• silicon.com
• Technology
• Hardware

By Kate Hanaghan, 19 March 2002 16:05

NEWS IT departments are incurring huge costs for their companies by not properly testing business critical systems before they are unleashed on users. According to research commissioned by Sim Group, a software testing specialist, business managers are not getting the IT systems they think they are. It shows a vast gulf between what business managers assume is going on in the IT department and the reality. Ninety two per cent of UK business managers questioned by Sim Group assume their business critical systems are fully tested. However, only half of IT managers questioned perform the final testing procedure, User Acceptance Tests (UAT) - considered to be the most critical kind of testing before a system or application hits desktops. Bob Bartlett, managing director of Sim Group, said it doesn't make economic sense to place the final and most important piece of testing right at the end of the developmental cycle. Discovering and fixing an error at this point is 10,000 times more costly than picking up on it earlier. This developmental cycle is typical in the UK and the US. Bartlett told silicon.com: "There is little or no understanding among IT managers that testing could be completed in parallel with development." The situation is quite different in Germany, Bartlett said: "For the Germans it's quality at any price." He added: "The Germans are more efficient, taking fewer problems to the final stage." The implications of this are that IT departments in Germany aren't left with glitches to fix at the end of the cycle that could cause launch delays. Bartlett added that despite this extra attention to testing, there is no indication that German projects take longer to deliver than UK equivalents. Bartlett is also a senior CSSA councillor and has been involved with the establishment of a new special interest group to promote testing in the IT market. He concluded: "UK IT departments test not because they are committed to quality but because of the implications of what could go wrong. There is an unwillingness to enter into the mechanics of quality."