Customer Super-Profiling Demonstrator to Enable Efficient Targeting in Marketing Campaigns

Abstract

Difficulties lie with identifying the right customers to engage in successful marketing campaigns. Thus far, segmentation has been a popular marketing method for selecting customer groups for targeted campaigns. However, each segment can be further exploited by performing customer profiling. In this paper, we explain the on-going development of a proposed simulator and demonstration tool that incorporates big data analytics to uncover hidden patterns within the customer dataset, thereby generating a customer super-profile. A developed toy problem and a large, realistic problem demonstrate segmentation, via clustering, and create customer profiles to enable marketers to identify appropriate marketing strategies. The proposed framework serves as the basis for enhancing customer relationship management by providing improved customer profiles for marketing campaigns. OPSOMMINGDaar is uitdagings by die identifisering van die regte kliënte vir suksesvolle bemarkingsveldtogte. Tot dusver is segmentasie 'n gewilde bemarkingmetode om te gebruik om kliëntgroepe te kies vir teikenveldtogte. Elke segment kan egter nog verder ontgin word deur kliënteprofilering uit te voer. In die artikel beskryf ons die ontwikkeling van 'n voorgestelde simulator en demonstrasiewerktuig wat groot data-analise inkorporeer om verborge patrone binne die kliëntedatastel te ontbloot, en daardeur word 'n kliëntsuperprofiel geskep. 'n Eenvoudige asook 'n groot, realistiese probleem demonstreer segmentasie deur groepering, en ontwikkel kliëntprofiele om bemarkers in staat te stel om geskikte bemarkingstrategieë te identifiseer. Die voorgestelde raamwerk dien as 'n basis om kliënte-verhoudingbestuur te versterk, deur verbeterde kliëntprofiele vir bemarkingsveldtogte beskikbaar te stel.

Customer Profiling for Malaysia Online Retail Industry using K-Means Clustering and RM Model

“…For instance, Maryani and Riana [6] measures the Euclidean distance among clusters to justify each cluster contains unique characteristics and is not overlapping with other clusters. Similarly, both Walters and Bekker [21] and Christy et. al.…”

Log-Based Session Profiling and Online Behavioral Prediction in E–Commerce Websites

“…Once customers' descriptions have been created, they are usually grouped using either clustering methods [13], [15], [16], [19], [23], [26] or classification methods [17], [18], [25]. As a result, the application of these techniques generates a set of clusters that must be subsequently interpreted to understand the particular behaviors of each class of customers.…”