Companies around the world have spent billions of dollars in recent years implementing enterprise applications to better integrate their corporate data and increase the overall value of existing information. With high-quality data, they hope to realize the significant savings that come from consolidated data, such as increased operational efficiencies, enhanced compliance initiatives and improved customer relationships.

These organizations have invested heavily in data-driven initiatives such as customer relationship management (CRM) and enterprise resource planning (ERP) applications, hoping to build more profitable relationships with customers and hold product-related expenses in check. Despite good intentions, however, most of these applications have faltered due to inefficient, outdated data. In fact, industry estimates show these projects fail or go over budget up to 75 percent of the time.

The very foundation of CRM and ERP systems is the data that drives these implementations. Beginning a data-driven initiative without first understanding the existing, underlying data is like repairing an automobile without first understanding the problems inside the engine. To repair the engine, the mechanic first has to understand the breadth and depth of the problem.

Successful data quality begins with a clear understanding of the integrity of your current data. Data profiling, also called data discovery, gives you the diagnosis of your existing data to begin building a successful data improvement and integration effort through consistent, accurate and reliable data throughout your organization.

The Problems with Data

Data problems abound in most organizations. Some of the more common problems today include: outdated, inconsistent, missing, orphaned or duplicated data; data anomalies or outliers; and data that does not meet specified business rules. Before you begin any data quality improvement initiative, you should ask and address some key questions:

• Do you trust the quality of the data you are using in this initiative?
• Does the data for this initiative conform to the business rules monitoring process you expect to set up later?
• Will the existing data support the needed functionality?
• Is the data you are using complete enough to populate the needed data repository?

Engaging in any data initiative without a clear understanding of these issues will lead to large development and cost overruns or potential project failures. The effect can be incredibly costly. For example, one company spent more than $100,000 in labor costs identifying and correcting 111 different spellings of the company AT&T. Data problems within your organization can lead to sub-standard customer relations, wasted expenses, poor decisions, lost sales, and ultimately, failed businesses.