When used wisely within Customer Relationship Management applications data mining can significantly improve the bottom line. It will end the process of randomly contacting a prospective or current customer through a call centre or by mailshot. With the effective use of data mining a company can concentrate its efforts on targeting prospects that have a high likelihood of being open to an offer. This in turn gives the ability for more sophisticated methods to be used such as campaigns being optimised to individuals.
Businesses that employ data mining techniques will usually see a high return on investment, but will also find that the number of predictive models can quickly increase. Rather than just implementing one model to predict which customers will respond positively, a business could build a different models for each region and customer type. Then instead of sending an offer to all prospects it may only want to send to prospects that have a high chance of taking up the offer. It may also want to determine which customers are going to be profitable during a certain time frame and direct their efforts towards them. To be able to maintain this quantity and quality of models, these model versions have to be well managed and automated data mining implemented.
Human Resources departments can also make a valid case for using data mining. It will allow them to in identifying the characteristics of their most successful employees. Information gained from such as resource can help HR focus their recruiting efforts accordingly.
Another example of data mining, is that used in retail. Often called market basket analysis, it is, for example, when a store records the purchases of customers, it could identify those customers who favour silk shirts over cotton ones; or customers who bought certain grocery items would also also buy the same specific item as well. This is often highlighted in on-line stores when you are told that so many people who bought a certain book or CD also bought XX as well.
Although some explanations of relationships may be difficult, taking advantage of it is easier. The example deals with association rules within transaction-based data. Not all data are transaction based and logical or inexact rules may also be present within a database. In a manufacturing application, an inexact rule may state that 73% of products which have a specific defect or problem will develop a secondary problem within the next six months.
About the Author: Mike has more than 15 years ox experience designing and implementing Data warehouses based on Oracle, MS SQL Server, MySql, PostgreSQL and more he is currently working for DB Software Laboratory