The evaluation of information systems and health information systems in particular, is recognised as a complex and challenging activity. There is no agreement on a 'best way' to evaluate, on what and how to evaluate, who to involve and within what paradigm to proceed. This paper argues that evaluation is a situated activity that requires a more interpretative approach, broadly based on hermeneutic principles. The paper reviews a number of guidelines and frameworks for evaluation presented in the medical and information systems literature, focusing on two related approaches: the context, process and content framework, and structure, process and outcome framework. A synthesis of these were used in the case study described - the evaluation of an intranet pilot deployment within primary care. The literature review and research experience (including the case study) form the basis for a set of general recommendations presented in the conclusion section.