Bob Seiner thinks your Data Governance definition needs to have “some teeth behind it.” According to Seiner’s definition, Data Governance is: “The execution and enforcement of authority over the management of data and data-related assets.”
He prefers strong words like “execution,” and “enforcement of authority,” to make it clear that the Data Governance program is not just an optional add-on, but is a fundamental part of the enterprise.
Seiner is an author, an industry educator, a DAMA award winner, and the President and Principal of KIK Consulting and Educational Services, home of Non-Invasive Data Governance™, while he also publishes The Data Administration Newsletter (TDAN.com), an online publication that covers the entire scope of the Data Management industry.
In a recent DATAVERSITY® interview, Seiner shared best practices for starting a Data Governance program.
Engage Support of Senior Leadership for Governance
The first best practice that Seiner discussed is the importance of having senior leadership that support, sponsor, and understand the activities of Data Governance:
“If they don’t support, sponsor and understand Data Governance, the program is going to be at risk, because they’re not going to give you the time to do it. If I told you I used this best practice with 99 percent of my clients, I’d be lying, because it’s 100 percent.”
Senior leadership needs to “get Data Governance,” said Seiner. If they don’t really understand what the governance team is doing and how the program works, they’re not going to provide the essential support and they’re not going to continue their sponsorship in the long run.
Another best practice states that it is essential to dedicate staff resources for Data Governance. It’s necessary to find somebody who can run the program, who is capable, has relationships, and can basically administer the program, while also getting senior leadership on board. “It’s not going to run itself,” said Seiner. Most importantly, “you need to have somebody responsible for this stuff, and accountable for this stuff, and who has time available to do this stuff, because without that, you’re going to be at risk immediately.”
Seiner’s Non-Invasive Data Governance™ approach fosters buy-in from senior management while introducing Data Governance with as little disruption as possible. Rather than creating a Data Governance program and forcing employees to comply or adding governance tasks to the workload and hoping employees make time for them, Seiner’s method builds a Data Governance program based on existing roles and responsibilities.
“If you want complete coverage of your organization, Non-Invasive is really the only way to implement Data Governance. Otherwise, you’re going to be assigning every single person something that they don’t feel like they already have, and that’s not going to be readily accepted,” he said.
Within the Non-Invasive approach, Data Stewardship is defined as “the formalization of accountability over the management of data and data-related assets.” Data Stewards are recognized for their relationship to data rather than being assigned new responsibilities and are held accountable for those relationships.
“Somebody who defines data is going to be held responsible for how they define data – and that includes the metadata that’s necessary for that data definition. Somebody who produces data needs to know how that data is going to be used, so they have a role. Somebody who uses data has to follow the rules associated with usage of the data, whether it’s protection of data or compliance.”
Readiness Assessment: Developing a Roadmap
Seiner recommends a Readiness Assessment that compares your existing Data Governance practices with industry best practices. He suggests asking these questions:
- What are you presently doing that supports each best practice?
- Where is there an opportunity to improve associated with that best practice?
- What’s the gap between what we’re doing and the best practice?
- What’s the risk associated with that gap?
This process will help create a list of problem areas and can help uncover risks that can be mitigated by applying good governance. The Readiness Assessment often shows areas where roles and responsibilities associated with Data Governance need to be better defined and can serve as an action plan or a “roadmap for what your program needs,” he said. The Readiness Assessment will also uncover gaps in communication that can be used to develop your Data Governance Communication Plan.
Non-Invasive Governance: Ownership Without Tears
Seiner says that there is often resistance to Data Governance, based on assumptions grounded in a lack of understanding about the options available for implementation.
“Management thinks that Data Governance is going to be difficult, it’s going to be hard, it’s going to be complex, it’s going to interfere with things that are presently being done, and the truth is, it does not have to be that way. If you take a Non-Invasive approach, you’re looking at what responsibilities people already have and you’re not necessarily handing them things that are going to feel new to them.”
Senior leadership may also have resistance to Data Governance and it’s important that they understand that in some fashion, governance is already taking place:
“It’s just very informal. It’s very inefficient. It’s very ineffective. To take the Non-Invasive Data Governance™ approach is to formalize, to make efficient, to make effective the levels of authority that are already in place in your organization.”
Don’t Create Data Stewards, Recognize Them
Rather than assigning someone to be a steward and adding governance to their workload, formally recognize people for their relationship to the data and for what they already do or should be doing, he said. “You’re not going to go into a room with 100 people and point at ten [of them] and say, ‘You know, those ten people, they need to protect the data,’ when all 100 people are using the same data and have the same responsibility.” Anybody who uses data needs to understand how that data can and can’t be used.
The ‘teeth’ in the program, aside from supporting compliance for Data Governance practices by all users, provide support and confidence for Data Stewards because they know as they recognize that they have greater accountability, they are backed up by leadership, and a meaningful, enforceable policy
Data Stewards: It Takes All Kinds
Another tenet of Seiner’s Non-Invasive approach is that there can be more than one type of Data Stewards. Operational Data Stewards are people who define, produce, and use data on a daily basis. There can also be Tactical Data Stewards, that have responsibility for a particular domain or subject matter and they have an increased level of responsibility for that data as it pertains to the entire organization, he said. The title ‘Steward’ acknowledges a formal relationship to and responsibility for the data.
“It’s not a new title or a new role. The most difficult step for an organization to take is to formalize a role or identify a person or a group of people that have that increased type of responsibility.”
Tactical Data Stewards are going to facilitate conversations around a specific domain of data and work with others that are stakeholders in that data while their responsibility is to the specific data domain and not their own specific use of the data. Finding an individual capable of being responsible for a domain across the enterprise can be very difficult for some organizations, he said.
“I have a customer right now who is identifying that person for customer, product and vendor data. Think about how hard that is, to find somebody who has that knowledge or is willing to be the go-to steward person for that subject matter of data.”
If a domain is too much for one Data Steward:
“There might be sub-domains of a subject of data that may require Sub-Domain Stewards. Under ‘customer,’ there might be demographic information, financial information, transactional information, so you might have a sub-domain steward for each of those sub-domains.”
The Role of Data Governance in the Organization
Data Governance and Data Management should ideally fall under the purview of a Chief Data Officer (CDO) or someone in a similar position, but not all companies have a CDO. Size and corporate structure will have an effect, Seiner said. “In a smaller company, you’re not necessarily going to have Domain Stewards responsible for sub-domains. It depends on the organization.”
A lot of support is available, “if people take responsibility for educating themselves.” And that education piece is the key that brings the entire program together. Stakeholders at all levels need to understand the purpose, intent, and necessity of the Data Governance program for it to work and continue working well into the future.
Photo Credit: Hilch/Shutterstock.com