Making a Case for Legacy Systems

Moore’s Law would suggest that IT used to build systems today is superior to yesterday’s technology. However, new is not always better. Certainly, today’s IT is smaller and faster, but that doesn’t necessarily make it superior to the technology used in legacy systems. In fact, implementing new IT systems carry inherent risks that may outweigh the drawbacks of modifying or converting existing systems.

New IT systems offer organizations the opportunity to increase speed and efficiency of business operations. However, implementing these systems carries a significant level of risk. In the book, Implementing Information Technology Systems, authors Lomangino, Kaufman and Wills observe that,

“Information technology projects can and often do fall short of their objectives for a variety of reasons, including cost overruns, resistance from staff or external users, and failure of the technology to perform as expected.”¹

In implementing a new IT project, the organization is moving from the known to the unknown and that, in itself, generates the highest degree of risk.

The observations of Lomangino, et al., are supported by the findings of the Standish Group, an organization specializing in IT research². Their 1995 Chaos Report, found that 31 percent of all projects were cancelled before they began and of those that went forward, 53 percent had cost overruns amounting to 190 percent of their original estimates. Furthermore, only 16 percent of projects are completed on time and of the largest American companies, only 42 percent of completed projects possessed the features and functions that were originally specified. Finally, though manageable, employee resistance to the new system is inevitable and is a normal response to any major change³; adding another hurtle for management to overcome. So, based on this evidence, an IT development project is more likely to fail than to succeed. This should prompt management to reconsider the modification or conversion of their existing system.

In making the case for maintenance of legacy systems, Greg Enriquez, writing for Enterprise Systems, claims that, “Legacy systems cast in the proper role can deliver significant value, perhaps indefinitely.”⁴ In fact, organizations that reflexively make decisions about legacy systems based solely on their age will likely take them from a position of strength into a “period of needless cost and disruption”⁵. This is due to the organization’s lack of understanding their legacy system’s long-term viability. The main reasons for replacing legacy systems are increasing inflexibility and maintenance cost. These are legitimate reasons, but according to Enriquez “are not sufficient to justify new systems as soon as they’re available.”⁶

An organization developing a new IT system assumes significant risk. Evidence suggests that the new IT development project will incur cost and schedule overruns, employee resistance and will not attain the desired functionality. These risks make it necessary for management to understand the role their legacy system plays within the organization and the rationale behind the desired change. If the change is the result of a perceived obsolescence of the legacy system, management will likely regret their decision.

References