Data Mining: Digging Deep Turns Your Customer Information to Gold
by Jeff Yowell
Your customer data is either an organizational asset or an organizational expense. You capture, store and maintain customer information, but have you turned it into an organizational asset that allows company leaders to make better decisions? You can, if you “mine” all of the data available to make it tell a complete story about your customers.
What are the keys to success?
Successful data mining efforts rely heavily on data collection and data integration. Organizations often make two common mistakes: They fail to collect and manage everything they can from their customer interactions, and they leave their data in disjointed data silos, making mining efforts shallow, ineffective or misleading. Developing a proactive strategy to support these essential data mining elements is critical.
Once organizations have conquered collecting the right data and organizing it in an integrated fashion, they can begin getting more from their data. Data mining efforts typically focus on:
- Campaign analysis—Before beginning a sales or marketing campaign, develop a predictive model that will help you determine which of your customers are most likely to respond to the campaign offer. Begin by identifying customers who have responded to similar campaigns or bought products similar to those being offered. Compare the attributes of those customers to others in your database to determine your targets—the customers that are most likely to respond to the campaign offer.
Next, test your marketing approach with a portion of your target group to measure response to your messaging. This step validates the campaign’s probable response rate and allows you to refine as needed before you’ve incurred the expense of the entire campaign. It also increases your campaign’s profitability by only targeting customers who are most likely to respond, allowing you to reallocate the resulting savings to additional marketing efforts. Post-campaign analysis is important, too, to determine the campaign’s ultimate effectiveness.
- Customer analysis—The most common analytics are churn analysis and segmentation. Churn analysis shows why customers leave or do not make ongoing purchases. Segmentation separates customers into groups with similar attributes, allowing you to tailor your communications to each group.
- Product/service analysis—Customers often buy products in sets or bundles. Analyzing product purchases will permit you to identify natural product bundles, allowing simpler, more profitable sales. Similarly, a next logical product analysis, which determines the typical sequence of customer purchases, lets you determine how to steer customers to the next logical purchase.
How do I get started?
Remember that it all starts with data. Pay attention to your customer contacts and the data that is (or isn’t) collected from them. Is your company gathering data from sales activity, marketing efforts, service calls, billing inquiries or Web activity? Is the data integrated? In what customer interactions are you not collecting data? The more consolidated and complete the data, the easier and more powerful the analysis.
Next, don’t be afraid to start with simple business analysis. Begin with high-level overviews—such as iden-tifying your best and your marginal customers—and gradually work into details. Even if you aren’t a statistics guru, you can still observe trends or patterns, which is the foundation of more complex analysis and data mining.
How does it happen?
Any data mining project follows the same basic steps: Understand the related business objectives; collect, analyze and prepare the data; develop and evaluate the test and put it into action. Evaluate the test in-market and modify it as needed. Be prepared to educate customers on this process, because adhering to a well-defined process ensures the greatest chance of success.
Finally, periodically review your data mining strategy. Collecting and integrating customer data will evolve as your company evolves. Consequently, your data mining strategy must evolve to take into account new and changing data variables. By taking this approach, your data mining initiatives will gain success and, in the end, support better organizational decision-making.
Jeff Yowell is the CEO of DATACORE Marketing, Inc. He can be reached by phone at 913.748.0800 or by email at jyowell@datacoremarketing.com.