This paper presents research-in-progress. An extensive customer-centric data warehouse architecture should enable both complex analytical queries as well as standard reporting queries on customer data without performance restrictions for both requirements. This paper introduces a dichotomic approach, which brings together these contradicting tasks of a data warehouse. On the one hand, it elaborates on the qualities of customer data and their implications on data structures due to their change over time. The authors will present a data concept, that is specialized in gaining a realistic image of the customer, ideally over her entire customer lifecycle. On the other hand, the paper works out the role of the operational data store (ODS) in the light of CRM: it presents how the ODS supports its counterpart, the data warehouse, in the dichotomic approach for the maintenance of high performance and low response times.