Digital transformation (DX) continues to be a hot topic in government. But the transformation efforts of many gov CIOs quickly reach a fork in the road. That means a key choice must be made upfront, before fundamental transformation can get underway.
The fork is pretty basic. Should an organization stay on one path, continuing with its legacy systems? Or should IT planners take a different path and start over again — with a so-called "greenfield" initiative?
The choice may be basic, but the consequences loom large.
I believe that going greenfield is often a better choice, given the way Third Platform initiatives have changed the nature of IT sourcing. Third platform technologies also have helped cut solution costs while boosting long-term IT flexibility. So, in theory, starting over gives organizations a chance to build their system "the right way" this time.
But it's not fair to say the greenfield approach is always the best choice. Migration costs can be steep, especially if an agency is moving away from a highly customized system. And many government systems can have a range of dependencies, such as shared outputs, communal data stores, etc. More importantly, many of today's legacy government applications include business and workflow intelligence gathered during many years of code evolution. So DX efforts may consist of connecting new business processes and information delivery channels to the legacy system. In such cases, the transformation effort mostly means helping the legacy system become more easily adaptable and supportive of new business purposes across the enterprise.
I ran the question by an IDC colleague, Al Gillen, who said "digital transformation is not really possible unless you either replace legacy systems or incorporate them into the DX efforts." He said the way incorporation would work is to “wrapper” the legacy systems with APIs that allow modern apps to get data into and out of the legacy systems. It's not an unprecedented approach. Many years ago, when GUIs came along, legacy apps couldn’t integrate directly with that interface. Al pointed out that programmers developed methods to bridge that gap to allow older mainframe systems to continue functioning in the client/server world.
Whether it's ultimately chosen or not, it's fair to state that a greenfield option should always be part of a DX discussion. And to make this a well-informed discussion, a return on investment (ROI) calculation should be done for both the proposed legacy system improvements and the potential new greenfield approach. Long-term costs, including end of life-cycle expenses, must be calculated, along with staff costs, software licensing, network improvements, and a range of other variables.
Only by understanding the true costs for both a greenfield design and a legacy transformation approach, can the best choice be made.
That's why a great deal of due diligence must be done before any DX effort can be tackled in a meaningful way. For this to happen, CIOs may need to be in lockstep with agency CFOs to be sure that the right things are being measured, and that informed choices are being made. CIOs should take the time to explain to CFOs the realities of platform dependence for some solutions and the cost and effort needed for migrating legacy systems, vs taking a greenfield approach.
The End-game for DX
On a fundamental level, DX is about saving money, and the money issue encompasses time, information accuracy, system reliability, and the ability to leverage changes in a way that will have a significant on impact for citizens.
When people talk about DX today they usually are referencing a variety of ways to accelerate the transformation of business activities and associated IT processes. But here's a key issue: Virtually no organizations building a brand-new application today will choose to build it the same way they would have a few years ago. Today, agencies can spin up a cloud-based virtualized server at very low cost, and tap into key middleware solutions without having to license and install the software on site.
Government IT managers are well aware that the days of $30 million upgrades for legacy systems every few years is a thing of the past. In many cases, brand new solutions developed with modern tools in an externally hosted environment can be built cheaper. That's why greenfield must always be an option when upgrades are discussed.
Government agencies can look to the Portfolio State Model to help them make decisions about when to upgrade systems, and why. We don't hear as much about Portfolio stat as when it was launched a few years ago, but the tools and the process it offers remain viable.
Finally, pay close attention to these documents: OMB M-12-10, M-13-09 and M-14-08. These resources help highlight the IT portfolio management process, using data and analytics to make decisions on eliminating duplication, choosing the best long-term system migration path, and working with CIO-led capital planning and investment control processes.
Only by preparing for the fork in the road, and doing the right kind or research to help support clear choices, can true DX be achieved for government IT systems, whether it's greenfield or not.