Data mining techniques have the potential to greatly expand and improve customer relationship management. With the amount of information available on the internet now, companies are not limited to the data they obtained themselves about prospective and current customers. They can supplement their information with much more. However, having more information is not beneficial unless useful knowledge can be extracted from it and applied to business practices in an efficient way. The more data you have, the more the need arises to have an automated process to analyze the data for you, which is where data mining comes in.
It is not a straightforward endeavor to define and apply a mining process to a set of data or a specifiproblem. Each industry and even each company’s data and goals are different, and the process must be developed accordingly. This requires resources with insight into the data in terms of its possibilities and limitations as well as resources with knowledge of the analysis techniques themselves. In the end, the results obtained must be helpful in making real world business decisions or the time spent doing the analysis was wasted. Communication between IT and business while developing the analysis is essential because of this.
It could be classified in two ways: the aspect of CRM being improved and the type of data mining being applied. The CRM aspects being improved were divided into customer identification, customer attraction, customer retention, and customer development. The data mining techniques were divided into several categories as well: association, classification, clustering, forecasting, regression, sequence discovery, and visualization.
Each CRM area includes many different tasks, and each data mining technique has many different algorithms that can be used. Matching a task to an appropriate technique would be one of the first steps to take. Each technique has its own advantages and limitations that make it better for a certain problem or ill-suited for it. For example, customer identification involves finding potential customers or regaining past ones and includes target analysis and segmentation. These tasks require separating people into groups based on certain characteristics which lends itself to classification or clustering algorithms. To implement the classification you could choose to build a type of decision tree or to implement clustering you could choose a K-means clustering algorithm.
Along with selecting the appropriate algorithm, appropriate data must also be selected based on the business’ focus. Some data such as age can be useful across most industries, but others are only useful in specific ones. Whether or not someone owns a car is a highly-valued piece of information for the automotive industry, but is significantly less important in the cosmetics industry. Besides just being selected, the data may need to be cleansed and supplemented before applying the analysis. Often times certain information is incomplete or known to be unreliable in some cases. The algorithm can be tailored to fit these situations in order to provide better results. This tailoring can be revisited and refactored as needed to continue improving the process. Designing data mining algorithms is not an exact science and should not be treated as such.
All of this is applicable for non-profit organizations as well as for profit ones. Instead of customers, donors must be identified, attracted, retained, and developed. Instead of a promoting a product or a service, fundraisers and volunteer coordinators are attempting to connect with individuals and organizations and get them engaged in their mission. Although there is a different focus, the data still holds that potential knowledge about prospects and donors that can be leveraged.
Data mining is part art and part science and can help organizations better manage their relationships with their customers. It is not a ‘one size fits all’ solution however, and must be adapted and used thoughtfully in a way that provides benefit to those using it. With proper insight into the business as well as the data, data mining can provide new views of your customers and enhance your CRM system.
 Ngai, Eric WT, Li Xiu, and Dorothy CK Chau. “Application of data mining techniques in customer relationship management: A literature review and classification.” Expert Systems with Applications 36.2 (2009): 2592-2602.