CHAPTER - 2 LITERATURE REVIEW This chapter describes and examines the previous work on Developing Data Mining technique to find optimal customers using Association Rule Mining and Particle Swarm Optimization for beneficial CRM. Peppers (1999) summarizes the following as the basic strategies and objectives of CRM initiatives: Customer identification: The organization must be able to identify the customer via marketing channels, interactions and transactions for a period of time in order to provide value to the customer by serving his or her need at the right time with a right product or service. Customer Differentiation: Every customer has his or her own needs and demands and therefore from the organization s point of view, customers have their own lifetime value. Customer Interaction: One of the most important objectives of CRM by an organization is to keep track of customer behavior and needs over the period of time. This is because, from a CRM point of view, the customer s long-term profitability and relationship to the company is very important. This is the reason why a company should continue to learn about its customers and in a continuous manner. Personalization: This can be defined as treating each customer differently or uniquely and that is the motto or a major objective of CRM. Through the process of personalization, the organization can increase customer loyalty. 61
According to Stone (2000) there are two main objectives that influence the need for CRM technologies to support the completion of CRM strategies and initiatives. These are according to Stone (2000) as follows: The need for a higher quality in CRM in order to meet the needs of the customers. CRM systems according to Stone (2000) are increasingly being used to arrange companies resources in a proper order. The need for greater productivity in CRM. CRM systems are giving the possibility to automate work previously done by hand (Ibid). According to Deck (2004), the objective and the strategy of CRM is that it should help organizations to use technology and human resources to understand the behavior of customers and the value of those customers. If it works that way, an organization can: Provide better customer service Make call centers more efficient Cross-sell products more effectively Simplify marketing and sales processes Discover new customers Increase customer revenues (Ibid). According to Burnett (2001) CRM objectives and strategies can also be grouped under the following categories: Increased profitability margins: By knowing the customers better, efforts can be made to switch less profitable accounts to lower cost/service delivery channels. Decreased sales and marketing administrative cost: The decrease can occur if the organization specifies and has good knowledge 62
about its target segment customers. As a result, the organization will use its resources better when no effort is a waste of money or time. Improve customer satisfaction rate: The increase in customer satisfaction rate will occur because customers will find that the offer is more in line with the customers specified needs. Win rates: This will also improve because the company will withdraw from unlikely or bad deals earlier in the sales process (Ibid). According to Almotairi (2009), there is a general acceptance among researchers of the categorization of CRM components. CRM consists of three major components: Technology, people, and business Process. The contribution to each component varies according to the level of CRM implementation(almotairi, 2009). Technology Technology refers to computing capabilities that allow a company to collect, organize, save, and use data about its customer. Technology is the enabler for CRM systems to achieve their objectives of collecting, classifying, and saving valuable data on customers. Integration technology allows organizations to develop better relationship with customers by providing a wider view of the customer behavior. Thus, organizations are required to integrate IT to improve the capabilities of understanding customer behavior, develop predictive models, build effective communications with customers and respond to those customers with real time and accurate information. For an organization to integrate IT, concepts such as data warehouse, software customization, process automation, help desk and call centers, and internet influence should be addressed as Mendoza et al. (2007) in (Almotairi, 2009). 63
People Employees and customers are a key factor for successful CRM projects. CRM is built around customers to manage beneficial relationships through acquiring information on different aspects of customers. The main objective of CRM is to translate the customer information into customized products and services that meet the changing needs of customers in order to gain their loyalty. Nevertheless, a full commitment of the organization's staff and management is essential for an effective CRM implementation to best serve customers and satisfy their needs Business process. CRM is a business strategy that has its philosophical basis in relationship marketing. CRM success requires a change of business processes towards customer centric approach. As such, all business processes that involve both direct and indirect interaction with customers should be analyzed and assessed. Although CRM has an organization-wide impact, process that has direct interaction with customers it should be dealt with as a priority when integrating and automating business processes. According to Mendoza et al (2007) in (Almotairi, 2009) the main business processes that should be addressed in CRM implementation are: marketing, sales, and services. CRM architecture addresses the requirements of enhancing/enriching and changing the customer experience by providing the functionality required to effectively interact with the customer, during the Sales and Marketing process. According to Gray and Byun (2001) the following are the main benefits of CRM. They went on to say that, for an organization to get all these benefits, sales, marketing and service functions must work together: 64
To improve the company s ability to retain and acquire the customers To maximize the lifetime value of each customer To improve service without increasing cost of service. Again they argue that, proper identification of the customer helps the sales force to do cross selling. They further add that, this is through clean data about the customer and a single customer view. Furthermore, they say that, understanding the customer through differentiation can lead to cost effective marketing campaign, it could also reduce something like for example direct mailing cost. Also, they argue that customer satisfaction and loyalty through interaction could also lead to cost effective customer service. Moreover, they argue that, customer satisfaction and loyalty through personalization can also lead to lower cost of acquisition and retention of customer and thereby maximizing the share of wallet. Crosby (2002) argues that, by using customer information wisely to deliver what the customer needs, companies will create long-term, collaborative relationships with the customers. He further states that, this will bring many benefits since long-term customers are less costly to serve and smooth-running relationships are less resource intensive (Ibid). A survey of more than 500 executives in six industries, communication, chemicals, pharmaceuticals, electronics/high-tech, forest products and retail, believes that 10% improvement of overall CRM capabilities can add up to$35 million benefits to a $1 billion business unit. (Gray and Byun 2001). CRM is a very big tool that contributes so much to profit indicated by Newell (2000). Furthermore he stated that, if organizations could transform the customer data into knowledge and then use that knowledge to build relationships it would then create loyalty and thereby creating profit (Ibid). Turban et al. (2000) suggest that increasing customer satisfaction increases customer loyalty. 65
Swift (2001 pp. 28) argues that organizations can get a lot of benefits from CRM initiatives. The author goes on to say that, these benefits could be found in these areas: Higher customer retention and loyalty: The customer retention will increase when customers stay longer, buy more and buy more frequently. The customers take more initiatives that increase bounding relationship, and as a result the customer loyalty increases. Increased customer profitability: The customer profitability will increase when the customer wallet-share increases, the up-selling goes up as well as cross-selling and follow up sales and also more referrals come with higher customer satisfaction among existing customers. Evaluation of Customer profitability: When the organization gets to know which customers are profitable and which ones might become profitable in the future, that is the potential profitable customers and those who will never become profitable. This is a very important area because the key to any successful business is to acquire and focus on those customers who bring profit, when you get them, you do not want to leave them. Reduced cost of sales: The costs regarding selling are reduced due to the fact that existing customers are usually more responsive. In addition, with better knowledge of channels and distributors, the relationships become more effective, as well as costs for marketing campaigns are reduced. Lower cost of recruiting customers:when the cost of recruiting new customers reduces or goes down, there will be savings to be made on marketing, mailing, contact, follow-up, fulfilling, service and many more. 66
When the number of long-term customers increased, consequently the need to recruit many new customers will decrease (Ibid). Bose (2002) argues that, most organizations can use CRM. However he goes on to say that, there are some organizations that are more likely to get more benefits from CRM than the others. Furthermore he states that, those are the companies that accumulate a huge customer data when doing business and whose customer needs are differentiated. On the other hand, Bose (2002) says that, companies that rarely have any contact with their customers have a higher customer turn over and identifying customer needs are likely to get less benefit from CRM (Ibid). According to Newell (2002) organizations should undertake CRM initiatives where they will get the best possible return and benefits. He goes on to indicate that, companies should then focus on customers who are already profitable and those who will become the company s most profitable customers in the future (Ibid). Newell (2000) divides customers into three types according to the benefits derived from the organizations which is as follows: The top group: These consist of the top 10% and these are the customers with excellent loyalty and of high profitability for the organization. The middle group: These consist of the 40%-50% of the company s customers and they are the ones delivering good profits and also show a very good potential for future growth and loyalty. The lower group: These consist of the lower 40%-50% of the company s customers those who are only marginally profitable and they may become potential future profitable customers. Xu, Yen, Lin, and Chou (2002) argue that CRM initiatives do not only improve customer loyalty, but also improve the internal processes, which in turn increase as well. Again they indicated that, one of the benefits of CRM is 67
that, it identifies and target s best customers based on recent purchase or behavior, frequency and monetary scoring. Furthermore they stated that, it helps to manage marketing campaigns with clear goals and measurable or quantifiable objectives. More over XU, et al (2002) state that one of the benefits of CRM initiatives is that it creates and manages solid sales which lead to field and Telesales representatives. They further add the following as some of the benefits that the organizations derive from their CRM initiatives and solutions: It increases marketing and cross selling opportunities It enables tight and accurate targeting and one-to-one marketing It helps to increase return on marketing investment It adds more valuable knowledge already gained from direct customer interaction and this knowledge in turn helps to improve product development processes (Ibid). Xu, et al, (2002) summarize the following as the benefits of CRM initiatives from the sales perspective: It improves Telesales, field sales, and sales management through real time information sharing among multiple employees. It increases sales efficiency through the wire and Internet-based order entry. It improves territory management with real time account information updates It improves the entire sales force by capturing, distributing and leveraging the success and expertise of the highest performers. It increases revenue per call by focusing on growing the best accounts (Ibid). 68
According to Xu, et al (2002) the following summarize the benefits of CRM initiatives and solutions from the field service perspective: It ensures customer satisfaction and retention by solving customer problems quickly. It integrates the management of people and materials within the service organization smoothly. It again ensures customer satisfaction by allocating, scheduling and dispatching the right people, with the right parts at the right time (Ibid). Xu, et al, (2002) summarizes the benefits of CRM initiatives and solutions from the perspective of the customer support as follows: Shared relationships with personalized customer care based on specific customer history and preferences are strengthened. Through automated scripting based on known solutions, call center efficiency and help desk support s quality are improved. Support and service costs are reduced when Web-based support functionality is extended directly to the customer and this increases customer satisfaction. All customer contact for sales, support, field service and marketing are centralized. Ching-Hsue Cheng et al. Their study has proposed a procedure for successfully extracting meaning rules to improve these drawbacks by combining quantitative value of RFM attributes and K-means algorithm into rough set theory (RS theory). The proposed method has first obtained quantitative value as input features by utilizing the RFM model. Then, customer values have been clustered using K-means algorithm; and in the end, classification rules that assist enterprises driving an outstanding CRM have 69
been mined by employing a rough set (the LEM2 algorithm). The proposed procedure proved to surpass the methods listed in terms of precision rate irrespective of 3, 5 and 7 classes on output, and produce comprehensible decision rules by analyzing the experimental results[20]. Abdullah et al. have presented the most excellent combination of ERP with Customer Relationship Management (CRM). The three main parts present in their model have been outer view-crm, inner view-erp and knowledge discovery view. The CRM EPR and knowledge discovery have been used for the purpose of gathering the queries of customers, evaluate and combine the data and provide forecasts and recommendations for the improvement of an organization respectively. They have utilized MADAR data and implemented Apriori Algorithm on it for the practical implementation of the presented model[21]. Yu-Teng Chang s approach has revealed its capability in forecasting with reasonable precision by constructing a churn prediction model using customer demography, billing information, call detail records, and the service changed log in an analysis of the result from telecom provider[22]. Abdullah et al. have gathered the data from a central database in cluster format using characteristics and background of ERP which has been based on the steps taken against the queries created by the customers. In addition, Apriori Algorithm has used the clustered data to extract new rules and patterns for the improvement of an organization. Data mining applications have been comprehensively implemented in an ERP framework for forecasting the solution of future queries[23]. Alok Mishra et al. have discussed that Customer relationship management (CRM) could help to maintain competitiveness in the present economy by enabling organizations to handle customer interactions more effectively. Effective CRM implementation has been an intricate, costly and 70
seldom technical project. In a trans-national organization with operations in diverse sections, they have effectively implemented CRM from a process point of view. Their analysis has been expected to assist such organizations in comprehending transition, constraints and the implementation process of CRM[24]. Adele Berndtet al. have concentrated on the implementation of a oneto-one program especially in the financial service atmosphere inside a developing economy. They have examined the phases in the implementation of CRM as proposed by Peppers, Rogers and Dorf (1999b) and analyzed the consequence of customer service in a developing market. The results have revealed that constructive relationships exist between these phases and customer service[25]. Carol Xiaojuan Ou et al. have described that the past research has frameworks and fragmented evidence of factors that result in effective customer relationship management (CRM). Environmental factors that spread outside organizational boundaries have also been covered in addition to customers by CRM systems. Thus, in addition to appropriate organizational factors, extra-organizational environmental factors must also be addressed by CRM initiatives[26]. The applicability of the work system (WS) framework as a guide to evaluate the CRM initiatives carried out by Shanghai General Motors (GM) has been validated by their framework. Their observations have suggested that comprehensive management of the individual slices of the WS framework and their interaction and interdependencies are essential for effective CRM operation. The analysis has also suggested that the success of CRM depends remarkably on the cultural values of not only the organization, but also its customers. Mohd H.N.M. Nasir et al. [10] have discussed the causes of failures of the CRM system. To rectify them, they have proposed a CRM prototype using Human Computer Interaction (HCI). For the purpose of capturing user's 71
requirements, they have acquired and analyzed the background, current conditions and environmental interactions of a multi-national company. The analysis mainly intends to determine the relationship between the stages of patterns and internal-external influences. Interviews, naturalistic documentation and studying user documentation were also done to gather blended data. Using all these data, the prototype had been developed with the incorporation of User-Centered Design (UCD) approach, Hierarchical Task Analysis (HTA), metaphor and identification of users' behaviors and characteristics. The performance of technique was measured using usability. Mon Fong has discussed about different data types and mining algorithms [28]. He says that all mining algorithms are not best suited for all data types. Selecting a mining algorithm not only depends on the pattern to be extracted but also on the data types of the data on which the mining algorithm has to be applied. Data is transformed into the required format. For data transformation data type s generalization process is used. Yen Ting talks about the ontology driven data mining on a medical driven database[29]. The database contains information about patients undergoing treatment for chronic kidney disease. Ontology was used as an aid to provide facts about the attributes of the database and also used for controlling vocabulary and the attributes with more impact are selected manually. Association Rule mining is applied on these selected attributes and antecedents are selected. Then the antecedents are taken as new class variables and association rule is applied. The final output was good when compared to that of naive method of association rule mining. Pawel [31] has done a Clustering analysis based on the ontology. Ontology is used to describe all the compared objects. Three dimensions were taken into consideration to calculate the similarity between individual objects and they are taxonomy similarity, relationship similarity and attribute similarity. Similarity was calculated through a formula using these dimensions. 72
Using protégé tool, owl file is generated to describe the objects. The similarity measures were calculated and a matrix is produced. This matrix is further processed using external tools. Wang explores the data generalization concept of data mining as a way to hide details Information, rather than discovering trends and patterns. Once the data are masked, standard data mining techniques can be applied without modification. This method used in this paper not only hides the sensitive data but also provides an appropriate mining result [32]. Jafari reveals that certain specific sensitive association rules are hidden by decreasing its support or confidence than the pre-defined minimum support and minimum confidence. To decrease the confidence of a rule X Y, either increases the support of X, i.e., the left hand side of the rule, but not support of X Y, or decrease the support of the item set X Y. In the second case, if it only decreases the support, the right hand side of the rule, it would reduce the confidence faster than simply reducing the support of X (Y [32]. Homburg, C et. al.,say that customer satisfaction research has focused primarily on the disconfirmation of expectations model, which states that feelings of satisfaction arise when Consumers compare their perceptions of a product's performance to their expectations. As per the marketing concept, any product should be considered as a total product that includes the core benefits from the product as well as auxiliary dimensions associated with that product. The customer is satisfied when he/she feels that the product's performance is equal to or more than what was expected (confirmation). Hence, customers have higher retained wills to communicate with enterprises and to develop a future relationship. Customer satisfaction should therefore be understood as relationship specific and individualized [34]. The RFM analytic model is proposed by Hughes (1994), and it is a model that differentiates important customers from large data by three variables 73
(attributes), i.e., interval of customer consumption, frequency and money amount. The detail definitions of RFM model are described as follows: (1) Recently of the last purchase (R). R represents recently, which refers to the interval between the time that the latest consuming behavior happens and the present. The shorter the interval is, the bigger R is. (2) The frequency of the purchases (F). F represents frequency, which refers to the number of transactions in a particular period, for example, two times in one year, two times of one quarter or two times in one month. The many the frequency is, the bigger F is. (3) The monetary value of the purchases (M). M represents monetary, which refers to the consumption money amount in a particular period. The much the money is, the bigger M is [35]. Data mining can help organizations discovering meaningful trends, patterns and correlations in their customer, product, or data, to drive improved customer relationships and then decrease the risk of business operations. The basic data mining techniques include classification, clustering, association rules, regression analysis, sequence analysis, etc. Other data mining techniques include rule-based reasoning approach, genetic algorithms, decision trees, fuzzy logic, inductive learning systems, statistical methods, and so forth. Generally, no tool for data mining in CRM is perfect because there are some uncertain drawbacks in it. For example, in decision trees, too many instances lead to large decision trees which may decrease the classification accuracy rate and do not clearly create the relationships which come from the training examples. In artificial neural networks, the number of hidden neurons, the number of hidden layers and training parameters need to be determined, and especially. ANN has long training times in a large dataset Moreover, ANN served as black box which leads to inconsistency of the outputs, is a trialand-error process. In the genetic algorithm, GA also has some drawbacks such 74
as slow convergence, a brute computing method, a large computation time and less stability. In association rules, major drawback is the number of generating rules which are huge and may be a redundancy. For solving the problems of the previous paragraph, two methods, K- means algorithm and RS theory, are worth to be explored in this study. K- means is one of the well-known algorithms for cluster analysis and it has been used extensively in various fields including data mining, statistical data analysis and other business applications. Cluster analysis is a statistical technique that is used to identify a set of groups that both minimize withingroup variation and maximize between-group variation based on a distance or dissimilarity function, and its aim is to find an optimal set of clusters (35). The particle swarm optimization algorithm was introduced by Kennedy and Eberhart in 1995 [37]. The algorithm consists of a swarm of particles flying through the search space. Each individual in the swarm contains parameters for position and velocity. The position of each particle represents a potential solution to the optimization problem. The dynamic of the swarm is governed by a set of rules that modify the velocity of each particle according to the experience of the particle and that of its neighbors depending on the social network structure within the swarm. By adding a velocity to the current position, the position of each particle is modified. As the particles move around the space, different fitness values are given to the particles at different locations according to how the current positions of particles satisfy the objective. At each iteration, each particle keeps track of its personal best position, as much the better. Depending on the Social network structure of the swarm, the global best position, gbest, and/or the local best position, lbest, is used to influence the swarm dynamic. After a number of iterations, the particles will eventually cluster around the area where the fittest solutions are [37] 75
The Agraval etal define the model in the following way. Association Rule Mining (Basket Analysis), also referred to as Association Rule Learning, is one of the unsupervised Data Mining methods CRM benefits from. It has found applications predominantly in the retail industry for instance, supermarkets or online retailers where the products are purchased in various combinations. The method uses the purchase transactions data from the retailer's database in order to find out which products are consistently purchased together, as well as the direction (antecedent and consequents) in the resulting sets. Pioneering work on this model has been done by R. Agrawal, who defines it in the following way [52]: Let I 1, I 2,..., I m be a set of n binary attributes called items. Let T be a database of transactions. Each transaction t is represented as binary vector, with t k 1 if t bought item I k, and t k 0 otherwise. There is one tuple in the database for each transaction. Let X be a set of some items in.it can be say that a transaction t satisfies X if for all items Ik in X,t k 1.Association rule is an implication of the form X I j,where X is a set of some items in, and I j is a single item in that is not present in X. The rule X I j is satisfied in the set of transactions T with the confidence factor 0 c 1 iff at least c% of transactions in T that satisfy X also satisfy Ij. The classical Association Rule Mining process consists of two distinct steps: 1. finding frequent itemsets Y I 1, I 2,..., I k, k 2 (groups of items that appear together in the percentage of transactions set by the threshold titled minimal support), c% of transactions in T that satisfy X also satisfy Ij. The classical Association Rule Mining process consists of two distinct steps: 76
2. finding frequent itemsets YI 1 I 1, I 2,..., I k, k 2 (groups of items that appear together in the percentage of transactions set by the threshold titled minimal support), 3. generating the rules for each Y comparing the proportion of support of Y to support of X Y, the potential antecedent of the rule (a subset of the itemset that contains all the items in the itemsets excluding the potential consequent) with the minimal confidence level threshold c set by the miner. All the inferences of the type X Y- X are the resulting association rules that satisfy confidence c. The confidence measure can be described in terms of conditional probability: it reflects how many transactions contain item I i given and they also contain Ij [59]. Agrawal and his colleague R., Srikant,`says the Association Rule Learning uses Apriori algorithm, which essentially performs the breadthfirst search of the transaction database in order to find all itemsets containing products that are purchased together and determine the itemsets in left hand side and right hand side of the association. The algorithm is described in the pseudocode in Listing 1.1 [60] Yan-hua WANG Xia FENG The Optimization of Apriori Algorithm Based on Directed Network, (2009 Third International Symposium on Intelligent Information Technology Application).This paper gives an experiment to analyze and compare the difference between the two (Apriori algorithm and proposes an improved algorithm based on the directed network) algorithms and the result shows that the improved algorithm promotes the efficiency of computing. In this paper, algorithm is improved based on directed network. Yiwu Xie, Yutong Li, Chunli Wang, Mingyu Lu The Optimization and Improvement of the Apriori Algorithm, Through the study of Apriori algorithm discover two aspects that affect the efficiency of the algorithm. One 77
is the frequent scanning database, the other is large scale of the candidate item sets. Therefore, Apriori algorithm is proposed that can reduce the times of scanning database, optimize the join procedure of frequent item sets generated in order to reduce the size of the candidate item sets. In this study, it not only decrease the times of scanning database but also optimize the process that generates candidate item sets. Soeini & Rodpysh, 2012, Evaluations of Data Mining Methods in Order to Provide the Optimum Method for Customer Churn Prediction: Case Study Insurance Industry This study using Clementine software and the database which contains 300 records of customers in insurance Company in the city of Anzali, Iran will be collecting a questionnaire. First, determine the optimal number of clusters in K-means clustering and clustering customers based on demographic variables. And then the second step is to evaluate binary classification methods (decision tree QUEST, decision tree C5.0, decision tree CHAID, decision trees CART, Bayesian networks, Neural networks) to predict customers s churn. Findings: Better performance than other techniques CART decision tree technique, perhaps that algorithm shows a better performance but due to the fact that the data collection results are not far-fetched. Patterns were extracted by decision tree and shows that most churn customers are in officers or engineers. Results of data mining methods provide an opportunity for managers and marketing professionals to make decision and choose suitable strategies to prevent churn of customers and let them go to other companies. 78
Farooqi & Raza, 2011," A Comprehensive Study of CRM through Data Mining Techniques". This study attempts to bring a new perspective by focusing the issue of data mining applications, opportunities and challenges in CRM. It covers the topics such as customer retention, customer services, risk assessment, fraud detection and some of the data mining tools which are widely used in CRM. Findings: Application of customer relationship management tool in business gives a new dimension. It proved beneficial but applying data mining in customer relationship management was further more beneficial. Data Mining would fasten up the process of searching large databases so as to extract customer buying patterns, to classify customers into groups which also make databases to be handled efficiently. Modeling those customers who have defected to identify patterns that led to their defection. These models are then applied to the current customers to identify likely defectors so that preventive actions can be initiated. Chopra, Bhambri & Krishan, 2011," Implementation of Data Mining Techniques for Strategic CRM Issues" This study throws light on the underlying technology and the perspective applications of data mining in customer relationship management. Findings: Data Mining techniques can be of immense help to the organization in solving business problems by finding patterns, associations and correlations which are hidden in the business information stored in the data bases. 79
Organizations can use these techniques for acquiring new customers, fraud detection in real time, providing segment based products for better targeting the customers. Organizations can use data mining techniques to analysis of the customers purchase patterns over time for better retention and relationship, detection of emerging trends to take proactive approach in a highly competitive market, adding a lot more value to existing products and services and launching of new product and service bundles. By using data mining techniques, the organizations will be able to offer the right product to the right set of customers through the right offer and through right delivery channel, which will in turn lead to better customer relationship management. Dhman 2011, "The Effect of Customer Relationship Management (CRM) Concept Adoption on Customer Satisfaction Customers Perspective, Case Study: Coastal Municipalities Water Utility CMWU- Rafah Branch". This study aims at investigates the effect of applying the concept of customer relationship management (CRM) on customer s acquisition, satisfaction, retention and decreasing customer s loss. Findings: CRM concept in the CMWU was significantly correlated in positive direction with reaching customer acquisition, satisfaction, retention and decreasing customer loss. CRM has effect on decreasing customer loss more than the other tested factors, where proportional mean for the customer loss was 82.24%, yet the mean for the other factors together was around 76.00%. 80
Zavareh 2007, "The Role of Analytical CRM in Maximizing Customer Profitability in Private Banking, Case Study : Two Swedish Banks" The main objective of the study is to investigate the role of analytical CRM in maximizing customer profitability in private banking. Findings: The analytical CRM system had been implemented and actively utilized by both banks. The Internet was found to assist collection of more precise data, to increase the analytical ability and to create faster degrees of performance. The results also indicate that customer profitability was highly considered by both banks and tactical measures were exercised to augment the customer profitability, particularly among the core customers, with providing them extra and personalized services. 81