The National Program for IT, (NPfIT), renamed “Connecting for health”, was launched in 2002, looking to improve the UK’s health service for decades to come, offering a more efficient service based on better data sharing and National Health Services (NHS) process harmonization. The program is a collection of about twelve initiatives going from a secured network to applications dedicated to managing patient histories, medical files, digital prescriptions and so on. All of these rely heavily on a detailed database containing 50 million medical records.
Initially, the work was divided among four large Systems Integrators: Accenture, CSC, Fujitsu and British Telecom. However, over the years, the program turned into a political, legal and economic nightmare, which was not in the interests of taxpayers, NHS staff or even the integrating firms; two of which left the sinking ship during the course of the project, having become frustrated by the lack of any tangible outcome. After many changes in people, governance and project scope, the situation remained turbulent until the program was officially abandoned March 31st, 2013.
In the aftermath, many of the studies carried out by both private analysts and by the government itself highlight problems that are actually pretty typical for any project. These include: vendor selection, contract agreement process, conflicting interests among project participants, lack of clarity in the definition of scope and goals, an increasing range of technical and functional issues to be addressed, no target system alignment, execution before task specifications had been defined and an organic resistance to change.
That being said, these studies do not teach us how to avoid making the same mistakes in the future. One piece of research, however, focuses directly on what went wrong. At the National Audit Office’s request, analyst firm QuinetiQ conducted a study in 2005 that quotes two key issues in regards to the ISO norm:
QuinetiQ found that these two critical activities were underestimated, lacking attention and resources in the context of a project of this scale.
I fully agree with this point of view. The ability to express the needs and to define the envisioned solution in a formal, structured way, should be critical success factors (CSF’s) for any major program, whether the project be organizational, IT or material-oriented. These CSF’s are required to synchronize the various stakeholders on a regular basis, to arbitrate between scenarios and to organize and plan for the delivery of work. In a nutshell, having a shared methodology to Stakeholder Requirements Definition and Architectural Design helps ensure progress towards a viable pragmatic solution.
Major projects, by nature, spread out over time and therefore face many environment changes. A working framework, together with reference design material, is necessary to ensure the consistency and continuity of work. In addition, such programs require the ability to adjust transformation scenarios based on technological, organizational or financial changes.
The NPfIT program is by all means out of the ordinary, both in its scale and in terms of the many setbacks it suffered. Obviously the design and definition efforts would need to be sized in light of the complexity of the initiative. However, we can be sure of two things: that a solid architecture approach would have helped overcome the many difficulties that this project encountered; and that mastering transformation is, in the end, the primary return on investment for enterprise architecture.