Organizations struggle with how to determine the appropriate level of authority to give to the people that run their data governance program. If your data governance team does not have the authority to tell business and technical areas that they must participate in governed data practices, you may consider following a federated method to implement data governance.
Notice that I called “federated” a method rather than an approach. A method refers to how the data governance program will be operationalized across the organization, rather than the philosophy behind how the program is designed.
What does it mean for something to be federated? My favorite dictionary definition (freedictionary.com) of the term federated is “to cause to join into a league or similar association.” The term federated focuses on bringing together. A league of sports teams could be considered a federation because they are autonomous bodies that are governed and guided by a central office. DAMA International could be considered a federation as it includes local chapters that are provided guidelines and that are governed by the international office.
In many situations, the central Data Governance Office, often referred to as the DGO, doesn’t have the authority to tell the business or technical participants that they must follow procedure. If your DGO has a higher level of authority, the method you are following is not that federated.
Federated Data Governance begins with a central DGO, or a Data Governance Team (DGT) or, even just a single person as a Data Governance Manager. This person or group of people is responsible to somebody for data governance in the organization. The size and complexity of the organization often influences the size, and often the name, of this central facilitation body.
The central body is focused on consistent and thorough governance across the entire organization. This is where the difficulties with this method start. The first task at hand is to address the following questions:
- How do we convince leadership that the federated method makes sense and that a central body is necessary?
- What value will the central body add to the organization, and how do we convince the business and technical areas (and their teams) of the same?
- What support will the central body provide to the “teams” as the “teams” take on the primary responsibility for their own governance?
Be prepared ahead of time to answer these questions when you suggest that the Federated method is the best method for your organization.
Why the Federated Method?
A successfully federated Data Governance Program requires that the central DGO understand the needs of entire organization and keeps them under consideration when selecting what the DGO will provide. The question becomes: how can you tell if you are considering the needs of everybody? Learning these needs is a result of gaining knowledge through conversations with people involved in many different facets of the organization (often projects, programs, and processes). You can only know the needs of the people that you speak with. The truth is that you will have a difficult time trying to match the needs of absolutely everybody.
So … What do you do then? The DGO takes on the responsibility of providing things that have been recognized by other organizations as important and effective in governing data as an asset. Federated Data Governance becomes a set of shared services that the DGO provides for business and technical areas across the enterprise.
Assistance from the Central Body
It is important to define the types of assistance that the central body will provide to the teams. There are several ways assistance can be provided. Each method of assistance requires significant planning and development before the assistance can be provided. The ways to assist can fall into several categories. The main categories that I have experienced have included:
- Data Governance Thought-Leadership / Education
- Enterprise Operating Model / Roles & Responsibilities
- Program Guidance / Facilitation
- Best Practices / Maturity Assessments
- Internally Developed Tools / Templates
- Enabling Technologies
- Standard Processes / Conventions
- Organizational Policy / Guidelines
- Communication Planning / Delivery
- Business Glossary and Metadata Management
Data Governance Thought-Leadership / Education
This type of assistance takes the form of research, development, and communications of the appropriate approach to govern data for your organization. The approach to data governance can be focused on 1) implementing command-and-control over the data, 2) implementing an optional program that you hope will be followed, or 3) implementing a non-invasive approach where accountability is formalized based on people’s relationship to the data (as definers, producers, and users).
Enterprise Operating Model / Roles & Responsibilities
This type of assistance includes the development of a sensible and understandable set of roles and responsibilities focused on governing the definition, production, and usage of data and that can be replicated across the organization at all levels. Therefore, it is commonly called an Enterprise Operating Model when using the Federated method.
Program Guidance / Facilitation
This type of assistance includes utilizing the Enterprise Operating Model of roles and responsibilities based on formalizing the accountabilities for the data relationships mentioned above. Definers must be helped with following the rules associated with defining data. Producers must be helped to understand the impact of the data they produce. Users must understand and follow the rules associated with governing and handling the data. These types of expectations require that someone guide and facilitate the organization toward formal data governance.
Best Practices / Maturity Assessments
This type of assistance includes providing the capability to conduct cross-enterprise (or individual part) comparison to industry or customized best practices focusing on lowering the risks associated with starting a data governance program and governing the most critical data in the area. Often these assessments result in maturity models and a detailed road map for implementing data governance in that part of the organization or the organization as a whole.
Internally Developed Tools / Templates
This type of assistance includes providing proven tools and templates, often in the form of internally developed instruments through Visio, Excel, and SharePoint, that help the federated teams to self-govern their data while staying consistent in the context of the enterprise. Tools and templates that provide this assistance include the common data matrix, governance activity matrix, communication plan template, questionnaires, and ways to measure the value that is coming from your data governance program. I have written in previous articles and spoken in my webinars about some of these templates. Please send me an email if you are interested in examples.
This type of assistance focuses on leveraging and enabling existing and new technologies that will assist you on the road to formally governing your data. Data Modeling tools can be leveraged to enforce business definition rules. Data Dictionary and Glossary tools can be leveraged to improve enterprise understanding of data. Data Quality tools can be used to profile and manage the quality of data in your information systems. Metadata Management tools can provide the central hub that is required to make the metadata most useful to your organization. A centralized Data Governance Office can aid in consistent and quality implementation of all of the tools that already, or will exist, in your environment.
Standard Processes / Conventions
This type of assistance includes the facilitation, development, and enforcement of standard data processes that can be used within a business function or across business functions. Standard processes include data quality issue resolution, request for access to data, and the managed integration of data from different parts of the organization. Conventions include naming conventions, definition conventions, standard use of business objects, and the rules to protect sensitive data. It is the responsibility of the DGO to document, record, and share standards, standard processes, and conventions.
Organizational Policy / Guidelines
This type of assistance focuses on assuring that the proper levels of policies and guidelines are in place and are made available to support the enforcement of authority over the data. Governance policies and guidelines often focus on regulatory compliance, the protection of sensitive data, and sharing data to improve effective use. The DGO or similar central body can provide templates for new policy, process for gaining approval of policy, assurance of access to policy, and assurance that the policies are communicated effectively across the enterprise.
Communication Planning / Delivery
This type of assistance includes the development of a communications plan that addresses the business and technical interests of the enterprise. Typical communication plans make certain that there is appropriate level of orientation to data governance, onboarding of people actively involved in processes associated with governing data, and on-going communications focused on delivering successful governance to all levels of the enterprise. The central body or DGO typically takes on the responsibility for providing consistent and quality communications about data governance across the enterprise.
Business Glossary and Metadata Management
This type of assistance focuses on helping the organization to improve its understanding, linkage, and lineage of the data from a cross-enterprise perspective. Organizations focus on business glossaries, data dictionaries, and/or metadata repositories, depending on their level of needs and requirements for managing the metadata associated with their most critical data. The DGO is often at the heart of developing and delivering the strategy, approach, and implementation of tools associated with governing data as an enterprise asset.
This brief article describes the reasons why organizations select to follow the Federated method of implementing their Data Governance programs. This approach has been shown to provide value to the organizations that develop strong Data Governance Organizations that offer multiple levels of assistance to the “teams” of people in the organization that can see value from data governance.
The Federated method doesn’t work perfectly in every organization. I started the article by describing what causes an organization to look at this method. Please feel free to share the method and approach that you used to implement Data Governance in your organization.