The key for successful targeting is the knowledge about the customers’ behaviour as well as their preferences. Based on this knowledge, optimal pricing strategies can be determined and thus, the profit can be maximized. Since it is very expensive to target each customer individually, a prerequisite for effective targeting is the detection of customer groups that behave similarly and have the same preferences. In order to tackle this challenge, a hierarchical cluster analysis is conducted before the optimal price (maximized profit while considering the willingness-to-pay) is determined.
The underlying data set was collected within the seminar “Customer Analytics and Customer Insights” in 2019/20 by the students. The survey deals with the preferences regarding the purchase of 10 different jeans with a focus on Levi’s, Tommy Hilfiger, Guess and a mystery product. After data cleaning, in total 780 respondents participated. For the analysis only a subset of 650 respondents is selected – depending on the matrikel number functioning as an unique ID.
All code was written in R within RStudio.
Author: Anna Franziska Bothe
Institute: Humboldt University Berlin, Institute of Marketing
Course: Customer Analytics and Customer Insights
Semester: WS 2019/20
.
├── data # folder with all data sets (see setup.txt for details) and the codebook
├── code # folder which contains the full code as well as all subparts in individual files
├── CACI1920_ABothe_576309.pdf # final paper including all texts
├── CACI1920_ABothe_576309.Rmd # final markdown including code and texts
├── README.md # this readme file
├── requirements.txt # contains all used libraries
├── setup.txt # describes execution of pipeline in detail