ICT-project failure in public administration: the need to include risk management in enterprise architectures

Abstract

The failure of large and complex ICT projects has gained the attention of politicians and public managers. Decision-makers are looking for ways to reduce the number of project failures. As one of the instruments, they are looking at architectures to mitigate project management risk in order to reduce ICT project failures. Enterprise Architecture (EA) has been heralded as a way to advance digital government developments. This has resulted in the development of building blocks, standards, principles, models and other architectural instruments.Despite the importance of this topic, little is known about the actual relationship between project failure and enterprise architecture. Although architectures might include risk management instruments, there is no insight in their actual use and experience. In this paper, the relationship between EA and project failure is investigated by creating a retrospective view on the use of EA in large and complex ICT-projects. A simple questionnaire is developed aimed at measuring the use of enterprise architecture in organizations and factors related to ICT project risks. During an interactive policy workshop, participants from various public organizations in the Netherlands were asked to fill in this survey for the ICT projects they were experience with. This was followed by a discussion about EA and risk management. The findings show that a disappointingly number of projects use either EA or risk management in a sufficient way. From the results, we conclude that architecture was hardly used as an instrument to mitigate project risks. This implies that one way of mitigating project failure is to ensure that the enterprise architecture and its use provide explicit attention to the subject of project risks. Six recommendations are formulated to improve risk management.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 12 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.