Model 2: Data Stewardship by Function – Data Democratization, Governance, and Security
By Sheila Simpson / February 8, 2024 / No Comments / Amazon AWS Exams, Azure and AWS, Azure Synapse and Its ETL Features, Microsoft Exams, Tools and Examples
Model 2: Data Stewardship by Function
As the name suggests, managing data stewardship by function, a.k.a. organizational data stewardship, focuses on managing data by lines of business or organization departments.
As there are multiple departments organized by function—e.g., marketing department, finance department, etc.—this data stewardship model focuses on the functional area. In our example, the marketing department in the retail industry can consist of data related to campaigns, customer details, promotions, purchasing history, and product details, as well as external data related to demography, customer credit scores, etc. The following are the benefits of this model:
• It is easy to establish a business glossary and data definitions by business function as they understand the context of the data. It becomes easy to assign and align data stewards by business function.There is good possibility of finding knowledgeable business data stewards.
• There is a good likelihood to get the sponsorship and the commitment of time from departments regarding aligned stewards.Data stewardship by function has the following risks:
• Putting a boundary of responsibility of data entities where there are multiple departments that may take ownership results in duplication of efforts and can become political. This may result in contradicting policies, procedures, and definitions of the same data. Getting a clear ownership matrix by entities will help in this scenario.
• In some cases, data stewards may feel reluctant to coordinate across department boundaries. Sponsorship from leadership may help regarding conflict resolution.
• This stewardship model does not work in companies that are looking for initiatives with a singular, consolidated, and standard view of data across the organization. For example, with a global organization,trying to create a singular financial data model for all countries—which have different tax structures, rules, and regulations—would be challenging to tie together within the same department. Strong data governance with a conflict-resolution process would be helpful in this scenario.