This article is part of an Essential Guide, our editor-selected collection of our best articles, videos and other content on this topic. Explore more in this guide:
1. - Best practices for a data governance program that works: Read more in this section
- Prioritize communication and purpose for effective data governance
- Five best practices for data governance from experienced users
- With enough grassroots support, data governance programs can thrive
Explore other sections in this guide:
Most data governance programs continue to be plagued by a lack of high-level executive interest and the resulting lack of funding, but proponents of the discipline say there are ways to combat the problem and eventually win greater support.
A recent SearchDataManagement.com survey of technology professionals found that about a third of organizations aren’t addressing data governance at all. And companies that are doing so tend to do it at a highly informal, departmental level.
Meanwhile, the people charged with enforcing data governance policies and procedures – data stewards – are often people with lots of other duties. About 39% of organizations that are addressing data governance report having only informal data stewardship roles in place.
According to technology analysts, vendors and end users, most of the problems associated with enacting “enterprise-wide” data governance policies and procedures are a direct result of an ongoing lack of executive interest and support.
They add, however, that data governance programs will persist, even if they’re starting from the bottom levels of an organization and occasionally producing poor or mixed results. They also point out that there are some steps to take that can increase levels of support and, ultimately, the success of a bottom-up approach to data governance.
“I think that oftentimes, practitioners want to start too big,” said C. Lwanga Yonke, an information quality professional and an advisor to the International Association for Information and Data Quality. “I think it’s important to define a specific problem one is trying to solve [and start there].”
Yonke added that data management professionals trying to get a data governance program off the ground should also take a hard look at the corporate culture – or, on a smaller scale, the departmental culture.
“First I would look at the organizational structure: How is the company structured?” he said. “How is it organized? I would look at the key data the company creates, uses and provides. What are the major data flows? If you don’t try to change the corporate culture about how the company sees data, then you will fail. I’m sorry.”
Support for data governance programs must cross boundaries
It’s true that executives today are highly concerned with budgetary issues and a weak economy. But even if they weren’t, data governance would probably still be a tough sell, said Rob Karel, a data management analyst with Cambridge, Mass.-based Forrester Research Inc.
Karel, who has researched data governance extensively, says there is almost an intrinsic feeling among executives that the discipline simply isn’t critical to success. And that feeling will almost always trickle down throughout an organization. The only people immune to the syndrome will be the database managers, data architects and others whose jobs are directly affected by untrustworthy data, he added.
The result, according to user interviews, is that data governance programs – if they exist at all – tend to be informal, bottom-up efforts that are focused on small areas and often get swept aside when “more important” issues pop up.
Vendor-sponsored research into customer behavior also backs up these findings. IBM’s Initiate Systems Inc., a Chicago-based provider of master data management (MDM) software, recently conducted a data governance customer survey and got similar results. IBM acquired Initiate last February.
Initiate found that most organizations have no formal data governance program, and those that do often produce ineffective results owing to mediocre policies and guidance. The Initiate survey also found that data governance programs are generally being led by mid-level IT managers and -- perhaps not surprisingly -- most executives beyond the CIO have little interest in the discipline.
“The thing that jumped out at me the most was that these programs are still fairly bottom-up efforts [that are] initiated by data architects and people that believe in data governance and data stewardship,” said Marty Moseley, chief technology officer at Initiate Systems. “But they still don’t know how to frame this from a value perspective where business leaders would see value in it.”
Forrester’s Karel points out, however, that the "success factor" may depend on getting executives to see the value of data governance. When executives see that the methodology actually is critical to success, he explained, data governance programs tend to be more successful.
“Paint that line of sight which says [that the organization needs data governance] to be successful because it is highly reliant on trustworthy data and most of the breakdowns and inefficiencies and risk exposure comes from bad data,” Karel said. “It’s also about actually making them recognize and understand how data governance is going to make them [personally] successful by allowing them to make better decisions more effectively and more efficiently.”
But for steadfast proponents of data governance initiatives, it’s not enough to get just one executive or manager on board, because data flows across all departments, geographic locations and information silos, Karel explained.
“For data governance to truly be successful,” he said, “you need a level of sponsorship and support and commitment that crosses these silos.”
National Life Group is one organization that is currently in the process of enacting a data governance program that crosses institutional boundaries. The Montpelier, Vt.-based collection of insurance, annuity and fund management companies began a major data integration and MDM project about two years ago and says that data governance is fundamental to the plan.
While analysts tend to recommend that organizations enact data governance programs before going forward with an MDM project, the group opted to do both at the same time, according to Jarugumilli Brahmaiah, enterprise data architect with National Life.
National Life now governs data by subject area. For example, financial data governance is governed by financial workers, and production data is governed by production workers. The company also has cross-functional boards and collaborative meetings that incorporate workers from different areas. Meanwhile, the company has no single data steward. Instead, each functional area appoints its own.
Brahmaiah said this approach -- while not entirely formal -- helps ensure that the people governing information really know their stuff.
“They not only know the data,” he said, “they also own the data.”