Governance is of concern to any enterprise, executive team or individual with an interest in how data is created, collected, processed and manipulated, stored, made available for use or retired. It is even more important to organizations that have invested in data warehousing to collect, analyze and use data to improve overall capabilities in the enterprise.

All users need to trust the data. If they make decisions based on the data, they want to know those decisions have ancestry; if they capture the data, they want to know the rules for storage; if they work for the organization and have signed a compliance and/or ethics legal form based on the captured data, they want to know they are personally protected. Consequently, understanding the focus and framework for governance is crucial.

Governance programs can differ significantly, depending on their focus. Some organizations pursue governance initiatives because of competitive or customer pressures or for compliance purposes. Whatever the focus, every governance program will have essentially the same three-part mission:

	Make/collect/align rules
	Resolve issues
	Monitor/enforce compliance

The challenge for most enterprises is to establish a context for their governance efforts while providing support to stakeholders.

Governance context
Governance activities occur within the broad definition of corporate governance and consist of three key areas:

	IT governance
	Data warehouse governance
	Data governance and its associated data stewardship

Many top-down organizations approach these areas of governance using the model shown in the left portion of figure 1 below. These companies cascade policies from corporate governance through IT governance to data warehouse governance to data governance. This approach is a technology-enabled view influenced by traditional organizational models.

Conversely, some organizations consider data and information as their most critical asset and will drive policies about their use down through the organization using the model shown on the right in figure 1. Either approach can be successful for an enterprise—the key is to understand how the hierarchy affects business operations and strategy.

Regardless of how governance is used in an organization, the overall message is clear: Rules, regulations and structure are important to maintaining the value of the enterprise data.

IT governance
IT is fundamental to the support, sustainability and growth of the business. To ensure that the enterprise's strategies and compliance requirements are followed, its values are adhered to and its objectives are met, an IT governance team must be in place.

The first of the three structures of the governance body, IT governance consists of the leadership and organizational structures and processes. With these in place, the effective application of an IT governance framework is critical in helping enterprises gain more value from the corporate data.

enlarge The two pyramids show different approaches to governance. The left pyramid is driven by corporate governance, while the pyramid on the right is driven by data governance.

Effective IT governance requires careful analysis about who makes decisions and how decisions are made in several critical domains: principles, infrastructure, architecture, investment and prioritization. Governance also requires risk mitigation of subjects such as:

	Business continuity and disaster recovery
	Regulatory compliance
	Information governance and information security
	Project governance
	Knowledge management
	Risk management
	IT service management

Data warehouse governance
Data warehouse governance defines the model the organization will use to ensure optimal use of the data warehouse and enforcement of corporate policies (e.g., business design, technical design and application security).

The function of a data warehouse governance approach is to oversee the data warehouse for the purpose of providing an environment that reaches across the enterprise and drives the best business value. It covers the integrated management of the entire data warehouse life cycle, from initial development through production to end-of-life. For this reason, data warehouse governance evolves as the business needs change.

A data warehouse governance approach requires:

	Setting strategic direction
	Determining priorities
	Committing resources
	Allocating funds
	Monitoring initiatives
	Communicating status
	Influencing key people

To ensure that the data warehouse is effective within the context of overall IT governance, the data warehouse governance must provide the necessary policies, process and procedures. These must be clearly communicated to the entire corporation, from the IT employees to the front-line personnel.

Data warehouses are not one-time projects; rather, they are long-term programs that include a number of projects and a significant amount of maintenance. The decision to deploy a data warehouse commits the business to a program that must be controlled to ensure that it provides business benefit and monetary value.

Organizations have different requirements for the level of detail and breadth of scope that should be covered by data warehouse governance. For example, some organizations will include coding standards, while others deem this excessive in terms of detail.

The key to designing a suitable level of governance within an organization is in understanding:

	What events affect the data warehouse environment?
	What is affected in the data warehouse by the impacting event?
	What type of organizational approach is needed to support the data warehouse?
	What policies and procedures are needed to control the data warehouse?

The answers to these questions will help develop the standards, processes and communications necessary to drive the data warehouse developments.

The processes, used in conjunction with the standards, define how decisions are made. Good processes enable quick, effective and well-communicated decisions and effective management of the data warehouse. Conversely, poor standards and processes lead to delays and unresolved issues that discredit the data warehouse and create significant cost or overhead that can eventually destroy the program.

Enterprises often deploy an enterprise data management (EDM) framework that defines how the people, processes and technology work together to support the data warehouse. An EDM approach is used to:

	Identify and apply the enterprise's core information to the business goals of the organization
	Manage data as an enterprise asset across the entire company
	Provide the organization with a central focus for establishing and documenting enterprise policies, procedures and standards, and for managing data resources (human, systems, data objects)
	Analyze the use and management of data and its related processes
	Encourage data management from a line-of-business and enterprise perspective

Data governance
Data governance refers to the overall management of the availability, usability, integrity and security of the data employed in an enterprise. How the data is used by the organization in its decision making is the outcome of a data governance system.

Functions covered by a sophisticated data governance framework are shown in figure 2 below and include:

	Policies/business rules
	Access/usage
	Risk management
	Quality/integrity
	Stewardship/owners
	Security/privacy/compliance
	Governing body/organization
	Accuracy/consistency
	Completeness
	Availability
	Update/freshness/refresh rate
	Audit/controls/metrics/management

With its guidelines established by corporate leaders and IT and communicated across the organization, the data governance oversight process prioritizes investments, allocates resources to projects within the data warehouse environment and monitors project results to ensure enterprise data is properly valued and used in alignment with corporate objectives. More specifically, data governance includes determinations of who can take what actions with what information, when the actions can be taken, and how those actions will be accomplished.

enlarge The sections of the framework show the various functions within data governance.

Data stewardship
The initial step in the implementation of a data governance program involves defining the data stewards—the custodians of the data assets in the enterprise. In a well-functioning data governance program, the data stewards are selected from each business unit within the enterprise.

A set of standards and procedures must be developed that defines how the data is to be used by authorized personnel. The policy should specify who within the organization is accountable for various aspects of the data, including its accuracy, accessibility, consistency, completeness and updating frequency.

The method in which the data is backed up, archived and restored, as well as the security components of the data, is also dictated by the data governance system. Finally, a set of controls and audit procedures must be in place to ensure ongoing compliance with government regulations.

A solid foundation
A solid governance foundation paves the way for users to trust and value the corporate data and to leverage the data to its potential. The foundation of the structures of governance established within the enterprise ensures the success of the enterprise and its use of data warehousing technologies. T

Lance Miller has been with Teradata for nine years as a director of services marketing, and is a subject matter expert in analytical applications and several areas of the software industry.

Photography by Getty Images

Teradata Magazine-March 2008