THE DATA ADMINISTRATION NEWSLETTER – TDAN.com
ROBERT S. SEINER – PUBLISHER

Start With a Good Definition - The Data Will Not Govern Itself

Rather than defining the terms data governance and data stewardship with each chapter of this series, I would like to begin the series by providing my most recent definitions here at the outset and refer back to them throughout. The definitions of data governance and data stewardship that I use have changed, albeit slightly, over the years of providing data governance and stewardship solutions.

These definitions are short and to the point however they are debated and discussed with each use (consulting & education). If you have a definition of these terms - that is good. It doesn't need to be the same as mine - but it should be the same throughout your organization. Consistency of message is important. There are a couple of things that I suggest you (at least) imply in your definitions.

1. Be direct and strong with your words. When the government makes the decision to put a law in place, they do not give you the option as to whether or not you follow the law. Break it and there are consequences. The same needs to be said for Data Governance. I like the word "execution" in my definition since it is solicits the "the definition is too harsh" response and it is necessary to "execute" on a plan. The data will not govern itself.
2. Stewardship is akin to accountability. Organizations understand this. Make certain your definition of Data Stewardship focuses on the formalization of the roles and responsibilities rather then the assignment of these responsibilities. Data Stewards are already there. They may not know they are stewards and the organization may not recognize them as such, but there are already people in your organization that have accountability for the management of data. The challenge is getting them to participate and operate more efficiently and effectively in these data management roles. The data will not govern itself.

Data Governance Definition

"Executing and enforcing authority" sounds pretty tough. And the truth is, it is. My definition is meant to be tough and it is meant to solicit the response. If someone tries to challenge it or questions the language, I use the situation as an opportunity to explain and educate that there is a need for putting "rules/laws that can be followed" in place. I use it as an opportunity to explain that someone high up in the organization has to say that data governance is "not optional" and that they will stand behind the steps that will be taken to make certain the data is governed. Without this type of management level support, the likelihood of establishing a maintainable program becomes slim. The data will not govern itself.

"Over the management of data assets and the performance of data functions" is the back half of my data governance definition. It is obvious that there needs to be executed authority over the management of the assets themselves - structured, unstructured data - but there also need to be executed authority over the data functions as well. The Data Stewardship Approach to Data Governance involves getting the "right" people involved in making the "right" decisions over the management of the "right" data at the "right" time for the "right" purpose. The first step of building a strong stewardship (people-based) foundation includes recognizing (identifying and recording information about) the people in your organization that already define, produce and use the data. After you know who the "de facto" stewards are, you can determine if these are the "right" people and you can start begin providing them with the education, mentoring, forums, communication tools, ... that will enable them to become more efficient and effective in their jobs.

Data Stewardship Definition

"Formalizing accountability" implies a few things as well. To identify who in your organization is already accountable for the definition, production and usage of data is a valuable first step toward applying stewardship to your organization's data. However, it is not the only step.

To formalize is "to declare or making legally valid".1 Once you have identified the data stewards, it will be necessary to make it official. Here are several ways that organization's have made the steward designations official:

1. By creating a system of record of the names of the people and the data that they steward. I often refer to this as the Stewardship Repository or a meta-data base that links information about the people in the organization to the data bases and systems in which the data resides.
2. By writing the steward responsibilities into their steward's job descriptions. This
3. By evaluating the people on their executed ability to be responsible for the data.

This series of articles (chapters) will focus on a practical and pragmatic approach to Data Governance and Data Stewardship. I call it the Stewardship Approach to Governance because, even though Governance begins and ends with policies and procedures, it ultimately depends on the people that are executing on the plan. The stewards are the "eyes and ears" into the data of your organization. They are the people that, in normal course of action, are involved in the definition, production and usage of data. Not only are these people involved, they are THE people that are making the daily decisions about the data. "We need to know who these people are?" or "Who owns the data?" are common war-cry's in many organizations.

Those of you familiar with my Data Stewardship 3-D Approach ( http://www.tdan.com/i022fe01.htm ) know that I focus on identifying the "De Facto" stewards. In the next chapter of this series I will go back and revisit the 3-D approach ("de facto", "discipline" and "database"). The truth is that Data Stewards already exist in your organization. Most companies do not consider "hiring" data stewards to be a best practice because they recognize that new people can not have the level of knowledge that an existing employee has.

1. The American Heritage® Dictionary of the English Language, Fourth Edition copyright ©2000 by Houghton Mifflin Company. Updated in 2003. Published by Houghton Mifflin Company. All rights reserved.