Claudia would like to thank Jonathan Geiger for his contribution to this month's column.

We often hear about the importance of the data warehousing group forming a partnership with the sponsoring business unit. This partnership helps to secure and retain business support and commitment. It also ensures that the data warehouse is aligned with the business goals. However, this is not the only important partnership. The business units must partner with each other, and the data warehouse group must partner with the other groups within information technology (IT) as well. These lesser-known partnerships often account for the difference between success and failure. See Figure 1 for the partnership types that should be of concern.

Business Unit Partnerships

The data warehouse needs to provide an enterprise view of the data. In today's customer-oriented world, the big push from business users is to build data warehouses that consolidate the critical customer data. What might happen if the marketing department sponsored a data warehouse project but didn't involve the sales department in its design? They would propagate the data silos commonly found in their current systems into databases that provide sophisticated reporting capabilities for them ­ they wouldn't have a data warehouse that provides an enterprise view of the customer data.

Figure 1: Partnership Types

Development of a data warehouse that stores comprehensive customer data must involve all (or at least the major) business units responsible for managing and nurturing customer relationships. The sales department is responsible for the point-of-sale systems that contain the customer transactions. Marketing keeps track of customer responses to its campaigns. Customer service is responsible for the customer complaint systems. Fulfillment is respon-sible for the order processing systems. Each system contains different views of the customers as shown in Figure 2.

Figure 2: Conflicting Views of Customers

Without resolving these differences, how can you answer even basic questions such as: How many customers do we have? How quickly is our customer base growing/shrinking? More importantly, without resolution of these differences, we are not in a position to assess the value of each customer or treat our customers in the fashion that is appropriate for them.

The quickest way to build what some may call a data warehouse is to select one system as the source of strategic data for use by a department's functional group. Even so, this approach will probably not resolve the conflicting views. We, as data warehouse practitioners, need to find a practical way to incorporate each functional group's per-spectives without unduly lengthening the implementation cycle of a project.

The first step in this process is to involve representatives from other business areas and departments. However, involvement alone is not enough. We need to obtain their commitment to the overall data warehouse project's success.

To gain commitment, each of the key business units must understand the value of the integrated approach to the enterprise's customer data. A cross-functional steering committee is one way of obtaining this commitment and partnership. Even if only one department sponsors the project, you should include key decision-makers from the other departments in the requirements-gathering interviews. During that process, we suggest you ask the attendees questions such as:

  • What is our strategy for customer relationship management?
  • What are the key steps and components needed to support that strategy?
  • What difficulties (if any) are you or your staff experiencing in collecting customer data?
  • If these difficulties were addressed, how would that impact the bottom line?
  • What problems have you experienced because of customer data quality problems?
  • What is a customer?
  • If we knew the profitability of each customer, what would we do differently?
  • How many customers do we have? (If the attendee provides a number, ask him or her how it is derived.)

These questions probably won't elicit all the answers needed to help you design the solution. However, the interviews should help the business leaders understand that the overall goal of the data warehouse environment is aimed at addressing more than a single department's business needs. These interviews should help them realize the broader business benefits to be derived from the effort.
Once you engage these partners in the process, make sure to maintain communication with them. Provide them with information and testimonials that demonstrate the need for consolidated customer data, keep them informed of similar activities within the industry and, most importantly, provide them with testimonials from successful applications within your company.

Partnerships within IT

Obviously, the data warehouse group does not operate in a vacuum. Within IT, the data warehouse group depends on contributions by operations, technical support, telecommunications, application maintenance, application development, database administration and data administration, to name just a few. You must have their support!

Some of these groups may view the data warehouse as a threat or rival. The data warehouse enables business users to gain access to data quickly; application development and maintenance stand to lose some control over how, when and who gets to use the corporation's data. Database administration designed databases that performed well with minimal redundancy; many data marts require star schemas, which require significant data redundancy. Telecom- munications and technical support need to provide consistent response time; data warehouse usage is often unpredictable with long response times and the return sets may be very large. For our data warehouse environment to be successful, we need help from these groups. The key ways to do this are through education, involvement and sharing the credit for the data warehouse's success.

Education. The data warehouse group should assume a responsibility to educate others within IT about data warehousing, the infrastructure requirements and the business benefits that the company will realize. It may be advisable to send some of the key people to courses sponsored by well-known industry organizations. For most, an in-house forum will be more appropriate, and this can be created and delivered by either the data warehousing group or an external party.

Involvement. Involve the appropriate groups in the data warehouse effort. Database administration, for example, should help design the databases; application development and maintenance can assist with source system analysis; telecommunications and technical support must understand the difference between OLTP and OLAP activities. Data administration can support the modeling activities.

Sharing the Credit. The data warehouse effort will deliver business value, and the project's success relies on contributions by people outside the core data warehouse group. When the effort is completed, be sure to credit all those who made significant contributions to the project's success ­ regardless of whether they report directly to the project team or are ancillary team members.

No one is an island, and no data warehouse group can succeed without partners. Successful data warehouse projects require teamwork; and the team must consist of the data warehouse group, business representatives and other participants within IT. Sponsorship is just the beginning of support. A full partnership in several areas is mandatory to ensure the success of your project.

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