Crowdsourcing is used for problem solving in different domains. A promising key to optimal solutions is collaboration among crowd workers. However, due to the distributed and asynchronous nature of crowd work, often with a large number of heterogeneous, anonymous and varying workers, crowd collaboration support is challenging. Thus, platform providers and crowdsourcers still struggle with or refrain from making full use of the potential of collaboration. The current state of research on this field explores this topic mostly for a specific domain, such as idea contests. This paper widens this scope and aims to validate a general process design framework for collaboration in crowdsourcing across various domains in an ongoing design science research project. To achieve this, we analyze current projects on crowdsourcing platforms with a conceptual process structure and a corresponding set of criteria for effective crowd collaboration support. We conduct a content analysis of ten real world projects to gain insights on their collaboration support features and collaborative interactions of crowd workers. This paper contributes to crowdsourcing research and practice by deriving recommendations for advancing the collaboration process design framework as well as for improving the conclusiveness of collaboration support on crowdsourcing platforms.