This portal is no longer updated. Aalto University School of Business Master's Theses are now in the
Aaltodoc publication archive (Aalto University institutional repository)
School of Business | Department of Marketing and Management | Marketing | 2009
Thesis number: 12176
Enabling the implementation of behavioral targeting through data mining: Case e-commerce service provider
Author: | Mäki, Antti |
Title: | Enabling the implementation of behavioral targeting through data mining: Case e-commerce service provider |
Year: | 2009 Language: eng |
Department: | Department of Marketing and Management |
Academic subject: | Marketing |
Index terms: | markkinointi; marketing; tietämyksenhallinta; knowledge management; tiedonhaku; information retrieval; kohderyhmät; target groups; arviointi; evaluation; mittarit; ratings; markkinointitutkimukset; marketing research |
Pages: | 125 |
Key terms: | data mining; behavioral targeting; segmentation; marketing measurement; data louhinta; käyttäytymisperustainen kohdentaminen; segmentointi; markkinoinnin mittaaminen |
Abstract: |
The objective of this thesis is to see how the current established segmentation and marketing measurement practices and theories fit and can be translated to digital channel. In addition an objective was also to support an e-commerce company in its service development in general and its implementation of behavioral targeting practices in
particular; viewing the topics both theoretically using literature and empirically using data mining for the company’s customer data.
The literature review was conducted on a set of topics relevant to the research questions; How the established segmentation theories can be utilized in digital channel? How the methods of data mining could be utilized in e-commerce service development towards the implementation of behavioral targeting? What is behavioral targeting and what are the benefits it offers? How is marketing and digital marketing measured? From the insight provided by the literature review, a segmentation framework was applied using clickstream data as the basis of segmentation with purchase data as descriptors. The individual web behavior data (click-stream data) was used in a number of quantitative analyses. The following quantitative tools were used: logical regression to predict purchase with (parallel with a neural network for comparison), self-organizing map (SOM), a neural network for clustering, and principal component analysis to reduce the click-stream data in order to depict the web behavior in a two-dimensional map. The analysis of data consisting 2770 observations concluded that the tools of data mining can be utilized to assist and support the e-commerce service development in general and in behavioral targeting implementation in particular. It also demonstrated that clickstream data can be used as a basis for customer segmentation and that models estimating purchase can be built. In addition, the company was given recommendations on how to improve their analytics practices, an analytics shortlist, and data collecting practices within the company and how these could be utilized in decision-making. |
Master's theses are stored at Learning Centre in Otaniemi.