Publication detail

Customer Satisfaction Modelling in Digital Marketing Era

SCHÜLLER, D. PEKÁREK, J.

Original Title

Customer Satisfaction Modelling in Digital Marketing Era

English Title

Customer Satisfaction Modelling in Digital Marketing Era

Type

conference paper

Language

en

Original Abstract

This paper deals with the use of data-driven marketing as the important part of digital marketing. Data are valuable and necessary input for sellers as they are able to make decisions in many fields of digital marketing. This study is focused on the customer satisfaction modelling whereas the main aim is to find out models that optimally describe total customer satisfaction and to determine the importance of individual factors and discover their impact on the total satisfaction. The process of modelling is demonstrated in the field of free-time dance industry. The emphasis is placed on the course attendants rated their satisfaction in particular areas as well as their overall satisfaction with the dance courses. The authors employed regression analysis to measure the importance of individual factors on overall satisfaction to find out what factors should be focused on. To do that, two separate models were developed, both with a clustered and non-clustered version. The first model uses individual factors as inputs, the second works with semantically differentiated factor groups. Both models perform poorly in the non-clustered case. The first model significantly improves its explanatory power when we apply it to clustered data obtained by clustering process using Ward's criterion, for which the factors are ordered according to their significance. When customer clusters are compared to each other, the most important are course attendance, price and instructor expertise for each cluster respectively. These findings can be used as the input for customer relationship management module to design appropriate products for individual customer groups or to improve overall quality of product portfolio. Moreover the clustering method provides more precise data and should be an integral part of customer relationship management module so that marketers can take more efficient measures.

English abstract

This paper deals with the use of data-driven marketing as the important part of digital marketing. Data are valuable and necessary input for sellers as they are able to make decisions in many fields of digital marketing. This study is focused on the customer satisfaction modelling whereas the main aim is to find out models that optimally describe total customer satisfaction and to determine the importance of individual factors and discover their impact on the total satisfaction. The process of modelling is demonstrated in the field of free-time dance industry. The emphasis is placed on the course attendants rated their satisfaction in particular areas as well as their overall satisfaction with the dance courses. The authors employed regression analysis to measure the importance of individual factors on overall satisfaction to find out what factors should be focused on. To do that, two separate models were developed, both with a clustered and non-clustered version. The first model uses individual factors as inputs, the second works with semantically differentiated factor groups. Both models perform poorly in the non-clustered case. The first model significantly improves its explanatory power when we apply it to clustered data obtained by clustering process using Ward's criterion, for which the factors are ordered according to their significance. When customer clusters are compared to each other, the most important are course attendance, price and instructor expertise for each cluster respectively. These findings can be used as the input for customer relationship management module to design appropriate products for individual customer groups or to improve overall quality of product portfolio. Moreover the clustering method provides more precise data and should be an integral part of customer relationship management module so that marketers can take more efficient measures.

Keywords

digital marketing era, customer satisfaction, linear regression modelling, clustering, classification of factors

Released

01.09.2018

Publisher

STEF92 Ltd.

Location

Albena Bulgaria

ISBN

978-619-7408-65-2

Book

SGEM INTERNATIONAL MULTIDISCIPLINARY SCIENTIFIC CONFERENCE ON SOCIAL sCIENCES AND ARTS

Edition

5

Edition number

1.5

Pages from

233

Pages to

240

Pages count

8

BibTex


@inproceedings{BUT160555,
  author="David {Schüller} and Jan {Pekárek}",
  title="Customer Satisfaction Modelling in Digital Marketing Era",
  annote="This paper deals with the use of data-driven marketing as the important part of digital marketing. Data are valuable and necessary input for sellers as they are able to make decisions in many fields of digital marketing. This study is focused on the customer satisfaction modelling whereas the main aim is to find out models that optimally describe total customer satisfaction and to determine the importance of individual factors and discover their impact on the total satisfaction.  The process of modelling is demonstrated in the field of free-time dance industry. The emphasis is placed on the course attendants rated their satisfaction in particular areas as well as their overall satisfaction with the dance courses. The authors employed regression analysis to measure the importance of individual factors on overall satisfaction to find out what factors should be focused on. To do that, two separate models were developed, both with a clustered and non-clustered version. The first model uses individual factors as inputs, the second works with semantically differentiated factor groups. Both models perform poorly in the non-clustered case. The first model significantly improves its explanatory power when we apply it to clustered data obtained by clustering process using Ward's criterion, for which the factors are ordered according to their significance. When customer clusters are compared to each other, the most important are course attendance, price and instructor expertise for each cluster respectively. These findings can be used as the input for customer relationship management module to design appropriate products for individual customer groups or to improve overall quality of product portfolio. Moreover the clustering method provides more precise data and should be an integral part of customer relationship management module so that marketers can take more efficient measures.",
  address="STEF92 Ltd.",
  booktitle="SGEM INTERNATIONAL MULTIDISCIPLINARY SCIENTIFIC CONFERENCE ON SOCIAL sCIENCES AND ARTS",
  chapter="160555",
  doi="10.5593/sgemsocial2018/1.50",
  edition="5",
  howpublished="electronic, physical medium",
  institution="STEF92 Ltd.",
  year="2018",
  month="september",
  pages="233--240",
  publisher="STEF92 Ltd.",
  type="conference paper"
}