That system everyone is afraid to touch? It doesn't need to be replaced overnight. It needs a smarter plan.

Almost every company has one. A system that's been running since before anyone can remember, held together by a handful of people who know exactly which buttons not to press. It works - mostly - but it slows everything down, and nobody dares change it.

So it stays. Year after year. While the business grows around it and the workarounds multiply.

Here's the thing: the fear is understandable, but it's based on a false premise. The choice isn't between "leave it alone" and "replace everything at once." There's a smarter way to think about it - and it starts earlier than most companies realise.

The best system is the one that never becomes legacy

Some companies never face this problem at all, because they never let their systems go stale. Instead of leaving software untouched for years and then facing a painful catch-up, they keep working on it continuously: small updates, steady improvements, ongoing attention.

We've been doing exactly this for ours clients for 30 years. The system has never had to "catch up" because it was never allowed to fall behind. That's the ideal scenario: prevention. A system that's maintained continuously simply never becomes legacy.

Not every company gets that chance from day one. But it's worth stating clearly, because it reframes the whole problem: legacy isn't inevitable, it's what happens when maintenance stops.

Why companies freeze

When prevention hasn't happened, systems don't fall apart dramatically. They erode slowly, silently, expensively. The real cost shows up in time lost to manual workarounds, reports that take three people to produce, and good staff who leave because their tools make the job harder than it needs to be.

By the time action feels urgent, the system is so embedded that a full replacement seems like the only option. And full replacements are scary. They take longer than planned, cost more than expected, and the old system's complexity is always underestimated. That's why so many of these projects stall before they finish.

"The software that most urgently needs replacing is exactly the software that feels most impossible to touch."

The false choice holding you back

Most companies that hesitate aren't being irrational, they've just framed the problem wrong. A system that hasn't been maintained properly feels like a ticking clock, and when something feels urgent and risky, the instinct is to act big. Rewrite everything. Start clean.

But that instinct is what causes the failures. Once a system has become legacy, you have options - and they're not all equal.

The right approach, step by step

There’s no one-size-fits-all approach to modernisation. The right path depends on how far a system has drifted and how much risk the business can absorb.

1. Prevent

The best modernisation is the kind you never need. Continuous maintenance keeps systems current by design, just as we’ve done for our clients  for three decades.

2.Assess before you act

Map dependencies, identify the real pain points and understand what could break. Start with the highest business impact, not the biggest ambition. Involve the people who know the system best.

3. Modernise piece by piece

Replace what hurts most first. Introduce new technology where it adds clear value, while keeping the business running. Try, to the extend possible, to make every change reversible or backwards compatible for a certain time. It will take more effort in the short term, but will pay-off in the long run. For the same reason, keep the change in every step manageable.

4. Big bang only as a last resort

A complete, big bang, replacement brings the highest risk, with the least room for error. It is a recipe for trouble and therefore in the long run, the most expensive. Long production interruptions, supply chain issues or customer complaints are no rare exceptions in these big bang projects, reputational damage neither. Avoid it whenever gradual modernisation is possible.

Good modernisation is quiet. The business keeps running, users barely notice the change until one day, everything simply works better. That’s the goal.

Where to begin

You don't need a grand plan. You need an honest look at where the friction is worst: which processes eat the most time, which workarounds have become permanent, what would become possible if one specific thing worked better.

Start there. Build confidence. Then move to the next thing. Done well, modernisation is invisible: the old system quietly disappears, piece by piece, and the business never had to stop to make it happen.

Coteng has been doing this since 1991. 

If your system is overdue for a change but you can't afford the risk. 

We'd like to talk.