A Continuous Delivery Framework for Business Intelligence

In the context of business intelligence, data warehousing is often perceived as an integral component of concrete business intelligence solutions. Since the nature of a traditional data warehouse is accumulative – data from operational systems is fed in to the system when it is ready for inclusion – and as the data from different component systems is interrelated, operations involving data warehouses have been traditionally considered tedious and delicate. Distinct steps take place one after another in a predefined, next to unalterable sequence. In this paper, we present an alternative model for dealing with data warehouses, where the goal is to apply principles of continuous software engineering in the domain of business intelligence. To validate the methodology, we present a tool chain that has been used in a real-life implementation of a business intelligence solution, together with experiences from its operations.