But perhaps our definition of legacy now needs to change.

A system doesn’t have to be old to be legacy. A relatively modern platform can create many of the same problems if it makes change difficult, restricts access to data, limits innovation or constrains what an organisation wants to achieve.

In other words, and something I argued at a recent international pension event, legacy is increasingly a characteristic of technology, rather than its age.

This distinction matters because life and pensions organisations are operating in an environment where the ability to change is becoming increasingly important.

Reform continues to accelerate with unprecedented levels of change in UK pensions. Expectations around data quality and interoperability are growing. Customers increasingly expect better digital experiences, I mean really good experiences. Operating models need to adapt faster and remain best-in-class, while organisations are also considering how technologies such as AI can genuinely improve the way they operate.

Against that backdrop, modernisation is no longer simply a technology discussion.

The more important question is whether technology enables the organisation to respond.

Can a provider launch or amend products without a significant technology programme? Can it change rules, fees, customer journeys and workflows as its strategy evolves? Can it connect easily with other components and services? Can it make effective use of its data and emerging technologies?

If the answer is consistently no, the age of the underlying platform becomes almost irrelevant.

Historically, the response to legacy has often been transformation.

Organisations replace one large technology estate with another through significant, multi-year programmes. There will undoubtedly continue to be situations where major transformation is necessary.

But there is also another way to think about modernisation: as something continuous rather than episodic.

That means moving away from an assumption that organisations periodically need to replace large parts of their technology estate and towards an architecture where individual components can evolve independently.

This is where modularity becomes important. The presentation identifies API-first architecture, loosely coupled components, independent deployment and upgrades, and best-of-breed ecosystem integration as characteristics underpinning a modular approach.

Instead of treating modernisation as a destination reached every decade or two, organisations can begin to treat it as an ongoing capability.

There is also a wider strategic point.

In a non-modular monolithic environment, strategy can become constrained by technology. Change becomes expensive, innovation slows and organisations have less choice.

That relationship needs to be reversed.

Technology should be capable of responding to business strategy rather than forcing business strategy to accommodate the limitations of technology.

For pension providers, that means retaining greater control over areas such as products, rules, fees, member journeys and workflows.

This is particularly important because the pension proposition itself is evolving. Providers need the flexibility to respond to regulatory developments, changing customer expectations and new ways of serving members without each change becoming another large technology project.

The significance of modular architecture therefore goes beyond software design.

It creates the possibility of a different operating model.

Rather than one technology decision determining the shape of the entire organisation for years to come, components can be more loosely coupled, integration can be API-led and different elements can evolve independently or completely replaced.

That also changes the conversation around best-of-breed technology. Organisations do not necessarily have to choose between an entirely integrated technology stack and a fragmented collection of specialist systems. A modular architecture can provide a foundation through which specialist capabilities can form part of a broader ecosystem.

The goal should ultimately be optionality: the ability to change one part without unnecessarily disrupting everything else.

No pension provider can confidently predict what its regulatory requirements, customer expectations, products or technology landscape will look like in ten or twenty years.

That makes the idea of buying a platform that is “future-proof” questionable.

Perhaps organisations should instead focus on being future-adaptable.

The strongest technology platform may not be the one claiming to anticipate every future requirement. It may be the one that makes responding to those requirements easier when they emerge.

And that brings us back to legacy.

The defining characteristic of legacy technology is no longer simply that it is old. It is that it stands in the way.

If technology slows change, restricts innovation, limits access to data or forces organisations into major transformation programmes simply to move forward, then it is creating a legacy problem, regardless of when it was implemented.

The ambition for pension technology should therefore be bigger than replacing old systems with newer ones.

Technology should unleash change, innovation and growth, not restrict them.

Dan McLaughlin, UK Country Head at Festina Finance