What is churn? An overview

What is drudge? An overviewChurn is the phenomenon where a node switches from adept run to a competitors dish out (Tsai Chen, 20092). There be two chief(prenominal) types of seethe, namely voluntary drudge and involuntary fag. Voluntary churn is when the customer initiated the service statusination. Involuntary churn means the lodge hang the customers service and this is usually be puddle of non-payment or service ab role.Companies, in various industries, have recently started to realise that their customer set is their virtually valuable asset. Retaining the animate clients is the best marketing strategy. Numerous studies have confirmed this by showing that it is more profitable to keep your existing clients satisfied than to constantly attract new clients (Van hideout Poel Larivire, 2004197 Coussement Van Den Poel, 2008313). concord to Van Den Poel and Larivire (2004197) successful customer retention has more than just financial benefits Successful customer ret ention programs free the organisation to focus on existing customers gets and the building of relationships. It lowers the need to find new customers with uncertain levels of risk. Long term customers tend to deal more and provide positive advertising through word-of-mouth. The company has better knowledge of long term customers and they are less expensive with lower uncertainty and risk. Customers with longer tenures are less likely to be influenced by competitive marketing strategies. Sales whitethorn slack if customers churn, due to lost opportunities. These customers as well need to be replaced, which apprize cost five to six times more than simply retaining the customer.1.1.Growth in Fixed-line Markets harmonise to Agrawal (2009) the high growth phase in the telecommunications market is over. In the future, wealth in the industry allow be split betwixt the companies. Revenues (of telecommunication companies) are declining around the world. name 2 shows Telkoms fixed-l ine customer vile and customer growth rate for the previous seven years. The weigh of lines is used as an estimate for the moment of fixed-line customers.Figure 2-Telkoms fixed-line annual customer base (Idea adopted from Ahn, Han Lee (2006554))With the lower customer growth worldwide, it is change state vital to counteract customers from churning. 1.2.Pr lawsuiting Customer ChurnThe two basic approaches to churn management are shared out into untargeted and targeted approaches. Untargeted approaches rely on superior products and mass advertising to decrease churn (Neslin, Gupta, Kamakura, Lu Mason, 20043). Targeted approaches rely on directing customers who are likely to churn and wherefore customising a service plan or incentive to prevent it from happening. Targeted approaches can be further divided into proactive and reactive approaches.With a proactive approach the company identifies customers who are likely to churn at a future date. These customers are then targeted with incentives or peculiar(prenominal) programs to attempt to retain them.In a reactive targeted approach the company waits until the customer cancels the account and then shots the customer an incentive (Neslin et al., 20044).A proactive targeted approach has the advantage of lower incentive costs (because the customer is not bribed at the last minute to catch ones breath with the company). It in like manner prevents a culture where customers threaten to churn in order to negotiate a better deal with the company (Neslin et al., 20044).The proactive, targeted approach is dependent on a addressive statistical technique to predict churners with a high accuracy. Otherwise the companys funds may be wasted on unnecessary programs that incorrectly identified customers.1.3.Main Churn PredictorsAccording to Chu, Tsai and Ho (2007704) the briny contributors to churn in the telecommunications industry are price, coverage, quality and customer service. Their contributions to churn can b e seen from Figure 3.Figure 3 indicates that the primary reason for churn is price related (47% of the exemplification). The customer churns because a cheaper service or product is available, through no fault of the company. This means that a perfect retention strategy, base on customer satisfaction, can only prevent 53% of the churners (Chu et al., 2007704).1.4.Churn Management FrameworkDatta, Masand, Mani and Li (2001486) proposed a five stage framework for customer churn management (Figure 4). The first stage is to identify commensurate data for the fashion sham process. The quality of this data is extremely important. Poor data quality can cause large losses in m iodiny, time and opportunities (Olson, 20031). It is also important to determine if all the available historic data, or only the most recent data, is going to be used.The second stage consists of the data semantics problem. It has a direct link with the first stage. In order to fire the first stage successfully, a complete understanding of the data and the variables entropy are required. data quality issues are linked to data semantics because it much influences data interpretation directly. It frequently leads to data misinterpretation (Dasu Johnson, 2003100). Stage three handles lark selection. Cios, Pedrycz, Swiniarski and Kurgan (2007207) define feature selection as a process of finding a subset of features, from the original set of features forming patterns in a given data set. It is important to select a sufficient physique of diverse features for the mold phase. fraction 5.5.3 discusses some of the most important features found in the literature.Stage four is the predictive model development stage. There are many alternative methods available. Figure 5 shows the number of times a statistical technique was mentioned in the papers the author read. These methods are discussed in detail in Section 6.The final stage is the model test copy process. The goal of this stage is to ensur e that the model delivers accurate predictions. 5.5.1Stage one Identify dataUsually a churn indicator flag must be derived in order to define churners. Currently, there exists no standard accepted definition for churn (Attaa, 2009). One of the usual definitions state that a customer is considered churned if the customer had no active products for three consecutive months (Attaa, 2009 Virgin Media, 2009 Orascom Telecom, 2008). at a time a target variable is derived, the set of best features (variables) can be determined. 5.5.2Stage two Data semanticsData semantics is the process of understanding the context of the data. Certain variables are difficult to interpret and must be carefully studied. It is also important to use consistent data definitions in the database. Datta, et al. (2001) claims that this phase is extremely important.5.5.3Stage three device characteristic selectionFeature selection is another important stage. The variables selected here are used in the modelling stage. It consists of two phases. Firstly, an initial feature subset is determined. Secondly, the subset is evaluated based on a certain criterion.Ahn et al. (2006554) describe four main types of determinants in churn. These determinants should be included in the initial feature subset.Customer dissatisfaction is the first determinant of churn mentioned. It is driven by network and call quality. Service failures have also been identified as triggers that reanimate churn. Customers who are unhappy can have an extended negative influence on a company. They can spread negative word-of-month and also appeal to third-party consumer affair bodies (Ahn et al., 2006555).Cost of switching is the second main determinant. Customers produce their relationships with a company based on one of two reasons they have to stay (constraint) or they indispensability to stay (loyalty). Companies can use loyalty programs or membership cards to encourage their customers to want to stay (Ahn et al., 200 6556).Service consumption is the third main determinant. A customers service usage can broadly be described with minutes of use, frequency of use and total number of distinct numbers used. Service usage is one of the most popular predictors in churn models. It is still un slip by if the correlation between churn and service usage is positive or negative (Ahn et al., 2006556).The final main determinant is customer status. According to Ahn et al. (2006556), customers seldom churn suddenly from a service provider. Customers are usually suspended for a sequence due to payment issues, or they decide not to use the service for a while, before they churn.Wei and Chiu (2002105) use length of service and payment method as further possible predictors of churn. Customers with a longer service history are less likely to churn. Customers who authorise direct payment from their bank accounts are also expected to be less likely to churn. Qi, Zhang, Shu, Li and Ge (2004?2) derived different growt h rates and number of abnormal fluctuation variables to model churn. Customers with growing usage are less likely to churn and customers with a high abnormal fluctuation are more likely to churn.5.5.4Stage four Model developmentIt is clear from Figure 5 that decision tree models are the most frequently used models. The second most popular technique is logistic regress, followed closely by spooky networks and option analysis. The technique that featured in the least number of papers is discriminant analysis. Discriminant analysis is a multivariate technique that discriminateifies observations into existing categories. A mathematical function is derived from a set of continuous variables that best discriminates among the set of categories (Meilgaard, Civille Carr, 1999323).According to Cohen and Cohen (2002485) discriminant analysis makes stronger modelling assumptions than logistic regression. These include that the predictor variables must be multivariate normally distributed and the within-group covariance matrix must be homogeneous. These assumptions are rarely met in practice. According to Harrell (2001217) even if these assumptions are met, the results moderateed from logistic regression are still as accurate as those obtained from discrimination analysis. Discriminant analysis give, therefore, not be considered.A neural network is a parallel data processing structure that possesses the ability to learn. The concept is roughly based on the humane brain (Hadden, Tiwari, Roy Ruta, 20062). Most neural networks are based on the perceptron architecture where a weighted running(a) combination of inputs is sent through a nonlinear function.According to de Waal and du Toit (20061) neural networks have been known to offer accurate predictions with difficult interpretations. Understanding the drivers of churn is one of the main goals of churn modelling and, unfortunately, traditional neural networks provide limited understanding of the model.Yang and Chi u (2007319) confirm this by stating that neural networks use an internal weight system that doesnt provide any insight into why the solution is valid. It is often called a black-box methodology and neural networks are, therefore, also not considered in this study.The statistical methodologies used in this study are decision trees, logistic regression and survival analysis. Decision tree modelling is discussed in Section 6.1, logistic regression in Sections 6.2 and 6.3 and survival analysis is discussed in Section 6.4.5.5.5Stage five Validation of resultsEach modelling technique has its own, specific establishment method. To compare the models, accuracy go away be used. However, a high accuracy on the training and validation data sets does not automatically result in accurate predictions on the population dataset. It is important to cook the impact of oversampling into account. Section 5.6 discusses oversampling and the adjustments that need to be made.1.5.Adjustments for Target Level ImbalancesFrom Telkoms data it is clear that churn is a rare event of great interest and great value (Gupta, Hanssens, Hardie, Kahn, Kumar, Lin Sriram, 2006152).If the event is rare, using a sample with the same equaliser of events and non-events as the population is not ideal. Assume a decision tree is veritable from such a sample and the event rate (x%) is very low. A prediction model could obtain a high accuracy (1-x%) by simply assigning all the cases to the majority level (e.g. predict all customers are non-churners) (Wei Chiu, 2002106). A sample with more balanced levels of the target is required.Basic sampling methods to decrease the level of class imbalances include under-sampling and over-sampling. Under-sampling eliminates some of the majority-class cases by randomly selecting a lower percentage of them for the sample. Over-sampling duplicates minority-class cases by including a randomly selected case more than once (Burez Van Den Poel, 20094630).Under-sampling has the drawback that potentially useful information is unused. Over-sampling has the drawback that it might lead to over-fitting because cases are duplicated. Studies have shown that over-sampling is ineffective at improving the recognition of the minority class (Drummond Holte, 20038). According to Chen, Liaw Breiman, (20042) under-sampling has an edge over over-sampling.However, if the probability of an event (target variable equals one) in the population differs from the probability of an event in the sample, it is necessary to make adjustments for the prior probabilities. Otherwise the probability of the event will be overestimated. This will lead to score graphs and statistics that are inaccurate or misleading (Georges, 2007456).Therefore, decision-based statistics based on accuracy (or misclassification) manage the model performance on the population. A model developed on this sample will identify more churners than there actually are (high false alarm rate). Without an ad justment for prior probabilities, the estimates for the event will be overestimated.According to Potts (200172) the accuracy can be adjusted with equation 1. It takes prior probabilities into account.With the population proportion of non-churners the population proportion of churners the sample proportion of non-churners the sample proportion of churners the number of true negatives (number of correctly predicted non- churners) the number of true positives (number of correctly predicted churners) the number of instances in the sampleHowever, accuracy as a model susceptibility measure trained on an under-sampled dataset is dependent on the threshold. This threshold is influenced by the class imbalance between the sample and the population (Burez Van Den Poel, 20094626).