A data warehouse is the major store of an organization’s historical data, and its corporate memory. It contains the raw materials to support managerial decisions. According to Greenfield 2005, a data warehouse is a copy of transaction data specifically structured for querying and reporting.
The data warehouse can be a relational database, multidimensional database, flat file, hierarchical database, object database, and others. The data often gets changed. However, data warehouses often focus on a specific activity. (Greenfield, 2005).
A data warehouse contains operation data that help in the daily functioning of business and information useful to business analysis. In data, warehousing creates a store of informational data. The data is extracted from operational data and then transformed for decision-making (Greenfield, 2005). Data warehouses are often at the center of the strategic reporting systems used to help manage and control businesses. It helps to consolidate and reconciles information from across disparate businesses units and their systems and provides a context for reporting on and analyzing various issues like profitability and corporate performance management.
Many companies find it necessary to set up data warehouses. There are usually basic reasons that push companies to undertake this project. Mostly the advantages that come with the creation of data warehouses motivate the company stakeholders to set up the data warehouse.
Companies usually set up systems for processing transactions that do tasks of the server that are related to querying and reporting on servers and not used by the transaction processing system (Greenfield, 2005). This helps transactions to be completed within a stipulated duration. In this way, an acceptable response duration can be met. This is less expensive. The best way of achieving acceptable transaction processing response time is to set up a data warehousing architecture that uses separate servers for some querying and reporting (Greenfield, 2005). This motivates companies to set up data warehouses.
On other occasions, Companies aim at using data models and server technologies that hasten to query and reporting and those that are not appropriate for transaction processing. Some ways of modeling data enhance the speed of querying and reporting and may not be appropriate for transaction processing because they complicate the processing. Other server technologies enhance transaction processing but slow down queries and reports (Greenfield, 2005). However, regardless of the impacts of server technology, processing varies according to the situation of use.
Data Warehouses provide an environment where limited technical knowledge of database technology is required to write and maintain queries and reports by technical personnel. They provide transaction processing systems data that can be reported against and that does not necessarily require fixing the transaction processing systems. (Greenfield, 2005).
They make it easier on a regular basis, to query and report data from multiple transaction processing systems and from external data sources and from data that must be stored for query/report purposes only (Greenfield, 2005). This has helped much since many firms have been writing data extracts and then merge logic to combine the extracted data and then running reports against the merged data
They provide a system of processing transaction data from a longer span of time than can efficiently be held in a transaction processing system and are able to generate reports as of a previous point on time (Greenfield, 2005). Older data are purged from transaction processing systems managing the expected response time better
Companies also set up warehouses for security reasons. They prevent individuals who intend to query and report transaction processing system data from having any access to transaction processing system databases and logic used to maintain those.
Companies, therefore, accrue a lot of benefits from setting up a data warehouse. The benefits enjoyed making all stakeholders gain a lot. The company administration and owners are major beneficiaries of this undertaking. The company management team is the major winner of these undertakings. The above reasons for setting up warehouses if fulfilled have a major impact on the company’s running (Greenfield, 2005)
However as companies enjoy these benefits, there are other factors that greatly affect the value of data warehousing for some organizations. These factors are mostly business and cultural in nature.
Data warehousing systems mostly store historical data that have been generated in internal transaction processing systems. This is a small part of the available data to manage a business. Sometimes this part also has limited value. There can be limited interest in any in-depth data analysis since a business is sometimes simply run that a data warehouse is not necessary.
Data warehousing systems in some cases complicate business processes significantly. It can complicate easily created reports whose reason for being is quickly forgotten, while individuals strive to process these reports. This happens if data warehousing is unchecked.
In some cases, data warehousing may not help much in a business. This is in cases where the business needs are to report on data in one transaction processing system or if all the historical data needed are in the system or are clean (Greenfield, 2005).
Data warehousing has a learning process that may be too long for impatient companies. Sometimes it takes time for an organization to identify ways of changing its business practices to get a necessary return on its investment. The intensive analysis of the return on data warehousing implementers’ investments finds a longer average payback period.
Sometimes the setting up of a data warehouse is a costly undertaking. The cost to get data, clean it up and deliver it in a format and time frame that is useful for users is costly. The cleaning activity requires special skills that developers might not possess (Greenfield, 2005). Strategies are often made which end up compromising the value of the information in the data warehouse.
The data warehousing systems require a high level of maintenance which many companies cannot support. It then becomes easy for users to abandon using a system if the personnel does not support it. There are also a limited number of employees and consultants with experience working with the full data warehousing system. The system needs a great amount of time to develop fully (Greenfield, 2005). Experience, therefore, takes a longer period than anticipated. Many strategic applications of data warehousing have a short span and require developers to quickly attempt a system that is technically low.
Considering the intensity of these limitations of venturing into data warehousing some companies are reluctant to enter into this field. The administration, therefore, stands to lose when enough research is not carried out. When the data warehouse performs well the company is the winner. The company gains a lot and overpowers other competing firms which subsequently lose.
The most crucial part of a data warehousing project is the design of the data architecture. There should be planned room for future expansion and growth. This should be kept in mind from the beginning as a company tries to design a perfect system. (Thearling, 2000). The design of data architecture involves internalizing how the data is interrelated. Afterward one can then consider the reports of interest. The power of a data warehouse becomes exceptional when one finds a link between data associated with various organization parts.
Data warehousing projects can be derailed by non-technical factors like politics within the company and the politicized nature of the data itself. (Demarest, 1997). Politics can cripple data warehousing activities. Organizational energy can be deviated to deal with political issues rather than the most pertinent issues. This affects the management of the organization (Demarest 2000). Political issues therefore should not be left to control the affairs of a data warehouse running and management.
Mostly, decisions to build warehouses are spearheaded by non-human resource needs. The supply chain, sales, and marketing departments of organizations have spearheaded the creation of enormous corporate data warehouses in various places. Improving the efficiency of the supply chain and competition for customers rely on the tactical uses that a data warehouse is meant to provide. The human resource and the administrative departments of companies should be involved in the creation and running of data warehouses (Thearling, 2000)
The management led by the project directors should be able to establish functioning guidelines to manage and control the data warehouse. Directors should work with developers to repackage existing frameworks into a more superior organization (Demarest, 1997). The data warehouse management team who includes a competent project manager, data administrator, business analyst, and a few programmers could turn an entire project to be able to deliver and achieve its real value.
Greenfield, L. (2005). The Case for Data Warehousing. The Data Warehousing Information Center. Web.
Greenfield, L. (2005). The Case Against Data Warehousing. The Data Warehousing Information Center. Web.
Demarest, M. (2005). The Politics of Data Warehousing Web.