This approach also allows businesses to offer that same set of data across numerous channels such as mobile, connected devices, reporting and so on. The list can be endless since all that needs to be done is to create a data consumption and exchange point.
According to Gartner, we can define legacy systems as: "An information system that may be based on outdated technologies, but is critical to day-to-day operations”. These legacy systems tend to have grown within organizations over several decennia. Not only making the system hard to maintain, but also implying that a large part of the contained functionalities are quite likely now obsolete, unused and most often very poorly documented.
Legacy systems have a few characteristics that make them easy to identify, but at the same time hard to replace or re-evaluate:
• Total cost – Legacy systems have typically been developed over a longer time span, making the total investment in knowledge retention, resources and functional oversight immense.
• Integration – Functionalities offered by the legacy system were most likely put in place to tailor for and to specific other systems. Making the system closely connected and very rigid.
• Silo – Data contained is typically hard to exchange and modify since the complexity of the system makes it hard and dangerous to modify without knowing the exact implications.
• Functionalities – Due to a lack of detailed functional description and the changes that were introduced to the system over its lifespan make it nearly impossible to list all functionalities offered by the system.
Why change at all?The tell-tale-signs described above make it clear that legacy systems introduce not only security risks, they also require high maintenance costs, both in personnel as well as in intellectual knowledge and documentation & testing. These higher costs can however, not mitigate the loss of business opportunities. Since a number of business requirements simply cannot be offered by the legacy systems (such as easy data manipulation and exchange), new functionalities cannot be implemented without making the system even more convoluted.
Organizations know that big data has become remarkably important. However, when your organization is stuck using legacy systems at its very core, that data (if at all collected) is stuck due to the lack of data interoperability. Meaning that the large quantity of data that is contained in your legacy system can simply not be put to use in order to realize new business opportunities.
When faced with the prospect of replacing a legacy system, IT and management typically look back at the total cost of the creation and maintenance of the system and use this as a benchmark for a new system or at least the cost thereof.
This approach is however far from correct. Evaluating the cost of replacing a legacy system should be done based off the value such a new system could bring. So, instead, evaluate on the realization of that backlog of business requirements, customer experience improvements and improved reporting.
In short, “If it ain’t broke, don’t fix it” – would also apply here. The legacy-system is most likely broken in terms of flexibility, offering an undersized feature-set and a mismatch with the current cloud-enabled mindset that would make fixing it an unsurmountable task.
How to move forward?
Because legacy systems lack a clear, service-oriented architecture, they typically reside in a siloed environment. Make sure that when you are selecting a new system, it provides an architecture that focuses on ease of connectivity, in order to survive in the modern world of application and APIs.
Consider the fact that a newly developed or implemented system might, over time, become your legacy application of the future. In order to prevent this, make sure that the new application can easily be interchanged without severely affecting other systems. Create a clear API-based/layered architecture and build up your IT strategy around this very concept.
More importantly, listen to the needs of your business and customers and let those dictate the choice for internal systems and processes. Avoid the inverse, where legacy systems tell you what is possible and what not.
Need our help?