The concept of evaluation policy usually refers to "the way to do evaluations" in line with SIDA's (Swedish International Development Cooperation Agency) strategy for conducting evaluations (SIDA 1999; 2012). This way of framing evaluation policy is in my experience, the most common way to think of the concept. A strategic and written plan for how evaluations should be conducted. That is, policy is the product of authoritative decision making and the policy work contains the implementation of this decision. However, this is not the only way to view evaluation policy, and a few years back New Directions for Evaluation devoted a special issue to highlight evaluation policy. With scholars such as Mark, Cooksy, Trochim and Chelimsky (2009) arguing for the need of a "fresh look on evaluation policy" and especially the intimate relation with the evaluation practice (Trochim 2009). Although this sounds like an attempting and promising opening for a debate, these scholars have not adequately addressed the issue of evaluation policy by focusing solely on the hierarchy of policy-making, thus ignoring the vertical dimension of evaluation policy, which is highly relevant to develop a deeper understanding of the making of evaluation policy. My paper addresses the issue of making evaluation policy and with special attention to the vertical and horizontal dimension, emanating from the understanding of policy as a contested, interactive process and activity of many hands. Specifically, in my PhD-project, I will be looking at the making of evaluation policy in the education field, a frequently evaluated domain. This paper will be devoted to the discussion of the concept of evaluation policy and the different dimensions of policy, that is the vertical and horizontal aspect (Colebatch 1998) with examples of evaluation policies from the educational field. This also involves evaluation steering which affect and constitute the conditions for how the evaluation practice can evolve (Hanberger 2013; Foss Hansen 2013). Hence, this sheds light on the constant question of power in evaluation policy. I argue that evaluation policy is an important concept that needs a different or more inclusive approach emerging from the literature on policy work. It also has to be recognized that the practice of evaluation policy is not so much about implementing an already formulated evaluation policy, but rather about creating evaluation policies which is a process of social construction, of collective sense making. In conclusion, this project, by closely examining evaluation policy, sheds new light on the processes of making evaluation policy, and how this can be researched.