The core banking replacement nobody wants to fund

Every bank I work with knows its core system has to go. The question that stops the program is never whether, it is why this year. A core replacement costs more than almost anything else on the technology agenda, takes longer than any executive’s planning horizon, and its benefits are mostly risks that never happen. That is a hard case to put in front of a board.

So most banks wait. They wrap the old core in new interfaces, add a digital front end, and push the decision forward another budget cycle. Each cycle adds integration debt, and the eventual migration gets harder and more expensive.

Stop selling efficiency

The business cases that fail lean on efficiency: lower run costs, fewer manual steps, faster product launches. Those benefits are real, but they arrive late and are easy to dispute. Boards discount them, and they are right to.

The cases that succeed lead with exposure. What does it cost us when the vendor ends support? How many people still understand the code? What happens to our product plans if we cannot change pricing logic without a six-month release? These are risk questions, and boards are built to answer risk questions.

A core replacement is not a technology investment. It is the cost of staying in business on your own terms.

Make it reversible

The second reason boards hesitate is that a core replacement feels irreversible. Once you start, there is no way back. The best programs I have seen design that fear out: they migrate customers in waves, run old and new ledgers side by side, and keep a tested way to roll back each wave. That turns one enormous bet into a series of smaller ones, each with a decision point the board can see.

Show the cost of waiting

Finally, put the cost of delay on the same page as the cost of doing it. Price another three years of wrappers, workarounds and specialist contractors. In my experience that number is rarely smaller than the first phase of a staged migration, and it buys nothing.