“A year ago, in April 2016, Leiden University’s Centre for Science and Technology Studies (CWTS) and Elsevier embarked on a project to investigate open data practices at the workbench in academic research. Knowledge knows no borders, so to understand open data practices comprehensively the project has been framed from the outset as a global study. That said, both the European Union and the Dutch government have formulated the transformation of the scientific system into an open innovation system as a formal policy goal. At the time we started the project, the Amsterdam Call for Action on Open Science had just been published under the Dutch presidency of the Council of the European Union. However, how are policy initiatives for open science related to the day-to-day practices of researchers and scholars? With this report, we aim to contribute to bridging the gap between policy on the one hand, and daily research practices from a global perspective on the other hand. As we show, open data practices are less developed than anticipated, with the exception of fields where data practices are integrated in the research design from the very beginning. While policy has high expectations about open science and open data, the motive force comes not from the policy aims, but in changing practice at the grass roots level. This requires we confront the harsh reality that the rewards for researchers and scholars to make data available are few, and the complexity in doing so is high….”