🏗️ Building Trust at Scale: How Razorpay Survived Near-Death and Conquered India's $180B Payments Market

📌 Introduction: From Oil Fields to India's Payments Giant

Few companies embody the journey from zero to generational winner quite like Razorpay. As the first Indian company ever backed by Y Combinator (Winter 2015), Razorpay now powers the largest payments platform in India — processing $180 billion in gross merchandise value annually, up from a market that was just $60 billion when the company started in 2014.

What makes this trajectory remarkable isn't just the scale. It's the grit, the crisis moments, and the founder conviction required to navigate regulated markets, existential threats, and a rapidly transforming financial landscape. Harshil Mathur's journey from techie in the Middle East to fintech founder offers timeless lessons on trust, timing, and building in hard spaces.

"If I knew what I knew today, maybe I would have not started. But the hard part — if it's hard for other reasons, it's not really a problem. It's actually a moat." — Harshil Mathur

🚀 Origins: A Side Project That Exposed a Systemic Problem

Harshil didn't set out to build a fintech empire. Graduating from IIT, he landed a job at an oil company in the Middle East — a role he never enjoyed. Weekends and evenings were spent coding side projects. One of those projects, a social crowdfunding platform, required payment integration. That's when he hit a wall.

The problem was glaring: accepting digital payments in India was harder than accepting cash. Banks weren't built for startups or developers. The onboarding process was arcane, slow, and inaccessible to small businesses. For a techie used to democratization through technology, this felt backwards.

Harshil spent time in Bangalore startup Facebook groups and saw the same pain point echoed everywhere: payments are brutally hard in India. The insight was simple but profound — when a digital system is harder than a physical one, it's broken.

That contradiction became the foundation of Razorpay.

🎓 The Y Combinator Paradox: Three Months Without a Single Transaction

Razorpay applied to Y Combinator with a plan to target educational institutions for fee collection. The logic seemed sound: billions in tuition payments were being collected offline, and digitizing that flow could unlock massive value.

Reality had other plans.

During the three-month YC batch, Razorpay didn't process a single live transaction. They had received an in-principle approval from banking partners, but the final green light took nearly a full year after YC ended.

In most tech businesses, you can start selling from your bedroom. In payments, you wait for licenses, certifications, compliance reviews, and approvals. The regulatory lag was brutal — but it taught an early lesson:

"Yes, there are a lot of hurdles. But everyone coming after us will also face the same hurdles. The regulation becomes a moat over time."

While waiting for approvals, the team pivoted away from education. Schools and universities didn't care about smooth digital payments — they had captive customers who would pay however they were told to pay. Startups, on the other hand, desperately needed better payment infrastructure. That pivot, born from boots-on-the-ground customer feedback, set the trajectory for everything that followed.

💥 The Near-Death Moment: When the Bank Pulled the Plug

Just two weeks after Demo Day, with about 50 live merchants finally onboarded, disaster struck. The bank that had enabled Razorpay's payments infrastructure pulled the plug — completely and without warning. A single customer complaint triggered the shutdown.

For a payments company, this is existential. Customers rely on you to collect money. When you can't deliver, trust evaporates instantly.

The team had a choice: go silent and avoid the fallout, or face it head-on.

They chose the latter.

Six or seven team members — including Harshil — sat in a room and started calling every single customer. They explained what happened, why it happened, and what they were doing to fix it. Customers were furious. Many vented with intense anger and frustration. But Razorpay kept picking up the phone.

"B2B is a business of trust. And trust, at the end of the day, nothing replaces the human touch point. Just having somebody on the other side saying 'I am with you, I am managing it' gives a lot of trust."

Within four to five days, Razorpay secured a new banking partner and brought merchants back online. Remarkably, many of the customers who had initially been furious remained with Razorpay — some are still customers today.

This crisis forged a core company principle: never stop picking up the phone. Even as AI transforms support functions across the industry, Razorpay maintains that human escalation remains non-negotiable. Trust isn't just about solving problems — it's about showing up, especially when things break.

🎯 Capital Efficiency as Strategy, Not Accident

Between 2015 and 2020, Razorpay grew at an extraordinary pace — scaling 40x with record capital efficiency. After raising an $11 million Series A, the company's monthly burn rate stayed below $200,000. In fact, interest earned on fixed deposits exceeded their burn rate, making Razorpay operationally profitable.

Their investor wasn't thrilled. This was an era when aggressive cash burn was expected — even celebrated. Consumer companies were lighting capital on fire to acquire users. But Harshil held firm:

"In B2B, it's very simple. You add value to a business, the business pays you for it. There's no point burning money for engagement. The business uses you today because you add value. If you stop adding value tomorrow, they use somebody else."

This discipline wasn't just philosophical. It was structural. B2B businesses don't benefit from irrational subsidy wars. Value delivery, reliability, and compounding trust matter more than CAC-fueled growth hacks.

The result? Razorpay built a durable, profitable engine long before the broader market woke up to the importance of unit economics.

⚡ The UPI Bet: Moving Early When Others Hesitated

In April 2016, India launched the Unified Payments Interface (UPI) — a real-time payment system that would eventually transform the country's financial infrastructure. But in its early days, UPI was anything but a sure bet.

The two largest banks in India hadn't even integrated UPI until November 2016. Most payment gateways dismissed it entirely, assuming that without major bank participation, adoption would stall.

Razorpay saw it differently.

In September/October 2016, Razorpay became the first payment gateway in India to go live on UPI — months before the big banks joined and before demonetization created a massive overnight shift toward digital payments.

When demonetization hit in November, demand for UPI exploded. No other gateway was ready. Razorpay was.

Within weeks, major platforms — Zomato, Swiggy, BookMyShow — integrated Razorpay specifically to access UPI functionality. It took competitors six months to catch up. By then, Razorpay had secured relationships and market position that would have been impossible to win through conventional sales alone.

"The worst that could happen was UPI doesn't take off — fine, we're still fighting the battle anyway. But if it does take off, it creates a unique wedge for us."

This is classic startup asymmetry: small players can afford to make bold bets that incumbents can't justify. Speed and conviction become structural advantages.

🤖 Rewiring for AI: Acting Like a Startup, Not an Incumbent

Razorpay recently launched a fundamentally reimagined platform — rebuilt from the ground up with AI at its core. The process began with a simple but radical question:

If we were starting Razorpay today, how would we build it?

The team went back to the whiteboard. They re-evaluated onboarding, integrations, support, and the entire user experience. The goal wasn't incremental improvement — it was reinvention.

Harshil is candid about the risk of incumbency:

"We disrupted a lot of incumbents to get where we are. The reason we succeeded was because those incumbents waited to respond. The market changes faster than you can respond. If you wait until the market starts changing, you're already too late."

AI, in Harshil's view, is compressing time-to-build to near zero. The moat is no longer in the build itself — it's in deciding what to build and moving faster than the market expects.

Razorpay is not treating AI as a feature. It's treating it as a platform-level transformation — one that requires rewiring teams, rethinking workflows, and accepting short-term friction for long-term positioning.

Notably, while AI is being deployed across the business, customer support remains a human-first function. Trust, especially in financial services, still requires a voice on the other end of the line.

🧠 Founder Mode: The Journey from Builder to Manager — and Back

Like many founders, Harshil went through the classic arc: early-stage builder → hiring executives → stepping back into "manager mode."

For a couple of years, he operated on the assumption that great leaders could run their domains independently. He focused on managing leaders rather than staying deep in the product, strategy, and vision.

It didn't work.

"Nobody's going to care about your company as much as you do. That's never going to change — not on day one, not on year 10, not on year 20."

Harshil realized that while delegation is necessary, founders must stay deeply involved in the things that define the company's direction. Product vision, strategic bets, cultural decisions — these can't be fully outsourced.

This isn't micromanagement. It's founder mode — a term Harshil resonated with when Paul Graham's essay came out. The lesson? Hire great people, empower them — but never abdicate the decisions that only a founder can make.

🔥 Advice for Builders: Find a Problem You Can Spend 10 Years Solving

AI is making it easier to build — but it's not making it easier to build a company.

Harshil's closing advice cuts through the noise:

"The core aspect of entrepreneurship is the ability to connect with a problem deeply and spend all of your time and effort solving that problem. With AI, my worry is that because it's easier to build, it's easier for people to latch onto problems they don't really care about."

The question every founder must answer: Is this a problem you can spend the next 10 years of your life solving?

AI will compress build cycles. It will lower the cost of experimentation. But company-building still takes a decade — and it still requires obsession, resilience, and the ability to show up every single day, even when things break.

✅ Key Takeaways

• Regulation as moat: Hard industries create defensibility — everyone behind you faces the same barriers
• Trust > efficiency: In B2B, especially fintech, picking up the phone beats AI chatbots
• Capital discipline works: B2B companies don't need to burn — they need to deliver value and get paid for it
• Bet early on transformations: Razorpay's UPI bet created a wedge that incumbents couldn't match
• Founder mode matters: Delegation is necessary, but abdication is fatal
• AI changes build speed, not company-building timelines: Choose a problem you can obsess over for 10 years

Razorpay's journey from a frustrated side project to a $180 billion payments platform is a masterclass in grit, timing, and conviction. The lessons here aren't just for fintech founders — they're for anyone building in hard spaces, navigating crises, or trying to move faster than the market expects.