The Inflection: From Payments to Credit
India crossed a line. UPI didn't disrupt banking. It unbundled it. Twenty billion monthly transactions mean transaction rails are commoditised. The next wave isn't faster payments. It's credit for people outside the formal system.
One hundred twenty million Indians have formal credit access today. Three hundred million are lendable. The gap is 200 million. Each represents $125–150 in annual fintech revenue by FY30. That's $25–30 billion in annual opportunity. No Indian fintech market is larger than credit right now.
Why Now? The India Stack Advantage
DigiLocker solved KYC without branch visits. UPI created transaction history on 500 million phones. ONDC is opening merchant cash flow data. Together, these rails let you underwrite people with no credit bureau score.
China crossed this exact inflection in 2008–2010. Ant Financial and Tencent Finance were born here. Neither was a bank. Both used transaction data and alternate signals to serve 400 million people banks couldn't reach profitably. India is three years into the same inflection.
The timing is acute. Credit infrastructure for the next 200 million users must be built in the next 36 months. Later, it gets regulated. Earlier, and you're building in a desert.
The Opacity Problem: Data Beats Narrative
Here's what the market reality looks like on the ground.
Twenty-two of 24 fintech companies tracked in this sector report zero operational metrics publicly. No loan disbursals. No default rates. No unit economics. No repeat customer rates. Only narrative.
One listed fintech showed 4 percent data completeness. Another had zero disclosed metrics, then faced institutional block deal sell-downs exceeding 50 crores. The direction of institutional movement contradicts the conviction of the pitch.
This matters because credit requires discipline. You cannot fake unit economics. Bajaj Finance proves this: it compounds because gross margins exceed 40 percent, loss ratios stay sub-3 percent, and the model works at every scale. That's not narrative. That's physics.
Zerodha and Groww demonstrate the opposite principle. Both are profitable, capital-light, and transparent about numbers. When you remove narrative friction, the model is bulletproof. When you hide metrics, the model is broken.
Three Betting Zones for Founders
Zone 1: Alternate-Data Lending
Serving 200 million Indians with non-traditional income requires underwriting you've never tried. Rent payment history. UPI transaction patterns. Merchant settlement flows. AgriStack data. ABDM health records.
The physics is brutal: cost of capital for this segment is 15–18 percent. If your cost of underwriting plus servicing plus default loss exceeds 12 percent, you don't have a business. Few companies this young can achieve that. But the ones that do will scale to 10 million customers.
Zone 2: Embedded Credit
Embedded fintech is a 30 billion dollar opportunity. Credit embedded in rent, e-commerce, health, and education platforms. The platform provides the moat (trust, data, repeat customers). You provide the capital and underwriting.
This is where the energy is moving. Flent in rental payments. Cashfree in bill payments. Health fintech partners in medicine procurement. The pattern: low customer acquisition cost, high repeat usage, platform benefits from financial gravity.
Capital-light and defensible. This is Zerodha model thinking applied to credit.
Zone 3: Protection and Wealth for Underserved Affluent
Affluent-plus Indians (income >15k annually) number 30–35 million by 2030. They drive 70 percent of deposits and 55 percent of retail investable assets. Yet insurance and mutual fund penetration among this segment is 8–12 percent.
Insurance companies have 200-year data. Their hit rates are precise. Protection and wealth products for this segment have 25–35 percent gross margins and sub-2 percent loss ratios. The model is proven. The segment has trust issues and access gaps. That's founder territory.
The Regulatory Crocodile
Dream Money shut down inside one year. Fibe is preparing an IPO after aggressive growth. Regulators are watching personal lending platforms closely. The RBI doesn't move fast. When it moves, it moves hard.
Institutional capital is walking backward. Block deals signal quiet de-risking. Not panic. Realism.
Founders building here should assume regulation will tighten inside 12 months. If your unit economics require loose rules, you don't have a business. If your moat is high-risk customers, you're betting against regulatory direction.
The Investor's Dilemma
The sector shows severe public data scarcity. You cannot underwrite what you cannot measure. Fintechs with zero reported metrics and premium valuations are pricing in flawless execution and regulatory tailwinds.
Zerodha and Groww are the exception, not the rule. Both report numbers. Both are profitable. Both earned their valuation multiples. The sector doesn't do that. Most fintechs are raising on narrative.
The institutional block deal data is telling. Smart money isn't staying for the IPO. It's exiting early. If your investor base is moving to the exits before you've proven the moat, that's a flag.
What Actually Compounds Here
The fintech value by FY30 is projected at 400 billion dollars across payments, lending, wealth, and insurance. That's real. But 200 billion of it lives in credit and protection for underserved segments.
Founders compounding here do three things:
They underwrite using India Stack rails, not legacy data. UPI and DigiLocker and AgriStack as the foundation.
They achieve unit economics that work in a regulated environment. 12 percent blended cost of acquisition plus underwriting plus default. Not 18 percent with hope.
They report numbers. Not revenue. Disbursals. Default rates. Customer acquisition cost. Repeat rate. The discipline founders in this space show early is the discipline they'll show at scale.
The best fintech investments don't start with a slide deck. They start with a founder who survived one bad quarter and still compounded.