Flowers, S.Software failure: management failure. Amazing stories and cautionary tales. Chichester, New York, etc.: John Wiley, 1996. 197pp. ISBN 0-471-95113-7 £19.99

As long ago as 1975 Henry Lucas wrote a book called, Why information systems fail but it seems that it takes a long time for us to learn the lessons, hence this book, which looks at specific cases of information systems failure, mainly in the UK.

Flowers embarks on a task similar to that which Lucas set himself, but with the advantage of many more years in which systems failed, failed at enormous expense, and failed, in some cases, catastrophically. Readers in the UK will be familiar with such disasters as the London Ambulance Service's computerised despatch system, which after a week of intermittent use:

...slowed down and then locked up altogether. Attempts to reboot the system failed to correct the problem and, when the backup system failed to cut-in, the control room staff had no alternative but to revert to a fully manual paper-based system. (p.74)

and with the case of the Wessex Regional Information Systems Plan, which failed after spending at least £43 million and which became a national scandal that has probably put back the development of effective information systems in the National Health Service for a decade, since everyone is now terrified of becoming 'another Wessex'.

However, entertaining though these stories of failure may be, Flowers's purpose is more serious: it is to derive from the failures some general understanding of why information systems fail. In the final chapter he develops the concept of critical failure factors and shows, quite conclusively, that there is more to failure than systems failure: indeed, that the causes of failure lie as much with failures of management and project direction as with the software.

A number of factors appear to interact in allowing failure to recur:

• senior management's lack of understanding not only of the technology but also of the general systemic nature of their organizations - the lesson that technological change means organizational change and cultural change is difficult to learn;
• the pressure of time, particularly in public sector organizations, where spending the money before the end of the financial year can result in too little attention to project definition and control;
• a failure to begin the process of systems definition with an organization-wide participative process of defining the needs, identifying the existing systems and their problems, and preparing staff (and management) for change;
• a failure to invest in planning the project within the organization, and buying a 'solution' instead.

I have been working recently with two public sector organizations that are trying to overcome these problems and, at the very least, the methods employed, training workshops, participative needs definition, for example, can reveal the complexity of the issues to senior management and encourage a cautious approach to systems development. Cautious means slow, but better a slow and successful implementation than a quick fix that fails.

Prof. Tom Wilson
Publisher/Editor-in-Chief