Supporting Data Selection for Decision Support Systems: Towards a Decision-Making Data Value Taxonomy

Data are ubiquitous and generate unprecedented opportunities to help making decisions within organizations. This has led so-called data-driven Decision Support Systems (DSS) to become critical, if not vital, systems for most companies. The design of such DSS raises important methodological challenges, since data-driven DSS should expose only useful information to decision makers, but data available in a company's database are numerous and not equally supportive. Failing to provide the right data to the right decision-maker may reduce the usefulness of a DSS, and can lead to lower quality decision outputs. This is particularly striking in the case of Self-Service Business Intelligence (SSBI) where users build DSS outputs themselves. In this paper, we elaborate on this idea of data profusion and propose a data selection criterion, namely the decision-making data value. To do this, we discuss the concept of value and its application to data and decision making, we review existing literature and propose a taxonomy of the dimensions of data value in the context of decision making. We also validate this taxonomy with semi-direct interviews and discuss the future research we plan to conduct as a way to apply this approach for the specification of high-value data-driven DSS.