Research data collections are tremendously import-ant and thus need good curation. However data collections are significantly different to publication repositories and so we need to ensure that these differences are taken into account when managing research data. We believe that a good way of approaching this problem is to articulate the needs of research data stakeholders – particularly users and creators. Consequently we have described an analysis of these needs and then examined costs in the light of these varying needs – it is important to note that costs are often incurred by different people to the beneficiaries. We finish the paper by showing practically how incurring software costs can provide valuable savings for both data creators and data managers.