Data mining helps companies understand the customer better and keep a step ahead of the competition, says Dinkar Sathe

Indian enterprises are at a critical juncture in their growth curve. With intense competition, the need to differentiate themselves from others within their industry and to continuously maintain the leading edge has become critical. While focussing on the quality of their products and services, companies are placing strong emphasis on understanding the behaviour and needs of customers to reduce time-to-market and to react quickly to any evolving market changes.
To stay competitive and keep abreast of industry dynamics, Indian enterprises are using IT to generate, store and analyse mass-produced data not only for operational purposes but also to enable strategic decision-making.

Customer data is typically stored in database management systems. But ultimately data is only valuable to the extent that it facilitates better business decision-making. Current approaches to exploit business data, such as query and report writers—EIS or OLAP (On-Line Analytical Processing) applications—do not bring out important elements such as trends and current behaviour hidden in the data. Here, applications such as data mining take a leap forward in catalysing timely and strategic business decision-making.

Data mining is the appropriate set of technologies that exploit patterns of information from massive customer-focussed databases. It is the process of selecting, exploring and modelling large amounts of data to uncover previously unknown patterns of data for business advantage.
Here are some of the typical business problems that can be addressed successfully with data mining:

An established insurance company is concerned about the relatively high number of customers they have lost over the past three years (policy-holder attrition). The company is currently faced with increased competition in the marketplace due to foreign insurance companies entering as a result of deregulation. The company needs to know in advance the likelihood of a policy lapsing at the end of the policy term. Data mining can help reduce policy-holder attrition.
A credit card company experiences a drastic increase in the rate of credit card theft and therefore credit card fraud. With millions of transactions daily, the company has no way to check the validity of each individual transaction. It is looking for an automatic way to detect fraudulent transactions that can be used for all transaction data as soon as it comes in. Data mining can help in the identification of fraudulent cases.

An established bank is looking to penetrate the market with its product portfolio. The bank is interested in acquiring ‘good’ customers. Unfortunately, the bank has sketchy information about its customer base and does not have an understanding of the pockets of segments it has within its portfolio. The management has been tasked to identify these profitable pockets inherent within the portfolio. Using data mining, banks can identify these pockets, and therefore the profile of customers within these segments.

A large retail company is interested in understanding the types of products customers buy together. They are interested in testing if a top-selling product can be bundled with another product which is not doing too well.

Other possible applications for data mining include database marketing, sales forecasting, call behaviour analysis and churn management in telecommunications; forecasting of demand for utilities such as energy and water; simulation of chemical and other process reactions; finding critical factors in discrete manufacturing (aerospace, automobile, electronics); CPU usage and forecasting.

Data mining is often referred to as analytical intelligence. As is evident from its key elements, it typically involves the use of predictive modelling, forecasting and descriptive modelling techniques.

By using these techniques an organisation can proactively manage customer retention, identify cross-sell and up-sell opportunities, profile and segment customers, set optimal pricing policies, and objectively measure and rank which suppliers are best suited for their needs. Enterprises can also use data mining to minimise purchasing costs; score suppliers by rating the quality of their goods and services; identify the most effective promotions; and address numerous other organisational needs including fraud detection, failure analysis, predictive maintenance, risk management and demand forecasting.

As is evident, data mining solutions have a wide variety of applications across industries. Although its benefits look great in theory, the proof of the pudding is really in its application to real-life situations. When an insurance company is able to spot a lapsing policy before its time, when a credit card company is able to contain fraud, when a bank is able to identify profitable customers, and when a retailer is able to effectively develop its product strategy—it’s data mining at work.