Continue in 2 seconds

Measuring Success

  • September 01 2004, 1:00am EDT

There is much conversation about building data warehouses, cost justifying data warehouses and different forms of database design, but there is little talk about what management should expect when the warehouse has been successfully built. Exactly what should management look for when the money has been spent, the consultants have gone home and the data warehouse has been declared a success?

The first payoff a manager should see is that when a request is made for information, it should be answered in a short amount of time. If the request is a study looking at huge amounts of data with a statistics package, the results may be returned in a week or two. However, if the request is for a summarization of all Northeast sales for the past three years when there has been a blizzard, the results should be returned in a matter of hours (or occasionally even less time than that).

The second thing the manager has the right to expect is that analysis involving history should become easy and natural. In the past when management wanted to look at history, most organizations did not have any historical data beyond two years. Or, if they did have the data, it was stored in an archival bin that had never been used. The data warehouse provides the ability to easily access historical data.

The third benefit that management should expect is the ability to access data in an integrated manner. Prior to data warehousing, management was always at odds with "stovepipe systems." In a stovepipe environment, you can see inventory data and you can see shipment data, but you cannot see inventory and shipment data together. Similarly, you can see sales data by product or by region and salesperson, but you cannot see sales data by product and region. With a data warehouse, you can see sales data by product and by region; furthermore, the numbers that the manager sees are able to be reconciled.

The fourth payoff for managers is a reconcilable foundation of information. Without a data warehouse, when reports are submitted to management from sales and marketing, the numbers are often very different. Sales would report that the past quarter's revenues were $1,000,000 and marketing would report the past quarter's revenues as $1,500,000. Without a data warehouse, there simply is no ability to reconcile the information. With a data warehouse, sales will have their perspective and marketing will have their perspective, but because there is a single store of integrated detailed information, the manager can see that sales is reporting booked revenue for the calendar quarter and marketing is looking at committed revenue. There will still be differences of opinion from various departments in the data warehouse environment, but those differences are reconcilable because the data ultimately is traceable to a single granular integrated source.

The fifth management expectation with a data warehouse should be that information comes in waves. A data warehouse project is really never completed; data warehousing is an ongoing activity. When embarking on the data warehouse experience, that data should be ready and usable within six months of the commencement of the data warehouse exercise. Not all data will be ready, of course, but some data should be available. Over time, increasingly more data becomes available.

The sixth thing that management should notice with the advent of data warehousing is that the costs of information processing drop dramatically in two ways. The costs of getting any unit of information decrease. When an analyst wants new information or a second opinion about other information, it is inexpensive to access the information through the data warehouse. However, the costs of information processing at the departmental level also drop. Because of the existence of the data warehouse, each department does not need to build its own information infrastructure. One of the problems with tracking this drop in expenditures is that at the central IT level, it will appear that the costs of information processing have risen significantly because the data warehouse is on the central IT budget. However, from the perspective of the corporation, each department's budget decreases significantly when there is a data warehouse.

The seventh management payoff is that with a data warehouse, the corporation can easily support data mining and exploration processing. Prior to data warehousing, data mining and exploration warehousing were difficult to support. Gathering data, integrating data and cleansing data is an activity that must precede data mining. In many cases, these activities consume far more resources than the data mining itself. With a data warehouse, all of that activity is already done. The data is awaiting access and analysis. This means that a corporation is in position to start using data strategically once the data warehouse is in place.

Management should see these benefits and others once the data warehouse is in place.

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