軟體專案的高失敗率，一直是軟體業的夢魘。為了解決這個棘手的問題，1968年北大西洋公約組織 (NATO) 提出軟體工程的概念，發展至今，已逐漸變成一門學科，軟體開發的成功率卻沒有顯著的提升。軟體工程讓軟體開發變得有組織性、秩序，但是，軟體開發除了規範流程外，還有一些內外部因素，是值得我們探討。目前描述軟體開發案例的文獻太多是以正面的角度出發，分享成功的經驗，相反的，失敗的分享卻屈指可數，容易造成開發團隊依據「最佳實作 (Best Practices)」進行開發流程，而忽略了要檢視其專案的弱點和降低風險。本研究個案為一開發客製化軟體的團隊，最後專案以失敗收場。本研究藉由文獻輔以蒐集內部資料和與團隊成員進行訪談，經由分析、歸納後整理出失敗原因並提出相關建議，避免其他團隊再重蹈覆轍。且研究個案使用的開發方法為，目前在國外大廠被廣泛使用的敏捷方法，但在台灣尚屬萌芽階段，藉由失敗經驗的分享，希望能協助敏捷團隊提高成功率，也希望能對台灣的敏捷開發方法能達到拋磚引玉的效果。The high failure rate of software projects has always been a major problem in the industry. In order to resolve this sticky issue, the North Atlantic Treaty Organization (NATO) first raised the term “software engineering” in 1968. Although software engineering has become a popular academic research topic since then, the success rate of software projects has not been dramatically improved. Software engineering is supposed to provide a disciplined and systematic approach to the development of software. Meanwhile, there exist other factors that are worth investigating.Literatures addressing the success of software development projects are abundant, but very few focus empirically project failure. Consequently, software development teams usually focuses on “Best Practices” without paying attentions to the threats of the project, thus fails to cut down the risks.This case study looks into a failed software customization project. The project team adopted the Agile methodology for this particular project. Although Agile has been widely employed in the west, it is still in its embryonic stage in Taiwan.It is a case study based on information collected through internal documents and in depth interviews with project team member. The goal of this study is to gain more practical in sights into the reasons for failure. Recommendations are made to avoid repeating the same pitfalls. We hope this research not only helps agile teams to increase the success rate, but also help to make Agile a popular method in software development.
