Investigation of project failures in the software industry : an empirical research of software development companies in Malta

Abstract

Software projects are famous for their high failure rate. Literature reports failure rates of up to 70%. In the Maltese scene, projects in software development also face many problems. However due to the small market, Maltese firms do their best for not having to halt a project half way, even if this means that it will reduce profitability in the short term. The aim of this dissertation is to find what factors lead to project failure and how companies tackle projects in order to reduce failure. 11 project leaders working in local companies were interviewed. The companies were divided into two main segments: small/medium companies and large companies. The study included comparing different perceptions of which attributes have to be met for a project to be successful. Apart from the traditional four factors that a project success is judged on: Time, Cost, Scope and Quality, project leaders mentioned client satisfaction and client achieving business advantage. The project leaders were also asked to describe two cases of software project; one successful and the other a failure. From these cases, a comparison of what factors lead to success and what factors lead to failure was compiled. The most factors attributed to project success are having a good team and having good communication facilities. The client and the end user must be involved into the project and be updated frequently on the status of the project. On the other hand, the factors attributed to project failure are bad estimate, problems in scope management (wrong requirements or requirements change) and problems with skills required by the project. The research also included an investigation of how projects going out of track are tackled, where the general agreement is to keep a continuous close watch on the project to detect problems the earliest possible. Also the dissertation investigated what post mortem exercises are being carried out. While all companies agree on the usefulness of post mortem exercises, some of the companies admit of not doing them due to lack of time.