What is a legacy system and why do we call it that?
In many organizations, the word legacy triggers strong emotions. For some, it’s synonymous with “something old and problematic,” while for others, it’s a critical part of the business that simply has to work. To have a meaningful discussion about legacy systems, it’s worth starting with the basics: what they actually are and where the term comes from.
What is a legacy system?
A legacy system is an IT system that:
- was created in the past (often many years ago),
- is still used in everyday business processes,
- plays a key role in the organization’s operations,
- but no longer keeps pace with current technological or business needs.
One important point:
👉 Legacy does not automatically mean “bad” or “broken”.
In practice, these systems are often:
- stable,
- well-proven,
- responsible for critical processes (e.g., settlements, accounting, core operations).
The problem begins when maintaining or developing them becomes increasingly difficult.
Why do we call them “legacy”?
The word legacy means heritage – something passed down from the past.
In the context of IT, it’s a very fitting metaphor.
A legacy system is:
- the legacy of technological decisionsthat were once the best possible choices,
- the result of past business, regulatory, and organizational realities,
- the outcome of years of modifications, workarounds, and compromises.
No one built legacy systems expecting them to become a problem. On the contrary – at the time they were created, they were modern and effective.
When does a system become a legacy system?
There is no single time or technology threshold. A system does not become legacy simply because it:
- is “old”,
- runs on older technology,
- has been in the company for many years.
We start calling a system legacy when signals such as the following appear:
- Difficulty in development
Every change takes a long time, is expensive, and carries significant risk. - Lack of flexibility
The system does not support new business models, integrations, or automation. - Dependence on a small group of experts
Knowledge about the system exists in the heads of a few people (sometimes just one). - Out-of-mainstream technology
No vendor support, difficulty recruiting specialists, lack of updates. - Rising maintenance costs
The system “works” but it consumes more and more time, money, and energy to keep it running.