Research on intergovernmental organizations (IGOs) has progressed considerably over the past twenty years. Much of this progress was enabled by what has become one of the most widely used datasets on IGOs, the Correlates of War (COW) Project’s data on IGOs. Like most large-n datasets, the COW IGO data is not without limitations. It includes organizations that do not meet the requirements for inclusion as cases of IGOs in light of the dataset’s stipulated criteria. In addition, basic IGO characteristics, such as start and end dates, as well as whether IGOs are independent of other organizations, and thus unique observations, are often coded inconsistently and in ways not supported by available evidence. These inaccuracies introduce measurement error and potential bias in analyses that use these data to generate dependent and independent variables of interest, such as IGO creation, state membership in IGOs, and co-membership of two states in IGOs. In this paper, we introduce new data that allows researchers to identify and correct several coding inaccuracies in the COW IGO data. We identify the most common inaccuracies and relate them to theoretical debates in the study of IGOs, including IGO death and informal governance. We also indicate the utility of the new data by demonstrating how established results are affected when using the new data. Overall, the data is intended to complement, not replace, the COW IGO data and, in doing so, contribute to a more accurate understanding of IGOs and how they shape world politics.