Welcome Guest. | Log In| Register | Membership Benefits
News In Review

May 24, 1999

Print this story
Print this story
Warehouse ROI

continued...page 2 of 3

Related links:
  • Maximize Software ROI

  • Software Resource Center
  • And from our sister publications:
  • TechWeb What's The Investment Worth?

  • InternetWeek Measuring ROI For The Top Line Of The Business
  • Another reason that some IS managers have begun calculating the ROI of data warehouse projects is simply because they can. Data warehousing is no longer the mystery it was in the early to mid-'90s. The technology has matured, and there is more industry experience with assembling data warehouses. That can make it easier to identify costs and benefits. "Data warehousing has entered the mainstream phase of its existence, so it has to justify itself by showing reduced expenditures or increases in income," McKnight says.

    Calculate The Costs
    So when is calculating the ROI of a data warehouse appropriate? Raden divides all IT projects into three areas: operational, informational, and strategic. Operational projects--when technology is used to automate business processes and make them more efficient--is where hard ROI numbers should be developed, since the whole idea of operational systems is to reduce costs. But for strategic IT projects--technology and processes that must be implemented for a company to gain a competitive advantage, retain market share, or even stay in business--Raden says the benefits should be so self-evident that a company that wastes time calculating ROI could find itself calculating the cost of bankruptcy.

    While data warehouses can be operational and strategic, they are most often informational, Raden says. Informational systems include all business- intelligence, decision-support, data mining, and reporting systems. If the project is small, a level Raden defines as projects that don't require approval from a company's board of directors or top executives, then management shouldn't worry about ROI, he says. But IS managers undertaking larger data warehouse projects that require approval from top managers ought to go through some kind of ROI exercise.

    The ROI equation requires identifying the project's costs and financial benefits--in hopes that the latter will outweigh the former. Even though CH2M Hill undertook no formal ROI exercise, figuring out the benefit of its data warehouse was easy. The project cost $100,000 in hardware, $125,000 in software, and a couple of worker-years worth of labor--for a total bill of less than $500,000. But the data warehouse allowed the company to stop mailing out hard-copy reports to its managers across the company, for an annual savings of $3.5 million in Federal Express expenses. "It was obvious to us," Crownover says of the return.

    Identifying costs is relatively easy. It's a matter of adding up the expenses associated with the project: Hardware, software, personnel, consulting services, etc. Future ongoing costs also need to be considered, as does any custom coding or potential hitches in implementation.

    The benefits can be trickier. Quantifying "improved customer satisfaction," for example, can be almost impossible. McKnight suggests looking at the proposed data warehouse as part of a process, such as inventory management or product returns, to help identify tangible benefits. If a data warehouse can be used to reduce inventory carrying costs or the number of products returned due to erroneous order processing, then it's possible to attach a dollar figure to the potential savings. The use of a data warehouse to identify potential customers for catalog mailings, rather than using expensive mass mailings, is another example of how savings can be quantified.

    Who Gets The Benefits?
    One potential problem is that savings or efficiencies generated through a data warehouse may be claimed by the operation or line of business that implements it, Raden says. Shipping, for example, may try to take the credit--and the budgetary benefits that follow--for inventory-reduction savings sparked by a data warehouse. To ensure that the IS department gets credit when it is due, IS managers should alert CIOs and other top executives about expected data warehouse benefits.

    If a data warehouse is to be available on an enterprisewide basis, IS might also consider implementing charge-backs to users, in which other company operations are charged for use of the data warehouse. Measuring how much a data warehouse is being used can help put a dollar amount on its value to the business.

    In some cases, an IS department that uses a data warehouse to capture and process information such as customer demographics can even become a "data brokerage" and sell that valuable information to other companies, thus generating hard returns.

    continued...page 3
    return to page 1



    Back to This Week's Issue

    Send Us Your Feedback

    Top of the Page