There is an increasing concern that information systems (IS) are not delivering anticipated value and benefits. There is a push for the development and adoption of improved evaluation metrics in an attempt to better quantify IS benefits. This has led to a growing number of well-developed methods for assessing returns. In this paper we take stock of the current situation and ask whether improvement lies not with the development of better quantitative methods, but rather by better understanding the experiences of multiple IS stakeholders. Using case material and current literature in IS/IT evaluation we draw predominantly upon the work of Heidegger and Suchman to explore the concept of IS evaluation as a highly complex social process. The analysis leads to an understanding of situated (context dependent) IS evaluation which suggests that interpretive evaluation methods may play a key role in helping practitioners and academics understand the complexity surrounding this area.