International Business Machines Corporation

Steve Colten

Paritosh Desai

Sean McCauley

Suzanne Noel Valentine

Rob Parkin

Jason Lee Gunnink

Sean Kervin

Howard Yihzan Wu

Phil Delurgio

Craig Silverman

Paul Algren

The present invention relates to a system and method for customer segment analysis. The system receives customer transaction data for the generation of segments, including point of sales data and customer identification information. Errors in the data may be resolved. Then, individual customers may be segmented by statistically relevant groups. The segmentation of consumers may be accomplished by comparing data of known customers to known segments. Unknown customers, new customers and point of sales data which is missing customer data may also be segmented via statistical similarity to known segments. Using the customer segments, segment wide point of sale data may be generated. This data may then be aggregated by consumer groups. Consumer groups may include by household or other communal purchasing entity. The aggregated segment data may be validated and transformed for outputting to the optimization system. The price optimization system may the use the segment data for generation of preferred prices. Feedback from the optimization engine may be utilized by the customer segmentation system to generate customer segment specific promotional activity.

Analyzing customer segments

The present invention relates to a system and method for analyzing product roles. The system receives a listing of products for classification into roles. The system receives volume data for each item, as well as demand coefficient. Elasticity of the products may be determined from the demand coefficients. Product volumes and elasticities may then be compared against one another by graphing the product by its volume versus elasticity. From this comparison the products may be classified into one or more roles. These roles include image items, niche products, assortment completers, and profit drivers. The assortment completer role is populated with products which have high relative elasticity and low relative volume. Niche product role is populated with products which have low relative elasticity and low relative volume. The image item role is populated with products which have high relative elasticity and high relative volume. And lastly, the profit driver role is populated with products which have low relative elasticity and high relative volume. This comparison may also include generating an 'image value' for the product.

Product role analysis

Growing community of inventors

Reno, NV, United States of America

Steve Colten