The reporting and analysis demands of the specific business departments were covered by the integration of the necessary operational data sources into a departmental data warehouse (data mart).. Each department created its own models, freely selecting and designating performance indicators, reporting dimensions, and their hierarchical levels. Master data required by almost every department, such as customer or product data, were duplicated by the departments and then reused without central control.

And what was the typical result? The various departments switched from Excel lists to modern OLAP tools, true; but they retained the performance indicators as defined in the past in terms of both content and designations. So a colorful assortment of indicators with the same names but differing content, with different names but the same content, and many indicators with similar content was carried over from the company’s previously decentralized reporting solutions and Excel worksheets into the world of OLAP. Virtually no one was willing to pay attention to the calls for coordination of and agreement on standard content of performance indicators and for formal description of their content from a general company point of view. Moreover, the software tools to support such a project had not yet reached the required maturity level.

The situation could be summarized like this: analysis and reporting solutions implemented at the business department level represented great progress in the eyes of these departments. Nevertheless, the diversity found in performance indicators favored by this decentralized development led to a lack of transparency and consistency from a company point of view. 

“Yes” to the required diversity in indicators – “No” to arbitrariness in content and designation 

One of the reasons for the proliferation of the performance indicators can be found in the departments’ differing viewpoints and information requirements. For example, a margin or sales figure can be used with or without a surcharge, a rebate, a price discount, tax, and so on. Taking all of these variations into account means introducing various indicators whose designation normally should be an indication of their content. Another factor leading to this enormous diversity results from the various dimensions used to look at the performance indicators. Typically, such dimensions have a hierarchical structure which allows drilling into more detailed data views. Even if the dimensions used in the various company divisions are given the same name, the aggregation levels within the dimensions are normally specific to the department or division. Moreover, the structure of dimensions can be highly complex.  Frequently, even multiple hierarchies are required within one dimension. Typical “problem childs” are the dimensions sales and sales channel, customer, company, product, or the dimension time. All these factors cause a diversity which is necessary, but makes it hard to integrate departmental reports into one central company reporting system.