When planning data warehouse design, I’d first focus on the business questions the warehouse needs to answer. Next, determine what data sources are needed, how frequently the data needs to be refreshed, and what reporting tools will be using the data. Other key aspects of good planning involve selecting the appropriate schema, establishing data quality rules and future growth considerations. It is better to have a first version which is practical, than to model something at once. If architecture sounds complicated, it’s a great idea to look at the work of a service such as Datalance to see how various design options apply to actual projects.