A Data Management Primer

You would think that every large enterprise and even small and medium sized businesses would have a data management strategy locked up tight, right? Sadly, that's not necessarily true -- there are still IT managers and even some CIOs out there who don't have a formal strategy for managing all this big data that's flowing through the pipeline. Plus there are people moving into roles where they need to better understand analytics and data management. Here's a quick primer on data management basics if you're one of the unlucky without some formal structure for how to manage your data.

For starters, a data management strategy can be distilled down to three really (some might say dangerously) basic components: the organization, management, and preservation of your data. Let’s take a brief look at each one.

Organization: It’s important to be ready to roll up your sleeves and get dirty with this one. Unfortunately, many companies view their database servers and data warehouses as junk drawers where all the data odds and ends go (along with useful data) with the thought that they could be important. So that means someone has to go through, sort, and classify all this data, get rid of all the bits that have no real business value, selectively retain and archive the stuff that’s valuable, and free up space for more, better-organized data. Only by slogging your way through this task can you identify truly valuable data assets and focus on managing, protecting, and storing them properly.

To save yourself from having to periodically repeat this process, make sure you use consistent data organization; for example, a spreadsheet format in which each file represents a complete record and the columns represent all the parameters that make up the record. Use plain text asci characters for all file names and data -- this will ensure that your data files will be readable by the widest variety of software. Finally, assign descriptive file names people can understand and that can stand alone as file descriptions for even better identification.

Management: To make managing your data as simple and easy as possible across platforms, business units, and even enterprises, use stable, non-proprietary software and hardware if at all possible. Nothing’s worse than trying to maintain a database and distribute data using some unique, non-standard technologies that are difficult to access and worse to upgrade. Conventional, best-in-class OTS technology is not only easier to support and more reliable to use, it also tends to last a lot longer, too. No one ever went broke betting on Oracle or Microsoft databases.

Preservation: Once you have your data organized and under control, you need to ensure it’s preserved, protected, and available for use, now and in the future. Start by associating metadata with your data -- that’s data that describes your data. Identify relevant industry standards for data and metadata content and format and use them to ensure your data is accessible and understandable by others.

Once your data is prepared, make multiple copies and store them offsite and in different systems. This is critical for business continuity and ensuring that your organization can be up and running quickly in the event of a systemic or natural disaster that can bring your IT infrastructure down.

Know the regulations regarding data storage for your particular industry. It might be affected by the Health Insurance Portability and Accountability Act of 1996 (HIPAA), the Sarbanes-Oxley Act of 2002, or other federal policies and regulations. Compliance is not just important -- it’s mandatory.

So there you have it, the very basics of data management for those unfamiliar with them.