Data governance for an IT techie:
- Managing the content of the data is as important as managing the container the data is in. For many techies, administration and system management is more important than the actual content and meaning of the data under management. This mindset must be changed to successfully implement data governance. Yes, you still need to backup/recover, reorg, and tune… but you also need to protect, retain, acquire and manage metadata, and improve the quality of your data.
- There is no magic bullet software that will deliver data governance. Data governance requires commitment and must include people, processes, software, and executive buy-in and support to succeed.
- Data governance practices and procedures must be integrated into the overall IT infrastructure in a repeatable and measurable fashion in order to achieve success. This most likely means changing most of your policies and procedures to incorporate better data management into them.
Data governance for the business manager:
- It is not enough to simply say "We treat data as a corporate asset" in order to start truly treating data as a corporate asset. Think about how you treat other corporate assets (personnel, financials, etc.). You model them, manage them and tune them continuously. You put infrastructure in place to ensure they are accurately and effectively managed and accounted for. If your books are one penny out of balance you work until you find that penny – do you do the same for your data quality? If an employee is out of the office for any reason, you know why – do you do the same for your data? You pay the rent, utilities, upkeep, etc. on your office building – but do you spend enough on data and metadata quality? Seriously think about these things.
- A data governance practice will improve the quality of your data which will add to the bottom line of the organization. Some estimates show that poor data quality can impact the bottom line by as much as 30% annually. Consider working to implement a data care policy where employees are tasked with treating data carefully and doing whatever it takes to ensure the accuracy of the data that they come in contact with.
- Data supports the business and there is no reason to save and manage data without the business. As such, a business must take the lead in establishing policies that care for the business's data. Consider creating a data council whose responsibility will be to oversee the data/information practices and policies of the organization. The council should include business and IT leaders, as well as legal representatives to interpret regulatory needs.
And finally, for everyone:
Consider implementing a data help desk that is staffed by people who know both the business and IT aspects of data. Encourage the organization to treat the data help desk in a similar manner to a tech support help desk. Users should be able to call the data help desk to ask about data quality issues, where to find the data they need, help with data definitions and so on.
- For more information on data governance, read "Data governance: Information ownership policies and ownership explained."
This was first published in March 2007