Legacy System: "Never change a running system." The most expensive mistake in mid-sized businesses.
The system is not running anymore. It is limping. And it is dragging the entire operation down with it.
Your employees are not unmotivated. They are just working with an interface that looks like Windows 95.
"Never change a running system" is one of the most dangerous phrases in German mid-sized business. It sounds like pragmatism. It is actually deferred surrender.
The system in question is not running. It is limping. It needs workarounds on top of workarounds. It relies on employees who have built up secret knowledge over years that is documented nowhere. And it costs energy every single day that would be better spent elsewhere.
On top of that: when sales or customer service staff have to do unnecessary extra work every day because the system is painful, and the solution gets pushed to "in 24 months", the best people leave. Rightfully so.
Three situations that arise when the legacy system becomes a permanent solution.
We see these situations in almost every company that comes to us.
The ticket nobody processes
Sales needs a small adjustment to the quote generator. Ticket to IT. Reply: "Not possible right now, we are in the middle of the big ERP rollout. Earliest in 12 months." Sales builds quotes with workarounds for another year. Frustration level: 100%.
The result: The best sales person hands in their notice. Not because of salary.
The knowledge that lives nowhere
The system is operated by one employee who has been there 15 years and is the only one who knows how to get certain things done in it. No manual. No documentation. When they leave, the knowledge leaves with them.
The result: The company is held hostage by a system nobody understands anymore.
The mega project that freezes everything
The standard answer: rip everything out, new ERP for a million euros, duration two years. While IT works on the new project, no capacity remains for daily adjustments. Two-year freeze for a system that is already outdated by the time it launches.
The result: A 30-month project blocks not just the budget, it kills innovation.
The oil tanker or nothing. Why that is wrong.
The IT world has had one answer to legacy systems for a long time: rip everything out, big platform project, new ERP. Holistic. Future-proof. Integrated.
Sounds good. The reality looks different.
A 30-month project freezes all IT capacity. Every request from daily operations gets the same answer: no capacity. And when the new system goes live after two years, it is sometimes already the latest from yesterday.
Then there is the fear of island solutions. "If we use different specialised tools, we suddenly have ten systems that do not talk to each other." That used to be a legitimate problem. Today almost every modern solution has an open API. Systems that connect via Make.com, Zapier or n8n, or with a small Python script. The oil tanker is no longer a mark of quality. It is a decision from the past.
Best-of-breed instead of oil tanker. And sometimes two days are enough.
There is an alternative to the mega project. It is called best-of-breed: sales gets the best sales tool, service gets the best service tool, and both are connected via open APIs.
The advantage: the business unit can evaluate, implement and adapt itself. Without the central IT being blocked for months. Specialised partners help when you get stuck. No vendor lock-in. Risk is distributed across multiple systems. If one tool fails or gets worse, you swap it out. Not the entire foundation.
And sometimes you do not even need a new system. Sometimes we lock ourselves in with IT for two days. At the end something has been achieved that was not supposed to happen for another two years. Not because we are wizards. But because nobody had asked the right questions before.
No mega project. Two days. A problem solved that has been waiting three years.
A sales team is working with a CRM from 2014. The system cannot do what sales needs. IT says: we will replace it completely, new project, earliest next year. The sales team has heard this three times now.
We spend a week looking at what is concretely painful. Finding: one single manual process that costs an hour a day can be solved with an API connection to a specialised tool. No new ERP. No year-long project.
- Problem analysis in two days
- API connection to specialised tool built
- Manual process of 60 minutes daily eliminated
- IT capacity not affected
- Business unit can further develop the tool itself
- Next problem area identified: also solvable without a mega project
“Sometimes we lock ourselves in with IT for two days. At the end something has been achieved that was not supposed to happen for another two years.”
What you usually ask us about legacy systems.
STUCK IN A MAJOR PROJECT? LET'S TALK.
Tell us briefly where the shoe pinches most, we will tell you in one conversation what is realistic and quickly solvable.