Impact of classes of development coordination tools on software development performance: A multinational empirical study

  • Authors:

  • Amrit Tiwana

  • Affiliations:

  • Iowa State University, Ames, IA

  • Venue:
  • ACM Transactions on Software Engineering and Methodology (TOSEM)
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Although a diverse variety of software development coordination tools are widely used in practice, considerable debate surrounds their impact on software development performance. No large-scale field research has systematically examined their impact on software development performance. This paper reports the results of a multinational field study of software projects in 209 software development organizations to empirically examine the influence of six key classes of development coordination tools on the efficiency (reduction of development rework, budget compliance) and effectiveness (defect reduction) of software development performance. Based on an in-depth field study, the article conceptualizes six holistic classes of development coordination tools. The results provide nuanced insights—some counter to prevailing beliefs—into the relationships between the use of various classes of development coordination tools and software development performance. The overarching finding is that the performance benefits of development coordination tools are contingent on the salient types of novelty in a project. The dimension of development performance—efficiency or effectiveness—that each class of tools is associated with varies systematically with whether a project involves conceptual novelty, process novelty, multidimensional novelty (both process and conceptual novelty), or neither. Another noteworthy insight is that the use of some classes of tools introduces an efficiency-effectiveness tradeoff. Collectively, the findings are among the first to offer empirical support for the varied performance impacts of various classes of development coordination tools and have important implications for software development practice. The paper also identifies several promising areas for future research.