The term metadata carries a mystique with it that adds to the difficulty of deploying a metadata management environment. The view of the metadata management environment needs to change. We need to think of it as a major application system that exists for the purpose of managing the enterprise asset of data. The human resource system manages data about employees; the general ledger system manages information about money; the inventory management system manages information about inventory. Similarly, the information management system (a.k.a. metadata management system) manages information about information.

Many companies are experiencing problems because they lack concise definitions of the fields in their reports. Their analysts waste a lot of time because of this one issue, and that time could be better used to generate revenue. This is one of many symptoms of the lack of an effective system (application and business processes) for capturing, maintaining and disseminating information about information. Changing the way we think of metadata is a critical first step toward improving the situation. We should stop calling it metadata management. Once we change our thinking, we can address development of the information management system as we would any other major application system.

As with any system, we should start by defining the scope. That scope can include both long-term and short-term goals so that we can build the environment incrementally. The long-term goal may be to ensure that information about the information resources is available to enable businesspeople throughout the enterprise to understand and apply the available data and to leverage it for additional business value. The initial short-term goal would focus on a set of immediate needs, such as providing a consistent set of definitions for a set of business terms. We can describe subsequent iterations as well. For example, our next effort may deal with capturing, maintaining and disseminating information about the quality of the data.

As part of the scope definition, we would identify the key participants, critical information about the program and would also create our project plan. Once the scope is defined, we continue with the subsequent application development steps, just as we would with any other project. Some aspects of this project appear to be unique, but really aren't. Three examples follow.

Roles and Responsibilities

IT has two roles when it comes to this system. On the one hand, IT is the development department, and it provides the resources to do the development. On the other hand, IT is the organizational unit accountable for managing the information resource and, as such, is the primary business unit for this project. As with other systems, additional business areas act as end users, and they need to be involved as well. For this system, for example, a person in finance will establish the definition for a financial term, and eventually, someone will want to access the term. The business areas, therefore, help define the business process for creating, maintaining and accessing the definitions. IT uses this input to design and build an application system that supports the process.

Buy versus Build

The buy-versus-build decision is not unique to metadata. In many companies, any major system replacement or addition entails an analysis to determine whether the company will build it from scratch or use a commercial off-the-shelf (COTS) software package. While the specific evaluation criteria for the information management system are unique, the decision process is not.

Technical Complexity

The information management system environment is considered to be very complex. People creating this environment need to deal with interfaces with multiple tools and with constraints imposed by those tools. While the specific technologies are unique, this is conceptually no different than any other major application system. When a major new system is implemented, it needs to interface with existing systems and must abide by constraints imposed by those systems.

Business Process

The business process for determining the content to be populated into the information management system differs from the traditional business process. Typically, a business process has a single recognized business unit responsible for it, and that unit dictates what data is captured and how it is captured. Information cuts across organizational boundaries, and several business units often claim authority for specific data elements.

Population of the information management system is dependent on addressing the horizontal management implications of data. This is often done through data stewardship, which was described in my November 2006 column.

Metadata management is really about managing information about the data asset of the enterprise. Companies need to recognize that the processes and application development techniques used for managing other assets apply to the information as well. When we do that, we can remove the mystique around metadata and help our companies leverage the information assets to increase revenues, decrease costs and meet other corporate goals. 

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access