A daily problem CIOs face is how to deal with legacy systems—systems built decades ago in what now feels like prehistoric times. Buried deep down in the depths of organizations live these scary, monolithic fossils—Mainframe-otops, ERP Rex, COBOL-odons and more—created in a much different world than the one we live in today. The challenge has now hit the radar of CEOs, as company leaders realize that they cannot innovate fast enough thanks to the cost and complexity of legacy systems.
When legacy systems were first built in the 1950s, they were cutting-edge. Mainframes were the only solution for large organizations looking to run high volumes of transactions and, as a result, captured a large part of the market. It was the right solution for that time in history. And in some regards, legacy systems have been a victim of their own success.
However, over the past 20 years and following the advent of the internet, technology has impacted our world in unprecedented ways. The easy access to capabilities such as cloud, SaaS, mobile, containerization and more is propelling businesses forward at staggering speeds and opening up the market to new, legacy-free entrants. These disruptive entrants are altering consumer expectations in every industry across retail, healthcare, banking and more. As a result, they’ve created an instant gratification economy, where consumers expect personalized 1:1 journeys where they want, when they want and how they want, whether shopping for new clothes, visiting a doctor, or paying a mortgage. Disconnected consumer experiences no longer cut it.
Now, with the dire need to connect legacy backend technology with new-age frontend technology, larger organizations are finding that legacy systems have quickly become a costly burden, slowing them down. And while it’s a natural human tendency to bury problems in the dust for another day, we no longer live in a world where the slow and steady survive. This is enterprise Darwinism, where the fast and agile survive, meaning legacy systems can no longer stay in the ground untapped burying rich data and capabilities. In order to stay relevant and innovate at the edges, organizations first need to evolve their core and bring their legacy systems out of the IT Jurassic Age.
It starts with a mindset shift
Contrary to what one might believe, legacy modernization starts with a mindset shift, not technology. When pulling data from legacy systems, IT teams often resort to ad hoc shortcuts in the form of point-to-point integrations in order to innovate as quickly and cheaply as possible. However, these shortcuts typically only get used once and end up creating tight dependencies between applications. As a result, the broader organization does not get any leverage out of existing assets and ends up redoing most of the work when the company or market shifts—which we know in today’s world is often.
CIOs, instead, need to play a strong role in shifting the way people think about the problem and need to create a proper strategy for modernization. This includes everything from educating developers on how to design APIs for reuse to encouraging the broader organization to discover and consume these reusable assets. That way, when one person taps into a legacy system to pull valuable data—say order history or billing information—the work is already done for the next person looking to leverage the same data. With IT owning the evolution of the core, CIOs need to lead this charge in moving away from a traditional project delivery approach and toward an approach that allows IT to build reusable, self-serve assets that are consumable by the broader organization.
Sparking life into legacy systems
Once the right mindset is in place and IT is ready to move beyond point-to-point integration, it’s time to cultivate an API-led approach. With an API strategy, IT can productize APIs and open up access to appropriate teams, while still protecting the integrity of the systems and maintaining security and governance. As a result, more of the organization is empowered to freely innovate, deliver better customer experiences, and quickly introduce new products and services to the market. According to a recent survey, 58 percent of IT leaders were able to increase productivity by leveraging APIs, while 31 percent were able to increase revenue as a direct result.
It’s important to emphasize that modernizing legacy systems doesn’t mean giving everyone under the sun access to these business-critical systems that house sensitive data. It also doesn’t mean giving everyone read-write capabilities—a common objection. Fortunately, with modern APIs, IT can control who has access to specific data on an individual level. Modernizing legacy systems is about leveraging existing systems—no matter how antiquated—and making them usable in a more modern IT infrastructure. It doesn’t mean relinquishing control and letting the data run amok.
Europe’s largest manufacturing and electronics company Siemens, for example, recently adopted an API-led approach to unlock siloed data residing in its legacy systems. Charged with rolling out 60 million smart meters to accommodate a UK climate change regulation, the company needed a more efficient way of managing its complex network of devices, vendors and suppliers. As a result of its API-led approach, Siemens was able to build an application network that helped maintain the downstream integrity of its legacy systems through throttling and rate limiting API policies, as well as reuse its APIs to expose energy consumption data to regulatory authorities in real time—eliminating the need to manually prepare and submit reports.
With the right mindset and approach in place, legacy modernization can be a strategic enabler, providing the business with long-term flexibility to meet rapidly evolving needs. So, ask yourself: Are your legacy systems still living in the Jurassic Age? If so, how are you going to spark life into them? Rest assured, when you do spark life into your dinosaurs, they won’t grow teeth and chase you down the hall, unlike Steven Spielberg’s version in Jurassic Park.
This article first appeared on CIO.com.
