Implemention Technology
Business Intelligence
  DataSage Customer Analyst
  Oracle Express Analyzer
Warehouse Engine
  Oracle
Hardware
  Dell Dimension

Founded in 1950, Dick's Supermarkets offers a wide selection of products, and we pride ourselves on our innovative, exceptional customer service. Our stores serve customers in predominantly rural areas of southwestern Wisconsin and northern Illinois. Dick's Savings Club, a customer loyalty card program, was introduced in 1994 and now has more than 60,000 active cardholders. Approximately 90 percent of total store sales are recorded in the database using the card program. Our mission is to build market position by differentiating ourselves from our competition and to continually work at strengthening relationships with our best customers.

Dick's has implemented a wide range of marketing programs including reward programs, tiered pricing, direct mail to 28,000 of our best customers and limited newspaper advertising. While we had varying degrees of success, the missing elements in all of these programs involved having easy access to data (creating targeted program initiatives) and easy analysis of the results. Our goal is to improve customer profitability and loyalty. By utilizing DataSage's Customer Analyst, Dick's customer data is analyzed on an individual basis, allowing us to develop targeted marketing programs that enhance loyalty and produce tangible results that positively impact our ROI.

Dick's was the first beta site for DataSage's Customer Analyst software. The implementation, which included project support and training by DataSage consultants, took approximately two months until full operation. Installation began last October, with the task of loading 18 months of data. User training began in November, and the system was fully functional in December. Customer Analyst is an open system, object oriented application. The platform runs on Windows NT 4.0, and the database is supported by Oracle. Dick's is using a client/server architecture with Dell Dimension as the server running the data mining and OLAP functions. On the client side, Oracle Express Analyzer enables the reports to be viewed as it pushes the reports to the user. Oracle Express features multidimensional OLAP (MOLAP) that allows queries to be automated. This automation makes the queries "invisible" because any information the user would request has already been stored within the cubed-dimension of the MOLAP. Approximately 100 megabytes of data from Dick's data warehouse are crunched weekly, with a total of two gigabytes stored each week. Currently, the warehouse is holding 20 gigabytes of data.

We found that Customer Analyst is ideal for meeting our business objectives. It is self-sufficient and able to handle the following tasks: maintaining customer data, accessing customer information easily, creating targeted promotions, developing reports easily, automating weekly processing and using available information to improve profits. To date, the product has produced measurable results, and the impact on our bottom line has been immediate with clear revenue and profit increases in certain product categories.

Using Customer Analyst, we can go beyond traditional measures, finding significant patterns so we can take actions to influence customer buying behavior. For example, we are able to track such variables as price elasticity, product affinities and customer retention scores, predicting the likelihood of defection to other stores on a category basis.

To test price elasticity, Dick's selected 75 to 100 SKUs in categories such as peanut butter, orange juice, canned vegetables and baby foods. Three stores were designated as test sites alternating control stores (no changes) with price-change stores. In one example, the price was lowered on a certain brand of orange juice, and the price was raised on a brand of canned vegetables for a period of four weeks. Comparing the control stores with the price-change stores, Dick's found an increase in sales of 29 percent for the orange juice and 20 percent for the vegetables.

Another interesting analysis entailed testing product affinities and promotional displays. For example, the software revealed that yogurt and granola bars have 10 percent attachment rate ­ that is, customers are likely to purchase one item in conjunction with the other. In the store, we put full retail-priced granola bars near a discounted yogurt display, which resulted in a 60 percent lift for granola bar sales. In another test, full retail-priced canned milk was placed near a discounted display for pie fillings, and the test stores noted a 24.5 percent revenue lift for those items.

Customer Analyst helps us to combat customer defection by analyzing category drops and closing the loop on promotional activity. As a result of lost cereal sales, Dick's pinpointed existing customers that showed a decrease in cereal sales compared to previous visits. The group was segmented into customers with a high or low propensity to respond to promotions. A personalized letter was sent with coupons to each person in the "high propensity to respond to promotions" group. Not only was the promotion successful in changing purchase behavior for those customers, but also coupon redemption exceeded 20 percent in certain categories.

One important factor retailers need to focus on when utilizing any data warehouse and data analysis technique is the integrity of the data itself. A tool, such as Customer Analyst, that can capture true data and hard-core numbers and provide accurate analysis benefits the entire organization because it facilitates business decision making which ultimately impacts a company's profit. Without data integrity, the data is misleading and is of little value. To avoid obtaining worthless data, it often is best for an organization to implement a data warehouse and data analysis application using expert third-party vendors. Good third-party vendors will offer support and technical expertise to run an effective and productive system. Running a successful data warehouse requires that the appropriate solutions be in place and the third-party vendor have a crystal-clear understanding of the business goals and objectives.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access