The Complete BI Professional
A well-rounded BI pro combines depth of knowledge of a specialty with a broad knowledge of the field -- but knowing those disciplines is less important than understanding their relationships.
By Dave Wells, Consultant, InfoCentric
Business intelligence (BI) is the pinnacle of information management; it is the place where information is directly applied to inform decision processes and make substantial contributions to business value. BI has grown to be a broad and complex field that includes many specialties, and BI professionals have become similarly specialized. Specialization certainly has advantages, most notably depth of knowledge and skills, but specialization also has drawbacks -- in particular, loss of the "big picture" perspective.
A well-rounded BI professional combines depth of knowledge in one or a few specialties with broad knowledge of the entire field of information management -- a broad field that encompasses at least these 14 distinct disciplines:
- Business analytics
- Business intelligence
- Content management
- Data governance
- Data integration
- Data mining
- Data modeling
- Data quality
- Data warehousing
- Enterprise information management
- Master data management
- Metadata management
- Performance management
- Predictive analytics
These terms surface in many different ways -- as technologies, as the next "IT silver bullet," as corporate initiatives, as formal business or IT programs, and more. Every one of them, however, can only make a real difference when undertaken as an information management discipline. That is a carefully chosen word; discipline implies a degree of structure, rigor, professionalism, and clarity of purpose.
In practice, however, there is no benefit in treating each of these as distinct and separate disciplines. Separating them and listing them alphabetically is a convenience to understand the scope of information management but does little to represent the complexities. Each discipline has relationships to and intersections with several other disciplines.
Some of the connections are obvious: data quality, for example, is influenced by data governance; there is a very strong association between business analytics and business intelligence. Some connections are less obvious but equally important. Also, consider how the disciplines are interconnected in other significant ways:
- They all operate against the same collection of facts
- They have different roles in the same information acquisition and distribution processes
- They involve many of the same stakeholders
- They share a common purpose and enterprise goals for information management
Ultimately, it takes several disciplines to comprise a complete information management program for any enterprise. One of the ways to bring organization and structure to the very broad field of information management is to group disciplines based on the very strong affinity among them; this provides clarity not seen in an alphabetical list. I organize them into six groups:
- Data modeling and metadata management are both related to defining and describing and understanding data. Modeling is one means of collecting metadata, and metadata is one of the products of data modeling.
- Content management and enterprise information management are the disciplines of managing information supply and demand. They are the means to achieve an enterprise view of information that is needed, information that is available, and the gap between the two.
- Data quality and data governance focus on data and information utility -- usefulness and reliability of the information resource.
- Data integration, data warehousing, and master data management address in various ways the issues of connecting disparate data and resolving conflicts among multiple sources to create a consolidated data resource.
- Business intelligence, business analytics, and performance management address business-focused aspects of information management. Business intelligence is concerned with information-enabled business capabilities, business analytics with measurement-based investigation and insight, and performance management with application of business measures to decision processes and management effectiveness.
- Data mining and predictive analytics are closely related in their statistical modeling foundations and their technological implementations. The capabilities of data mining to discover patterns in data and the ability of predictive models to draw inferences from those patterns go hand-in-hand to provide business insight and foresight from information.
Now consider these groups as the levels of an information management stack. Understanding the data (modeling and metadata) is the foundation, with insight and foresight (BI, predictive analytics, etc.) at the top of the stack. The layers between describe a chain of dependencies. Thus every BI professional, no matter your specialties, depends in some way on several related disciplines.
Learning about those disciplines is sure to increase your effectiveness as a BI practitioner and your value as a BI professional.
|Insight and Foresight:
Data mining, predictive analytics
|Measurement and Monitoring:
Business intelligence, business analytics, performance management
|Consolidated Data Resources:
Data integration, data warehousing, master data management
|Data and Information Utility:
Data quality, data governance
|Information Supply and Demand:
Content management, enterprise information management
|Understanding the Data:
Data modeling, metadata management
Dave Wells is a consultant, mentor, and teacher in the field of business intelligence. He can be reached at email@example.com