Why it’s worth taking a fresh look at legacy applications right now
Anyone in a leadership role within a building society’s IT department today has a long list of strategic issues on their plate. The modernisation of core systems, DORA, cybersecurity, artificial intelligence and digital sovereignty all demand time, budget and staff. However, in many discussions, I’ve noticed that another area often only receives attention when problems have already arisen or are on the horizon: legacy applications.
These are not applications that are being further developed on a daily basis. On the contrary. They are often precisely those systems that have been running stably for years. They perform their tasks reliably, are rarely the focus of projects and are therefore easily overlooked.
In the building society sector in particular, this applies to applications that process historical tariff generation, bonus interest models, housing construction grants or legacy contracts. Many of these business-specific features have been part of the institutions’ operations for decades and will continue to exist for years to come. That is why such applications will remain part of the IT landscape for a long time to come.
Stable does not automatically mean non-critical
The fact that an application has been functioning reliably for years is, at first glance, good news.
The challenge often lies elsewhere.
Who really knows the application anymore? Who understands its specific business requirements? Who can explain why certain calculations were implemented in exactly that way?
In many organisations, this knowledge is now concentrated among just a few COBOL or host specialists. Some of them will be retiring in the coming years. At the same time, there is growing pressure to deploy experienced staff on new projects – ranging from the modernisation of core systems to AI initiatives.
As a result, legacy applications find themselves in a situation that I am observing more and more frequently in projects: technically, they are running stably. Organisationally, however, they are increasingly becoming a risk.
Not every application needs the same level of attention
In my view, it is therefore worth considering a simple question. Does every application need to be maintained in the long term using the same internal resources as projects that contribute directly to future competitiveness?
Many legacy applications rarely undergo functional changes anymore. At the same time, they must run reliably, meet regulatory requirements and remain traceable at all times.
For such applications, specialised application management can be a sensible option. Not because these systems are any less important, but because they have different support requirements to projects that are continuously evolving.
Whilst operations are safeguarded, internal teams gain the freedom to focus on areas such as digitalisation, AI, regulatory requirements, the further development of core systems or the implementation of new business cases.
Safeguarding knowledge whilst it is still available
Another point is often underestimated: legacy applications are rarely just about source code. What is crucial is the technical and domain-specific knowledge behind them.
Why does a tariff behave in exactly this way? What exception was built in twenty years ago? What impact does a change have on other processes?
Such interrelationships are often not recorded in any documentation. This is precisely why knowledge preservation should not only begin when experienced staff leave the company.
In my view, it is now an integral part of any professional application management practice.
AI can provide valuable support here
It is precisely at this point that the role of artificial intelligence is also changing. Whilst AI is often associated primarily with software development or automation in many discussions, it opens up entirely different possibilities in application management. It helps to analyse existing applications, create documentation and systematically organise specialist knowledge that has accumulated over the years.
This can make knowledge transfer considerably easier, particularly with mainframe applications – naturally, always whilst taking into account data protection, digital sovereignty and the regulatory requirements of the financial sector.
My conclusion
Legacy applications will continue to play an important role in modern IT landscapes for a long time to come. This applies in particular to building societies, with their historically evolved range of tariffs and contract portfolios.
That is why it is worth giving these applications the same attention as major transformation projects. Not because they are spectacular, but because they must function reliably every single day.
Organising operations, documentation and knowledge transfer at an early stage achieves more than just stability. It creates scope for the issues that will shape building societies’ IT in the coming years.
