Brodbeck (2001) stated in his article titled Communication and performance in software development projects that “in summary, there is insufficient empirical evidence to support either view of internal communication as detrimental, facilitative or neutral to software development (SD) project performance. In fact the evidence is contradictory.”
The above statement and finding is very surprising. Communication has been known to improve team work and clarify issues where existing information is insufficient or where there is a challenging task. Also it has been variously recommended that communication is relevant in IT due to the fast changing environment and high turnover of software and technologies.
Another strange finding is that in the early stage of software development that communication seems not to make any difference and that higher levels of communication appear to benefit SD in late stages. As a project manager, during the early stages of a project development both lateral and vertical communication is paramount for the success of the project. The required communication involves discussions and meetings within group as well as feedback to management. The early stages of a project involves initiation and development of the project charter which entails communications with various stake holders. In reality internal communication is important at all stages of SD
The result further supports the hypothesis that communication facilitates SD project functioning when the degree of standardization of method and tools is low. High standardization methods and tools communication seem not to make any difference in SD performance. According to Brodbeck (2001), this outcome contrasts with Keller’s (1994) study that the more general class of R&D projects shows detrimental effects of internal communication when task routine is high and facilitative effects when task routine is low.
In support of Keller’s findings, it has become a rational basis in modern business culture for designing organizations to determine the level of certainty or uncertainty of the organization and the information process. When the task is routine or the environment is stable (task routine high) then minimum amount information processing and subunit interaction is expected. A task can therefore be completed through the employment of a largely hierarchical structure, standard operating procedure and a formal information system. (Tushman & Nadler, 1977). However, where the environment is unstable (task routine low) or the level of uncertainty facing an organization is high, Tushman and Nadler (1977) suggest that the organization should adopt an organic system characterized by stratified structure, a lateral system and greater peer responsibility due to its effectiveness. There should be quick information exchange and faster communication amongst work teams.
It is also surprising for Brodbeck (2001) to state that user participation is quite strongly associated with reduced SD project performance and that it burdens the ongoing SD process with additional tasks. User requirement analysis is required when designing a software project. User input through beta testing and user testing has been known to enrich the whole software development process and makes for easy deployment and acceptance when the software eventually goes life.