UNIVERSITY OF GRONINGEN FACULTY OF ECONOMICS AND BUSINESS MASTER DEFENSE: MSC MARKETING INTELLIGENCE
JUNE 24, 2020
THE INFLUENCE OF REGULARITY AND CHANGE IN ACTIVITY ON
CUSTOMER CHURN
SANDY KLEIN (S3406105)
INTRODUCTION & THEORY METHODOLOGY
RESULTS
DISCUSSION
TABLE OF CONTENTS
THE BUILDING BLOCKS OF A FIRM:
“MANAGE CUSTOMER CHURN BY IDENTIFYING
THOSE CUSTOMERS AND TAKING APPROPRIATE
ACTION TO RETAIN THEM”
INTRODUCTION
INTRODUCTION
Analyzing customer churn matters:
- Helps firms to retain customers ;
- Acknowledges customer motivations;
- Creates a financially healthy company structure . However,
- There are areas of dissonance ;
- It is difficult to track customer motivations . Other researches focused on:
- Customer interaction choices and future behavior ;
- Timing patterns .
But, research is mostly done in non-contractual settings.
1
2
3
4
5 6
Footnotes explained in bibliography
WHAT ABOUT CONTRACTUAL SETTINGS?
INTRODUCTION
Different contexts may require different approaches. Therefore:
- This study focuses the influence of regularity and change in activity on customer churn in contractual settings
- Regularity and customer churn is moderated by different activity types
REGULARITY (-)
CHANGE IN ACTIVITY (+)
CUSTOMER CHURN ADMINISTRATION (+)
INFORMATION (-) MOBILE (+)
ACTIVITY TYPES:
AGE GENDER
TOTAL ACTIVITY
METHODOLOGY
METHODOLOGY
Dataset of website information (page types, churn) of an European insurance company with 15,000 customers.
- Model I: Logistic regression (cross-sectional)
Regularity measure across sessions (n = 9,735): measure of the complete customer journey and taking the standard deviation of every time-interval between sessions.
Moderating effects of activity types: Information, Administration, Mobile Control variables
- Model II: Individual effects (panel data)
Change in activity within sessions (n = 219,221): measure of a standard deviation of a rolling 9 window of the activity (based on AIC)
- Model III: additional analysis
Timing of customer churn based on change in activity Conditional probability
RESULTS
RESULTS
Model I
β Marg. Eff. Odds
Intercept .268 30.8%
Regularity .005* .001* 0.1%
Information .052 .009 0.5%
Administration -.037 -.007 -3.7%
Mobile -.046* -.009* -4.5%
R*I .001 .001 0%
R*A .001 .001 0%
R*M .004* .023* 0.1%
Age -.032*** -.006*** -3.3%
Gender -.122* -.022* -11.5%
Total activity -.003** -.001** -2.3%
Model I main results:
- Regularity has a positive effect on customer churn;
- The interaction effect of mobile and regularity has a positive effect on customer churn;
- Age has a negative effect on customer churn;
- Gender has a negative effect on customer churn;
- Total activity has a negative effect on customer churn;
- Forecasting performance of 77.04% (TDL = 3.23, Gini = .18).
Model II main results:
- Fixed effects better than random and pooled effects;
- Change in activity has a negative effect on customer churn.
Model III main results:
- Probability of churning increases over time when change in activity increases as well.
Model II
β st. Error P-value Change in
activity -.002 .001 <. 001
DISCUSSION
DISCUSSION
Striking but inverse results for regularity and change in activity…
- Customer motivations in contractual settings differ from non-contractual settings;
- Regular customers lead to churn, and changes in activity decreases customer churn;
- Timing of taking appropriate actions depends on changes in activity.
After a year, the probability of customer churn increases with 35%
Limitations
- Bug in package
- Quarterly effects in insurance companies Recommendations
- Managerial: not wake sleeping dogs;
- Research: study quarterly effects
Q&A
BIBLIOGRAPHY
Picutures: jcomp / Freepik
Drotski, A. (2011). Customer and organisational value added through customer experience differentiation. Journal of Economic and Financial Sciences, 4(1), 51-62.
Gupta, S., Lehmann, D. R., & Stuart, J. A. (2004). Valuing customers. Journal of marketing research, 41(1), 7-18.
Ryder, I. (2007). Customer experience. Journal of Brand Management, 15, 85-88.
Lemon, K. N. & Verhoef, P. C. (2016). Understanding customer experience throughout the customer journey. Journal of Marketing, 80(6), 69-96.
Richardson, A. (2010). Using customer journey maps to improve customer experience. Harvard business review, 15(1), 2-5.
Barwitz, N., & Maas, P. (2018). Understanding the omnichannel customer journey: Determinants of interaction choice. Journal of interactive marketing, 43, 116-133.
Ascarza, E., & Hardie, B. G. (2013). A joint model of usage and churn in contractual settings. Marketing Science, 32(4), 570-590.
Platzer, M., & Reutterer, T. (2016). Ticking away the moments: Timing regularity helps to better predict customer activity. Marketing Science, 35(5), 779-799.
Zhang, Y., Bradlow, E. T., & Small, D. S. (2015). Predicting customer value using clumpiness: From RFM to RFMC. Marketing science, 34(2), 195-208.
2 1
3 4 5
6