Data Governance provides a framework for dealing with the challenges around data compliance and regulation which aids in the management of the availability, usability, integrity, quality, consistency, and security of the data employed in an enterprise and helps organizations meet compliancy with legislative laws, regulations, and mandates.
Data Governance Framework
Organization :
- It requires representative participation and commitment from both IT and business stakeholders. Senior level executive sponsorship from both areas and active consulting practices to drive and champion Data Governance implementation.
- A dedicated and active Data Governance Board with having oversight of data assets that exist across the enterprise that is sanctioned through an approved charter that defines the board’s scope, objectives, authority, organization, procedures and measures
- Role of Data Gov Board :
- To set and authorize the direction of the Data Governance initiative
- Align business and IT goals
- Manage organizational data as a strategic asset
- Drive business priorities and compliance with regulatory mandates
- Define roles and responsibilities for data owners
- Create data policies, procedures and standards for the organization as a whole
- Direct how data should be used, managed, and monitored across the organization
Policies, Principles & Standards:
A policy must be developed for enforcing data standards and governance procedures that specify: who is responsible and accountable for various segments and aspects of the data, including its:
– accuracy
– accessibility
– consistency
– completeness
– how it is updated
RELATED READING:
Data Integrity & 21 CFR Part 11 Compliance Requirements
FDA’s Dec 2018 Guidance on Data Integrity
Processes, Practices & Architecture:
- Processes must be established and formalized to guide principles for how policies, processes, and standards are created, collected, modified, implemented, and distributed across the organization
- Without formalizing the process, IT will constantly find itself having to demonstrate its value add to business stakeholders
- Setting formal processes and practices helps identify and document how the organization manages its data
- The organization must define how the data is “to be” stored, archived, backed up, and protected
- Practice and procedures are also instituted to ensure compliance and government regulations and audits are met
- Data Governance processes and practices help organizations face challenges of enterprise level data integration concerns and include enterprise standardization for data and systems
Data Architecture:
- Addresses how the data is to be organized and integrated
- Includes enterprise data standards, data models, data flow diagrams, mapping spreadsheets, data definitions, and a metadata dictionary, in addition to security and privacy measures
- Is essential for determining requirements and preparing the organization for efficient and effective data integration
Data Integration:
- Involves the process of cleansing, transforming, merging and enriching data that is merged from multiple sources
- Addresses error handling, scheduling, process restart capabilities, data administration, gaps in data and audit
- Ensures data is integrated in the timeframes required by the business and outlined in the SLA
Data Cleansing:
- Identifies data model schema differences (data types, length, value)
- Validates rules based on business user roles and processes
- Recognizes duplication of data, behaviors, and functionality
Data Quality:
- Involves problems with incorrect and/or inconsistent data
- Requires creating and managing data models from the source system
Requires creating enterprise standards - Can be aided using a data profiling tool to allow for:
- Data to be assessed, identifying cross-system data overlap and making sure the data is consistent
- The collection of metrics and statistics that track the effectiveness of Data Governance across the enterprise
- Continuous improvement
MetaData:
It is structured information and business rules about data. It allows for an examination of duplicate definitions, dissimilarities of definitions, and identifying consistent inconsistency. It becomes information knowledge sharing where definitions, data types, entity layouts, and domains are published
This generally includes:
- data lineage
- business rules
- business term definitions
- ownership/stewardship
- transformation rules
- data mapping
- source systems
- structures of data
- systems of record
- data currency
- data access
When establishing a data model repository for metadata, keep in mind differences that may exist across the organization:
- Different applications and systems have been built using various platforms and databases
- The data contained across the different applications and systems might not be stored in a standardized method
- There might be: different meanings, different data types, different naming conventions and also various definitions may be applied to the same term.
- Some information may be captured in manual format
Investigation & Monitoring:
- Identify the data quality issues
- Prioritize the issues based on urgency, importance, dependency, and critical success factors
- Conduct root cause analysis to determine and identify the probable cause of the data issue
- Formulating a corrective action plan
- Decide on the next steps
- Implement the fix
- Monitor the results
Gap Analysis:
- Focuses on mapping the organization’s governance policies and processes against industry standards and best practices
- Allows the organization to have an understanding of where their organization is, what their target needs to be, and addresses plans to get there
- The Governance Board can then:
- Plan to identify the tools and technologies needed
- Make recommendations for best approaches in creating a standardized model that accommodates the requirements and organizational strategic objectives and initiatives
- Determine requirements for a “to-be” data architecture and enterprise information model, business impact for implementing Data Governance
- Identify possible actions to take to mitigate risk
- Identify methods of effective communication and training with the business stakeholders
Tools & Technology:
Includes tools to be considered and evaluated for use during and post implementation.
Examples include:
- data profiler
- data modeling tools
- modeling repository
workflow data management application to alert, track, notify, escalate and approve Data Governance standards and policies as the model matures.
In summary the industry standard Data Governance framework will help an organization identify the key processes and related systems required, along with the current gaps and challenges that provide opportunity for change