In the last decade, public sector agencies and departments have increasingly shifted their IT developmental approach from the traditional waterfall to agile/iterative methodologies, most commonly, scrum. While such moves were heralded with fanfare, challenges remain.
The public sector seeks to mimic the private sector, particularly Silicon Valley, to achieve transformation and product success, and while numerous successful examples of private-sector emulation have occurred, some ideas remain elusive. One such example is the concept of the minimal viable product, or MVP, which gained popularity after entrepreneur and author Eric Ries described it in his book, The Lean Startup.
The MVP is simply defined as a version of a product of which a segment of its user base finds valuable. It allows the product team to collect the maximum amount of validated learning about its customers to support an effort to enhance the product using the least amount of effort. The need for an MVP becomes even more imperative based on findings such as the Standish’s research and more recent estimates which indicate 20% of software features are used frequently while the remaining 80% of features are used infrequently or never. The MVP is meant to eliminate waste—save money and time that would otherwise be spent on fruitless ideas. However, agencies typically struggle with defining and implementing the MVP due to their unique situations and challenges.
There are multiple factors contributing to the challenge of defining, developing and deploying an MVP. Most agencies today are seeking to modernize their legacy systems (often mainframe systems) using the latest processes and technologies available. This results in an intense pressure to define an MVP for a product already used by their constituents. Such scenarios result in agencies defining their MVP as “parity” with their old system, but such an approach can be at odds with the true spirit of the MVP. It also prevents the agency from realizing the full benefits of modernization, such as cutting-edge technologies and improved customer experience, which lowers return on investment. Furthermore, such a “lift and shift” approach prompts rebuilding of existing waste and unused features.
A different, more effective approach is to think of the modernization as a new product altogether, while keeping existing challenges in mind to resolve them. Identify the most common path that a key user segment would take—a student borrower with parental support, a patent examiner’s common case types—that helps to define the MVP. Development can focus on these key processes while less common cases continue to run on the legacy system until they can be included in the new product, allowing the largest user group to start using the new system first—a win in and of itself. This Build-Measure-Learn, or BML, iterative feedback loop of the lean methodology can be leveraged to glean the benefits of implementing an MVP and helps developers jumpstart the fine-tuning of the system.
The BML cycle advises incrementally building the MVP while it is being tested by users to continually collect data to measure the most important assets of the product. This data is used to improve the system, which allows developers to identify and fix bugs and increase the value and usability of the system. With an increased focus on customer experience and usability across the government, this approach provides a perfect opportunity to understand usage and obtain feedback from frequent users that could significantly decrease helpdesk calls leading to cost reduction and pleased constituents.
Using the BML approach makes it easier to implement the remaining requirements as developers are continually learning user needs and pain points to build better while incrementally reducing legacy maintenance costs. This also develops buy-in for the new system before the old system is entirely replaced, leading to higher user acceptance over time.
Public sector organizations should leverage the key advantages of implementing a MVP: allowing organizations to start small and iteratively build up to produce a better, more polished product that satisfies constituents and takes advantage of user intelligence and experience to make the best decisions to build the right, fit-for-purpose product while reducing waste.
Khyati Nayak is a senior manager for digital transformation and management with Grant Thornton Public Sector LLC.