The Data to Make Decisions.

Data warehouses help companies keep up with their customers' needs.

There has never been a better time to invest in data warehousing. Today's dynamic business environment is driven by financial-services globalization, industry consolidation, increasing customer demands and ubiquitous technology, to name a few. To succeed, companies must leverage their information assets not only for corporate strategy development but also to enable tactical decision-making in the field for underwriting, claims and sales.

Data warehousing is evolving. Successful data warehouses deliver business value on an iterative and continuous basis. Each iteration builds upon its predecessor to increase business value. Early data warehouses focused on providing strategic decision-making in home office functions-such as marketing, strategic planning and finance-which resulted in access to information that dramatically increased the decision-making quality and corporate-strategy development. But a great strategy is nothing without effective execution. The emerging generation of data warehousing improves the execution of a business strategy through dynamic measurement and management.

There are five stages that are most common in the maturation of decision support within an organization: reporting, analysis, prediction, operations and disintermediation.

* Stage 1, Reporting: What Happened? Stage 1 typically focuses on reporting from a single source of truth within an organization. Tremendous value results simply by integrating disparate sources of information into a single repository to drive decision-making across functional and/or product boundaries. The resulting well-integrated information for decision-makers becomes the foundation for all subsequent stages of data-warehouse deployment.

* Stage 2, Analysis: Why Did It Happen? In stage 2, decision-makers focus less on what happened and more on why it happened. Analysis activities are concerned with drilling down beneath the numbers on a report to slice-and-dice data at a detailed level. Hands-on ad hoc analysis with iterative refinement of questions in an interactive environment is required, as questions are not known in advance as they were in stage 1.

* Stage 3, Prediction: What Will Happen? As an organization becomes experienced in quantitative decision-making techniques and understands the "what" and "why" of its business dynamics, the next step is to leverage information for predictive purposes. Understanding what will happen next is imperative to managing the strategy. Data-mining tools and methods are required for building predictive models using historical detail. Advanced methods often employ complex mathematical functions-such as logarithms, exponentiation, trigonometric functions and sophisticated statistical functions-to obtain the predictive characteristics desired. Access to detailed data is essential. Typical users are sophisticated analysts with market research, statistical or actuarial backgrounds.

* Stage 4, Operations: What is Happening? Stage 4 starts to bring us into the realm of active data warehousing, focusing on tactical decision support. Tactical decision support focuses on supporting the people in the field (e.g., regional and branch claims, underwriting, marketing, sales managers) who execute it. This means continuous data updates to the data warehouse for decision-making capabilities to be relevant to day-to-day operations.

* Stage 5, Disintermediation: What Do I Want to Happen? The larger the role an active data warehouse plays in the operational aspects of decision support, the more incentive the business has to automate its decision processes-both for efficiency and consistency in decision-making. E-commerce business models dictate automated decision-making when a customer interacts with a Web site. Interactive customer relationship management is all about making decisions to optimize the customer experience through individualized product information and offers, quotes and policyholder or claim services. Increasingly, decisions will be executed with life triggers (birth, graduation, etc.) and other events (buying a home or car, moving) to initiate fully automated decision processes.

The underlying philosophy of an active data warehouse is to increase the speed and accuracy of business decisions. Its goal is to allow decision-making to take place as near to real time as necessary to deliver maximum business value and enable decision-making to manage the business quickly and effectively. It's all about integrating advanced decision support with day-to-day, minute-by-minute, decision-making that increases the quality of those customer interactions, encourages customer loyalty and drops right to the bottom line.

Patricia L. Saporito is a senior insurance industry consultant for Teradata, a division of NCR Corp., New York.