Sign in   |   Register

How to Build a Great Data Governance Team

  |     |   Bookmark    
Posted March 19, 2015 By Staff     Feedback

One data governance team does not fit all. Your data governance team must fit your organization's business structure and data priorities.

By Michael Collins, BackOffice Associates

The quality of data has become a priority as organizations strive to gain efficiencies in their information systems. As a result, more organizations are building data governance teams to manage data, measure quality and attribute costs to poor data.

According to Gartner, organizations will lose significant money and opportunity due to bad data. Gartner also has stated that more than 25 percent of critical data in the world's top companies is flawed. What steps can you take to address these challenges?

Your data governance organization's charter should be to improve the quality of the data and the processes involved in data management. It is intuitive to expect that better data will result in better outcomes, but the approach to building your governance team and where it fits in your company depends on your business structure and priorities. 

Understand Your Data Demands

Highly effective governance teams tend to include business analysts with an understanding and passion for the effects of data quality on the business objectives. The team could be as small as a single person or comprise large teams. No matter the team’s size, members should understand the requirements and interdependencies of data across organizations.

For example, while it may sound trivial to set up a raw material as a data element, the quality of this data affects many processes throughout an organization. The material must be purchased, maintained in inventory, consumed in manufacturing processes, possibly substituted for other materials, transferred to other locations and planned for replenishment. The importance of these interdependencies cannot be understated. 

We’ve seen customers who disable or abandon their advanced planning initiatives because the outcomes were unreliable or simply didn’t work, or they can’t meet customer demands due to lack of on-hand supplies. Often the root cause is related to inconsistent or unreliable data. Not only does this decrease the effectiveness of supply chain management, but there are ripple effects throughout production planning, and meeting customer demand. Your governance team should understand the demands on the data to support your processes. 

Help Data Governance Teams Evolve

Governance teams often evolve through a maturity process. As teams are formed, the initial steps include documenting the business processes and defining data standards to support those processes. These documents and definitions become the standard operating procedures (SOPs) that the company is expected to follow.

They may be published through the company’s internal communications and collaboration system. However, simply documenting processes and defining data standards does not represent true governance and should not be the end goal for the team. The business must realize value from improved processes and quality data to support the process. 

How do you know if the data supports the stated processes? To accomplish this, data requirements need to be verified, preferably automated by technology.

While not recommended, data verification can be manual in its initial form. When first meeting with enterprise governance teams, we often hear about the manual queries and manual reports they execute on a semi-formal schedule. The teams often aggregate results to create a sense for how "good" the data is based on their individual threshold. Some will accept a "hit" rate of 80 percent compliance, while others will strive to report Six Sigma-level results.

A better strategy is to automate data quality checks via passive data governance. PDG allows the governance team to be proactive in monitoring and measuring data, allowing them to find errors and make corrections before the error becomes a problem and costs money. Many organizations find that some errors need to be prevented in advance; there are different strategies to prevent errors from entering your system.

Data Governance Team Deployment Structure

Your corporate structure and business operations will determine how you deploy your governance team. Typical deployment structures fall into three general categories: global, local or a hybrid of both. 

Many businesses have a global team responsible for data across all business units. Depending on the size and complexity of the business, a single global team may be all that is needed. If the business is small enough, or operating on a single global ERP instance, then a single global data governance team may suffice.

As you consider your organization, you will see where a global team will work and where it may create challenges. Local teams may need to be deployed to help manage data across geographies or within independent business groups.

Like many things in business today, the data governance solution becomes a combination of people, process and technology. Data governance teams are no different.

The people should have an expertise and passion in the business processes and the data to support them. Following reliable and measurable methodologies such as PDG with executable data standards will give the team a proven strategy foundation. Finally, technology will enable the governance team to meet the growing demands of the business without adding too many people.

Michael Collins is a global vice president at BackOffice Associates, a worldwide leader in information governance and data migration solutions, focusing on helping customers manage one of their most critical assets – data.

Previously he has written about passive data governance and about building data assessment business cases for Enterprise Apps Today.

Submit a Comment

Loading Comments...

Thanks for your registration, follow us on our social networks to keep up-to-date