The process of managing data throughout its entire lifecycle, from creation / acquisition to archival / disposition and throughout the lifecycle, make data available and presentable in an intelligent fashion that takes it from pure data to knowledge and wisdom thereafter.
Data in its pure form, regardless of type, structure, source, or creator, goes through a lifecycle that is well known and defined.
The lifecycle involves operations, systems, and people at every stage. If a governance framework does not exist while data is stored, the lifecycle becomes broken since the steps following can not be completed as planned.
If data is stored using the wrong mechanisms, then the use, sharing and disposition of data becomes inaccurate or non-existing. This translates to loss of productivity which translates to loss of revenue, direct or indirect.
The bottom-up approach begins with raw data. Data is first ingested, and then structures, or schemas, are built on top of the data once it has been read. Governance rules, policies and quality controls are also added to the data set at this time. The advantage of this approach is its scalability; however, it can be difficult to maintain consistent quality control across a large volume of data.
In the top-down approach, data modeling and governance take priority and are the first steps in developing a data governance framework. The process begins with data professionals applying well-defined methodologies and best practices to data. The advantage of this approach is its focus on quality control; however, it can be challenging to apply in organizations with a large volume of data
The INFORMATECHS approach is to utilize a strategy and solutions that would allow us to use the bottom-up philosophy while at the same time applying strict quality controls usually found and utilized in the top-down approach.
When applied correctly, a data governance framework would serve as a tool enabling the transition of data throughout the steps of the pyramid in a business-oriented fashion that is controlled, audited, and secure, so that it can assist end users in their daily operations.
Since data governance depends greatly on the exact needs of the organization trying to deploy it, there is no one-size fits all approach to data governance. However, there is a set of best practices that can be utilized to reach the best possible framework.
1.
2.
3.
4.
5.
Accuracy, completeness, relevance, timeliness
Privacy, access, security, roles, Privacy Impact Assessments, User Consent/Disclosure
Common identifiers (e.g., PRI, DOB, Name, etc.), full applicant profile, “official” list of departments, etc.
Tags used for retention, disposition; tags used to describe content and format, attributes to help understand data
Data Inventory, Asset lifecycle planning, reports and data used internally and externally, client centered needs
Usefulness and purpose of data, ensuring data can be used across many systems for multiple user needs
Establishing standards and “step-by-step” procedures; systematic creation and iteration of building blocks used in queries and reports
Data Lake, Analytical Hub, Open Government Portal, etc. (technical infrastructure to support storage, staging, analysis, self-service, innovation), client-facing (needs-based) requirements
Data mapping, understanding flow of data
No strategy would work without the direct involvement and contribution of an organizations largest asset, the users
As per the Data Governance Institute, the following are the 10 universal components of Data Governance. A framework will contain all of them, with varying degrees and emphasis, based on organizational needs and requirements:
1. Mission and Vision
2. Goals, Governance Metrics
3. Rules, Policies
4. Decision Rights
5. Accountability
6. Controls
7. Data Stakeholders
8. Data Governance Office
9. Data Stewards
10. Flexible Data Governance Processes
INFORMATION, AUTOMATION, INTELLIGENCE