Combining visual customer segmentation and response modeling

Zhiyuan Yao, Peter Sarlin, Tomas Eklund, Barbro Back

    Research output: Contribution to journalArticleScientificpeer-review

    14 Citations (Scopus)

    Abstract

    Customer relationship management is a central part of Business Intelligence, and sales campaigns are often used for improving customer relationships. This paper uses advanced analytics to explore customer behavior during sales campaigns. We provide a visual, data-driven and efficient framework for customer-segmentation and campaign-response modeling. First, the customers are grouped by purchasing behavior characteristics using a self-organizing map. To this behavioral segmentation model, we link segment-migration patterns using feature plane representations. This enables visual monitoring of the customer base and tracking customer behavior before and during sales campaigns. In addition to the general segment-migration patterns, this method provides the capability to drill down into each segment to visually explore the dynamics. The framework is applied to a department store chain with more than 1 million customers.
    Original languageUndefined/Unknown
    Pages (from-to)123–134
    Number of pages12
    JournalNeural Computing and Applications
    Volume25
    Issue number1
    DOIs
    Publication statusPublished - 2014
    MoE publication typeA1 Journal article-refereed

    Keywords

    • Business Intelligence
    • Campaign-response modeling
    • Customer relationship management (CRM)
    • Customer segmentation
    • Visual analytics

    Cite this