As markets become more complex and demanding, corporate data warehouses are taking strain, reports Johan Cloete, CEO of Global Technology Business Intelligence. Enterprise data warehousing can help alleviate this strain.
The benefits of enterprise data warehouses are legion, and when properly used, they provide a real competitive edge in a world of constantly shifting technology paradigms.
Data warehouses are typically built to consolidate and report historical data on assets, products, services, customers, and revenues using standard SQL-generated reports. This data is then commonly posted against increasingly complex what-if scenarios, with the end product a prediction of a trend or a hypothesis on projections based on the data. However, the increasingly complex demands on manual query techniques pose a challenge with the increase in the amount of data being processed.
As a result, data warehousing users are turning to data mining as a way of breaking through the limitations of traditional query tools and analysis techniques. Data mining shifts the burden of analysis from the person to the computer, enabling the discovery of more subtle relationships in data that could not be found with manual query tools. Information produced in this way, in the hands of the correct corporate decision-maker, can lead to new insights and initiatives.
The three components of an enterprise data warehouse perform vital functions: the data warehouse itself is the source of quality data for building and validating models produced by the second component, the data mining application. Data mining creates analytical models that reveal knowledge; the third component is the ability to leverage all data sources from enterprise-wide databases and departmental or function-specific data marts.
Vital attributes
The most important function of an enterprise data warehouse is to create a central data repository which can provide a single account of the entire enterprise`s operations. To achieve this, an enterprise data warehouse must possess a number of attributes:
1. Scalability. An enterprise data warehouse must grow with increased data volumes, adapt to changing data access patterns, and support an increase in the number of users.
2. Detail data. The ability to answer any query lies in the application`s ability to organise all business data as a unit, instead of as separate departmental areas. Access to just a few aspects of the enterprise`s operations will lead to a potentially skewed overview and ultimately the wrong business decision.
3. Connectivity. The data that constitutes the data warehouse is drawn from the operational systems used by the enterprise. Critical to the creation of any data warehouse is the ability to find and quickly offload historical data from the operational system to the data warehouse.
4. Availability and maintainability. The system must be refreshable while live; if refreshed offline it must be done without affecting the availability of the data to business users. A good data warehouse will allow users either to access live data or update data without downtime and interruption of user access.
5. Business needs. Above all, the data in the data warehouse should be aligned with strategic business issues and objectives. The real value of such data will come to the fore only if it can produce knowledge and information that affect actionable events in an organisation.
Once a data warehouse has met these requirements, it can begin to deliver real value and strategic competitive advantage. Such competitive advantage is built through deep and far-reaching strategic thinking, something unobtainable without an enterprise-wide view of all aspects of a business.
The strategic ideas that support data warehousing are learning, manoeuvrability and prescience. Enterprise data warehousing meets the fundamental business needs in all respects.
Enterprise data warehousing is a prerequisite to engaging the market in a manoeuvrable manner. Survival will depend on the ability to change business direction continually, and focus requires judgement and knowledge. Experience provides the former; enterprise data warehousing provides the latter.
The strategic intent of enterprise data warehousing is to enable the business to win in the market every day, with every customer, and with every purchase. By repositioning operational data and combining it with selected third-party data, employees are empowered to satisfy all aspects of a customer`s wishes. Through enterprise data warehousing, data warehouses and departmental data marts are elevated to the point where they become a compelling and durable contributor to the sustainable competitive advantage of the business.
Share
