So you realize you have to manage tons of current (and possibly future) organizational data. You also realize that you aren’t sure where to start. Well, a data management strategy will provide you with a clear roadmap.
So how does one craft a compelling and relevant data management strategy for his/her organization? I recommend these three steps:
1. Always Start with Information Architecture: I suspect this example might be as a result of my background in Architecture, but no one (at least no one I know) would start building a house without first sitting down to create floor plans, elevations, and cross sections. In the same vein, the starting point of your data management strategy should be the design of your information ecosystem. Take the time to identify the current state of:
- Data Sources
- Data Formats
- Data Sizes
- Update frequencies
- Accessibility requirements
- Interoperability requirements – how different data formats should be able to ‘talk’ to one another, and
- Security requirements
Next, figure out how these different ‘moving parts’ can be put together to create information that is relevant, and reliable for your organization. This is the meat of your information architecture design. A key component of information architecture design is alignment. What I mean by this is that your design should closely align with, and be driven by the business strategies of your organization. This is the only way to derive true value from your design.
2. Create and Enforce Data Governance Policies: Your organizational data is only as good as its quality levels, and high quality levels cannot be sustained without the necessary data governance policies in place. Senior business leaders have to ensure that the data being generated by their team members conform to the standards and formats set by the organization. The emphasis here is that this initiative has to be driven by senior business executives, as they alone possess the political ‘capital’ to get this done.
3.Pay Particular Attention to Security: This is kind of like flogging a dead horse, but with incidents like Stuxnet, we cannot say too much about security. The danger of inadequately protecting your data includes the possibility of data corruption, which might include complete destruction, or data manipulation, which might reduce data reliability. Effective security protocols such as encryption, intrusion prevention and threat detection should be implemented where necessary. However, it is noteworthy that different data types in an organization require different levels of security. Personnel blog posts about the annual company christmas party would definitely not be protected like the company’s financial records.
Data management is important, and will continue to be so. We therefore cannot afford a scarcity of best practices on the subject matter. Do you have any ideas or recommendations? Please post them below.
Image by Corbis.com