Continue in 2 seconds

Information Semantics

  • March 01 2004, 1:00am EST

Do you fully understand the information you use from your data warehouse?

Many business intelligence (BI) users today are making incorrect and flawed decisions based on the information they retrieve from their corporate data warehouse. They run queries and reports against the data warehouse, thinking they understand the content of the information they retrieve. They will either rely solely on their own interpretation of the field descriptions on input pages plus column and row headers in reports to understand the information, or they may consult with a fellow colleague who is also misinterpreting the data. In some cases, there is a pool of "tribal knowledge" about reports and information that is passed from one employee to another, leading to further misinformation. They do not know the true, full context of the information they are about to use to make important tactical and strategic decisions for their firm.

The cause of this unfortunate scenario is the lack of use of meta data and a meta data repository in the data warehouse environment. Many data warehouses exist today without the adequate means to capture, maintain and deliver meta data to users who rely on accurate and concise information from the environment. This situation has been further proliferated by the growing use of packaged applications in corporations. These packaged applications do not have common means of providing meta data about their data stores. In many cases, corporate management questions the value and cost associated with developing/purchasing and maintaining a meta data store for use with the data warehouse –­ especially in times of negative or flat information technology (IT) budgets.

The question for management should be, "What is the cost and impact to the business of misinterpreting and acting on misleading information from our operational and data warehouse environments?" Without a full understanding of the context of the information you are using, you cannot accurately use its content.

If a user incorrectly misinterprets information, the resulting conclusions can lead to incorrect actions and responses, affecting the firm's performance. For example, consider the two human resource (HR) reports in Figures 1 and 2. Do you believe you could make informative and correct deductions based on the information presented in these two reports? A cursory review of the two reports seems straightforward –­ total head count and turnover percentages by region –­ but what do you really know about the categories and measurements presented?

In Figure 1, let's first look at the "Europe" region. What does this category represent? Without a meta data reference for the category, the user is left to guess whether the United Kingdom and/or Ireland are included in the European head count numbers. Can you tell from the information presented which region includes Australia and New Zealand head counts? Asia maybe? How sure are you?

Figure 1: Sample HR Report of 2003 Head Count

Let's look at a second report example ­ specifically the "full-time" category. Without a meta data reference tied to this category value to clarify the measurement, it is impossible to know whether full-time includes temporary or expatriate employees.

Now let us look at the turnover report example in Figure 2. Can you tell from the report whether the turnover rate has been calculated as a rolling average or annualized turnover? No. Without the business meta data about the field and how it was calculated, you are left to interpret or guess what the report's information is attempting to represent.

Figure 2: Sample HR Report of 2003 Turnover

This misinterpretation of information can go beyond business definitions, lookup values and calculations (business and technical meta data). By looking at operational meta data on extract, transform and load (ETL) processes from the corporation's various operational systems to the data warehouse, a user may find that not all data was included in the final tallies due to incomplete or erroneous data.

This is why a meta data repository is vital to the data warehouse environment. A meta data repository can collect and transfer meta data about your firm's entire enterprise. The repository allows companies to manage change and growth with their applications with reduced cost and resources. How does your company currently determine the effect, level of effort and downstream impact of business changes to an application? Do you know what data movement, data transformation, business processes/transactions and reports are affected by this change? With a meta data repository, you can immediately create an impact analysis of the change on the enterprise to get a total view. A meta data repository can also be used to supplement the operational and data warehousing information delivery methods so users have accurate knowledge on the definition, business rules, lookup values and definitions, calculations, operational statistics and other pertinent meta data on the information they are using to make decisions. The meta data from the repository needs to be integrated into the information delivery environment and readily available to users to reference through online help, drop-down, meta data reports or other means. By leveraging meta data from the repository, business users can confidently make informed and accurate decisions about the corporation which will benefit its performance and the bottom line.

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