|
useful cases from previous editions Ocean Spray: finding the news in sales results Ocean Spray Cranberries is a $1 billion fruit processing cooperative. Working with Information Resources, Inc. (IRI), it developed an extensive marketing database plus analysis capabilities using a fourth generation language called Express. The largest part of the database comes from grocery store scanner data and includes up to 100 measurements for 10,000 products in 50 geographical markets. Other data cover newspaper ads, flyers, in-store displays, and household purchase patterns. This is 10 to 100 times as much data as was available 10 years ago. Ocean Spray's mainframe computer with 10 gigabytes of secondary storage can process this detailed data, but users must figure out what to do with it. Looking at all of the data is impossible, and creating giant printouts and reports is impractical. IRI and Ocean Spray developed a number of ways to analyze the data and use models to support decision making by sales and marketing managers. One innovative tool is CoverStory, an expert system developed by IRI to generate automated news bulletins when the new data arrives. CoverStory grew out of a teaching exercise in marketing: "How would you summarize what is important in this data?" CoverStory uses a variety of models and analytical techniques to identify the important news in the latest scanner data. It reports the news as a brief memo containing data and graphs. Development is continuing as IRI and Ocean Spray gain experience with new situations. Questions:
Source: Schmitz, John D., Gordon D. Armstrong, and John D. C. Little. "CoverStory - Automated News Finding in Marketing." Interfaces, Vol. 20, No. 6, Nov-Dec 1990, pp. 29-38.
Back to Useful Cases from Previous Editions
|