Here at IData, we take an expansive view of data stewards, and we group a wide variety of tasks and responsibilities under the rubric of data stewardship. Without conscientious data stewardship, neither data governance nor data intelligence nor effective data management has any real chance of succeeding. So, we think it is important to situate data stewardship at the center of data governance practices, and to recognize that the job description for a data steward is broad.
We have had some recent conversations that reminded us that not everyone has the same understanding, and we thought we would talk about these narrow definitions, and why they are insufficient.
Data Steward as Gatekeeper
"Our data stewards mainly grant access to data," said one colleague to us last week. We assume that means either directly in one of their transactional processing systems, or to an existing report, or on an ad hoc basis as individual requests come in.
While data stewards are ultimately responsible for the security of the data in their domain, this particular description of their duties seems like a great way to maintain silos. If an organization has a robust data catalog and clear rules for accessing and using organizational data, then data stewards should only be responding to requests like this in unusual circumstances. Now, obviously, those organizational rules probably need to be developed and refined by data stewards, so they are not off the hook here.
Another aspect of this mindset is that reporting and analysis are things done by other people once we grant them access to "our" data. How does data then get understood in context? How are data quality issues noticed, much less addressed?
When data is siloed, and data stewards have minimal involvement in producing, sharing, or contextualizing analyzed data, then how do data stewards develop and apply data literacy competencies? At a more basic level, how do data stewards make basic but fundamental decisions about collecting, storing, using, and retaining data?
Business Analysts as Data Stewards
Another organization has named a group of people as data stewards to define terminology and relate definitions to data elements in a homegrown data warehouse. But these people are primarily business analysts, and they do not themselves secure or collect data, nor do they have much--if any--input into these data decisions.
We would argue that these people are really subject matter experts with a technical orientation, or potentially technical data stewards. But they lack the organizational authority to compel adherence, and they lack the visibility to model compliance. Without putting too fine a point on it, there is a pretty good chance that they are doomed to fail in this endeavor.
We love that they are focused on a data warehouse, which is an example of the just-in-time data governance with a customer service orientation that we have long championed. These users are knowledgeable about data, and can undoubtedly serve as a bridge between functional and technical documentation. Their input is absolutely valuable and should be sought throughout any business glossary development. But their exposure to data occurs midway through its lifecycle, and it has a compressed focus.
Data Stewardship as Data Classification
A client shared with us the job description given to their data stewards, and the top duties in that document are to ensure that data is stored safely and classified fully.
Data storage is probably an IT security policy at one end, and a user training/system usage concern at the other. Data classification is important, and it helps protect private data and demonstrate compliance with regulations, but rare are the occasions when classifying data is a necessary step for day-to-day functions.
Of course, if a question comes up about whether a particular data element contains sensitive or confidential information, or whether a data set or data product falls under a certain classification, then data stewards are ultimately the people with the answers. So this vision isn't incorrect, it's just incomplete.
Towards a Holistic Data Stewardship
People do not act as data stewards just for meetings, or just when they are providing data to auditors, or just when someone wants to see some data. People are data stewards every day, and their data-related activities should reflect principles, practices, and privileges of data stewardship.
We have written in this blog before about stewarding data throughout its lifecycle. Data stewards make decisions about data on behalf of their organization: what to collect, how to gather it, where to store it and for how long, when and with whom to share it, and in what level of detail. They also make decisions based on organizational data, so their understanding of data is paramount, their commitment to data quality is critical, and their ability to speak data fluently sets the standard for the whole organization.
If all you ask of data stewards is that they protect data, that they squirrel it away safe from prying eyes, that they do the bare minimum to comply with regulations, then that is probably all you will get. Here's a thought: why not ask for more?
We hope that this blog post gave you something to think about. Additional resources about data stewardship can be found here.
The Data Cookbook can assist an organization in its data governance, data intelligence, data stewardship and data quality initiatives. IData also has experts that can assist with data governance, reporting, integration and other technology services on an as needed basis. Feel free to contact us and let us know how we can assist.
Photo Credit: StockSnap_IVDECL6YQX_datastewardholistic_manholdingworld_BP #B1231
