The ideal scenario to derive the multidimensional conceptual schema of a data warehouse would entail a hybrid approach (i.e. a combined data-driven and requirement-driven approach). Thus, the resulting multidimensional schema would satisfy the end-user requirements and it would have been conciliated with the data sources. Currently, most methodologies follow either a data-driven or requirement-driven paradigm and only a few of them follow a hybrid approach. Furthermore, current hybrid methodologies are unbalanced and they do not benefit from all the advantages brought by each paradigm. In this paper we present a novel methodology that derives conceptual multidimensional schemas from relational sources bearing in mind the end-user requirements. The most relevant step within our methodology is the MDBE method that introduces three main benefits with regard to previous approaches: (i) the MDBE method is a fully automatic approach and therefore, it also handles requirements in an automatic way. (ii) Unlike data-driven methods, we focus on data of interest for the end-user. However, the user may not know all the potential analysis contained in the data sources and, unlike requirement-driven approaches, MDBE is able to propose new interesting multidimensional knowledge related to concepts already queried by the user. (iii) Finally, MDBE proposes meaningful multidimensional schemas derived from a validation process. Therefore, schemas proposed are sound and meaningful.