Data warehouse (DW) projects have become increasingly large and difficult to manage because of the growing use of analytical systems in which the data come from heterogeneous transactional systems. Usually, DW projects involve staff with different profiles, technological backgrounds and internal assignments within their institutions. Be- cause of these factors, many DW projects are slow to deliver effective results and often are not completed. According to a study by the Gartner Group in 2005 [12], about 50% of DW projects tend to fail due to prob- lems during DW design and construction. One of the most important causes is the long development time, which leads to delays in delivery of functional features to the end user. Often, when DW systems are finally available, some of the features implemented are already obso- lete, while newer needs end up being postponed until future phases of development.