Deciding whether to proceed with the development of a metadata system will eventually be based on two questions:
To answer these important and appropriate questions, planners engage in cost-benefit analysis to ensure that both the positive and negative implications of a metadata system have been considered. As an extension of cost-benefit analysis, "return-on-investment" (ROI) is a concept used to express the amount of benefit (return) relative to the amount of resources (investment costs) needed to produce the return. Based on thorough analyses, many organizations have found that the potential improvements to data quality and use are well worth the costs of developing and implementing a metadata system.
In addition to costs for hardware and software, staff and/or consulting, and other direct development requirements, planners should also expect indirect costs. These are often referred to as "unanticipated costs," although many of can them can be anticipated with careful planning. These types of costs include staff training (initial and ongoing), user support (help desks, tutorial development, etc.), system maintenance costs, licensing agreements, and ongoing system evaluation initiatives.
The absence of a market price for good data presents a challenge to cost-benefit analysis for metadata systems. However, some cost savings from improved data quality can be measured in the areas of purchasing, staff allocation, and maintenance and operations. Cost avoidance may also be factored into the analysis; this might include not needing to hire consultants or purchase products to revamp aspects of your data system.
Some of the benefits of a metadata system are easily quantifiable, but many are not (see our ongoing story, Solving the Case of the Inaccurate Dropout Count). Still, potential financial benefits can be estimated. We know, for example, that a robust metadata system can reduce redundancy in a data system; this, in turn, should decrease collection, access, and reporting burdens—each with a tangible cost. Similarly, metadata systems can make data more accessible, saving staff time. Metadata also improve data quality and use, and help users better understand the data they are analyzing. This can lead to savings associated with improved decisionmaking. Some financial implications of improved decisionmaking can be estimated as well. For example, a better command of information may improve purchasing choices; staffing decisions; and even academic preferences, such as curriculum selection, teaching assignments, and leadership decisions.
Exhibit 4.5 presents an example of several frequently recognized categories of costs and benefits (including cost avoidance) and the return on investment (ROI) that can accompany metadata solutions. Note that these categories are illustrative and may vary for individual organizations depending on a wide range of factors. These costs and benefits can be placed in a spreadsheet with columns to estimate dollar values.