The case for business meta data seems obvious. With business meta data, there is documentation of what is what inside the information system. More importantly, there is the chance that the end user will have business context to accompany raw data that turns that raw data into something useful. To illustrate the value of business context supplied by meta data, consider a vice president of finance who asks his department head to provide quarterly revenue figures. The department head's response is: "Quarterly revenues were $18,970,997."

At first, the vice president is pleased, but then he realizes that the number is quite a bit different than what was expected. Upon exploration the vice president finds that the number:

  • Is an estimate, not an audited number.
  • Was calculated on March 15, when not all business for the quarter was closed.
  • Included Canadian revenue, but not offshore North American revenue.
  • Was calculated using several sources of data, some of which had never been used before.
  • Converted European revenue at the exchange rate effective March 1 when the books needed to be closed as of March 30.

Suddenly, the quarterly revenue number becomes circumspect, not from the value that has been calculated, but from the context of the number. The context of the calculation is profoundly shaped by different forms of meta data that are associated with it; and, in the long run, the business context of information is as important as the actual number calculated.
From this perspective – that of supplying the context of information – it is easy to see why meta data is important. It is also easy to see why it should be obvious that the justification for meta data should be self-evident. Business meta data is some of the most important information the corporation possesses.

There is, however, another expectation for business meta data that is not met and is not reasonable, although at first glance it appears to be reasonable. To understand where business meta data is not useful, consider a president of a company who asks, "How much revenue did we have last month?" The finance spokesman says that revenue was $1,998,887. The marketing vice president says it was $4,778,330, and the accounting controller says it was $2,998,008.

The president complains, "How can we run a company when no one knows how much revenue we have? No wonder we can't make a decision around here." At this point, it is tempting to respond that the situation would be improved if the company had enough business meta data – that with the right meta data, the company could determine the correct revenue amount. However, this is an incorrect expectation for meta data. Revenue, like many corporate entities, has many different definitions. Accounting focuses on recognizable revenue. Marketing focuses on committed revenue. Finance focuses on received revenue. Sales focuses on forecasted revenue. Each of these types of revenue are different. There should not be a single number for corporate revenue. The business demands that there be different numbers – each of which are related, but each of which is nevertheless different. It is naive and disingenuous of the president to complain that the different departments just can't get it right.

A more sophisticated president would ask questions such as: What are the recognizable revenues? What are the forecasted revenues? What are the received revenues? All the business meta data in the world – all the business context in the world – is not going to solve the problem of trying to create a single corporate number for something such as revenue. That is not a meta data problem, but a basic business problem.

Of course, business meta data can be used to define a crisp value for recognizable revenue, and business meta data should be useful in defining and calculating what is received revenue. Business meta data, however, does not "resolve" the differences in the way corporations understand and use the term "revenue."

It is not just corporate revenue that defies description at the corporate level. What about the number of employees a company has? This seems like a simple question, but is it? If we are going to count employees, we need to take into consideration: full-time employees, part-time employees, consultants, etc. The truth is that for all but the smallest and the simplest of organizations, the number of employees is anything but a trivial calculation.

What about expenses? And what about products? There is work in progress, sold products, products in storage and so forth. Everything you can think of has shades of information. There simply is no "corporate number" that is valid for everyone. That is a myth, and it is a false expectation that business meta data is going to clear the air to the point that the corporation is at last going to have the definitive value for anything.

What is needed is a sophisticated understanding of meta data, where meta data describes what it describes and corporations look for resolution of basic departmental differences elsewhere. Of course, when those differences have been resolved and documented, business meta data should be used as the basis for documenting the results. Using business meta data as a basis for resolving the confusion in the corporation is simply not understanding the role and proper usage of meta data.

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