The Data Stewardship Approach to Data Governance: Chapter 6

This is the sixth article in a series by Robert S. Seiner. This chapter begins to address the KIK Data Governance Framework© that is demonstrated in the diagram below. This discussion starts with the top inside section (green) of the pyramid structure that I label as the strategic layer or the Data Governance Council. The articles in the coming months will address each level of the pyramid as well as the side-bars and the tower sticking out of the top. Please address all comments and questions to Mr. Seiner.

I have been asked many questions about the role of the “Data Governance Council” (or as one person put it, the “supreme beings”) in a Data Governance Program. Well … for one, are they the supreme beings? Do they sit at the top of the data food chain? Why do we need a ‘Data’ Council? How are members identified? What DO they DO?

This brief article will guide you toward answering these questions for your organization.

For the purpose of this article, a strategic decision is a cross-enterprise decision. We could spend the next few paragraphs defining ‘enterprise’ as well but, for my purposes, the enterprise is “all of the business that you care about for your data governance endeavor.” This can be as large as a group of companies or as small as all of the functional areas that will be feeding data into a data warehouse. Typically it falls somewhere between.

Strategic decisions need to be made when decisions cannot be made at the operational level or the tactical level. Strategic decisions require that the people making the decisions have the appropriate knowledge and documentation to help them to make the right decisions. This, in part, is what Data Governance is all about. Sharing, recording and using knowledge about data.

A data-oriented strategic decision can be as major as the definition of a customer, as complex as the decision to use the coding scheme from system A or system B in the warehouse, or as involved as deciding which data profiling tool will best fit in our environment. These are data-related decisions that will have an impact across the organization. These decisions need to be made. Somebody (or some group) needs to be in a position to make those decisions. My suggestion is that the group in mention is the Data Governance Council.

In a Name

Before we jump into the questions, we should consider the name that is used to define a strategic level of decision-making individuals in our organization. These individuals are asked to convene regularly to make decisions by representing their division, business unit, company, and so on. Ask yourself, does one of these groups exist for the enterprise you defined above? Are the individuals of that group data savvy? Could we leverage an existing group or replicate a group-like-that with data savvy individuals that can understand the Data Governance Program and step in to make decisions based on sufficient data knowledge? If they exist, what are they called? Do we really need to change the name? Definitely not! Non-Invasive Data Governance©, the approach I follow, wouldn’t dare request a change to something that is already in place and working.

Ask yourself how you will be able to continue to operate in the “Information Age” without a group of people like that. The rest of this article addresses the questions in the opening paragraph. Hopefully it will steer you down the path of clear definition for your Data Governance Council (or whatever you call them).

Are the Data Governance Council Supreme Beings?

It depends. What do you mean by supreme? The buck stops with these individuals when it comes to strategic decisions being made around the definition, production and usage of enterprise data. The fact that these people are at the strategic level implies that they may not be hands-on involved in day-to-day operations. Often times, they are only as informed about the daily definition, production and usage of data as they want to be or have time for. Some strategic thinkers are very hands-on. Some are less so. This group is typically not involved in day-to-day operations because they have VPs (in some cases), Directors, Mangers, Supervisors and so on that work for them to handle those responsibilities.

In successful data governance programs where I have been involved, only a small percentage of data related decisions (sometimes as few as one percent), get escalated through the operational and tactical ranks to reach the Data Governance Council. By the time the decision making reaches the supreme beings, the knowledge about the issue (cause and effect, source(s), threat, to name a few) should be recorded and prepared for presentation in a to-the-point manner. This is the responsibility of the group that administers the Data Governance Program (the Data Governance Program Team in the diagram above) as well as the operational and tactical stewards involved in decision activity.

Thrones and crowns are not required at the regular meetings. Often the meeting attendees are technologically brought together and, as often as needed, virtual meetings can be held when an issue requires attention when a regular meeting is not close. Timely communications with the Data Governance Council requires significant attention when rolling out a program.

Is the Data Governance Council the Top of the Data Food Chain?

The simple answer to this question is “Yes.” The only level higher, the Executive Level (see the diagram above), includes the sponsors, the senior-most management of the organization, the individuals that are so far removed from daily operations (typically) that they do not have time to become involved in data-oriented decisions. The Executive Level may set priorities and may squash projects and programs that they do not understand (hint!), but the strategic-level decision making decision is often at the next level down, at the level of the supreme beings, or the Data Governance Council.

Why Do We Need a Council?

I define Data Governance as “the execution and enforcement of authority over the management of data and data-related assets.” Of course, this definition can be and has been debated in several settings. The fact is that there needs to be a level of authority over how data is managed. Somebody has to be responsible and accountable for making the tough decisions where the enterprise is concerned. This group should be formal. This group should include all necessary parts of the enterprise. This is not always the case.

Again, it makes sense to have a group of individuals, identified is the answer to the next question that has the top gun responsibility to make strategic data decisions. This group may already exist in your organization without the data component, or with a completely different name. At a recent client, a university, the name of this group was the Administrative Systems Group (implying IT but it was not). Another recent client called them the Data Council, another the Technology Review Board (implying IT and it was).

Search for that group before starting a new group. But you should have a group.

How are Members Identified?

This is typically the easiest question to answer. I typically answer the question – one person per division, business unit, however you break your enterprise down at the highest level. A simple example – At the university I just mentioned, they had five people on their council representing the five divisions of the university (Academic Affairs, Student Affairs, and so on). At a bank the divisions were Human Resources, Finance, Risk Management, and so on. At a manufacturer implementing SAP it was individuals representing the four companies that were being pulled together in the same SAP instance.

Often I suggest that each division provide a back-up or an alternate representative that may or may not have voting (decision-making) capabilities for their division. In the case where the representative could not attend a meeting or become involved in a decision, the alternate has the responsibility for bringing the information forward to the member.

Expect the question “How much time will be required of these people?” The answer to that question is more difficult. In the early phases of Data Governance Program development and roll-out, you may consider holding meetings with the Data Governance Council to explain the basis and drivers of the program, the Non-Invasive Governance© key concepts and best practices, the organization, the policy, … and so on so that they feel a part of the program. In some organizations, the Data Governance Council is asked to or required to approve those items just mentioned.
When a program matures, the Data Governance Council meets regularly with the individuals responsible for administering the program typically setting the initial agenda. Often times, the program administrator chairs the meeting and actively engages all members of the council.

What Does the Data Governance Council Do?

This question is also easy to answer – and to demonstrate that I will give you Bob Seiner’s “List of the Top Eight Responsibilities of the Data Governance Council”:

• Become interested in Data Governance because you recognize shortcomings in how your enterprise manages data
• Become educated in what Data Governance means, how it can and will work for your enterprise
• Become educated in what it means to embrace Data Governance and activate your enterprise data stewards
• Approve things that need to be approved – i.e., Data Policy, Data Role Framework, methods, priorities, tools, …
• Push Data Governance into their areas by actively promoting improved data governance practices
• Make decisions at a strategic level in a timely manner given the appropriate knowledge to make that decision
• Meet regularly (or send alternate), read minutes to stay informed of Data Governance Program activities
• Identify and approve of pivotal data governance roles including cross-enterprise domain stewards and coordinators

Initially this was going to be a top ten list (i.e., Letterman) but the truth is that the Data Governance Council tasks only came to eight. From my experience, I have found it dangerous (to the Data Governance Program and to authority in general) to over-burden the Data Governance Council. As a client once told me, “these people have day jobs.” The idea of the Non-Invasive Data Governance© approach is to get people doing the right thing around the management of data. Sometimes it involves strict authority, sometimes it is just “I need to know the right thing to do.”

Consider leveraging existing organizational structure to handle the responsibilities of the Data Governance Council. Also consider limiting the number of issues that require the Data Governance Council’s decisions. The second point will require additional data governance structure that exists at the tactical level.

The next article in the series (August 2007) will dig into the tactical level of the KIK Data Governance Platform. Please feel free to contact me via email to discuss this article in greater detail or to find out how to implement a Non-Invasive Data Governance program at your organization.

© Copyright 2007 – Robert S. Seiner – KIK Consulting & Educational Services, LLC – All rights reserved.