Data mining lets you analyze large amounts of historical data – millions of transaction records -- to find hidden patterns and trends that can be used to predict future results. You can’t do that in spreadsheets or even multidimensional queries.
For example, you want to predict which customers are most likely to churn, using the data from your billing and customer service systems. You also want to know whether income, age or marital status affects churn. Without data mining, you could spend months on this analysis – and never be sure of an accurate answer. But data mining can solve this problem easily, if all the data is available and in the right format.
Complex Questions, Quicker Answers
With a well-designed data warehouse, you already have the information in the right format for data mining. You just have to choose the right model (such as a decision tree), select the data you want to study, and run the model. In a short time, you can answer a very complex question: Who are your most at-risk customers?
Until recently, data mining required highly skilled statisticians. Most data mining projects were one-off efforts. Today data mining can be performed by “power users” and can be used as the key element of ongoing marketing programs.
E.J. Barry data warehouses are designed for both multidimensional analysis and data mining. If you already have a data warehouse, we can add a data mart designed specifically for data mining. We also can build data mining models that you can deploy on your own systems.
Contact us to find out what data mining can do for you.
Web Analytics: Tailor content and product recommendations based on demographics and/or Web site behaviors.
Customer Acquisition: Segment and profile customers to determine who is most likely to buy from you and remain a loyal customer.
Retention/Churn: Predict which customers are most likely to defect to a competitor, or to not pay.
Up-Sell/Cross-Sell: Determine what products your customers are likely to purchase next based on what they just purchased. Amazon is a great example of how this works.
Sales Forecasting: Determine sales next month for your top 10 products for each of your stores.
