How Stripe Turned 7 Lines of Code Into a $107 Billion Company

One of the topics I find fascinating is startup ideas.

This is only natural, as anyone who has ever thought about starting a business has thought about it.

Paul Graham has a whole series of blog posts about it:

Want to Start a Startup? (Oct 2005)

"Treating a startup idea as a question changes what you're looking for. If an idea is a blueprint, it has to be right. But if it's a question, it can be wrong, so long as it's wrong in a way that leads to more ideas."

Organic Startup Ideas (April 2010)

"The best way to come up with startup ideas is to ask yourself the question: what do you wish someone would make for you?"

Frighteningly Ambitious Startup Ideas (March 2012)

"The biggest startup ideas are terrifying. And not just because they'd be a lot of work. The biggest ideas seem to threaten your identity: you wonder if you'd have enough ambition to carry them through."

How to Get Startup Ideas (November 2012)

"The way to get startup ideas is not to try to think of startup ideas. It's to look for problems, preferably problems you have yourself."

Now, Paul Graham is not looking for small, lifestyle business startups. He's looking for big, VC-backable, crazy ambitious startups. But it's interesting nonetheless.

Because many of the largest companies today, or at least the hottest ones: Stripe, Vercel, Supabase, Databricks, Snowflake—the list goes on—would not have been obvious to me.

And so I like studying and understanding where these ideas came from.

That's what this series is going to be about: studying some of these big companies, what exactly it is they do and how they differentiate themselves, and how the idea came about. What came before them? Why are they a better solution? How did they market themselves? Did anything change along the way?

So, with that said, the first company in the series we're going to take a look at is a Y Combinator darling: Stripe.

If you've ever bought something online — a subscription, a course, a SaaS product — there's about a 1 in 3 chance Stripe processed that payment. They handle hundreds of billions of dollars a year for millions of businesses.

But what's interesting to me isn't the scale. It's what came before.

What Stripe Actually Does (The Simple Version)

If you're not technical, here's the deal: when you type your credit card number into a website and hit "buy," something needs to connect that website to the banking system. That something verifies your card, moves the money, handles fraud checks, deals with currency conversion, manages refunds, and does a hundred other things you never see.

Stripe is the layer that handles all of that. For the business, it's a few lines of code. For you, it's a checkout form that just works. Behind the scenes, it's an absurdly complex piece of financial infrastructure.

What "Connecting to the Banking System" Actually Means

Here's something I didn't appreciate until I dug into this: when you swipe or type in a credit card, the transaction passes through a surprisingly long chain.

It goes: merchant → payment gateway → payment processor → card network (Visa, Mastercard) → your issuing bank. Then it comes all the way back through that same chain with an approval or denial. The whole thing takes about two seconds.

Every step in that chain has its own protocols. Its own security requirements. Its own compliance rules. PCI-DSS alone (the security standard for handling card data) is a 300+ page document.

Before Stripe, a developer had to understand and integrate with multiple parts of this chain. You needed relationships with banks. You needed to handle sensitive card data yourself (or find someone who would). You needed to speak the language of each intermediary.

Stripe's move was to abstract the entire chain into a single API call. You send card info to Stripe. Stripe handles every step after that. You never touch the banking infrastructure directly.

For someone like me — mechanical engineering background, not a payments expert — this is the kind of simplification that makes you go "why didn't this exist sooner?"

The World Before Stripe

I wasn't building software in 2010, but I've integrated enough payment systems at this point to appreciate just how painful it used to be.

PayPal was the default. Founded in 1998 as Confinity, they launched their first electronic payments in 1999 and merged with Elon Musk's X.com in 2000. Their big innovation was letting people send money via email — you didn't need to be a merchant or have any special banking relationship. Just an email address. That was genuinely new.

They rode the eBay wave hard. For years, if you sold something on eBay, you used PayPal. That's how they went mainstream.

If you wanted to accept money on your own website though, you either used PayPal buttons (ugly, limited, redirected users away from your site) or their API (which was...not great). PayPal's documentation read like it was written by a committee that hated developers. Their merchant support was infamous — frozen accounts, held funds, Kafkaesque dispute processes. But they were the biggest game in town, so you dealt with it.

One thing worth noting: PayPal's early team produced an absurd number of future founders. They call it the "PayPal Mafia" — Elon Musk went on to Tesla and SpaceX, Reid Hoffman started LinkedIn, Chad Hurley and Steve Chen founded YouTube, Peter Thiel co-founded Palantir, Jeremy Stoppelman started Yelp. It's probably the most influential alumni network in tech history. Says something about the caliber of people who were drawn to payments as a problem space early on.

Authorize.net was the "serious" option. If you don't recognize the name, here's why it matters: Jeff Knowles founded it in 1996 and basically invented the concept of the "payment gateway." They were the first company to let merchants process credit card payments over the internet. Their innovation was connecting websites to the existing banking and card network infrastructure — the chain I described above.

Their own documentation literally said that connecting to payment processing networks is "exceptionally difficult and typically beyond the expertise and technical resources of most merchants." They weren't wrong — they just weren't fixing it either. Integration took weeks. You needed a merchant account from a bank, a payment gateway account, SSL certificates, PCI compliance audits, and a developer who'd done it before and could navigate the maze.

Visa acquired them in 2010 for $2 billion, which tells you the payment gateway concept was clearly valuable. The execution just hadn't evolved to match what developers actually needed.

Braintree was the best of the bunch before Stripe. Better API, less hostile than PayPal. Founded by Bryan Johnson, who sold it to PayPal for $800 million in 2013. Bryan Johnson is now famous for something completely different — he's the longevity guy spending millions on his Blueprint protocol, trying every possible method to reverse aging. But Braintree was genuinely good for its time. Still complex enough that "integrate payments" was a multi-week project on most engineering roadmaps, but a step in the right direction.

The whole ecosystem had this weird acceptance of friction. Payments were just... hard. Everyone knew it. Nobody was doing much about it.

Two Brothers From Rural Ireland

In 2010, Patrick Collison was 21 and his brother John was 19. They'd grown up in Dromineer, a village of about 200 people in County Tipperary, Ireland. Both were the kind of precocious that makes you feel bad about your own teenage years — Patrick won the Young Scientist of the Year award at 16, John was building and selling software before he could drive.

They were studying at MIT and Harvard respectively, and they'd been building things online for years. They kept running into the same problem: accepting payments was way harder than it should be.

Their insight was deceptively simple: developers are the ones choosing payment tools, not finance teams. And developers care about one thing above all else — how easy is it to integrate?

So they built something that took 7 lines of code.

That's not a metaphor. The original Stripe integration was literally 7 lines of JavaScript. Copy, paste, deployed. What used to take weeks of integration work became a 5-minute job. You didn't need a merchant account. You didn't need to handle PCI compliance. You didn't need to understand the banking system. Stripe handled all of it.

How They Actually Pulled It Off

Stripe didn't rebuild the banking system. They didn't create new payment rails or start their own bank (not initially, anyway).

They partnered with existing payment processors and acquiring banks. Early on, they worked with Wells Fargo as their acquiring bank. All the plumbing underneath — the card networks, the processors, the compliance infrastructure — stayed the same.

The genius was the API layer on top.

Think of it like this: the banking system is a series of old pipes running underground. They work fine, but connecting to them requires digging, permits, specialized tools, and an engineer who knows the pipe layout. Stripe built a single, clean faucet that connects to all the pipes. You turn the handle, water comes out. You don't need to know what's underground.

For someone with an engineering background, this clicks immediately. It's abstraction. Same principle as why you don't need to understand combustion thermodynamics to drive a car.

Why "Just Make It Easy" Was a $107 Billion Idea

This is the part that fascinates me from a product perspective.

The underlying technology wasn't revolutionary. Credit card processing existed. APIs existed. The banking rails were the same. Stripe didn't invent a new way to move money. They invented a new way to access the existing system.

It's like the difference between a car engine and a car. The engine existed. But most people couldn't build a car around it. Stripe built the car — and made it a Toyota, not a kit car. Reliable, turnkey, just works.

A few things made this stick:

The documentation was a product. I know that sounds weird. But Stripe's docs were so clear, so well-organized, with real code examples you could copy and run, that developers chose Stripe because the docs were good. In a world where most payment companies had documentation that felt like it was written under duress, this was a genuine competitive advantage.

Bottom-up adoption. PayPal and Authorize.net sold top-down — enterprise sales teams pitching to CFOs. Stripe sold bottom-up. A developer would try it for a side project, love it, and then push for it at their company. By the time the finance team got involved, the integration was already done.

Instant onboarding. No merchant account applications. No waiting weeks for bank approval. Sign up, get API keys, start processing. For a startup that needs to validate an idea quickly, this was transformational.

The Expansion: From Payments to Financial Infrastructure

Here's where Stripe gets really interesting. They didn't stop at payments.

Stripe Billing handles subscriptions — recurring charges, usage-based pricing, invoicing. If you run a SaaS product (I do), this alone saves you from building a billing system from scratch.

Stripe Connect powers marketplaces. When you pay a driver on Lyft or a seller on Shopify, Stripe Connect is splitting that payment between the platform and the service provider. It handles the complex multi-party money movement that would be a nightmare to build yourself.

Stripe Atlas lets you incorporate a company in Delaware — from anywhere in the world — in a few days. Over 50,000 companies have used it, and they're collectively generating $5 billion a year in revenue. For a founder in Nigeria or India or Brazil who wants to set up a US entity to accept payments, this removed a massive barrier.

Stripe Identity handles KYC (Know Your Customer) verification. Treasury offers banking-as-a-service. Radar does fraud detection with machine learning.

They've quietly become the AWS of financial infrastructure. Payments was the wedge. The platform is the business.

The Moat: Why Competitors Can't Just Copy the API

But if Stripe's original advantage was "payments in 7 lines of code," why hasn't someone just done the same thing and undercut them on price?

Some have tried. But Stripe's moat has gotten deeper over the years, and it's not just the API anymore.

Switching costs are real. Once you've built your billing, checkout, subscription logic, webhook handlers, and reporting dashboards on Stripe's APIs, migrating to a competitor is a months-long engineering project. Most companies won't do it unless they have a very strong reason. And if they're a startup that's grown, the devs in that startup probably chose Stripe at the beginning for their ease-of-use (I've seen this in many companies).

Network effects in fraud detection. Stripe's Radar product gets better as more merchants use the platform. More transactions means more data. More data means better fraud models. Better fraud detection attracts more merchants. It's a flywheel that's hard to replicate without the transaction volume.

Platform expansion creates lock-in. Each new product — Billing, Connect, Atlas, Identity, Treasury — gives customers another reason to stay. If you're using four Stripe products, you're not switching to save 0.1% on processing fees.

R&D reinvestment. Stripe reportedly spends a higher percentage of revenue on R&D than most competitors. They keep shipping, which keeps them ahead.

Brand trust. There's a version of "nobody gets fired for choosing IBM" happening with Stripe. If you're a startup choosing a payment provider, Stripe is the safe choice. Your investors have heard of it. Your developers want to use it. Your board won't question it. That kind of trust takes years to build.

The Competition: Who's Coming for Stripe

Stripe isn't operating in a vacuum. The payments space has gotten crowded, and some of these competitors are genuinely strong.

Adyen is probably the strongest direct competitor. Dutch company, publicly traded, valued at over $50 billion. Their differentiation: they built a single platform that handles both online and in-person payments across the world, all on one stack. No third-party integrations, no stitching together different providers for different countries. Spotify, Uber, Microsoft, and McDonald's use them. Where Stripe grew bottom-up from startups and developers, Adyen went top-down for enterprise from day one. They don't even offer a self-serve signup — you talk to their sales team.

Square (now Block) — Jack Dorsey's company. Their wedge was the opposite of Stripe's: physical payments. That little white card reader let any coffee shop or food truck accept cards with zero setup. Their differentiation is the full ecosystem: point-of-sale hardware, Cash App for consumers, Square Online for e-commerce, banking, payroll. They're building a closed loop where money never needs to leave their platform. Stripe is developer-first infrastructure; Square is merchant-first tools.

Paddle and LemonSqueezy solve a fundamentally different problem. They're "merchant of record" platforms — meaning they legally become the seller. You don't deal with sales tax, VAT collection, tax filings, or payment compliance in 200+ countries. They handle all of it. The tradeoff: higher fees (5%+ vs Stripe's ~2.9%) and less control. For a solo founder selling a SaaS or digital product who doesn't want to think about international tax law, that's a real differentiator.

Regional players win on local knowledge. Razorpay dominates India because they support UPI, net banking, and dozens of local payment methods that Stripe still doesn't handle well there. Flutterwave does the same for Africa (mobile money, bank transfers across fragmented markets). MercadoPago covers Latin America with deep Mercado Libre integration. The insight: payments aren't actually global. Every country has local payment preferences, regulations, and banking quirks. These players know their markets better than Stripe does.

The real threat to Stripe isn't any single competitor. It's that basic payment processing is becoming commoditized. Moving money from point A to point B is table stakes now. That's exactly why Stripe has been building the platform — their moat isn't payments alone, it's everything around payments.

What This Tells You About Building Products

I think about the Stripe story a lot when I'm building products. A few patterns stand out:

1. The "already solved" problem. Payments worked before Stripe. They just worked badly. The biggest opportunities aren't always in new categories — they're in making existing categories 10x easier. When everyone in an industry accepts that something is "just hard," that's a signal, not a fact.

2. Ride a rising tide. Stripe launched in 2011. Online commerce was exploding. Mobile payments were just starting. The number of businesses needing to accept payments online was growing every year. They weren't fighting over a fixed pie — the pie itself was getting massively bigger. This matters more than people realize. If you're building in a space where the overall market is expanding fast, even average execution gets pulled forward. Stripe had great execution and a rising tide. That combination is hard to beat.

3. Developer experience is a moat. This applies beyond dev tools. Any product where the buyer is technical — or where the person choosing the tool is different from the person signing the check — can win on experience. Stripe didn't win on features. They won on friction removal.

4. The wedge → platform play. Start with one thing that's undeniably great (payments). Get adopted everywhere. Then expand into adjacent problems (billing, identity, banking, incorporation). Each new product is easier to sell because the customer already trusts you. This is the playbook I see everywhere now — Vercel did it with hosting → edge → analytics. Cloudflare did it with CDN → compute → storage.

5. Bottom-up beats top-down for technical products. If developers love your tool, they'll bring it into their companies. You don't need an enterprise sales team on day one. You need a product that's so easy to try that people integrate it on a Saturday afternoon.

The Numbers

As of late 2025, Stripe processes over $1 trillion in payments annually. They're valued at roughly $107 billion. They handle payments for millions of businesses across 195+ countries.

All because two kids from a village of 200 people thought payments were harder than they needed to be.

I'm not a developer by training — I studied mechanical engineering. But I've set up Stripe for my own products, and the experience of going from "I need to accept money" to "money is flowing" in under an hour still feels like a small miracle. Especially when I hear stories about what it used to be like.

The best products don't feel like products at all. They feel like the way things should have always worked.

This is part of a series exploring what billion-dollar dev tools replaced and why the old thing died. Next up: Supabase, Vercel, Railway, and more.