3 min read

Why Relying on Legacy Code Like COBOL Is a Disaster Waiting to Happen


In April, New Jersey Governor Phil Murphy held a press conference regarding the state’s response to the coronavirus pandemic. Among other announcements and updates, Governor Murphy made a strange request: an open call for programmers familiar with COBOL, a development language that dates back to the 1960s. Of all of the things needed to respond to COVID-19—ventilators, masks, retired doctors to help out overwhelmed hospital staff—why did Murphy ask for coders?

Due to COVID-19, businesses all across the nation have had to furlough employees or close their doors entirely, which has resulted in record-high unemployment claims and an unprecedented strain on public sector systems. During the first week of the outbreak, unemployment claims in New Jersey shot up 1,600%, and more than 362,000 New Jersey residents filed for unemployment during the following two-week period. The surge of claims overloaded the state’s COBOL-based unemployment system, and many claims failed to go through. This prompted Governor Murphy to urgently seek out COBOL-versed programmers to help repair the system.

While the circumstances surrounding the ongoing pandemic are extreme, they help illustrate what happens when we rely on legacy systems for too long. In this case, a reliance on legacy code not only means shouldering the burden of expensive maintenance, but also scrambling for a solution in unexpected situations. It’s a disaster waiting to happen. 


Click above to see Unqork’s platform in action.

How Legacy Technology Holds Industries Back

New Jersey residents are in a tricky situation because their current system forces them to rely on COBOL—but there simply isn’t enough COBOL talent out there to support it. The IT skills shortage has made finding quality tech talent difficult across the board, but finding COBOL talent is nearly impossible. The average COBOL programmer is currently between 45 and 55 years old, which means an already limited talent pool will continue to shrink as older programmers reach retirement age and younger programmers bypass COBOL for more modern languages. At this rate, the pool of COBOL talent will run dry before many organizations move on from using COBOL in critical systems.

COBOL has been sufficiently obsolete since the 1980s, meaning most universities don’t even include it in their computer science curricula. Unfortunately, it hasn’t been as easy for enterprises and organizations to phase it out. The New Jersey government isn’t the only institution that relies on decades-old technology to power its legacy systems—as of 2017, there were 220 billion lines of COBOL in widespread use. 43% of banking systems, including those of major institutions like JPMorgan and Bank of America, run on COBOL, and 95% of ATM swipes still rely on it. 

Relying on legacy code places these institutions in a tight spot. They struggle to find the right talent—and pay a premium when they do. In addition, the significant portion of their IT budget that goes toward legacy systems is spent paying for maintenance, not innovation. An outdated technology stack also makes them unattractive to young talent, who are trained on the latest coding languages and other value adds like social media, big data analytics, or agile development.

COBOL is only one example, or rather, one cautionary tale. There are numerous outdated languages in use that will place enterprises in a less than ideal situation in a matter of years. At best, relying on legacy code will hold an enterprise back from innovation and add millions to their technical debt. At worst, relying on legacy code will critically limit an organization’s ability to react quickly to crises and provide reliable services when people need them most.  

Disaster-Proofing Your Organization with No-Code

Organizations might be able to find COBOL talent in a pinch, but this is only a temporary solution to stave off total collapse during a pandemic or other unpredicted event. To better prepare systems to withstand these issues in the long term, the public sector (and industries in general!) should choose a solution that’s cost-effective, eliminates legacy code, and fosters innovation. No-code is that solution.

“No-code allows you to move away from stopgap measures, break away from outdated technologies for good, and start developing software with the future in mind.” 

No-code allows you to move away from stopgap measures, break away from outdated technologies for good, and start developing software with the future in mind. With Unqork, you can extract existing rules manually and recreate entire systems with ease, using templates and self-validating drag-and-drop components. What’s more, Unqork is so intuitive and user-friendly that anyone, regardless of formal technical experience, can be trained to put together complex workflows. 

In short, not only does no-code help you find talented developers by making it easier to find people compatible with your stack, but it also helps you save money on maintenance. Most importantly, it frees you from the trap of legacy code and allows you to quickly and cost-effectively provide a better user experience for your end-users—whether they’re your constituents or your customers.  

Take a self-guided tour of Unqork’s Codeless-as-a-Service (CaaS) platform

Take the tour!