Technology has the power to change your life or ruin your day, and can sometimes do both. The possibilities offered by IT are often so alluring and the potential returns so great that many organisations are prepared to invest heavily.
Unfortunately, many of these investments are either lost or do not deliver the returns expected. This month's postponement of Japan's third generation (3G) phones, the quadrupling in cost of Railtrack's west coast link to the north, and the abandonment of the Home Office immigration IT system are just the latest additions to the Icarus list - IT projects whose ambition exceeded their ability to deliver.
Considering the length of the Icarus List - recent additions to the public sector category alone include the abandoned project to pay benefits through post offices, the flawed passport system and the cancelled MoD communications system -the wonder is not that so many fail, but that any ever succeed. But succeed they do, although often not in the way that was originally envisaged.
One major factor in the challenges faced by IT projects is the level of novelty that must be managed. Many IT projects are highly complex, high-risk ventures that aim to deliver innovative outcomes by employing novel clusters of technologies in new and often evolving organisational contexts. Figures published by the Standish Group indicate that only 28% are successful, the remainder being either abandoned, delivered late, over- budget, or with lower functionality that envisaged. With those kind of figures, embarking on a major IT project is not just a triumph of hope over experience, but also very good business for lawyers.
However, the current "joined-up government" initiative depends fundamentally upon the ability of the public sector to deliver both profound IT-enabled organisational change and a large number of new, complex and interlocking IT systems. The mismatch between the ambition and reality was a key factor in the adoption of a more rigorous and comprehensive approach to IT procurement that is now in place. This approach, outlined in the recent McCartney report, is a key part of a broad reform of public sector procurement that is overseen by the Office for Government Commerce. The systems and procedures now appear to be in place to prevent further public sector IT disasters and improve overall performance.
The Organisation for Economic Co-operation and Development (OECD) has launched a project aimed at creating a better understanding of the issues affecting public sector IT projects.
But although the UK is struggling, there is little evidence that other govern ments are doing much better. For example, key elements of the software for the Bibliotheque Nationale de France were delivered nearly two years late with a budget overshoot of more than 40%. Even when it did go live, the problems were such that staff walked out, complaining of verbal abuse from readers waiting for books. The Tress 90 unemployment benefit system within the Norwegian Ministry of Labour shared the problems of poor project scoping and management, timetable slippage and escalating costs common to many public sector IT projects.
The challenges facing governments in delivering successful IT-enabled change are much the same as those faced by all organisations involved in IT projects, albeit in a more acute form. However, the very open nature of public sector IT failures diverts attention from the private sector IT projects suffering many of the same outcomes. The challenge common to both sectors is to manage the conception and delivery of the IT systems that will underpin the dot.com, dot.gov world, systems whose complexity is increasing dramatically.
These characteristics are not unique to IT but are found across a range of hi-tech industries. Such highly complex systems, and the challenges faced in their creation, together form an unprecented new class of product. They may be the technological wonders of our time, but there is much to be understood about the process of their creation.
The work of improving the understanding of such highly sophisticated products is taking place at all levels. At the industry level, collaborative work with research groups is delivering insights that have significant implications for critical activities across the development lifecycle. In contrast, the Human Centred Process Improvement Group, a collaboration of practitioners from a range of organisations, is taking an end-user perspective on the effectiveness of systems.
The strength of such work lies in the diversity of the industrial collaborators and the multidisciplinary nature of its approach, enabling important research findings to play a pivotal role in the development of best practice. But while such collaboration is making an important contribution to many of the UK's leading companies, they are not necessarily learning the same things, or applying them in the same way. Nor should they, since the needs of industries differ dramatically. Which brings us back to IT.
The IT industry, more specifically the bit that develops bespoke software such as the Home Office Immigration System, still has a lot of growing up to do. The bespoke software business and its clients are, consciously or not, evolving towards a state in which both the ends and the means of applying technology to organisational processes will be understood more clearly. However, as the growing Icarus List shows, there is still some way to go.
•: Steve Flowers shf@brighton.ac.uk is a principal fellow in the ESRC Complex Product Systems Innovation Centre, University of Brighton
