Analytic Techniques: Data Mining Applications  Use the index below to go directly to a section or use the back/next buttons         Data Mining Applications Simple Profiling Cross-tabulation Regression CHAID Other Techniques   Typical Data Mining Process

Data Mining Applications

Data mining can help you reduce go-to-market costs and increase profits. It can be a powerful tool for Customer Relationship Management (CRM) as well as many other business applications. Here are just a few examples of CRM and non-CRM applications:

CRM Applications

Improved prospecting - Identify the key characteristics of the 20% of your customers who bring you 80% of your profits, so that you can more easily find more of them in the prospect population.

Better market segmentation - Develop clearer, more powerful segmentations than you could with traditional spreadsheet analysis or other simple statistical techniques.

Increased customer loyalty - Easily distinguish between customers who are most likely to remain loyal and those who are at greatest risk of attrition. Create data-based, personalized outreach efforts to retain potential disloyals; and save time and money by focusing future prospecting efforts on those prospects who most resemble current loyals.

Clearer customer relationship definitions - If your customer base includes a mix of businesses and consumers, there is often a subgroup of customers that are difficult to classify clearly as either businesses or consumers. Data mining can help you sort through and clarify your relationship with these customers, so that targeted business/consumer outreach efforts can better address their needs.

More successful cross-selling and upselling - Sophisticated data mining techniques can help you identify the next product to sell to each individual customer, whether by direct mail, telemarketing, customized Web site content delivery or some other vehicle.

Risk management (bad-debt or insurance-risk prediction, fraud detection, etc.) - By developing and applying models using historical data, data mining can help you reduce the future incidence of these problems.

More effective and efficient broad-reach media spending - Data mining can be useful not only for improving highly personalized communications strategies, but also for optimizing broader-reach media spending across markets. For example, sophisticated data mining techniques can supplement and enhance standard media-analytic techniques to better optimize TV spending by DMA or ADI.

Beyond CRM

In addition to the myriad applications of data mining to CRM efforts, there are many ways to apply data mining to other areas of your business. Here are a few examples dealing with cost reduction, engineering, etc.:

Failure analysis - Using specialized analytic techniques to identify how maintenance schedules and various operational influences affect a component's time-to-failure.

More effective and efficient bundling of product features - Satisfy customers' needs while reducing unnecessary product-line complexity. By better understanding the relative frequency with which customers prefer various combinations of product features, significant cost reduction can be achieved by eliminating less profitable configurations.

Manufacturing problem identification and correction - Applying sophisticated data mining techniques to the analysis of statistical process control data can help identify and fix production-line problems.

In general, wherever data exist, powerful data mining techniques can help reveal important data patterns that would otherwise remain unnoticed when using spreadsheet or other simple types of analysis. This enhanced understanding can significantly improve a company's bottom line.

					Data Mining Applications
Simple Profiling	Cross-tabulation	Regression	CHAID	Other Techniques		Typical Data Mining Process