This article originally appeared on the BeyeNETWORK.
Data has never been well managed because of the priority given to application functionality. My last article addressed treating data assets more like physical assets. Another important aspect of treating data as an asset is creating a managed data environment to manage enterprise data assets.
A managed data environment is a cornerstone for business effectiveness today. I use the term “managed data environment” to distinguish it from the current IT focus on data stores and data availability. While focusing on data as an asset is about maximizing the return on assets (which tells you what earnings were generated from the assets), the managed data environment builds from data asset management to mitigate business risk.
Did you notice the story about the computer stolen from a Veterans Administration official earlier this year? The laptop contained personal data, including social security numbers, on 26 million veterans. This and other stories have found their way into the press to tell about lost, hacked, or stolen data at financial services, retail, and other companies. In all of these, the bottom line is that the price of lost or stolen data, in legal fees, liability, and data recovery, let alone for regaining customer confidence, is high for any company or organization.
The root cause behind lost and stolen data is basic: data is not managed well. Over the years, the focus has been on applications and functionality of application systems for the business. A growth in personal tools such as spreadsheets has extended the capability of business people to do their jobs without IT support. Consequently, data is provided for their use to help them and to alleviate the need for IT support. Data is distributed with limited controls, not managed.
What is a Managed Data Environment?
A managed data environment satisfies a few basic, but far from simple, criteria. It comes down to always knowing the answers to a few key questions about your data:
What is your data?
You cannot fully secure or manage your data if its identity isn’t known. The data asset inventory is one way to respond to this question.
Where is your data and where did it come from?
This includes, but is not limited to, the data store where it originally resided. It needs to address the personal computers where it is also stored and used and should also include the problematic and extremely difficult problem of external hard and flash drives that contain it as well. A data asset classification capability can respond to this question.
Who is using your data?
With regulatory restrictions on financial, medical, and other data and its distribution, knowing who is using data and what they are doing with it is an essential component of risk management. It is more important than ever to know that data is in the right people’s hands. Also, note that the amount of data usage is another way to determine data asset valuation. Knowing the answer to this question is important for controlling data access, ensuring compliance with policies on data access and usage, and managing risk.
When is the data being used?
Knowing when data is being used allows you to determine if it’s being used at the correct time, which is an important component of fraud detection. Simply who is using the data is important, but knowing when it is used as well can be a factor in evaluating whether the use is appropriate or not. Knowing the answer to this question is important for tracking the usage of data across time.
Why is your data being used?
Understanding the role data plays in the business allows you to evaluate the answers to the previous questions and identify potential problems. Ultimately, it is essential to know that data, as with other business assets, is used for the benefit of the business. Knowing the answer to this question is important in order to focus on delivering business value from data assets.
How is data delivered?
Knowing that data is delivered in a report or query or as a riskier data extract allows you to adopt appropriate security measures and follow up for each. It is preferable to deliver data in a form that cannot be changed or easily replicated (for example, as a PDF rather than in Excel). This is important for managing risks associated with extracted and replicated data.
A managed data environment manages data as an asset, ensuring its security, appropriate use, and value optimization. As such, establishing a managed data environment is a best practice.
When considering the questions above, it’s interesting to note that business intelligence and data warehouse (BI/DW) environments, especially those with strong data tracking, security, and administrative capabilities, may satisfy the criteria for a managed data environment, depending on the policies and use of operational, business, and technical metadata. But even where BI/DW environments are the strongest and most capable, the majority of essential enterprise business data used in an organization or company remains unmanaged.
The challenge every organization faces is how to extend data tracking, administration, and security to this currently unmanaged data. In part, it is a technology problem; but by beginning with practices for managing data as an asset, an organization can establish managed data functions for data tracking, security, and administration as well. The challenge is to undertake the effort.
The managed data environment goes beyond a data warehouse, its reports, OLAP analytics, and queries. It requires an operational environment separate from the hodge-podge of existing applications and proliferated data marts that exist in businesses today. Existing separately from the application environment frees data assets from applications that use them and the reporting and analytic processing used for transaction processing. It enables enterprise data management.
But what about the additional costs for storing data in both the managed data environment and the application environment? This is not the major issue. If the cost of redundant data is a concern, there is more of it in the existing application environment than will be introduced by a managed data environment, so eliminate it from the applications through master data and SOA-compatible data delivery services. Also, the cost of storage and computer memory continues to decline rapidly, reducing the cost impact. And finally, a managed data environment protects and uses data assets while at the same time allowing applications to focus solely on transaction processing. You optimize both managed data assets and transaction processing.
A managed data environment cannot be built overnight. It generally requires a data strategy and a road map of a sequence of incremental implementations that build out the managed data environment one step at a time. Successfully establishing the managed data environment and using data as an asset requires solid data management and architecture practices and extending them into the wider realm of digital data and its delivery in the business. These practices are not a mystery but they need to be applied in a disciplined manner across critical data assets, those assets that will provide the greatest value. What is clear is that the best way to maximize a return on data assets is through the rigor of establishing a managed data environment.