There are main 2 strategies to design DW.

1. Bottom up design : In this approach, data marts are first created to provide reporting and analytical capabilities for specific business processes or departments, data marts provide atomic data or summarizes if needed. These data marts are then integrated together to create enterprise DW. The most important management task in this approach is to ensure that the dimensions among the data marts are consistent.

2. Top down design : In this approach, the EDW is first created using a normalised enterprise data model. Atomic data i.e. data at the lowest possible level, are stored in EDW, dimensional data marts are created from EDW for specific business processors or departments are created from EDW. This approach generates highly consistent dimensional views of data across data marts since all the data marts are created from the central re position. it is also proven to be robust against the business changes.