In part 1, data governance was defined as a process and structure for formally managing information as a resource. Data governance ensures the appropriate people representing business processes, data and technology are involved in the decisions that affect them. Data governance supplies the structure, roles and processes that provide venues for interaction and communication paths for gathering appropriate input, making decisions, identifying and resolving issues, escalating when necessary, implementing changes and communicating actions. Data stewardship is an approach to data governance that formalizes accountability for managing information resources on behalf of others and for the best interests of the organization. I covered the need for data governance because of changes in business requirements for integrating information across the company. The problem is that our applications and business needs for information are integrated, but our behavior has not changed to work effectively in this world. That is where data governance and stewardship comes in. Part 2 continues with a look at structure, roles and responsibilities.
Structure, Roles and Responsibilities
The data steward's role. This role called "data steward" is used casually and in a way that assumes everyone has the same meaning for it. A few months ago, I attended a conference session on data governance. Following are a few of the comments from the audience and panel members in response to the question, What is a data steward's role?
- "The data steward should be a subject matter expert."
- "The data steward is at the data name and definition level."
- "The data steward is a strategic role with responsibility for a data subject area across business processes and applications."
- "The data steward is at the application level."
- "The data steward's role is to fix data."
It is apparent there is no standard definition for data steward. You need to clearly define the role of the data steward in your company. In order to do this, you need to determine what issues data governance will be resolving and put together a structure to support that. Then, responsibilities need to be articulated for all the roles (including the data steward) within the governance structure.
To determine the data governance structure and the data steward's role, the real first questions are, what questions will data governance answer, and what issues will data governance resolve? The answers to these questions are based on your business needs, but here are a few ideas to get you started.
- Identify, own and facilitate resolution of enterprise-level data issues, such as system of record, reporting, security/privacy and integration.
- Understand the current state of data across the enterprise.
- Be responsible for data quality for a given subject area (across business processes and applications).
- Track/coordinate a roadmap for planned applications.
- Define all data standards, including names, definitions and value domains.
- Determine who should be involved in decisions regarding data used across business processes, e.g., the manual process for synchronizing the product hierarchy and finance hierarchy is too time-consuming and, ultimately, ineffective. Who should be involved in developing a new process?
Data governance structure. A data governance structure typically has multiple levels of roles and responsibilities. The structure and processes must provide venues so appropriate people can interact and make decisions. The processes must provide the ability to gather input, make decisions, identify and resolve issues, implement changes and communicate actions (see Figure 1).
Note that some form of information or data steward is listed in three of the four levels. For example, an enterprise information steward may be more at the strategy or policy level and a data steward at the execution level. You may have more than one type of data steward, with different stewards focusing on data definition, data creation or integration. Whether you are trying to implement data governance at an enterprise level or within a large business unit, there are the same natural divisions of responsibility for strategy, data governance program management and execution. You decide where the various roles and the associated responsibilities belong.
Appropriate representation. Business processes, data and technology must all be represented in the governance structure. When looking for people who can represent different areas or levels in the model, realize that a person may represent more than one area of knowledge or more than one level of responsibility in the governance model (see Figure 2). However, it would be unusual for someone to be able to appropriately represent more than two levels. For instance, the same person might represent both the data governance and execution level of responsibility, but it is unlikely that one person can represent the strategic, data governance and execution levels. One person might have knowledge of and represent both business processes and data or data and technology. But again, it is unlikely for one person to have the appropriate knowledge to represent business processes and data and technology.
Design your data governance structure with the titles and the levels that work for your company. Use these ideas and translate them to fit your culture and company. These concepts apply whether you are starting data governance at an enterprise, business unit or organizational level.
Following are suggestions for identifying participants:
- Ask: Who has the knowledge? Who has the authority? Look for those with authority both by position/title (assigned responsibility) and by influence and visibility. The data governance participants will be most effective if they have the position/title, but you shouldn't underestimate influence. Find those who are already filling the data governance responsibilities as part of their current job. Bob Seiner refers to "de facto stewards" and that you recognize them, not assign them.1 He says data stewards are discovered, not hired. Cheri Mallory uses the term "incognito stewards" to indicate someone already doing data steward-type activities and who understands the data, but who has another job title and another full-time job.2
- Avoid the tendency to put in people who are not experienced. I often hear, "This is a great job for our new college hire." This will not work. You need those who are experienced in the areas they represent and knowledgeable about your company.
- It is common to have those participating in governance to be in that role part time. Often those with a role in the business gain more respect, and a part-time approach takes fewer resources away from the business. However, making time for governance responsibilities is usually difficult. Ensure that you have their manager's support and governance activities are a written, established, recognized part of how their time is spent.
- Look for logical representation. For example, a data steward for customer data would logically come from sales or marketing, not manufacturing. However, also look for those who have the ability to represent all data usages - not just those within their business area.
Finally, be able to answer the following key questions:
- What is the motivation for the individuals participating in data governance?
- What will they get from participating?
- What will they give to participate?
- Do participants have support from their managers?
- Is their work with governance acknowledged and rewarded by their managers as an important part of their job responsibilities?
- In addition to the responsibility given, do they have the authority to represent their areas and make decisions?
Data governance is the process and structure for formally managing information as a resource. The goals of data governance and stewardship are to:
- Ensure the appropriate people representing business processes, data and technology are involved in the decisions that affect them.
- Supply the structure, roles and processes that provide venues for interaction and communication paths for gathering appropriate input, making decisions, identifying and resolving issues, implementing changes and communicating actions.
- Robert K. Seiner. "Stewardship in 3-D: de facto, discipline & database." The Data Administration Newsletter, 4 Jan 2003.
- Cheri Mallory. "The Genesis of Data Quality, the Emergent Data Steward." Business Objects, 3 April 2006.
Register or login for access to this item and much more
All Information Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access