TDWI Report Excerpt: Turning Business Intelligence into an Enterprise Resource (Part 1 of 2)
What organizations must do to transform BI from a departmental resource to an enterprise one.
- By Wayne Eckerson, Cindi Howson
- August 17, 2005
Once they understand the benefits of enterprise BI, many executives want to deploy a solution immediately. Unfortunately, it often takes several years for an organization to gain the experience and skills and build the technical infrastructure required to deliver an enterprise solution and reap its benefits. TDWI’s Maturity Model defines six stages that organizations pass through on their way to a mature BI deployment. Only when organizations reach stage five—the “adult” stage—can they begin to roll out BI on an enterprise scale (see Note 1).
Although a discussion of all the steps required to achieve BI Maturity is outside the scope of this report, we will highlight a number of tasks that organizations must undertake to transform BI from a departmental resource to an enterprise one. Much of the advice below involves aligning the BI solution with business strategy and making sure the business, not the IT department, owns the solution and guides the IT project teams during implementation and beyond.
[Editor’s note: We present the first half of this task list this week. The second half will appear next week.]
1. The Business Must Recognize the Need
The most important—and often the most overlooked—requirement for creating an enterprise BI solution is to make sure the business understands the need for such a solution. Although the IT department and select business managers may recognize the need for consistent information and standard tools, technologies, and processes, the project won’t get off the ground unless key executives share the vision and lead the charge.
Executives often aren’t motivated until they feel substantial pain from the lack of an enterprise BI solution. “Our executives got frustrated because it took five to six days at the end of the month to get a revenue report,” says Sapana Patel, director of application services at Interval International, a membership-based travel services firm in Miami that recently deployed a series of executive dashboards and other reports based on a new enterprise data warehouse and standard set of BI tools. “Now, when they arrive in the morning, they can quickly check the key metrics driving their business and know what issues they need to track or address during the day.”
2. Strong Executive Support
Once the business feels sufficient pain to take action, a strong and committed executive needs to spearhead the initiative and evangelize the BI solution. The executive needs to have sufficient clout, be well respected, have many allies and few enemies, and make a clear business case. The executive needs to appoint a trusted business driver to oversee the project who devotes 50 percent of his or her time to executing the business plan.
“I’m fortunate to have strong backing from our top executives,” says Interval International’s Patel. “They drove a lot of our information requirements and mandated a standard solution in the face of some internal resistance.”
3. The Business Must Own the Initiative
Besides evangelizing the BI solution, the business sponsor needs to assume accountability for the outcome of the project and recruit (or strong-arm) other business executives and managers to support the initiative. At this point, executives should avoid making the mistake of turning the BI project over to the IT department. Although IT needs to administer and support the BI solution, the business needs to own and drive it.
Many organizations establish a variety of business-driven steering committees to manage an enterprise BI initiative. For example, Alstom Power created a data warehouse council in 2004, comprising business executives and managers, to set the direction and strategy for a new enterprise data warehouse, which consolidated numerous data warehouses, reporting repositories, BI tools, data marts, and Access databases into a single enterprise resource.
“We needed to move ownership away from the IT department to the business. Now, the IT group only implements and manages the system on behalf of the business,” says Alstom’s Michael Sykes.
4. Gather Requirements and Set Strategy
To deliver an effective enterprise BI solution, organizations need to treat the project strategically. They need to interview a cross section of users, establish priorities, and create a project plan that delivers the desired information, functionality, and applications in an incremental fashion. Unfortunately, many organizations believe the only thing they need to do to implement a BI solution is select and purchase a BI tool, which is actually the last step in a long process.
Interval International was heading in this direction until Patel attended a TDWI conference and learned the importance of first establishing a BI strategy and project plan that aligns with business goals and objectives. “After the conference, we reassessed our project and selected a BI vendor whose offerings met the information requirements of the entire organization. We established a partnership with the vendor, who helped us develop a road map and strategy and validated our tool selection process.”
5. Standardize and Validate Data
One of the most challenging aspects of delivering an enterprise BI solution is getting various business groups to agree on definitions for common data elements, such as “customer,” and rules for calculating common metrics, such as sales or net margin. This is often a politically charged task that requires the executive sponsor to referee between various business groups and establish enterprise standards.
“We have two distinct businesses—commercial and government—and the measurements each uses are very different, which makes it very challenging to develop corporate-wide standards,” says John Monczewski, manager of Balanced Scorecards at Booz Hamilton Allen. “We’ve had strong backing from our CEO to make this work and we’ve made a lot of progress. But even with that, it takes a lot of time. Our partners have decided to postpone trying to resolve some issues until a later time.”
Once definitions are standardized, IT must maintain them centrally and update tools in the BI environment if and when the definitions change. Unfortunately, there is no good way to automate the management of this metadata, so the process is manually intensive. The IT group must also validate incoming data and ensure that proper rules and checks are in place to preserve data integrity and quality in the data warehousing environment.
6. Standardize Architecture, Processes, and Success Criteria
Once an organization agrees on definitions, it needs to establish a standard BI architecture and processes for delivering a consistent view of information to end users. This involves defining the flow of data from source systems through a data integration infrastructure to reports and end users. It also involves defining standards for the use of technology to manage these data flows. Enterprise Business Intelligence
Boeing Corporation uses a BI center of excellence to promulgate enterprise standards for BI technology and processes as well as provide resources, services, and advice to groups that want to implement BI solutions. The center is staffed by 35 people with various BI skills who are loaned to different groups throughout Boeing on a temporary basis. The key is to intervene before groups get too far into the process and deviate from corporate standards and best practices.
“Every group faces time and resource constraints that cause them to want to revert to the tools and techniques with which they are familiar despite our enterprise standards,” says Harvey Kriloff, information architect at Boeing.
7. Make BI Tools Conform to Users
The hardest part about deploying a BI solution is getting employees to use the tools. BI tools should conform to the way users want to work and not force them to conform to the way the BI tool works. In this regard, a mistake that many organizations have made is to distribute a single BI tool or solution to all users and expect it to work. This is usually a recipe for disaster, since different types of users (executives, managers, power users, frontline staff, customers, and so on) use information in very different ways. Unfortunately, many organizations have learned the hard way that one size does not fit all.
“Our BI tools were a mismatch for our users,” says Alstom Power’s Michael Sykes, who admits that his group originally issued a single BI tool to all users when they first deployed a data warehouse. It wasn’t long before they purchased other tools to meet user requirements, but that still didn’t work.
“The tools were overbearing for what most users are trying to accomplish. So, now we carefully classify how users interact with information and match new and existing tools to these modes of use,” says Sykes.
Next week: Six more tasks.
- - -
1. Wayne Eckerson, “Gauge Your Data Warehousing Maturity,” TDWI’s Case Studies & Solutions, August 2004. http://www.tdwi.org/publications/newsletters
- - -
This article was excerpted from "Enterprise Business Intelligence: Strategies and Technologies for Deploying BI on an Enterprise Scale" by Wayne W. Eckerson with Cindi Howson, published by TDWI. The full report is available online (short registration required). Report sponsors include Actuate, Business Objects, Cognos Inc., Hyperion, Information Builders, SAP America, Siebel Systems, Inc., and Temtec.