Every business intelligence or analytics initiative needs to have the foundation of accurate, well-managed data that comes from a robust data governance program
Many corporations experience significant business benefits using business intelligence and analytics. Users report gains in market competitiveness (increased revenue) and data / information management (reduced costs).
Foundation of Business Intelligence and Analytics
The foundation of most business intelligence and analytics data can be found in a data warehouse. A data warehouse is a separate architecture used to maintain critical historical data that has been extracted from operational data storage and transformed into formats understandable to and usable by the organization’s analytical community and management.
Data integrity is a major issue within most organizations, and the development of a data warehouse can be a vehicle to improve data quality significantly. Accuracy in data can mean realized savings in areas such as marketing, customer service, and finance. Many studies by organizations such as Gartner Group and Innovative Systems point to the savings obtained from a 4% increase in the integrity of data in many diverse companies.
A data warehouse serves as the focus for analytical and decision making querying and reporting. Therefore, building and maintaining the data warehouse needs the attention to data requirements across the enterprise that a robust data governance program provides. Data warehouse / BI initiatives require organizations to make many decisions that involve data from several sources, to enable the cross-application analysis. In addition to these foundational data challenges, a data governance program for a data warehouse also can provide analysis for external data that is brought into the warehouse, and can offer the oversight to enforce standards and rules after the analytics system becomes operational.
Data Governance Principles and BI / Analytics
The principles that drive a data governance effort usually involve components such as data integrity, data standardization and metadata, standardized change management, and audit capabilities. These components are especially important in any cross-organizational effort and are essential in business intelligence and analytics.
Although any data governance initiative starts with the proven foundational principles, developing a business intelligence / analytics-oriented data governance program may focus on the following points:
- Identification of the initiative’s stakeholders and their decision-oriented data requirements.
- Analysis of data integration needs to achieve the cross-application / cross-functional decision-making expected in a data warehouse.
- Improvement of data quality and data integrity, including standardization of data elements for the data stored in the data warehouse.
- Creation of data definitions for master data (common data) and development of standard codes sets for common (master) data used in the warehouse, including appropriate algorithms and calculations.
- Creation of common reporting requirements for analytical data, based on stakeholder requirements.
- Creation and organizational enforcement of data creation / modification and deletion standards for analytical data.
- Development of recommendations to reduce data redundancy and encourage appropriate data reuse.
- Oversight of management and development of metadata repositories.
- Development and enforcement of data quality metrics for business intelligence and analytical data.
- Analysis of whether data is fit for its intended use, including completeness and business-rule compliance.
- Implementation of processes to cleanse, transform, integrate and enrich fresh data across subject areas.
- Development of security and privacy requirements across integrated subject areas in the data warehouse / data mart.
- Reporting results on data management to appropriate senior management.
Starting a BI / Analytics Data Governance Program
Many organizations start a business intelligence or analytics effort to examine data across functional or subject areas, often using data from a data warehouse or data mart. Every data warehouse should have a data governance program that is focused on managing the improvement of the data and metadata quality and controlling risks to avoid its continued degradation. These controls may be preventive or investigative, and they may be completely automated or consist of technology-enabled manual processes, or a combination of both approaches. There are many products available to examine and monitor data and metadata quality, both in transaction and decision support systems, and many processes for improving data quality. No data warehouse should be without a data governance program that does not include the ability to manage and monitor its data and metadata quality.
Data marts need data governance as much as data warehouses; in fact, the creation of one or more data marts shows the need for an organizational approach to data governance. Doing so will provide the oversight and guidance for decision-oriented data across business units, and enable each data mart data to be useful across the divisions. As with many robust data governance initiatives, a business intelligence / analytics data governance program may start small and grow to an enterprise approach as data governance is accepted and sustained across the organization.
Conclusion
A data governance program can help ensure valid data is in the hands of business users in every department and business function. The results for a data governance program as part of a business intelligence / analytics initiative include: more informed decisions; reduced redundant data and colliding definition and calculations; statutory and regulatory reporting using accurate and consistent data and an integrated approach to data management and usage throughout the enterprise.
