Software conception or inception?
INDEX | ORGANISATION | TIME LINE | PEOPLE | STORY | RESEARCH METHOD | COMMENTARY | FINDINGS | HELP | BOTTOM
This section considers the decision to build an internationally distributed database system for the Multinational aspects of X's business. It asks the question: were the foundations for this software development based upon rational decisions reflected in the IT Strategy and Requirements Study or were there other influences?
So, why did X-Group decide to build a system for the Multinational Operations element of the business? What was the scope of the system and how was this established? In particular was the system designed to meet the operational needs of Services and underwriters or management information needs? Why did they decide to adopt a client-server architecture based on PCs and an internationally distributed database? Why did they decide to adopt a prototyping approach for phase I but a structured approach in phase III? Why did they choose to implement initially in the UK, Holland and the States and how was the division of work into phases arrived at?
C.1.1 Why build a system for Multinational?
C.1.2 What affected the choice of system boundary?
C.1.3 What affected the choice of systems architecture?
C.1.4 What affected the choice of process model?
C.1.1 Why build a system for Multinational?
The formation of X as a separate company and the arrival of Tim as MD provided the impetus needed for X to develop its own system and the IT Strategy really just reinforced this decision. Market considerations and the inadequacies of existing information systems indicated the need for a new system for Multinational even if this was difficult to justify in cost benefit terms. However the nature of the business, with special requirements for each large client, and the attitude of senior management explains why there had been no previous successful large scale computerisation of this area of X-Group's business.
There was a view that the IT Strategy study was just a formality to get the money. In support of this, the strategy looks very like Colin's existing vision, the points in the strategy about data accuracy were not heeded and when the strategy document became out of date it was not reviewed.
As with many management information systems, it was actually difficult to justify the system on cost benefit grounds. The projected cost of the system was less than the notional cost of buying the service from XUK but the benefits were hard to quantify especially when compared with the cost benefit analysis for the Marine system. In practice, partly because of the Multinational system, there were significant increases in income without increases in staff.
Tim, the MD for the new company was a driving force for computerisation. He got approval from X-Group for the expenditure despite financial constraints and problems with demonstrating benefit.
The creation of X to focus on underwriting worldwide insurance risks for multinational customers meant there was a need to draw together information from around the world. A system for Multinational was regarded as critical to getting the business going.
The overall information systems for X were poor and prevented X from achieving its market potential or managing the business properly; the quality of overall management information for X actually went down when X was formed. X depended for its IT on local operating companies (ie XUK in the UK) but the nature of X's business was quite different from XUK and the UK mainframe system was inappropriate for Multinational business. New systems would allow X to assume more direct control over its own IT direction.
Where the mainframe systems were inadequate 'home built' PC systems were used. The IT Strategy study expressed concern that these business critical systems had not been subject to 'conventional system disciplines'. The Q&A system was such a system but it was in use for well over three years and was not overly criticised by its users although it was limited by its technical architecture.
C.1.1d Market considerations
|
Domain
 |
 |
In 1991 the international insurance market was well defined with few players and competition was on the basis of quality. This and the difficult state of the market meant that IT was critical in meeting the competition when establishing a new company such as X.
C.1.1e Nature of business
|
Domain
 |
 |
The complexity and one-off nature of the business mitigated against computerisation of Multinational. Multinational business consisted of a small number of long-term complex cases (about 6000 policies worldwide), processed in many countries with service a key factor. As a consequence, information had to be tailored to each client's requirements making automation difficult.
Despite a strong case from both market considerations and the incompleteness of existing information systems, the drive to computerise Multinational Business at management level was not particularly strong. There was an attitude that getting the business was more important than controlling the business and so senior management could not see the need for a computer system. Perhaps it was the one-off nature of X's business, which made it different from the rest of the X-Group, that had allowed this attitude to persist.
In contrast, Keith and before him another underwriter, felt so strongly that computerisation would enable them to do their jobs better that they developed the first two successful Multinational systems despite the fact that they were neither experienced software developers nor was it within the remit of their job. One of the factors that influenced Keith was his previous experience of computerised offices and the technical management job he had had in Indonesia. Despite an absence of organisational commitment, Keith continued to develop and enhance the Q&A system beyond his immediate requirements. Many of the underwriters used the Q&A system but the management reports produced from it were ignored.
Top
C.1.2 What affected the choice of system boundary?
The original focus for the system as presented in the IT Strategy was for management information for Multinational but the Requirements study extended the system boundary to include much of the functionality of the existing operational system, Q&A. Unfortunately the operational needs of Services were not considered, perhaps because this group had only just been set up. It was not until the system had been in use for over a year that this deficiency was addressed. The scale of the system varied according to who was driving the vision. Under George's influence, the Multinational system became key to the success of X despite the serious problems with the use of the system highlighted in the Post Implementation Review.
C.1.2a IT Strategy and Requirements studies
|
Method
 |
 |
The IT Strategy report recognised the need for both operational and MIS type systems and suggested that in the short term the Q&A system could continue to be used for data entry. Three systems were identified for immediate work: Financial Consolidation and Reporting, Multinational and Non-Marine Reinsurance. However the Requirements study does not appear to have been constrained by the recommendations in the Strategy and the resulting requirements were larger and included reinsurance and much of the MIS. The Q&A system was replaced but this does not appear to have been made explicit and not all features of the Q&A system were included in the requirements for the new system.
C.1.2b Development of Multinational Services Unit
|
Organisation
 |
 |
The Services unit for Multinational had only just been established when the Requirements study was undertaken and prior to that the underwriters had done all their own data processing. This was probably why the operational needs of Services were not really addressed in the development of the system. It is unlikely that the different location of Services was a barrier because X-Group were very used to Services operating at physically different locations.
The actual implementation of phases I and II highlighted the limitations of the system for Services and in phase III an exercise was undertaken to decide how best to improve the efficiency of their work.
The scale of the Multinational system and its relationship with the mainframe systems of other companies in X-Group varied through the life of the project mainly as a result of who was providing the vision of the system. At one end of the spectrum, the Multinational system was a core system for X replacing existing mainframe systems making X more independent of the X-Group; at the other extreme, the Multinational system was just a management information system drawing data from the mainframe systems.
Initially Colin in the strategy work favoured the replacement of mainframe functionality but when Gordan took over he proposed drawing data from existing mainframe systems.
With George's arrival in phase II the vision for the system expanded; IT was taken much more seriously in X, leading to a long term aim of seriously reducing X's dependence on mainframe systems of the X-Group. So, while under Gordan's influence, the Financial Accounting sub-system was a fairly simple extension for Holland because there was no business benefit in replacing the accounting systems in the UK and US, a year later, the Financial Accounting sub-system was being designed to be installed in the UK.
C.1.2e Post Implementation Review
|
Method
 |
 |
The Post Implementation Review had a curious impact. Instead of the senior management saying we want less to do with IT and Gordan which is what might rationally have been concluded from the report, they promoted Gordan and decided that they wanted more IT and to become independent of the XUK mainframe systems.
By mid 1994, X had changed into a more serious established company and with this came an acceptance that IT was key and a recognition that success with the Multinational system could make the company number 2 in the world. IT had become both a driving and supportive force within X as the systems raised people's expectations of what could be achieved.
Top
C.1.3 What affected the choice of systems architecture?
With the benefit of hindsight, a client-server architecture seems entirely appropriate but at the time (1991-92) this technology was not so prevalent. One of the driving forces seems to have been the choice between the PC and the mainframe. The mainframe solution represented a loss of control by X to XUK, while the PC meant a quick system, owned by the business and developed by GIS. However technically inappropriate the PC solution, it would have been attractive to X, particularly to Tim who was trying to establish X as a new business independent from XUK.
The key decision to use a PC rather than Unix-based server for phases I and II was influenced by the increasing power of PCs, the team's familiarity with PCs and the relative costs, as use of a PC server allowed phase I to go in under budget.
The nature of X, operating in a global market out of many locations, favoured a distributed database solution. The change implicit in a new company was one of the reasons for choosing an open systems architecture.
C.1.3a IT Strategy and Requirements studies
|
Method
 |
 |
The original shape of the system using a PC LAN appears in the very first part of the strategy. The decision to use a relational database and fourth generation tool came, at least in part, from Tim but the choice of tool was fully investigated in parallel with the work on the Requirements study. This work involved careful investigations into hardware, communications, database and development tool.
The PC LAN model originated from Colin during the early work on the strategy study. Prior to the formation of X, XUK was working on a mainframe-based solution for Multinational Business. Tim is reputed to have said 'none of this mainframe nonsense we need a dedicated PC system'. The decision to stop the mainframe development also encouraged GIS to take a different approach and consider using a fourth generation language.
On X's part there was a great desire to go for a 'dedicated PC system' possibly as a backlash against the 'nice IBM mainframe' or because they saw it as a way of getting control of their computing. Viewed dispassionately, the decision between a mainframe and a PC should depend upon the power needed for the problem (RAM and mips), but there was a belief that a mainframe would take longer (the earlier requirements study took one year) and would have difficulty coping with the complexities of multinational insurance whereas a PC would be fast, flexible and owned by the business.
The team chose client server architecture because they believed it would enable them to build something quickly to compete in the market and that use of PC LANs would allow incremental development at reasonable cost. The influence of PC technology is also reflected in the prominence given to the GUI interface in the business objectives part of the IT Strategy study. One of the outcomes of the project was that it introduced the concept of a PC network infrastructure into X offices.
The development team was probably influenced by a move in GIS away from packages back to building systems using the new powerful PC-based development systems. Although a mainframe own build approach was rejected during the IT Strategy because it was too expensive, they adopted an own build client server solution even though there was no evidence that it would be cheaper. The development technology facilitated the use of prototyping, but also made prototyping more necessary because the developers did not know its capabilities.
The Multinational system was developed at a time when the power of PCs was developing fast. The increase in power was such that they did not have to use Unix for the server in phase I as planned but did have problems running graphical applications on the client even though by this stage they were using 486 chips instead of 386 chips as planned.
One of the problems with early attempts at computerisation was that the XUK computing people were used to the XUK way of doing things (high volume, low value cases) which did not match the cases handled by X. GIS (Group Information Services) had wider experiences because they serviced a number of X-Group companies. In terms of architecture, GIS favoured PC based systems because they serviced the smaller companies in X-Group while XUK preferred mainframes because they were used to large operations.
The choice of a PC based LAN rather than Unix was initially a temporary measure while they continued to evaluate the Unix options but they were already familiar with PC technology and at least one of the key programmers did not like Unix.
Unix was also expensive (almost half the original hardware/software budget for phase I) and if they had used Unix for phase I it would not have come in under budget.
C.1.3g X needed to operate as a global company
|
Domain
 |
 |
The system architecture had to support the movement of information between offices and countries. A Multinational programme typically involved policies in many countries each reflecting local conditions, local collection of premiums and claims submitted world-wide. All this information had to be consolidated in the issuing country. Operating internationally also involved mixed currencies and national financial regulations. This was a fairly specialised market so there was not an off-the-shelf package and X would not have wanted to use a competitor's system because they hoped to use IT to get ahead of the competition.
To handle these requirements the system consisted of distributed databases with each database containing information on programmes controlled locally and business serviced locally. Initially information was transmitted electronically overnight but following the installation of the Unix servers they planned to use the Sybase Replication technology. The Multinational system thus had to link to a wide variety of local systems and operate within different IT cultures.
Not only did X operate out of offices in many different countries, the organisation was dispersed within a country. Despite the inconvenience from an IT perspective, the global and widely dispersed nature of X was one of its strengths. Moreover the organisation was used to working remotely and transmitting information between physical locations and the autonomy of many of the local offices was conducive to a distributed system.
C.1.3i Allowing for change
|
Change
 |
 |
That change was part of life for X and that this would affect IT solutions was recognised in the IT Strategy study. While there was not much evidence that this influenced the way the system was built, it certainly was an issue in the choice of system architecture and was one of the reasons for adopting an "open systems" product such as Sybase. The resulting system seems to have been resilient to change.
Top
C.1.4 What affected the choice of process model?
GIS had little experience of using rapid development approaches and was naturally risk-averse but despite this the Requirements study established the evolutionary prototyping approach that was used in phase I. The success of the Requirements study using a small committed team and the personal relationship between Gordan and David contributed to this decision. The Requirements study was also responsible for the ambitious decision to install in three countries in phase I based on business demands not feasibility.
Full evolutionary prototyping was not proposed for the Financial Accounting sub-system because of its complexity but the appointment of Colin as project leader led to a very traditional approach to development. The appointment of a methods expert as Multinational project manager and the organisational reaction from the technical problems of phase II, resulted in them adopting a more traditional approach to the whole of phase III.
C.1.4a GIS development approaches
|
Method
 |
 |
GIS had a development method, which included traditional, structured and rapid development approaches, but this was regarded as outmoded in some areas and projects were expected to develop their own procedures. Multinational was regarded as a modern forward-looking project and the team chose to use rapid development techniques which had not been used by GIS for a major external project. This decision seems to be inconsistent with GIS's reported fear of failure as prototyping was quite new for its time and would not have been regarded as risk free.
C.1.4b IT Strategy and Requirements studies
|
Method
 |
 |
The IT Strategy set the methodology, an aggressive timescale and costs but these were based on producing a different kind of system 'using existing and off the shelf IT components'. It was the Requirements study which established that evolutionary prototyping would be used for the development. The study also divided development into two phases and decided that UK, Holland and USA would be included in the first phase. The size of the system seems to have been driven by a desire to 'get as many things that seemed important into the first stage' and data dependencies were not realistically considered.
The IT Strategy also correctly predicted the importance of data accuracy and the need for the right operational procedures and warned that 'significant management effort and commitment' would be needed to address these issues.
The decision to include UK, Holland and the US in the first phase was based on the fact that these were key countries in X's business and did not take account of the huge cultural differences in the way the business was conducted in America compared with the UK and Holland.
The Requirements study was completed in a remarkably short time. Key to this was a small team of committed and enthusiastic individuals. In choosing the development approach they would have wanted to match this success.
One of the key factors in planning the Multinational project was that it should be managed as an equal partnership between the business and IT people. This allowed the business to have ownership of the project but avoided the problems there can be in handling the technical aspects of a project which is managed solely by the business. A senior person, David, was seconded full-time from the business side who had the additional advantage of having been involved in the previously failed attempt at computerisation by XUK. Senior people were also involved on a part-time basis from Holland and the USA. One reason for seconding someone as senior as David was that it demonstrated organisational commitment. One of the key reasons for continuing the partnership during the development of phase I was the very good, equal personal relationship between Gordan and David.
The complexity of the Financial Accounting sub-system was why it was initially decided to use a more traditional development methodology with throwaway prototyping. There were some difficulties in managing this different approach when it interfaced intimately with the rest of the Multinational system, which was still being developed using a more flexible approach. It was also difficult to manage user expectations.
When Colin was appointed as manager of the Financial Accounting sub-system of Multinational he seemed to be determined to take a very traditional approach to project management introducing all the conventions of a traditional life cycle (such as signing off specifications, programming to a module spec). Although the Financial Accounting sub-system was running behind schedule, I feel this change had more to do with personalities than just the nature of the problem.
Although Jane was not recruited because of her strong methods background it was very influential in her subsequent management of the project. It was a key factor in the decision to introduce a more structured approach to both method and team organisation in phase III.
The change in process model seemed more organised and formal but Gordan summed up the impact as 'It's all traditional, sequential, boring.'. Unfortunately I did not continue to follow the project for long enough to observe the impact the change in methodology had in practice.
C.1.4h Organisational experience of development
|
Organisation
 |
 |
The development method in phase II seems to have been less well defined. I think they just continued to develop in the same way but without the prototyping workshops to give the process structure and a timetable. A contributing factor to the problems with phase II was the change in emphasis in development strategy that was not recognised clearly, or early, enough by the development team.
The Post Implementation Review of phase I recommended a more throwaway approach to prototyping during phase II. Gordan was still enthusiastic about the evolutionary approach but the analysis for the extensions to phase II was undertaken as a series of mini studies with limited user consultation. By phase III a more traditional development approach was being used, not necessarily because it was more appropriate, but because their fingers had been burnt in phase II and so they could not take chances.
Top
INDEX | ORGANISATION | TIME LINE | PEOPLE | STORY | RESEARCH METHOD | COMMENTARY | FINDINGS | HELP | TOP
© Clare Tagg 2000