3 Fundamental Principles for BI Success
Amid all the hype, three core principles will help ensure your success with a data warehouse.
By Laura L. Reeves, Principal, StarSoft Solutions, Inc.
We’ve seen dramatic advancements in the tools and methods used to develop data warehouses. However, we continue to struggle with delivering true value to the business. In the midst of the increasing rate of technological change, there are three core principles that can be applied to help ensure the success of your data warehouse:
- Partner with the business community
- Gather and leverage true business requirements
- Organize the data so that the business can understand it
Although each of these warrants an entire chapter of a book, an introduction to each of these is a good place to start.
The need for IT to partner with the business is a theme that hopefully you have heard before. It almost seems like common sense that you need to work together to deliver a successful solution. However, it is much harder to put into practice.
Intentions are always good when a project is launched. Often there is input from both business and technical perspectives to set up the project. However, this is when many organizations see the IT and business communities begin to drift apart. What can be done to prevent this from happening?
IT needs to:
- Take the time to learn more about the business. This can be accomplished through the requirements-gathering process we describe in the next section.
- Work and communicate in business terms. When possible, keep technology out of discussions.
The business needs to:
- Learn basics about data warehousing and business intelligence.
- Think about actual objectives and purpose rather than getting bogged down in reports that may or may not deliver what you really want.
Together you need to:
- Actively collaborate by working together on project deliverables.
- Jointly scope projects and set priorities that deliver business value and are achievable.
Business should drive the purpose and direction; IT should determine how this can best be accomplished.
- Make conscious choices to invest in the partnership. This starts by changing your own behavior.
- Most important -- listen!
This partnership will last throughout the life of the project and into the ongoing support and growth of the data warehouse. One important way to build this relationship is through the requirements-gathering process.
Gathering Sound Business Requirements
The primary purpose of gathering business requirements is to understand what the business wants to accomplish. It is important to look at data the business has collected and architect a solution that allows it to make more informed decisions.
A secondary purpose of gathering requirements is to build a relationship between the business community and the data warehouse team. Business users begin to have a vested interest in the data warehouse when they help define what it will do. Use the requirements-gathering process as an opportunity to connect with users.
Gathering requirements is the most critical step in any project, and certainly for a data warehouse project. A data warehouse should deliver an environment that empowers the business community rather than an application system that accurately performs specific business functions. This difference in approach will change the way you go about gathering requirements.
True business requirements are not a list of data elements, data sources, or even report specifications. These requirements are more fundamental to the business itself. Ask questions to understand:
- What are your business goals and objectives?
- What could prevent you from meeting those objectives?
- How is business performance measured?
- Challenges facing the organization today and for the next several years
- What it the impact on the business if you do/don’t reach these goals?
You must ultimately develop detailed specifications for the ETL process and calculations on reports. However, it is critical to learn more about the rationale for what the reports are intended to help the business accomplish. This provides a backdrop that will guide decisions so that what is delivered will help today and into the future.
Organizing the Data for the Business
While working on requirements, it is fairly easy to keep the business community engaged in the process. However, as you move to detailed data analysis and modeling, the enthusiasm and interest of the business representatives begins to wane. How can we keep them involved? We need to continue to communicate in business terms and organize that data so that it makes sense to them. This is also how data is presented through a BI interface.
The business dimensional model (BDM) is a method that allows you to capture the business perspective and carry it forward to the implementation stage. There are many different ways that data is physically stored for analysis, such as a classic star schema, a multi-dimensional cube, a columnar database, or a data warehouse appliance.
The BDM captures the dimensions visually to validate hierarchies or drill paths. In addition to the visual image, it is critical to also document business definitions and sample values of the data. The BDM also visually captures the facts, including the specific grain for each dimension that is relevant. Business descriptions and default aggregation rules are needed to complete the fact group documentation.
You can learn more about Business Dimensional Modeling by attending the Dimensional Modeling Beyond the Basics: Intermediate and Advanced Techniques session at the upcoming TDWI World Conference in Chicago (June 6-10, 2011) and reading about it in my book, A Manager’s Guide to Data Warehousing (John Wiley and Sons, Inc., May 2009).
Let’s not discount the power and capabilities of today’s tools. It is exciting to be able to explore the possibilities of how technology empowers us to build more sophisticated and powerful solutions. However, regardless of how great the technology is, we need to remember that we have to continue to strengthen our partnerships between IT and business communities, apply the technologies to solve real business problems and to organize and present the data in a manner that the business understands. These core concepts will help your data warehouse to be more successful today and for years to come.