Dangerous Labels
From Shutterstock
"The word legacy first appeared in English in the late 14th century and derives from Latin legatus meaning 'ambassador' or 'delegated person' (in relation to a will). Over time its meaning subsequently shifted to refer to 'property left in a will'. With the advent of computers in the 20th century, legacy also converted to use as an attributive adjective, and in this sense describes software or hardware that has been superseded but is difficult to replace because it is used so widely." - MacMillan Dictionary
I went to several insurtech events recently and had many discussions on what is an insuretech and what the various types meant, which lead to an upcoming blog on defining the types of insuretechs. Another word I heard a lot is "legacy" used as an adjective, not just for technology but also for businesses, which led me to dig into the origins of the term (see above).
These days, we’re all about labels. Conservative, progressive, inclusive, retro, vintage, basic and more. The labeling has spilled into industry, too. One label that I think is dangerous for insurers to use is “legacy.” Disclaimer: I have done this in the past myself and will try to avoid it in the future to be more clear in my communications.
Legacy was a noun denoting a generally positive thing for centuries. It became an adjective in the 1960s specifically to refer to software or hardware that has been superseded but is hard to replace because it is widely used. Over the last 60 years, legacy has become a pejorative, conjuring visions of rusty systems that no longer meet the needs of the customer. Anyone running MS Windows knows that a previous (legacy) version of Windows that has been superseded by a new release is difficult to replace and is often still the best option at the time.
I am the first one to agree that insurers need to replace core systems to meet the needs of a changing market. Those decisions must be based on actual requirements including maintaining modern, supported technology, and an understanding that existing systems contain a lot of insurance IP that requires real work to replace on a generic, configurable platform -- even if it has APIs and microservices.
Unlike many other industries, the duration of insurance contract relationships, especially in the LA&H segment, requires core systems to have substantial longevity. Using the adjective “legacy” to describe an existing system that has been around for some time is not helpful in explaining why it needs to be replaced. It also fails to define the new requirements for the next system which will, by definition, become a legacy system after a few months or years.
Consigning a core system to legacy status assigns it to death by cancel culture without considering its intrinsic value and what specific gaps have led you to make a change. Worry less about labels like legacy and insurtech, and more about good business requirements, the best combination of insurance IP and a solid core platform with insurance in its DNA.
