Today, large organisations and regulated markets are subject to the control of external audit associations, which require the submission of a huge amount of information in the form of predefined and rigidly structured reports. The compilation of these reports requires the extraction, transformation and integration of data from different heterogeneous operational databases. This task is usually performed by developing a software ad hoc for each report, or by adopting a data warehouse and analysis tools, which are now established technologies. Unfortunately, the data warehousing process is notoriously long and error prone, and is therefore particularly inefficient when the output of the data warehousing is represented by a limited number of reports. This article presents “MMBR”, an approach that can generate a multidimensional model from the structure of expected reports as data warehouse output. The approach is able to generate the multidimensional model and populate the data warehouse by defining a knowledge base specific to the domain. Although the use of semantic information in data storage is not new, the novel contribution of our approach is represented by the idea of simplifying the design phase of the data warehouse, making it more efficient, by using an industry-specific knowledge base and a report-based approach.