India's banking sector sits on a paradox. It operates the world's most innovative real-time payments infrastructure — UPI processed 16.73 billion transactions in December 2024 alone — while simultaneously running core banking systems in some cases built in the 1980s and 1990s on COBOL and mainframe architectures never designed for this volume or pace of change.

The Reserve Bank of India's 2025 Financial Stability Report identified legacy IT infrastructure as one of the top five systemic risks to the Indian banking sector. Not because the old systems fail — most of them are extraordinarily robust — but because they make change prohibitively expensive and dangerously slow.

When a bank wants to launch a new product — a co-branded credit card, a BNPL facility, an API for account aggregator integration — the limiting factor is almost never market demand or regulatory approval. It is the time and cost required to build the capability into or around a core system that was not designed for it. Workarounds accumulate. Integration debt compounds.

Global banks have been grappling with this problem for a decade. Goldman Sachs spent years and significant capital migrating from COBOL to modern architecture. JPMorgan Chase runs over 50 technology modernisation programmes simultaneously. In India, HDFC Bank and Kotak Mahindra Bank have both made public commitments to core system modernisation programmes.