The first question we address in this paper is what do we really mean when we talk about the hybrid cloud? What is the difference between a genuine hybrid deployment, as opposed to a siloed or heterogeneous approach. And having answered that question, what sort of data should reside on-premises or in the cloud, or in a (genuine) hybrid environment? When doesn’t it matter? When does it matter? If it does matter, why does it? These are the sorts of questions that this paper addresses. There is a short section at the end discussing IBM’s support for hybrid cloud environments (IBM sponsored this paper) but the bulk of the discussions are generic and not IBM specific.