3 October 2005 Two thirds of IT managers believe that too many software projects are finalised before they have been sufficiently tested, according to new research.

A survey of IT and business managers based in the UK, The Netherlands and Sweden, found that 89% of respondents had experienced significant difficulties with systems they had implemented as little as 48 hours after going live.

But despite these recognised problems, the research indicates that inadequate testing is commonplace. The survey also showed that while the lack of resources was identified as a prime cause for insufficient application testing, there were also concerns about the availability of skills to carry out the tests.

The overwhelming majority of respondents saw failed or troublesome implementations as the responsibility of IT directors and the software suppliers – very few recognised the role of business departments in ensuring a successful roll out.

This reflects a failure on the part of business departments to recognise the damage to competitive advantage, brand and market value a poorly tested software project can cause, said Alex Garrick, general testing manager at IT service company Logica CMG, which funded the research.

“Software testing is not properly aligned with the needs of the business,” he added.