Starting with Data Governance
By 
No one wants more bureaucracy. It seems the hesitation of organizations to implement data governance usually comes down to a concern that it will just introduce an un-necessary level of red tape to getting BI projects completed. The truth is that data governance is a mechanism to improve user adoption and manage a corporate asset. But there is no question that there is a complexity to managing data. There is the question of ownership, appropriate definitions, data accuracy, data quality and data usage. The value of starting with data governance far exceeds any risks, some of the most obvious ones are:
- Improved Data Quality
- Consistent Definitions of Business Terms
- Decision Making Based on Data (Confidence in the Data)
- Collaboration Among Business Units
- Appropriate Use of Data
- Sharing Data Internally (Data Integration and Re-Use)
- Regulatory Compliance
- Simplified (and Known) Data Management Business Processes
Once the value is established the process for getting started revolves around creating a data governance council (or re-purposing an existing team), identifying your data stewards and creating their working group. Each group has different responsibilities.
Governance Council:
- Data Owners represented by key business area
- Create policies & procedures associated with BI and Data Management
- Enforce Policies & Procedures
- Ensure work of Data Management/IT/IM team is aligned with Corporate Strategy
- Champions BI and Data Governance to broader organization
- Review and approve recommendations from the data stewards regarding terms and definitions, quality standards and thresholds
Data Stewards:
- Subject matter experts in their field, aligned with business units represented in steering committee
- Reviews issues with data and makes recommendations
- Ensures work is aligned with policies & procedures
- Each Steward will ensure proper usage, accessibility and quality of the information within their data domain
- Establishing enterprise-wide standards for data quality and appropriate usage
- Ensure conformity, extensibility, security and compliance
Finally, the team will need to decide where to begin. The best approach for this is to not attempt to boil the ocean. Data governance programs can quickly get overwhelmed by the sheer volume of data that need to be harnessed and defined. If you have a BI project that is starting just begin with the data points that are applicable to that project.
Out of all of the blog posts for this series data governance is probably the most complex. This post just barely scratches the surface. To learn more on this post or any other post in our series please join our collaboration site (email me at lmadsen@lancetsoftware.com).
