The importance of social embeddedness: Churn models at mobile providers

Gábor Benedek, Ágnes Lublóy, G. Vastag

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

This article argues the importance of social embeddedness at mobile providers by examining the effects of customers' network topological properties on churn probability-the probability of a customer switching from one telecommunication provider to another. This article uses data from regional snowball sampling-the only practically feasible network sampling method-to identify groups with significantly different churn ratios for customers with different network topological properties. Clear evidence indicates that individual network characteristics (node-level metrics) have considerable impact on churn probabilities. The inclusion of network-related measures in the churn model allows a longer-term projection of churners and improves the predictive power of the model. With no possibility to carry out repeated sampling, sample stability was checked through simulation results. On the one hand, this article highlights the importance and effectiveness of the provider's tailored marketing campaigns by showing that customers targeted by direct marketing campaigns are less threatened by churn than nontargeted customers. On the other, this article shows that social embeddedness blocks the impact of the very same marketing efforts. This article forwards the idea that social embeddedness, also prevalent in vendor switching, can be extended to understanding the development of professional societies threatened by membership churn.

Original languageEnglish
Pages (from-to)175-201
Number of pages27
JournalDecision Sciences
Volume45
Issue number1
DOIs
Publication statusPublished - Feb 2014

Fingerprint

Marketing
Sampling
Telecommunication
Social embeddedness

Keywords

  • Cluster Analysis
  • Consumer Behavior
  • Customer Relationship Management
  • Network Analysis
  • Social Network Analysis
  • Supply Chain Risk Management
  • Telecommunications

ASJC Scopus subject areas

  • Management of Technology and Innovation
  • Strategy and Management
  • Business, Management and Accounting(all)
  • Information Systems and Management

Cite this

The importance of social embeddedness : Churn models at mobile providers. / Benedek, Gábor; Lublóy, Ágnes; Vastag, G.

In: Decision Sciences, Vol. 45, No. 1, 02.2014, p. 175-201.

Research output: Contribution to journalArticle

Benedek, Gábor ; Lublóy, Ágnes ; Vastag, G. / The importance of social embeddedness : Churn models at mobile providers. In: Decision Sciences. 2014 ; Vol. 45, No. 1. pp. 175-201.
@article{d2f5c602a82e493b9544634cdfb52ba0,
title = "The importance of social embeddedness: Churn models at mobile providers",
abstract = "This article argues the importance of social embeddedness at mobile providers by examining the effects of customers' network topological properties on churn probability-the probability of a customer switching from one telecommunication provider to another. This article uses data from regional snowball sampling-the only practically feasible network sampling method-to identify groups with significantly different churn ratios for customers with different network topological properties. Clear evidence indicates that individual network characteristics (node-level metrics) have considerable impact on churn probabilities. The inclusion of network-related measures in the churn model allows a longer-term projection of churners and improves the predictive power of the model. With no possibility to carry out repeated sampling, sample stability was checked through simulation results. On the one hand, this article highlights the importance and effectiveness of the provider's tailored marketing campaigns by showing that customers targeted by direct marketing campaigns are less threatened by churn than nontargeted customers. On the other, this article shows that social embeddedness blocks the impact of the very same marketing efforts. This article forwards the idea that social embeddedness, also prevalent in vendor switching, can be extended to understanding the development of professional societies threatened by membership churn.",
keywords = "Cluster Analysis, Consumer Behavior, Customer Relationship Management, Network Analysis, Social Network Analysis, Supply Chain Risk Management, Telecommunications",
author = "G{\'a}bor Benedek and {\'A}gnes Lubl{\'o}y and G. Vastag",
year = "2014",
month = "2",
doi = "10.1111/deci.12057",
language = "English",
volume = "45",
pages = "175--201",
journal = "Decision Sciences",
issn = "0011-7315",
publisher = "Wiley-Blackwell",
number = "1",

}

TY - JOUR

T1 - The importance of social embeddedness

T2 - Churn models at mobile providers

AU - Benedek, Gábor

AU - Lublóy, Ágnes

AU - Vastag, G.

PY - 2014/2

Y1 - 2014/2

N2 - This article argues the importance of social embeddedness at mobile providers by examining the effects of customers' network topological properties on churn probability-the probability of a customer switching from one telecommunication provider to another. This article uses data from regional snowball sampling-the only practically feasible network sampling method-to identify groups with significantly different churn ratios for customers with different network topological properties. Clear evidence indicates that individual network characteristics (node-level metrics) have considerable impact on churn probabilities. The inclusion of network-related measures in the churn model allows a longer-term projection of churners and improves the predictive power of the model. With no possibility to carry out repeated sampling, sample stability was checked through simulation results. On the one hand, this article highlights the importance and effectiveness of the provider's tailored marketing campaigns by showing that customers targeted by direct marketing campaigns are less threatened by churn than nontargeted customers. On the other, this article shows that social embeddedness blocks the impact of the very same marketing efforts. This article forwards the idea that social embeddedness, also prevalent in vendor switching, can be extended to understanding the development of professional societies threatened by membership churn.

AB - This article argues the importance of social embeddedness at mobile providers by examining the effects of customers' network topological properties on churn probability-the probability of a customer switching from one telecommunication provider to another. This article uses data from regional snowball sampling-the only practically feasible network sampling method-to identify groups with significantly different churn ratios for customers with different network topological properties. Clear evidence indicates that individual network characteristics (node-level metrics) have considerable impact on churn probabilities. The inclusion of network-related measures in the churn model allows a longer-term projection of churners and improves the predictive power of the model. With no possibility to carry out repeated sampling, sample stability was checked through simulation results. On the one hand, this article highlights the importance and effectiveness of the provider's tailored marketing campaigns by showing that customers targeted by direct marketing campaigns are less threatened by churn than nontargeted customers. On the other, this article shows that social embeddedness blocks the impact of the very same marketing efforts. This article forwards the idea that social embeddedness, also prevalent in vendor switching, can be extended to understanding the development of professional societies threatened by membership churn.

KW - Cluster Analysis

KW - Consumer Behavior

KW - Customer Relationship Management

KW - Network Analysis

KW - Social Network Analysis

KW - Supply Chain Risk Management

KW - Telecommunications

UR - http://www.scopus.com/inward/record.url?scp=84893140408&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893140408&partnerID=8YFLogxK

U2 - 10.1111/deci.12057

DO - 10.1111/deci.12057

M3 - Article

AN - SCOPUS:84893140408

VL - 45

SP - 175

EP - 201

JO - Decision Sciences

JF - Decision Sciences

SN - 0011-7315

IS - 1

ER -