The GL data mart scores well for Accuracy, primarily because it is a copy of the financial books, and therefore shows precisely how much the company is making or losing, as well as its assets & liabilities. Beyond the overall numbers, however, the GL data mart provides little integration back to the real operational entities (customers, suppliers, products & services, human resources, transactions, etc.) which drive these numbers. The GL data mart, in and of itself, is missing the core allocation logic required to elucidate the operational / financial connections. For this reason, the depth of analysis that can be performed within the GL data mart itself, unto actionable insight, is limited.
This limitation in insight, combined with the resultant downstream clutter created by having to cope with a morass of often poorly integrated allocation tools and spreadsheets so as to get to capturable value, pulls down the marks of the GL datamart for Value Derived as well as Scalability, Topological Efficiency, and Ease of Use. The GL datamart may itself be easy to use and scalable / efficient, but not when viewed thru the end-goal lenses of value capture, and the additional steps required so as to meet this pre-eminent goal.
It is important to note that all of the foregoing critique assumes that the GL datamart is used by itself, or in conjunction with a poorly integrated morass of other tools, in an attempt to deliver performance insight, absent a VDDW approach. The VDDW approach itself dictates the inclusion of the GL star schema as one of the key inputs to the ABC allocation engine (see the VDDW Architecture section for more details). So, for the purposes of this comparison, we are critiquing not the GL data mart itself, but the GL data mart by itself, that is apart from the VDDW.
