Legacy System

A legacy system is software the organisation depends on but no longer trusts itself to change: aging, poorly understood, indispensable. Every large organisation runs on several. The strategic questions are always the same — how long can it last, what will replace it, and what does the interim risk cost — and the answers are usually postponed until they become incidents.

What It Is

A legacy system is one still in operational use despite being built on outdated technology, often undocumented, with shrinking expertise (the people who understand it retire) and integration burden (everything newer connects around it). Legacy is not merely age — a well-understood old system is maintainable; legacy in the pejorative sense means change is slow, risky and expensive relative to alternatives. The strategic dilemma: replacement is a multi-year programme with its own failure statistics, while deferral accumulates risk — security patches ending (see End-of-Life), expertise vanishing, and the business bending its processes to what the system can do.

Why It Matters to You

Legacy strategy is a board-level topic disguised as an IT one, because the exposure is enterprise-wide: the core system that processes the pension payroll, the registry no one dares touch, the Excel-and-prayer layer between them. The executive disciplines: maintain an inventory with replacement horizons and interim risk owners; resist the twin temptations of indefinite deferral (risk compounds silently) and big-bang replacement (the graveyard of large programmes); and favour strangler-pattern modernisation — building the new around the old, migrating function by function, running both through the transition — which trades heroics for years of parallel running.

When You'll Meet It

In audit findings, continuity planning, and transformation business cases. Ask for the legacy inventory with last-risk-review dates; systems absent from that document are usually the most dangerous ones in the estate.