Lack of cash is halting investment in new technology, leaving many data processing managers struggling to maintain old, complex and poorly integrated systems, according to a new report by Price Waterhouse & Co. The Financial Services Information Technology Review 1992-93 comprises a survey of a number of industries and another exclusively centred on the financial services – a market sector that is 10% more likely to cost-cut than others, Price Waterhouse discovered. Information technology managers from over 60 UK-based banks, building societies, insurance companies and stockbrokers were questioned. More than 50% of respondents expect spending to rise in the next three years on application systems and systems infrastructure presumably when the replacement of old systems becomes unavoidable. But fewer than 30% of respondents expect an increase in data processing spending; fewer than 20% expect an increase in personnel spending; and under 50% expect an increase in spending on software. The need for improved planning and cost constraint causes dilemmas for many data processing managers. Some 70% felt that lack of resources prevented them from implementing quick fixes; nearly 80% felt the age and complexity of their current systems was the biggest difficulty a problem they were simply adding to. Packaged software vendors are well placed. Full 90% of respondents said they would choose shrink-wrapped applications in the future if they could, despite the functional and technical compromises entailed, because costly custom developments had failed too often. Such results show quite a change from the 1980s philosophy of competitive edge through leading edge technology.

Reliable, proven, flexible and maintainable

In the recessionary, cost constrained 1990s the desire for reliable, proven, flexible and maintainable systems now dominates. Competitive advantage has slipped to sixth place on the list of priorities in over two thirds of cases. For the remainder, who declared it their top consideration, technological advantage was more important for front office applications – and in many cases it was better implementation of packages rather than their inherent characteristics that was seen as the best way to achieve a lead. The switch away from new technology was also reflected in the ‘open versus proprietary systems’ debate where better management rather than new systems is key. The one area where increased sophistication and quality is becoming more attractive is that of packaged software. This was put down to the need for data processing departments to provide swift, cheap and proven services. One investment bank, for example, claimed that it had implemented new back office systems built on unmodified packages. Customisation is more likely for front office systems however, to integrate packages into existing systems. It is not suprising then that flexibility came out as the most important yardstick against which software options are assessed. The technical fit of software was considered the fourth most important factor, compared with flexibility, vendor reliability and risk of failure, suggesting that problems with systems integration are likely to disappear, the report says. It also identified a growing concern about the assessment of packages – as either plug-and-go systems or as tool kits needing extensive effort to integrate them. Where customising a package was particularly difficult, some directors said they would opt for a higher cost custom system. Nearly all respondents pointed to the need for improved planning. Over 90%, for example, felt that poorly specified requirements were the biggest hurdle for achieving short term needs and said they wanted to change the data processing planning process. Three quarters cited better management of user expectations and improved assessments of change as important. Two thirds wanted radical new approaches to analysis and design. And 85% said there was a need for data processing staff to be involved early in overall businesss planning. A return to quantitive and analytical rigour

was needed, the report suggests, to help integrate data processing with marketing, operational, financial and staffing strategies.