A New Model for the Calculation of Customer Life-time Value in Iranian Telecommunication Companies

International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 394 A New Model for the Calculation of Customer Life-time Value in Iranian Telecommunication Companies Reza Samizadeh Assistant Professor, Department of Industrial Engineering, Alzahra University, Tehran, Iran Hamidreza Koosha Assistant Professor, Department ofIndustrial Engineering, Ferdowsi University of Mashhad, Mashhad, Iran Soudabeh Namdar Zangeneh Assistant Professor, Department of Industrial Engineering, Alzahra University, Tehran, Iran Sahar Vatankhah 1 Department of Industrial Engineering, Alzahra University, Tehran, Iran Abstract In this paper, we proposed a new model to evaluate a customer's lifetime value, considering non-financial elements such as the customer’s churn probability, cooperation capability, willingness to refer, willingness to recommend, and innovation. We tested our proposed model on customer data from a mobile phone operator to evaluate the effect of each element on the customer's lifetime value. Four hundred and twenty questionnaires were distributed and 400 questionnaires were determined to be suitable for our study. We employed structural equation modeling using Smart-PLS software and we have found that the innovation, customer’s churn, willingness to refer, and cooperation elements have the strongest effect on the customer's lifetime value. Keywords: Customer life time value, customer innovation, cooperation potential, willingness to recommend, willingness to refer, churn Cite this article:Samizadeh, R., Koosha, H., Zangene, S. N., & Vatankhah, S. (2015). A New Model for the Calculation of Customer Life Time Value in Iranian Telecommunication Companies. International Journal of Management, Accounting and Economics, 2(5), 394-403. 1 Corresponding author’s email: Sa.vatan[email protected]m International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 395 Introduction With the advent of a more competitive economic environment, concepts such as customer orientation and customer satisfaction are considered a basis for business, and organizations that do not pay attention to them will be eliminated from the market. Nowadays, organizations do not rely on their products selling, but rather prefer to attract, and subsequently retain, profitable customers. Research demonstrates that some of the most successful organizations have a customer retention rate of more than 90% (Haenlein et al. 2007). The most important challenges in customer-oriented organizations can be expressed as identifying customers, understanding the differences between them and classifying them (Liang 2011). We cannot say that all customers will make a similar contribution to an organization’s success, so increasing the satisfaction of key customers is vital (Glifford 2005). In addition, organizations that claim it is not necessary to spend money to gain customers are, inattentive to customer profitability levels (Blattberg et al. 2001; Blattberg et al 1996) To determine which class of customers is more valuable than others, and which customers (assets) to target, while operating with a restricted budget, we must provide a plan to maximize the investment profit (choosing the best customers). A concept such as customer lifetime value (CLV), which focuses on customer behaviour, makes this possible. The main idea of CLV, first defined 30 years ago by Kotler, is to evaluate customers based on their profitability for the organization, determining the current value of the expected future income stream during a specific period of time, while communicating with customers (Han et al. 2012). The main purpose of calculating CLV is to estimate customer weighting and, based on this, assign them the relevant level of resources. Evaluating CLV provides a method for customer comparison and, in industry for example, makes it possible to provide distinct products and better serve customers with higher CLV. A great deal of research has been devoted to calculating CLV, with most based on the equation proposed by Berger and Nasr (Berger & Nasr 1996) which can be considered as follows. 𝐶𝐿𝑉 =∑(𝑅𝑖−𝐶𝑖) (1+𝑑)𝑖 𝑛 𝑖=1 (1) In equation (1), i corresponds to the time horizon, 𝑅𝑖 is the acquired income from the customer over a period i, 𝐶𝑖 is the total customer costs over a period i and n is the number of periods. There exist other models to calculate customer lifetime value. Hwang et al (2004) were the first authors to use customer churn probability in their model. Cheng and Chen (2009) and Liang (2010) used the RFM model to evaluate customer lifetime value. In this model, R derives from the word ‘recency’ that refers to the interval between the customer’s last purchase and now, F derives from the word ‘frequency’ that refers to the number of customer purchases over a certain period of time and M derives from the word ‘monetary’ that refers to the value of customer purchases over the same period of time in real. In the RFM method, we first calculate the parameters M, F and R for each customer. Chan et al (2010) and Donkers et al (2007) used a Markov Chain model to calculate CLV. Glady et al (2009) used a Pareto/NBD model. Han et al (2012) used non-quantitative elements and truth to calculate CLV and have proposed a new method to quantify these elements. Cheng et al International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 396 (2012) introduced a new model based on a Markov Chain model to calculate CLV. Chen and Fan (2013) considered customer dynamic purchase behavior in their proposed model. After reviewing the literature, we found that most of the studies considered only financial elements that affect CLV but didn’t pay attention to nonfinancial elements. In the current paper, we consider financial and nonfinancial elements simultaneously and quantify CLV based on these elements. Figure 1 shows the conceptual model of the relationship between defined elements of the research. This conceptual model demonstrates the relationships between variables in which accuracy is not examined using experimental data. CLV Cooperation Price Entertain Quality Intimacy Satisfaction Trust Change obstacles Churn Willingness to recommend Willingness to refer Innovation Figure 1 Research Conceptual model International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 397 Materials and methods Investigation methodology The present research according to its goal can be classified to applied researches category and according to its method is categorized as a survey research, because we use questionnaires to aggregate data. Based on our studies, researches, scientific papers and studies in similar context provided by reliable international Professors, researchers and scientists, to evaluate and measure each desired criterion and parameter, we need 2 to 6 questions to acquire scientific and documentary results. So in our research, we have used 2 to 6 questions for each variable to aggregate necessary data. The research questionnaire has included 36 questions, evaluating 3questions about customer churn, 3 questions about customer satisfaction, 6 questions about change obstacles, 3 questions about service quality, 2 questions about service cost, 2 questions about trustworthiness, 2 questions about intimacy, 3 questions about entertainment, 2 questions about referring willingness, 2 questions about recommendation, 3 questions about cooperation, 4 questions about innovation and finally 1 question about customer life time value (according to late acquired profit from customer). These variables are expressed in Table-1. The questions are graded by 5 points Likert Scale (in this scale, point 1 means very low, point 2 means low, point 3 means moderate, point 4 means high and point 5 means very high). The research statistical population is the entire set of customers corresponding to one of Iranian mobile-phone operators. 420 questionnaires have provided to sample peoples and 400 of these questionnaires determined as suitable to be the base of Statistical calculations. Figure 1 shows the conceptual model of the relationship between the various defined elements of the research. The conceptual model demonstrates the relationships between variables, although their accuracy is not examined experimentally. The main assumption of our research can be considered as follows: Nonfinancial elements affecting customer life time values consist of innovation, cooperation capability, willingness to refer, willingness to recommend and churn. We also have two secondary assumptions: - The churn element has a negative effect on a customer’s life time value. - The elements of innovation, cooperation, willingness to refer, and willingness to recommend have positive effects on a customer’s life time value. Table 1 Visible variables of the customer life-time value model Variable Symbol Indicators Source Churn 𝑦11 Unwillingness to use services Kim and Shin (2008) 𝑦12 Not recommending services to others Kim and Shin (2008) 𝑦13 Need to other organizations services Kim and Shin (2008) Satisfaction 𝑦21 Being satisfied from organization services Kim and Shin International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 398 Variable Symbol Indicators Source (2008) Liu et al (2011) 𝑦22 Meeting needs Kim and Shin (2008) Liu et al (2011) 𝑦23 Total satisfaction of Organization Liu et al (2011) Trust 𝑦31 Reliability of Organization Liu et al (2011) 𝑦32 Keeping promises Liu et al (2011) Change obstacles 𝑦41 Time consuming of changes Kim and Shin (2008) 𝑦42 Changes cost Kim and Shin (2008) 𝑦43 Difficulties of changes in organization Kim and Shin (2008) Liu et al (2011) 𝑦44 Difficulties in use of other organizations services Kim and Shin (2008) 𝑦45 Uninterestingly of other organizations Liu et al (2011) 𝑦46 Difficulties of information gaining about other organizations Kim and Shin (2008) Price 𝑦51 Organization main service costs suitability Kim and Shin (2008) 𝑦52 Sidelong service costs suitability Kim and Shin (2008) Quality 𝑦61 Satisfaction of organization personnel behavior Kim and Shin (2008) 𝑦62 Good sense about organization services Kim and Shin (2008) 𝑦63 Better service quality compared to other organizations Kim and Shin (2008) Entertainment 𝑦71 Delightfully degree Liu et al (2011) 𝑦72 Happiness degree Liu et al (2011) 𝑦73 Fun degree Liu et al (2011) Intimacy 𝑦81 Friendly relationship with personnel Liu et al (2011) 𝑦82 Willingness to friendly relationship Liu et al International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 399 Variable Symbol Indicators Source (2011) Willingness to recommend 𝑦91 Recommending organization as a good one to others Boles et al (1997) 𝑦92 Organization request acceptance to recommend it to others Boles et al (1997) Willingness to refer 𝑦101 Organization request acceptance to refer others to it Boles et al (1997) 𝑦102 Presenting of familiar who are not organization customers if the organization want Boles et al (1997) cooperation 𝑦111 Cooperation intent with organization Tang et al (2013) Rudolf-Sipötz (2001) 𝑦112 Having technology, product, software and … which are profitable for organization Rudolf-Sipötz (2001) 𝑦113 Having organization required specialty Tang et al (2013) Innovation 𝑦121 Interfering with evaluation and idea purge Ballantine (2003) 𝑦122 Being full of idea Ballantine (2003) 𝑦123 Testing prototype by customer Ballantine (2003) 𝑦124 Interfering with design and extension of products Ballantine (2003) CLV 𝑦13 Obtained profit from customer in the past years of his lifetime Blattberg et al (2009) Data analysis and research assumptions test Evaluating the suitability of the questionnaire Our intention from suitability is to ensure that the content of the tools or the questions provided in these tools must evaluate the variables and the issue under study in an accurate manner. Using unsuitable tools will result in aggregating unrelated information and destroying the discipline of the investigation stages and data analysis. In this research, in order to determine suitability, the questionnaires were examined by experts so that they could determine the accuracy level and relevance of the questions. Consequently, some of these questionnaires were presented to the participants and finally some of the questions have corrected. Evaluating the reliability of the questionnaire In order to evaluate the reliability of the questionnaire, we used the Cronbach's alpha method. This method calculates the internal consistency of measurement tools that International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 400 measure different features. We used SPSS software to calculate Cronbach's alpha. Considering that the Cronbach's alpha value for elements and the questionnaire is higher than 0.7, the reliability of the questionnaire and elements is confirmed. Discussion The resulting data from the questionnaires have been processed using suitable software. The results obtained from executing the model via the software, are expressed in Table 3. If the statistic value is more than 1.96, with more than %95confidence interval, then the acquired relation is meaningful. According to the table, the effects of trust on churn and of intimacy on trust have been rejected. In Figure 2, acceptable and meaningful paths are represented. R2Values of models element are represented in Table 2. Based on our calculated R2 value of 60% for customer life-time value, we propose that the model includes 60% of elements that can affect customer life-time value. Table 2R2 values for models element Dependent variable R2 CLV 0.60 Churn 0.782 Satisfaction 0.914 Willingness to refer 0.57 Willingness to recommend 0.51 Table 2 Result obtained from executing the model via Smart PLS Path Standardized coefficient t-value Result Cooperation-CLV 0.143 3.101 Accepted Innovation-CLV 0.285 2.97 Accepted willingness to referCLV 0.208 2.346 Accepted willingness to recommendCLV 0.215 2.526 Accepted churn-CLV -0.249 2.99 Accepted satisfaction-churn -0.554 6.243 Accepted trust-churn -0.143 1.11 Rejected Churn obstacles-churn -0.301 4.28 Accepted Price-satisfaction 0.382 3.346 Accepted Entertainment-satisfaction 0.299 2.03 Accepted Quality-satisfaction 0.399 4.626 Accepted Quality-trust 0.802 7.23 Accepted Intimacy-trust 0.109 1.36 Rejected Satisfaction-willingness to recommend 0.241 3.230 Accepted Satisfaction-willingness to refer 0.251 3.520 Accepted International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 401 Discussion In this research, we proposed a new model to evaluate customer’s lifetime value.Our model is different from the other models as it considers non-financial elements that affect customer lifetime value. These elements include: Customer churn probability, cooperation capability, willingness to refer, willingness to recommend and customer innovation. We tested our proposed model on a set of data aggregated from a mobilephone operator and determined the coefficient of each element in the customer lifetime value. For this purpose, the structural equation modeling was used via the Smart-PLS software. However the most important restriction in this research was the inaccessibility to the company’s database, which has concluded incorrect calculation of customer lifetime value. Future research can focus of the following points: 0.241 CLV Cooperation Price Entertain Quality Satisfaction Change obstacles Churn Willingness to recommend Willingness to refer Innovation Figure 2 Acceptable and meaningful paths 0.285 0.215 0.218 0.143 -0.249 -0.554 -0.301 0.251 0.382 0.299 0.399 International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 402 - Measuring the effects of demographic variables on the customer’s life time value, as this study disregards the effect of these variables. - Using a more precise method to calculate the structure of customer life time value: in the current research, the structure is calculated approximately and is based on distributed questionnaires. It should be calculated according to the customer’s database. This model should be implemented in other service organizations and firms. References Blattberg, R. C., Malthouse, E. C., & Neslin, S. A. (2009). Customer lifetime value: Empirical generalizations and some conceptual questions. Journal of Interactive Marketing, 23(2), 157-168. Blattberg, R. C., & Deighton, J. (1996). Manage marketing by the customer equity test. Harvard business review, 74(4), 136. Blattberg, R. C., Getz, G., & Thomas, J. S. (2001). Customer equity: Building and managing relationships as valuable assets. Harvard Business Press. Berger, P. D., & Nasr, N. I. (1998). Customer lifetime value: marketing models and applications. Journal of interactive marketing, 12(1), 17-30. Boles, J. S., Barksdale, H. C., & Johnson, J. T. (1997). Business relationships: an examination of the effects of buyer-salesperson relationships on customer retention and willingness to refer and recommend. Journal of Business & Industrial Marketing, 12(3/4), 253-264. Chan, S. L., Ip, W. H., & Cho, V. (2010). A model for predicting customer value from perspectives of product attractiveness and marketing strategy. Expert Systems with Applications, 37(2), 1207-1215. Cheng, C. H., & Chen, Y. S. (2009). Classifying the segmentation of customer value via RFM model and RS theory. Expert systems with applications, 36(3), 4176-4184. Chen, Z. Y., & Fan, Z. P. (2013). Dynamic customer lifetime value prediction using longitudinal data: An improved multiple kernel SVR approach. Knowledge-Based Systems, 43, 123-134. Cheng, C. J., Chiu, S. W., Cheng, C. B., & Wu, J. Y. (2012). Customer lifetime value prediction by a Markov chain based data mining model: Application to an auto repair and maintenance company in Taiwan. Scientia Iranica, 19(3), 849-855. Donkers, B., Verhoef, P. C., & de Jong, M. G. (2007). Modeling CLV: A test of competing models in the insurance industry. Quantitative Marketing and Economics, 5(2), 163-190