Implementing an enterprise-wide data governance program is not a trivial exercise. New methods for handling data can be met with resistance from various parts of an organization. It can be very challenging to attain the necessary level of cooperation from everyone involved in the program. Without this cooperation, the data governance program will not provide the intended results and may simply fail altogether.
An important point when discussing data governance is that it is an initiative designed to modify the behavior of the people who interact and use enterprise data. The data remains the same, but the way it is used is modified through data governance. Defining data governance as the execution and enforcement of authority over the management of data and data-related resources highlights the importance of people in the process.
Three Approaches to Data Governance
Selecting the right way to implement data governance in an organization can have a tremendous effect on the program’s success or failure. The corporate culture and structure need to be considered when determining the best way to approach data governance implementation. Psychology may play a major role in the employees’ acceptance or resistance to the program.
Command and Control
The command and control approach to data governance is a top-down methodology that is considered the most invasive type of data governance. Data stewards and data owners are assigned by the leaders of the initiative and are essentially told that they will do what the organization decrees as necessary. Only a subset of the enterprise’s population is involved in data governance, making it difficult to cover the entire company.
Problems that can occur when a command and control approach is chosen usually revolve around the assignment of new duties to employees who are potentially already overworked. People may consider the new responsibilities to be above and beyond their job description and not give them the attention they deserve. Without the proper buy-in by the parties involved in the data governance program, it is doomed to failure.
Traditional Approach
The traditional approach that many companies use when implementing data governance is similar to the command and control method with a few subtle, yet important, changes. Rather than assign individuals as data stewards and owners, they are identified and selected. The rationale behind their selection is that their help is needed to help the organization by implementing new data handling control procedures.
This approach can run into the same types of problems associated with the acceptance of new roles and duties for which individuals have been selected. The leaders of the initiative hope that the people selected will understand the importance of data governance and give it their full attention. The traditional approach also restricts data governance responsibilities to a subset of the organization.
Non-Invasive Approach
The non-invasive approach recognizes people as data stewards by their relationship to the data. Everyone in the organization is involved in this method of implementing data governance because everyone in the enterprise uses data in some way. In many cases, the non-invasive approach seeks to improve the way people are already using data by defining roles and codifying procedures.
Since the new responsibilities revolve around doing a better job of handling data they were already using, there is reduced pushback. With everyone involved in the process to some degree, nobody feels singled out for additional duties. This approach can foster a sense of shared responsibility that gives the program a greater chance of success
When viewed as a people-centric program, the difference in the terminology used when defining roles can have a significant impact on their acceptance by the individuals involved. Most people are more comfortable with being recognized for their potential contributions to data governance rather than being assigned or selected for a new role. This is where the corporate culture and psychology may come into play. You need to adopt an approach that is acceptable to the organization. It may be that a hybrid solution that combines elements of the various methods is preferable for your enterprise.
Focusing on the Important Components
Six core components of a successful data governance program need to be reconciled with the strategy selected for implementation.
- Data is the prime currency and determination needs to be made regarding which data is most important to who and to differentiate between data, info, records, and knowledge.
- Roles need to be defined and how that is done can have a big impact on the effort needed to implement governance successfully.
- Processes define how the roles are being applied to business activities.
- Communication is necessary to formalize accountability and educate the organization regarding rules and policies.
- Metrics are used to measure the impact of the initiative. There may not be a direct ROI associated with data governance but its effects will be felt in many areas of the enterprise.
- Tools are used to formalize accountability and improve the knowledge of rules and processes.
Tools to Help Implement Data Governance
An IDERA Geek Sync Webcast presented by Bob Seiner delves more deeply into the different approaches to data governance. Bob uses his experience and unique view of data governance to educate the audience on the best way to implement a successful program in their organization. He provides a sensible framework that can be used to identify the best way your enterprise should move forward when putting together a data governance program. It’s essential viewing if you are considering getting data governance working in your company.
Collaborative tools are an essential component of data governance. ER/Studio Enterprise Team Edition provides teams the ability to create multiple data glossaries and definitions and easily share them across the organization. Business concepts can be represented with full documentation and metadata can be cataloged to assist the governance program. The tool helps build the shared data language that forms the foundation of a strong data governance program. It can be used with whichever approach your organization selects as they embark on their data governance journey.