Tracking usage of research output and publishing statistics is a value-added service of Open Science research infrastructures. On the content provider level, usage statistics can serve administrators as a tool to evaluate the success of their platform. On the individual item level, they can demonstrate popular items, like datasets, software, or other research output to research community. In addition, usage statistics can indicate the relevance of a research output over the course of time and up to the present, and therefore they can be considered a useful indicator to analyze trends. Different organizations facilitate the above tasks by monitoring and analyzing usage data and exploit usage metrics like research items downloads and metadata views. The final outcomes are published via services like DataCite, IRUS-UK, OpenAIRE, or national aggregators like LA Referencia. A number of parameters are involved for tracking, e.g the identifier of the requested item, the timestamp of the request, etc., whilst publishing of usage statistics is realized by exploiting the COUNTER Code of Practice (CoP), that enables publishers to report and compare usage of resources in a consistent way. Although these practices are employed by most usage statistics services, there are several issues that need to be tackled, due to the complexity of the research products and their diversity. Definition of a set of guidelines, together with a framework that can facilitate and standardize the collection and exchange of scholarly usage events could alleviate those issues. The Open Science Usage Metrics Exchange (OpenSUMex) Guidelines and Framework is developed to cover these requirements by defining the guidelines and provide the framework to exchange usage metrics information. The resulting usage metrics could identify multiple counted events that could be comparable across scholarly infrastructures and exploited by a variety of stakeholders from research community, like repository managers, authors, funders, etc.