Every proposal to migrate a working system runs into the same objection, and it is a fair one: if it still does the job, why spend money replacing it? The status quo has a powerful advantage in any budget conversation — it is the default, and defaults win unless someone makes the case against them. The mistake is to argue back with risk and urgency, which sound like scaremongering to a sceptical board. The argument that works is quieter and harder to dismiss: a number. What does staying actually cost, over five years, once everything is counted?
The cost of staying on Oracle Forms is real; it is simply spread out and unlabelled. There is no line in the budget called Oracle Forms that a CFO can point to, which is exactly why it escapes scrutiny. Instead the cost hides across several other lines — licences, infrastructure, contractor invoices, delayed projects — and in risks that have not yet materialised. Staying feels free because nothing is billed for it directly. It is not free. It is a recurring, and rising, cost the organisation pays whether or not it ever names it.
Start with the costs you can invoice. Forms depends on WebLogic middleware, which carries its own licence. As the support deadlines approach, Extended Support surcharges are added on top of standard fees, and they rise. And keeping a client-server application built around a Java applet running in modern browsers increasingly demands custom infrastructure and workarounds — effort that exists only to keep an ageing architecture alive. None of these is large enough on its own to trigger alarm, which is precisely how they persist. Added together and projected across five years, they stop looking trivial.
Then there are the costs that never reach an invoice. The pool of developers who know Forms is shrinking and their rates are rising, so maintenance gets slower and dearer. Every change to the system is harder and riskier than it would be on a modern platform, so the business waits longer for less. The initiatives the system cannot support — mobile access, integrations, AI — are an opportunity cost paid in things that never happen. And after support ends, the compliance and security exposure becomes a risk with a real, if uncertain, price. These are the largest costs of staying, and the easiest to leave out of a spreadsheet.
A simple illustration shows how quickly the invisible adds up. Suppose maintaining the system needs a specialist contractor at a day rate that rises ten percent a year as the skill grows scarcer; suppose two or three planned initiatives stall each year because the system cannot support them; suppose the WebLogic licence and the Extended Support surcharge sit quietly in two unrelated budget lines. None of these alarms anyone in isolation. Compounded across five years, they routinely add up to more than the one-time cost of a migration — the very figure everyone was worried about in the first place.
The reason staying wins by default is an asymmetry of visibility. The cost of staying is paid in small, recurring increments that are easy to absorb and easy to ignore. The cost of migrating is a single, large, visible number that demands a decision. Set side by side in a budget meeting, the small recurring cost looks cheaper than the big one-time cost — even when, over five years, it is not. Winning the argument means correcting that asymmetry: making the recurring cost cumulative and visible, so it can be compared honestly against the one-time cost of acting.
That is what a proper business case does. It frames the decision as a five-year comparison: the full cost of staying — licences, surcharges, infrastructure, rising maintenance, opportunity cost, risk — compounded over time, set against the cost of migrating, which is largely one-time and followed by lower running costs. The comparison only persuades if its numbers are credible, which is where an evidence-based diagnosis of the system earns its place: it produces a defensible scope and cost, accurate to within roughly ten percent, so both sides of the comparison rest on evidence rather than assertion.
The version of this that lands with a CFO is not that the system is old and risky. It is: here is what staying costs us over five years, here is what migrating costs, and here is the point at which migrating becomes the cheaper option. That reframes the migration from a grudging expense into an investment with a measurable return and a payback period. It is the same logic the CFO applies to every other capital decision, and it lets the migration compete on the terms the finance function actually uses, rather than on fear.
There is one more number worth putting in front of the board: the cost of delay itself. Every year the decision is postponed is a year of the rising recurring costs paid in full, and a year less runway before the support deadline forces a rushed, more expensive migration. Delay is not a neutral holding position; it has a price, and that price climbs as the deadline approaches. Framed honestly, “wait and see” is itself a spending decision — one that happens to buy nothing.
Next step. A diagnosis turns the cost of staying into a defensible five-year figure you can put in front of the board, next to the cost of acting.
