Fundamentals of Data Stewardship
If organizations must have Data Stewardship, but struggle with deploying valuable Data Stewardship programs, how can companies mitigate the challenges and get better results faster?
A recent report reveals companies’ growing understanding in the value of Data Stewardship. The report shows that around 39 percent of companies have been practicing Data Stewardship from 1-5 years, while only 7 percent said the role of data stewards “is recognized and maintained.” An overwhelming 90 percent said that both Data Governance and Data Stewardship had more interest and were more important today than 10 years ago. Despite this growth, businesses lack both satisfaction in their Data Quality and energy in extracting data, as 68 percent said they were either neutral or disagreed when asked if they had a high level of trust in the quality of their data assets.
Andy Hayler, CEO of the analyst firm The Information Difference, said it becomes: “Hard to answer really basic but important questions—such as who are the most profitable customers and what are the most profitable channels—when different parts of the business may be classifying product, customer, and channels in different ways.”
No wonder some organizations try to be data-driven and get the most of their data assets yet see initiatives crumble or die. Freddie Mac took four or five attempts before getting a successful Data Stewardship Model. Andrew White of Gartner said: “More and more organizations are adding the role ‘information steward,’ but they are not asking for the right solutions of the vendors to help them do their job. Demand and supply is misaligned.”
What is Data Stewardship?
The Data Management Body of Knowledge (DMBoK) defines Data Stewardship as:
“The most common label to describe accountability and responsibility for data and processes that ensure effective control and use of data assets. Stewardship can be formalized through job titles and descriptions, or it can be a less formal function driven by people trying to help an organization get value from its data.”
This means that Data Stewardship falls under both strategies (plans of action designed to achieve an overall aim—or the “what”) and tactics (actions planned to achieve a specific end—or the “how”). The combination of Data Stewardship’s strategy/tactical decision-making patterns form models and frameworks. Data Stewardship can come under a formal Data Governance program or a Non-Invasive Approach™ (coined by Robert Seiner), where governance is applied to existing processes rather than redefining all of them. Either way, strategies and tactics need to blend into frameworks.
In his webinar Getting (Re)Started with Data Stewardship, Peter Aiken describes frameworks or models as a:
System of ideas for guiding analysis
Means of organizations projecting data
Means for prioritizing decision
Means of accessing progress
Case for continued funding
Like martial arts, Data Stewardship operates within different frameworks. Karate, kung fu, capoeira,a and Vigny’s canne de combat exemplify different systems. While there are some shared principles (e.g., deriving power from the core muscles) each requires a different mindset and consequently, different techniques. Likewise, Data Stewardship can be described by different systems. Data Stewardship success depends on the framework choice and implementation to the system strengths and constraints. See the examples below.
Ready, Set, Go
Jenny Schultz, from Freddie Mac, described how Freddie Mac developed the Data Stewardship framework “Ready, Set, Go,” to leverage data for competitive advantage and enhance business value. She included business and executive buy-in to this Data Stewardship, into the overall objective. Strategy/tactic sets include:
Ready: Laying the groundwork and establishing the program. During this stage, Schultz’s team consulted stakeholders, co-workers, the internal audit department, compliance, and the privacy office to determine their pain points. Much of this phase involved listening closely and narrowing down challenges to address. Also, at this time, Schultz’s department brought credibility and communicated about Data Stewardship through branding.
Set: Identifying and working with stewards and stakeholders. This stage can take a long time (say, 90 days). During this period, the company formed working groups to create data standards, including Data Stewardship roles and responsibilities. In times of conflict Schultz’s supervisor made the final call as needed, based on information presented. She set up a budget demonstrating business value/value standpoint.
Go: Executing the program. The work involves knowing who needs more “data therapy” by identifying and supporting detractors and sponsors. During the Go phase, Schultz maintained that top-down support, so the C-suite and VPs were on board. Go also means engaging in ongoing planning every three years, to assess, rethink, and update strategy/tactics across the system.
Fire Station Model
Peter Aiken’s Fire Station model acknowledges that many organizations spend some percentage of time fighting fire and cleaning up after them. This requires reactive and proactive activities. For example, keep the house from burning and save a patient’s life, but also plan in between calls. The author uses the example of a relative, who is an EMT. She spends time checking and addressing vehicle functioning, checking in with coworkers, restocking and updating supplies, training, and practicing. That way, the EMT has what she needs when she goes on a call. She also reports to a chief and a manager who organize the fire department. They report to a larger funding board, e.g. town boards or regulators.
How does this model work with Data Stewardship? Consider the fire station as the Data Architecture, bridging business strategy, and technical execution. A well-designed Data Architecture, like a well-constructed firehouse, is key to disaster response. In this model, people have specific roles, including the chiefs (CFO, CRO, CMO) and VPs. The chiefs guide the funding. The equipment is composed of data, facts, and information, as described below.
Strategies and tactics to fight fires and prepare well come from a combination of Data Strategy, Data Governance, and data stewards.
In the fire station model, progress can be measured by assessing the triage of data and information needs, the readiness of the equipment and the team capability to get the job done. Also, considered is the response time to successfully avert disaster and keep the business living and whole.
The Culture Campaign
Mary Levins, Data Governance Principal at Sierra Creek Consulting, and Cassie Elder, Co-founder, Principal, and Data Strategist at DataCraft Partners, initiated the Culture Campaign. This framework puts business culture front and center, supporting its aims without compromising security.
“Respecting culture across the Data Governance journey means considering factors such as people, processes, and shared values, beliefs, and norms.” These tend to be overlooked by many companies and comprise a huge stumbling block when implementing Data Stewardship.
As a first step, Levins and Elder advise an organization to learn about its culture type through observing behaviors in adherence to the corporate mission statement and conducting/reviewing employee surveys. Then the company gets a sense of the data culture personality. Levins and Elder describe two dimensions: collaboration—emphasizing harmony and team synergies—and competency—stressing the best talent and pride in the work done.
From understanding a business data culture, Data Governance, and Data Stewardship strategies/tactics can be defined. For example, in a high-collaboration culture, Data Stewardship needs to be designed around keeping things moving and the number and quality of decisions made. In a competency culture, Data Stewardship needs to be designed around better communicating real-world work results and engagement with people. This type of culture tends to frown on perceived make-work or pointless exercises.
The vendor erwin combines a cultural campaign with their technical solution erwin DM NoSQL as Data Governance and Data Stewardship stretches beyond technology to a cultural shift. Progress aligns with The Leader’s Data Manifesto, which was developed, in part, to develop cultures that put data on par with other assets.
FAIR: Findable, Accessible, Interoperable, and Reusable
FAIR describes a system in operation by the National Institutes of Health (NIH) through its Science and Technology Research Infrastructure for Discovery, Experimentation, and Sustainability (STRIDES) Initiative, part of NIH Common Fund’s New Models of Data Stewardship Program. This framework aims to speed up discoveries in biomedical research.
Usage limitations come from disconnected, incompatible, inaccessible, and expensive large data sets, says the NIH. Making information findable, accessible, interoperable, and reusable, the Fair Standards, promotes new scientific research.
The NIH is testing strategies and tactics with three data sets. STRIDES states:
“More details regarding broader availability of the efforts created by these partnerships are still evolving and will be shared after a series of pilot programs to refine policies and procedures for the initiative.”
Conclusion
Think of Data Stewardship like martial arts: while there is some overlap in best practices, approaches differ, depending on philosophy and strategic/tactic combinations. Each system has its strengths and weaknesses. Companies have many options when determining their best matching framework for Data Stewardship.
This article is authored by Michelle Knight and originally published by Dataversity. We received permission from the author to republish the article here for the ADCG community.