Access to Source Systems
Access to source systems is something analysts do not always have if a data warehouse has already been established. This access is usually not automated, which is the reason people often associate considerable time consumption with the use of information obtained directly from source systems. In addition to this, data quality can be quite variable, depending on what the information is used for in the source system. Furthermore, the source system itself might suffer in performance if accessed as a data warehouse. Developments in these cases will be that if information of significant value is identified in the source systems, a process will be started to ensure that this information becomes accessible in the data warehouse. If the business does not have a data warehouse, analysts always work directly with source systems, in spite of the weaknesses this entails. Over time, an analyst will usually want to carry out regular runs every month for reporting;the data set generated from this can therefore be seen as a data warehouse at the simplest level and as a quantity to be optimized.
In recent years, companies have come to realize the strengths and weaknesses of the Internet. There has been a growing understanding of the fact that, for instance, customer behavior on a company's Web site is relevant CRM information, too. It is not without relevance, for instance, for a telecom company to know whether a given customer has checked the company's Web site to find out how to cancel his or her subscription. If this information is then combined with information about the customer's value as well, it can be decided whether a "loyalty call" should be made to the customer in question. Long term, the company could structure its Internet portals so that the users' way of moving around these portals will affect the way this customer is treated. The Web site thereby becomes a questionnaire completed by the customer via his or her clicking, instead of merely providing information and automated services.
Generally speaking, though, Web log files are not yet providing the company with information about how customers and users use Web portals. In connection with commercials, however, there are exceptions, such as Amazon.com and a few search engines that are related to commercials that specifically and successfully use Web information as a strategic resource that can provide them with a competitive advantage. In these extreme cases, it seems possible to collect and use information to drive the company's strategy.
In this chapter, we have discussed how to store data to best support business processes and thereby the request for value creation. We also looked into the architecture and processes in a data warehouse.
BA is not possible without access to a combined data foundation from the organization's data-creating source systems. In fact, that is exactly what a data warehouse does: It increases the usability and availability of source data, as will be explored further in Chapter 6.
One central enterprise data warehouse ensures consistent, integrated, and valid data definitions across business areas and countries. This principle of one version of the truth is fundamental for companies to avoid spending much time with contradictory reports and deviating business plans (budgets).