Our Strategy : Addressing the correct problem

We have seen that the enabling data warehouse technologies and the methodologies for their implementation are not sufficiently powerful for the creation of corporate or enterprise data warehouses. The very technology that may be used to produce a successful line of business data warehouse may yield, at best, poor results when applied to an enterprise data warehouse.

This is because a line of business warehouse can be built successfully from the bottom up whereas an enterprise data warehouse can only be built from an enterprise model down, and these two constructs do not map to one another. These two forms of data warehousing are also neither pre-, nor co-requisites of each other, although for practical purposes the existence of well-constructed line of business warehouses may speed up the population of an enterprise warehouse. Similarly, the existence on an enterprise warehouse will not obviate the need for line of business - specific data warehouses.

In this paper we contend that the only sensible and useful method of creating an enterprise data warehouse is to define it to fit a single pyramid model of the enterprise. This model represents vertical key relationships, possibly with several hierarchies defined in parallel, and it is built on a single base level of granularity of the data, and the possible consolidations thereof.
The problem to be solved when designing an enterprise data warehouse is one of definitions. Firstly, an Enterprise Entity Model (EEM) has to be defined and secondly, data structures that will support this model have to be created. The definition of the model and the attendant data structures will define an internally coherent and complete schema which is a "first principals" definition of what the business sees as its business model, and this includes the constituent elements that comprise this model.

It does not necessarily hold that this enterprise entity model has any commonality with exiting systems or structures in the host organisation because the model is not an attempt to simply map the entities employed by the various lines of business into a single model. Rather, it will express the current business strategy of the enterprise, and the methods that the business will use to monitor the success of its strategy.
It follows that every time that an enterprise changes strategy, the enterprise entity model and the data warehouse supporting it will have to be changed, and perhaps even re-created in its entirety. This is a simple and normal cost in business administration, similar to the costs incurred when re-structuring branch networks, consolidating manufacturing processes and acquiring or starting new lines of business.

To follow...
  • MTN
  • Santam
  • Absa
  • Old Mutual
All Our Clients
Case Studies
  • Old Mutual
  • Santam
  • Absa
  • CSC
  • Uskotec
  • MTN
  • South African Airways
  • Vyinde
More Casse Studies
The 500 Year Delta

"The Information Age...is dead and gone, replaced by the information economy...The Information Age was about the building of databases. It was about the rise of…

Read More