Legacy systems are often cited as one of the top barriers to digital transformation success.
While enterprises pour millions into digital transformation and innovation initiatives, many find the commercial impact is missing while they remain tied to their aging legacy systems.
Leaders need to develop a modernization strategy for their systems that focuses on improving them and creating near-term customer value and business benefits—but getting organizational buy-in for even incremental changes can be challenging, since legacy systems come with more baggage than just outdated technology. Leaders have to debunk some common myths that will surface in their executive group’s list of objections before they can hope to move forward.
Getting organizational buy-in for even incremental changes can be challenging, since legacy systems come with more baggage than just outdated technology.
1. Legacy systems can’t be modernized
Although some organizations find success in completely replacing their legacy systems, for most it’s not a feasible approach because of costs, complexity, or because recreating all of the underlying business logic contained within their current systems seems near-impossible.
As an alternative, most organizations choose a more pragmatic approach: Using one or more methods to incrementally replace their aging systems, or to allow their existing systems to merge with new ones.
Incremental replacement
With an incremental replacement, the organization reduces their dependence on legacy systems over time. One of the more popular methods for doing this is called the Strangler Fig Application, where you gradually dismantle and re-architect the application to modernize it—for example, slowly replacing some of the services of your on-premise monolith with newer services in the cloud.
The Strangler Fig approach. Original diagram courtesy of Paul Hammant.
Augmentation
With the augmentation approach, organizations adopt a new system that runs in tandem with an old system—either independently or with an orchestrating service coordinating the old and new products or services.
This approach is often a good fit for organizations where the legacy systems are not exactly the problem, but the overall ecosystem prevents teams from building new products and services without friction.
Using modern technologies like GraphQL, it is possible to free the organization to create better customer experiences faster, while keeping all existing systems (regardless of their technology) integrated in a cohesive enterprise architecture.
Illustration of how a GraphQL Server streamlines communication between legacy systems and new cloud services to deliver a better customer experience.
Continuation
Another approach that can be employed for systems that must stay in service for the long term is continuation. Here, the older system is kept running with gradual improvements added for customer or business benefits, such as build automation.
However, the risk of this model is in obsolescence and missed opportunities. There might be cases where replacing old systems is not feasible, but this should be an extremely rare situation.
2. Legacy systems are a utility and don’t need modernization
Many executives still hold the idea that their legacy systems are utilities, and require little more than day-to-day IT maintenance and support. However, legacy systems represent a significant barrier to overcome on the critical path to successful transformation and innovation, and customer-centric digital solutions require core business systems that can adapt to support better digital workflow. They must cut across many different user experiences—not just what the system was built to serve when first created. The result? We often see clients with six or more legacy CRM systems, none of which can communicate easily with the others. Transformations are stuck, deemed too expensive, re-implemented multiple times, or avoided altogether.
There are three risks inherent in the mindset that these systems are merely a utility: 1) Replacing your workforce, 2) Rising maintenance costs, and 3) Cyber security threats.
Replacing your workforce
For many enterprises, systems were built and upgraded over decades, with older styles of programming and documentation methods (or no documentation at all). Much of the knowledge needed to maintain and operate these systems is kept by a few senior employees, or is hard to hire for as new programming languages gain popularity.
A recent example: As New Jersey’s unemployment numbers spiked during the Covid-19 pandemic, residents found they were unable to process online employment insurance claims as the system kept crashing. Governor Phil Murphy publicly pleaded for experienced COBOL developers to come to the aid of the state in order to keep the system up-and-running. When even vital services are dependent on outdated coding languages, the need for continually-updated systems becomes clear.
Rising cost of ownership
When leaders factor in employee salaries, system infrastructure, and maintenance, legacy systems become extraordinary financial sinkholes.
According to the U.S. Government Accountability Office, (GAO), the federal government spends close to $100B a year on “IT,” with 80% spent on the operations and maintenance of legacy systems—leaving only 20% for development, modernization and enhancement.
These unbalanced costs create a significant risk for non-digital enterprises when compared to their born-digital counterparts. Rather than spending significant resources on support and maintenance, most digital organizations can simply use those resources to create innovative customer experiences and improve their supply chain management, for example.
Compounding the effect, modern technologies like cloud computing are often much less expensive. Holding onto legacy systems can deny organizations the opportunity to use their capital in more efficient ways.
Cyber security threats
As the threat of security breaches rises, organizations must reduce the risk of their own systems becoming a target. Many legacy systems fall short of modern best practices for security—organizations must therefore consider that the potential consequences of a data breach will far outweigh the actual cost of modernizing.
According to IBM, the average cost to remediate a data breach in 2020 was $10.1 million CAD in the healthcare industry, $8.2 million in financial services and $7.1 million in technology. (Cost of a Data Breach Report 2020)
Some older technologies are impossible to protect against certain attacks, forcing companies to dedicate an immense amount of resources to setting up protective systems around them. For leaders seeking to sell the idea of digital transformation and legacy system modernization in their organization, resource utilization for these support systems and projection models for the cost of a security breach are an excellent argument in favor of modernization.
3. Legacy modernization projects take “forever”—and rarely achieve expected results
Fred Brooks, author of The Mythical Man Month once wrote, “How does a project get to be one year late?” The answer is, of course, rhetorical: “One day at a time”.
What Brooks meant was the tendency for a large number of small mistakes to accumulate, producing lengthy overall delays.
But modernization initiatives don’t have to take decades or cost billions. While every initiative is unique with varying degrees of complexity involved, leaders can dramatically reduce the time-to-value by following these key principles:
- Engage the entire organization by including both functional and technical expertise
- Find a balance between pragmatism and technical excellence
- Focus on modernizing to achieve real-world customer outcomes
- Define the relevant measures of success and work in short cycles with fast feedback loops
Consider that companies often miss out on the expected results of a digital transformation because they aren't also modernizing their ways of working. Change management and enablement must accompany new capabilities.
4. Modernization is just a side project
In some organizations, legacy modernization initiatives are viewed as just another tech upgrade. Without fully dedicated resources or defined customer outcomes. the project languishes and fails. The result is precious little to show for the effort beyond a few pilot applications that struggle to scale in real-world scenarios.
While much has been written about shifting from a project to product mindset, the tension between the “tyranny of now” and organizational complexity creates resistance to change that’s not easily scaled.
Instead, organizations should strive to continually assess and evolve their modernization strategy to address near-term customer needs. This requires plenty of collaboration between functional experts from across the business and dedicated technical teams with domain experience.
5. Our modernization plan has to be perfect
The goal of modernization should never be for its own sake. Losing sight of modernizing for near-term customer value risks placing too much focus on creating bulletproof technical solutions. A focus on technical perfection, at the expense of results, involves far too much upfront planning and navel gazing, stalling a modernization process that has already been funded and approved by executives—two major hurdles, the results of which should never be wasted.
With too heavy a focus on planning, teams can often accidentally rebuild the prior system because they focus on requirements and lose the North Star vision driving the modernization. For this reason, incremental modernization that focuses on delivering value can chart a better path than so much up-front planning.
Indeed, when modernization initiatives are planned in a lean way, with a focus on incremental improvements, organizations can reap benefits in the short term and adjust their strategy as customer expectations and market forces change.
What now?
Ultimately, your organizations must include legacy modernization as part of broader digital enterprise strategy. The results are worth the risk and investment.
Even in sensitive markets or mission-critical systems, when architectural, regulatory and business constraints are heavier, it’s possible to achieve modernization goals and maintain high levels of service, security and compliance, including ownership of the architecture.
While every modernization initiative is unique in its scope, challenges and constraints, leaders can overcome organizational inertia with a vision that focuses on creating near-term customer value and improving the efficiency of their IT units. Many modern frameworks, tools and technologies exist to help accelerate the work, plus partner consultancies like ours to help drive strategy and execution.
For more on making your digital transformation a success, check out my recent post: Planning for the future of your digital transformation? Focus on these 3 outcomes