Data warehouse projects bring together different communities of practice, with the primary objective

of producing one body of information which is capable of comparative advantages in business

analysis. Due to the number of involved communities and the complexity of their collaboration, data

warehouse projects are costly. In this paper we give a closer look at communication problems on

boundaries between participating data warehouse projects’ communities. Our analysis enlightens the

potential relation between the early creation of language communities of the involved communities

and lowering data warehouse project development costs. As today, there is hardly any methodology

available for analyzing and aligning mutual understanding between data warehouse project

participants. In this paper, we propose a data warehouse development scheme for project

improvement based on our discussion as a first step in a design science project.