Business Intelligence Best Practices -

Collaboration. Communication. Community.


Awards - 2007 Best Practices Categories

In an effort to reflect new user practices in the best practices awards, TDWI refines the award categories on an annual basis. New for 2007: Enterprise BI and Operational BI.

1. Enterprise Data Warehousing - The enterprise data warehouse (EDW) delivers a consistent set of data for the entire organization. The EDW is the sole source of integrated, atomic, and historical information that the organization uses to drive key analytical and operational processes. The EDW enables the business to work in an efficient, coordinated fashion since all users and applications work off the same set of information and rules.

2. Customer Intelligence - Many organizations use enterprise data warehouses to integrate customer data. There are many strategies to integrate customer data, but the end result is the same: to provide a 360-degree view of customers so that organizations can optimize customer interactions across all channels. This customer-centric information is used to attract, enhance, and maintain customer relationships.

3. Dashboards and Scorecards - Organizations use dashboards and scorecards to monitor the degree to which they are achieving key strategic objectives and goals. BPM applications translate top-level strategy and goals into measures and initiatives specific to every level of the organization. The applications then measure performance against those goals and provide timely information so users can act to change outcomes before it's too late.

4. Enterprise BI - Enterprise BI delivers information and insights to all users in an organization. But rather than provide a "one size fits all" approach, enterprise BI tailors the delivery and content of information to meets the unique requirements of different groups of users, from executives and managers to business analysts and power users to customers and suppliers. Organizations typically standardize on a variety of BI tools to deliver enterprise BI. Sometimes these toolsets all come from a single vendor with a comprehensive BI platform; in other cases, the tools come from a variety of vendors whose capabilities uniquely mirror organizational requirements.

5. Operational BI - Traditionally, BI solutions provide users with historical data collected in a batch process on a monthly, weekly, or nightly basis. Now, companies are beginning to deliver data to users for decision making on a hourly or near instantaneous basis using a variety of methods. In some cases, companies are updating their data warehouses or operational data stores using trickle feed techniques; in other cases, organizations are querying operational systems directly using query or enterprise information integration tools. Whatever the technique, operational BI solutions help business users get information in a more timely fashion so they can work more proactively.

6. Data Governance - Data governance (or data stewardship) is an activity that ensures executive support, funding, direction, and quality for BI/DW and other data environments. An effective data governance program focuses on the usability, reusability, quality, accessibility, and security of data. More to the point, however, organizations that excel at data governance foster a tight working relationship between business and IT, so that together they manage and leverage all data assets for organizational benefit.

7. Master Data Management - Master data management (MDM) is a best practice that achieves a single, global view of a business entity - usually customer or product. The single view may reside in a physical database, like a warehouse or other BI data store. But the trend in MDM is to use federated or virtual styles of data integration. Whether an MDM implementation focuses on data analysis, data quality, or integrating applications in real-time, it provides views that are more complete, accurate, up-to-date, and standardized than those of individual applications. Note that MDM capabilities may be embedded in software solutions, especially those for customer data integration (CDI) and product information management (PIM).

8. Unstructured Data Integrated with BI/DW - Corporate executives and other organizational leaders base their decisions on many sources of information, including documents of mostly natural language text, which includes word processor files, Web pages, scanned documents, and the text fields of database records. Yet, IT has only recently begun integrating so-called "unstructured data" into the strictly structured world of the data warehouse and other BI data stores. This newest frontier for data integration and data warehousing promises more informed and accurate decision-making for its end-users.

9. BI/DW on a Limited Budget - Many organizations face considerable financial constraints when building BI environments. This requires project sponsors and managers to come up with clever ways to stretch a limited budget to deliver real and sustained value to the organization. It also often requires setting realistic goals, rolling out functionality in a phased approach, and evangelizing and selling the organization on the value of BI at every opportunity.

10. Predictive Analytics - Companies are increasingly using data mining to unearth patterns or correlations in data to deliver business value. For example, established applications range from money laundering and fraud detection to customer-base segmentation and cross-sell recommendations. In response to recent edicts from homeland security and federal legislation, government agencies and corporations alike have ramped up predictive analytics to combat national security risks, automotive part failures, illegal trafficking, insider trading, and questionable accounting practices.

11. Radical Business Intelligence and Data Warehousing - Some organizations have gained a competitive advantage by bucking mainstream principles about how to construct BI/DW solutions. These organizations use radical new approaches or technologies to address problems and issues where traditional methods have failed or proven cost-prohibitive. Often these organizations are pushing the envelope of massive deployments or have significant cost constraints that force them to think "outside the box."

12. Government and Non-Profit - Government, education, and other not-for-profit agencies face unique challenges when implementing and managing BI/DW solutions. These organizations often must sell, justify, and sustain BI/DW projects using different tactics from those employed by commercial organizations. However, the impact of these projects can be far-reaching, often providing citizens and businesses much easier access to information and beneficial programs and services.