“It still works” is the most expensive sentence in your company
The most expensive sentence in most companies is “it still works.” It sounds prudent. It ends the conversation. And it is almost always said about a system whose true cost has never once been written down. Legacy is not, at root, a technical problem: it is a leadership one, because the decision to keep an ageing system is a financial decision that almost nobody actually makes. It gets made by default, every quarter, by not being made at all.
I have watched a migration that lost to feature requests for three years straight get funded in a single week. The argument never changed. The arithmetic did. The difference was that, for the first time, someone put a real figure on the system instead of a feeling.
The three lines that change the conversation
Take a fourteen-year-old monolith running order processing. Everyone “knows” it is a problem; nobody has ever costed it. So you cost it, on one page, three lines. The numbers below are rounded and openly estimated, but the magnitudes are the kind we see repeatedly:
- Workaround cost. The team burns time coding around the system, say ten hours a sprint. Loaded and annualised, that is real money leaving the business every year, until someone fixes it. This is the accidental complexity that an old system quietly taxes you for.
- Key-person cost. One engineer holds the billing logic in their head. Replacement plus ramp, if they walk, is a large one-off hit, and HR’s own attrition data rarely makes the odds of that comfortable over a couple of years. This is an organisational risk wearing a technical disguise.
- Hiring drag. Candidates decline after seeing the stack. Each re-opened search has a cost, and the role stays unfilled while the workaround tax keeps running.
None of those three lines lives in any budget. The sprint cost of migrating does. That is the entire problem in one sentence: the cost of change sits on a Jira board where everyone can see it, and the cost of standing still sits in nobody’s budget until someone resigns.
Be honest about the numbers: it still beats a precise zero
The numbers are not audit-grade, and pretending otherwise destroys their credibility. Workaround hours are self-reported and run hot. Attrition statistics are a population rate pointed at one individual. Hiring drag rarely has one clean cause. Say all of that out loud, to the budget owner’s face. Directionally right and openly estimated still beats the precise zero that “the cost of doing nothing” gets logged as by default, because zero is the number that loses every argument to the next feature request.
And it genuinely cuts both ways. “Revisit next quarter” is sometimes the correct call. Plenty of legacy should be left exactly where it is, and plenty of estimates land well over. The failure is not keeping an old system. The failure is never pricing the alternative, so the decision gets made by default instead of on purpose, which is the same way modernisation programmes fail in the decision phase long before any code is written.
This is not the CTO’s fight to win alone
Here is the part most engineering leaders get wrong. The CTO can sequence the migration, design the incremental approach, and de-risk the cutover. What the CTO cannot do alone is make the invisible number real enough to survive the next feature request. Only finance can do that. Workaround cost, key-person cost and hiring drag are financial quantities, and they belong on the same page as revenue and headcount, not buried in an engineering backlog where they are easy to defer forever.
This is a discipline European engineering culture tends to take seriously and one we bring to the organisations we work with across the UAE and the wider MENA region: treat the cost of standing still as a number a named person has to defend, not a vibe everyone agrees with and nobody acts on.
The three lines are right there: workaround cost, key-person cost, hiring drag. Loaded, annualised, one page. You can build them yourself; you do not need us for that. But if the most expensive system in your company has never once appeared on a budget line, that is the conversation worth starting. Talk to TrueForge and we will help you put the number on the page that finally moves the decision.