To begin with, you need to decide on what data you will store and share. The Digital Curation Center (DCC) "Five steps to decide what data to keep" guide and checklist could assist you in making this decision. Data sharing practices vary widely across research domains and geographical areas. Data published as supplementary material of journal articles is still the prevailing choice with data repositories option being on the rise. Data repositories can assist you in better managing your data, offer a persistent identifier for your data to be cited, help support discovery and long time preservation of your data. Nevertheless, finding and selecting the appropriate research data repository to submit one’s work can be challenging. This guide aims to help you in both.
You should start thinking about data selection for sharing as early as the planning phase of your research project. At that stage you should have a general overview of the expected outputs and their potential for re-use. As the project evolves this picture is refined, data are produced and processed and part of them will lead to a new scientific discovery. You should at all times be mindful of all policies and requirements (institution, funder, publishers) that apply to your research and also follow best practices and recommendations by your research domain community that may affect your choice for what, when and how to share.