Business Ignorance or Business Intelligence

The effective use of business intelligence by insurance companies over the next three years will differentiate market leaders from market laggards in capturing, servicing and retaining profitable market segments. The rising interest in the use of business intelligence (BI) is a reaction to a number of concerns, including a return to a soft market, competitive risk and differentiation, and operational efficiency versus cost reduction. Further, regulators and rating agencies have raised awareness of the need for more standardization, consistency and predictability in the business decisions that carriers make. BI is an emerging, high-interest topic for senior managers to find opportunities to differentiate their firms in the market, to protect against business loss to competitors and to maximize the financial results of core transactions.

Carriers are implementing BI primarily for three purposes:

1. Customer relationship management. Carriers are applying BI in CRM for marketing, cross-sell and up-sell during service calls, and managing and controlling consumer self-service through Web portals.

2. Enterprise risk management. Carriers are applying BI to analyze individual risk, manage entire portfolios of risk and manage internal risks that affect the entire organization.

3. Enterprise performance management. Carriers are using BI to analyze their performance metrics on a daily basis and, in some cases, an hourly basis to discover performance parameters that are outside the norm.

Enterprise Potential

BI solutions can be applied to a wide variety of business issues, such as channel preference, product offering, fraud and subrogation; often, a focused success in one area leads to adoption in others. While the use of BI in isolation provides valuable insight, the real power of BI is in the enterprisewide application of it and its use beyond a single business unit or functional transaction. As carriers evaluate the advantages of BI, it is important to have a clear understanding of the differences between data mining and predictive analytics, as capabilities vary widely and offer differing performance, from empirical to retrospective results.

The focus of data mining is generally on the past. Data mining is a means to search large databases to extract data that may be useful in determining what has happened and what caused it to happen. For example, data mining is useful in profiling customers to define common characteristics and for segmenting the customer base into risk groups.

Predictive analytics is the application of data analysis that focuses on the future. It combines analyses in a model to predict outcomes of various events. A predictive analytics query includes parameters, but fewer specifics than with data mining. Unlike static rules, predictive analytics utilizes sophisticated computer algorithms to identify subtle patterns across hundreds of fields of data and to adjust as patterns change or new ones emerge.

Best Practices

Best practices in BI include managing core data and external data, analyzing the data, and executing analytic models and scorecards on transactions being processed in real time. At the core of any BI platform is accurate, comprehensive and complete data that is error-free. From that, BI is developed in data warehousing and reporting, data mining and analysis, predictive analytics and scorecards, and real-time analytics and automated decisions. This hierarchy of business insight builds expertise and develops a higher level of capability with combinations of tools, experience and infrastructure. As the demand for data insight grows, so does the business value derived - along this hierarchy, business rules are better defined and insight into a company's potential develops. As senior managers apply their learning, opportunities to leapfrog over the competition become clear and a carrier's performance improves.

BI gives senior managers the keys to distinguish their firms and articulate their differentiating capabilities, which is imperative in a marketplace that is increasingly commoditized.