Studies of the transition process remind us of Stiglitz's comparison between pathology and the economic analysis of institutions (1989): asking what went wrong and what did not is the essence of understanding the functioning of social systems. Thus, carefully conducted empirical studies of the transition process itself may yield elucidative results applicable not only to theory but also to institutional policy changes in transition. This paper presents the key findings of KATO, a comprehensive research project focussing on the transition process of agriculture in Central and Eastern Europe. From 1997 to 2000, the project empirically examined processes of liberalization and market development, privatization and property rights changes, as well as restructuring and path dependencies in three Central and Eastern Europe countries: Poland, the Czech Republic and Bulgaria. Conclusions on three analytical levels are drawn: (1) the empirical design for analyzing rapidly changing and evolving institutions in transition economies, (2) the suitability of different theoretical approaches for understanding transition, and (3) policy recommendations targeting better governance and an improved institutional framework.