Despite a large body of research on corporate social responsibility (CSR), there has been little research on the actual implementation of CSR strategies in companies.Therefore, this article examined top-down and bottom-up approaches for implementing sustainability at company level. For our survey, we chose a large IT services company, which is one of the global leaders in the Software as a Service (SaaS) business. Based on theoretical foundations, namely the Social Identity Theory and the Social Exchange Theory, six interviews were conducted with employees and managers of the company. Our results show that the company's vision and values strongly influence the employees' sustainable work activities and behavior. In addition, the employees also have options for shaping the sustainability strategy within the framework they are given. Concluding, the article shows the advantages of integrating both top-down and bottom-up approaches, and identifies options for companies implementing sustainability.

Data science driven applications (e.g., big data and artificial intelligence) can support the transition to a green economy. However, this requires overcoming existing barriers and providing appropriate framework conditions. Based on an analysis of 295 German and US start-ups using data science to create positive environmental impacts, we identify six main obstacles to a greater use of data science for sustainable transformation, and propose six measures that can be used to formulate policy recommendations.This paper examines the intersections between the hoped-for shift toward a green economy and data science (various forms of big data analytics and artificial intelligence). It does so through an analysis of data science applications with environmental relevance developed or deployed by German and US start-ups. The majority of the data science applications identified seek to improve the efficiency of existing products and processes, or to provide information. Applications that support more fundamental transformations of existing production and consumption patterns are fewer in number. To increase the sustainability-related impact of data science, it seems necessary to adjust policy framework conditions. Based on our findings, recommendations for action are presented regarding sustainability-related changes of the legal and regulatory framework conditions.