid to Fire Customers Brex fired its SMB customers to save the business. It was PR suicide but strategic salvation. If a customer segment is unprofitable and distracting, you must be brave enough to cut it, even if it hurts your topline temporarily. ### 2. Pivot from Product to Platform Cards are a commodity (interchange is racing to zero). Software (Workflow) is a moat. Brex realized that to win, they had to be a software company that sells money, not just a money company. ### 3. The Founder Advantage As outsiders (Brazilians in the US), Henrique and Pedro questioned every assumption of US banking. Why does credit history matter for a funded startup? Why does international issuance take weeks? Sometimes, you need fresh eyes to see the obvious gaps.