An Empirical Analysis of Software Productivity

The aim of our research is to discover what factors impact software project productivity (measured as function points per hour) using real world data. Within this overall goal we also compare productivity between different business sectors and project types. We analysed a data set of almost 700 projects that have been collected by STTF from a number of Finnish companies since 1978. These projects are quite diverse type (new and maintenance projects), in terms of size (6 to over 5000 function points), effort (55 to over 60000 person hours), application domain and implementation technology. There are three main findings. First productivity varies enormously between projects. Second, project type has limited influence on productivity. Third, application domain or business area has a major impact upon productivity. Because this data set is not a random sample generalisation is somewhat problematic, we hope that it contributes to an overall body of knowledge about software productivity and thereby facilitates the construction of a bigger picture.