Sales and Marketing Customer Loyalty

Question: Discuss about the Sales and Marketingfor Customer Loyalty. Answer: Aim of the Project Customer loyalty and purchase trends can be analysed in a systematic way. Goods purchased at different periods by the same customers can be grouped into sequences. Methods of sequential pattern mining can then be used to investigate changes in customer consumption or loyalty, and suggest adjustments on the pricing and variety of goods in order to help retain customers and attract new ones (Bassi, 2009). By the end of the project, we shall be trying to answer the key questions that arise from this project. The key questions are: What does a Loyalty Card do for any type of retailer that cannot be observed from the level of details in the invoice and baskets? How can the success of these loyalty programs be measured? What data is to be mined for in the loyalty programs? What are the various data mining techniques? How to identify the loyal customers and support them from the loyalty program data? Project Background It is to be noted that any normal customer has to go through 5 different stages throughout their Lifecycle. These stages are being depicted in the figure 1 herein. Figure: 1 In each step, data analytics can be made use of as a tool to make the right decisions for the following: Right Customer / Prospect: based on the estimates from the Lifetime Profitability. Right Message: based on the resonance and connection with the targeted customer group. Right Channel: based on the estimates of ROI and the response rates. Right Promo / Offer: based on the estimates of ROI and the response rates. The entire process is often described to be a marketing process driven by the customers. At the core of the process lies the usage of the data that is available about the customers transactions along with the demographics and various other characteristics for segment targeting and personalising the marketing efforts of the organisation (Butler, 2005). Data Analytics and Methodology Customer Loyalty Systems The data, which is primarily to be used in the project, for understanding the marketing system oriented around the customer is generally taken from the Loyalty Management System of a retailer. Maximum of the retailers in the modern business world has adopted the above mentioned system (Czarniewski, 2014). This type of system is generally used for accumulation of reward points and its redemption. It is to be noted that Shoppers was the first organisation to step foot in this system with their initiative known as First Citizen and following the same, most of the major departmental stores have initiated similar loyalty program systems. The primary intention of the different loyalty programs were to be considered as additional incentive for the customers for increasing the level of purchases. With the passage of time, the different organisations started to accumulate data on the pattern of purchasing at the various levels of customers. Soon, the organisations learned the potential that the data had and the new business fronts that would open up based on these data (Dixon, 2010). The purchase pattern, which was primary gathering from the available data, shall be revealing the general preference for any customer, which is much more efficient than any data collected through different surveys, since: Customers can be hiding their true motive and preference in a survey. A survey is generally conducted over a very small scale population of the entire customer base. The cost of collating and fielding the results from the different surveys is very high. The different surveys have to be designed before their execution, thus the retailer needs to make some assumptions, which they are trying to validate or invalidate using the data from the survey. However, it is to be noted that the data from transaction does not have the similar disadvantages as stated above for data collected from surveys (Finlay, 2014). Moreover, the system that is being used for collection of data is already in operation and does not acquire additional high cost for leveraging it for the process of data analysis. Retailers will have the ability to understand preferences of the customers based on the demographic differences. It would be easier for any retailer to stock up on an item that would sell the most based on past purchasing behaviour of the customer base (Guenzi and Troilo, 2006). The ability of a retailer to cross sell and up sell at any POS turns out to be much more effective. The ability for targeting specific customer base with an approach that is focused on the customer instead of the concept of one size is fit for all. Business Brief The Client for the project is considered to be one of the leading lifestyle and apparel goods retailers, which is a subpart of larger conglomerate. The said business has been operating in the region for over 5 years and the previously mentioned client has over 30 stores all over the nation. It is to be noted that the Client for the project had initiated a program pertaining to Customer Loyalty in the mid period of the 2000s. During the period of time for the consideration of collection of data, the retailer had around 350,000 customers in its loyalty program (Inghilleri and Solomon, 2010). For the purpose of the project, two primary streams pertaining to direct marketing were used directing the same exclusively towards the members of the loyalty program. A half-yearly mailing of the brochure of the retailer, with emphasis being laid on the seasonal apparels. Different marketing campaigns focused on a single product, which is carried out by the usage of different campaign channels all through the year. Previously, the concept that one marketing scheme would be applicable for all was being utilised by the client. It was soon understood that this was not an ideal approach to the business. Thus the retailer sought to use the knowledge of data mining for customising their business plans and communication methods to efficiently transform their business and make the marketing process optimal in nature. Customer Segmentation The term segmentation is referring to the practices that help in identification of the homogenous groups that can be different from each other, yet same form the internal structure aspect. In terms of statistics, it I often referred to as groups having the least amount of distance within the group, while at the same time the same having the maximum distance between each other across the group once the identification of the different customer segments have been completed, then we have the task of marketing the best offers or the various promos based on the niche group from the segmented data La and Yi, 2015). Multiple analytical techniques have been used for the segmentation of the customer data, the most widely used technique being the Cluster analysis (Solomon, 2012). This method helps in identifying the segments that are naturally occurring within the entire data set or the customer base. It also helps in the identification of the major variables that are helping to drive the various segments. Moreover, there are advanced methods such as the Decision trees or Latent Class Analysis, etc for various situations where the algorithm for Cluster Analysis does not provide us with the intended results (Lawfer, 2004). Segmentation Basis The given can be based on a number of topics such as demography, latitudinal and even value based (Demographic information, 2000). The perfect basis for the different segmentation would be dependent on the ultimate objective of the business. The table below is an illustration of the segmentation scheme being dictated by the objectives of the business. Data The primary data source for the project is considered to be collected from the customer loyalty program, which is active within the clients business (Martirano, 2016). The system provided us with three primary streams of data. Customer Personal Details Loyalty Program Tenures Demographics Occupation, age, Marital Status, Gender, etc. Customer Transaction Details Basket Level Price of the different items that are brought together Level of Items in the basket The discount amount of the different items that have been purchased. Mode of payment preferred by the customer. Item Master Details Item Description and the Codes Hierarchy of the item across several levels. Store Master Details The size of the store, the period of operation of the store, intensity of competition and the area of catchment. Data Preparation The three major activities that shall be carried out within this section before the creation of the proper model for analysis are: Transformation of the raw data to the level of the customer by the aggregation of the primary key variables. Derivation of necessary variables dependent on the requirement of the scheme of segmentation from the basic primary variables (Merceron, Blikstein and Siemens, 2015). The total customer population is to be split into two sections. One for creation of the model and the other as a holdout sample for validation of the results from the created models. It is to be noted that at the level of the customer, the segmentation procedure would be applicable on the primary transactional variables such as the ones mentioned in the list herein below: Frequency, Recency and Monetary Variables. of visits by the customer to the retailer per year. of items that has been purchased upon every visit. Amount that has been spent on each category in every year. The behavioural patterns of a customer to chase after promotional values and offers and discounts. It is to be noted that the different variable for the customer profile are to be made from the basic information available, such as the preference in Lifestyle from the pattern of purchase across the different segments available for purchase (Mehta, 2008). Variables pertaining to the Life-Stage are also to be created by the use of marital status, age and the no. of children the customer has. Analytical Model After the data has been prepared in the previous step, an algorithm pertaining to the Non-Hierarchical method is to be used for determining the membership segment (Predictive Analytics: The Core of Data Mining, 2016). The process of modelling shall be looking into different combinations of clusters and the ideal one shall be the 5 cluster solution. The no. of exact segments was decided upon because of: Objective Statistical Criteria Primary Statistical Criterion such as the Pseudo F Statistic or the Cubic Clustering Criterion, RMSSTD, etc. Clusters having their own pattern over the modelling unit and the hold out sample unit. Sizes are same. The mean values of the different primary variables are same. Subjective Criteria Driven by Business The Size of the different clusters. The clusters should be large enough to give out meaningful action for the marketing solutions (Raphel, Raphel and Raye, 2005). Differentiation among the primary variables. The different clusters should be non-similar in nature on several factors such as, category wise purchase value, total purchase value, etc. Segmentation Results The Clusters that has been processed form the modelling process has been profiled based on the demographics of the customers for creation of a holistic description, which are given herein below: Loyal Families: 17% of the total customer population. Married Value Seekers (Loyal) Comprises of 18% of the total population (Stinson, 2008). Married Shoppers (Occassional) 34% of the total customer population Single Female (Pormo and Discount Hunter) Consists of 19% of the total customer population. Single Male (Value Seeker) Comprises of around 12% of the total customer base. Plan and Time Table The plan for the project has already been made. The data analytics and data mining model has been decided upon. A basic study of the retailer data has been made from the loyalty program. A further analysis of the same using the prescribed data analytics model would help us in finding out further characteristics of the different segments that has been mentioned herein above. Based on the different characteristics, the retailer would be able to focus their marketing process on the different segment and come up with different products, discounts and promotional offers for the different segments (Timm, 2001). Based on the plan a grant chart has been prepared depicting the various stages of the procedure and the total time it would take to execute the same. Task Days Planning the process Collection of Data from the Loyalty Program Designing the Analytical Model Data Analysis Recommendations References Bassi, F. (2009). Latent Class Models for Marketing Strategies.Methodology, 5(2), pp.40-45. Butler, T. (2005).Power conflict, commitment the development of sales marketing IS/IT infrastructures at Digital Devices, Inc.. 1st ed. Hershey, PA: Idea Group Pub. Czarniewski, S. (2014). Building Customer Loyalty on the Polish Market.ECONOMICS SOCIOLOGY, 7(3), pp.208-222. Demographic information. (2000). 1st ed. Oak Lawn, Ill.: The Village. Dixon, R. (2010).Methodology. 1st ed. Oxford [u.a.]: Oxford University Press. Finlay, S. (2014).Predictive analytics, data mining and big data. 1st ed. Basingstoke [u.a.]: Palgrave Macmillan. Guenzi, P. and Troilo, G. (2006). Developing marketing capabilities for customer value creation through MarketingSales integration.Industrial Marketing Management, 35(8), pp.974-988. Inghilleri, L. and Solomon, M. (2010).Exceptional service, exceptional profit. 1st ed. New York: American Management Association. La, S. and Yi, Y. (2015). A Critical Review of Customer Satisfaction, Customer Loyalty, Relationship Marketing, and Customer Relationship Management.Korean Marketing Review, 30(1), p.53. Lawfer, M. (2004).Why customers come back. 1st ed. Franklin Lakes, NJ: Career Press. Martirano, M. (2016). Transcendental Phenomenology: Overlooked Methodology for Marketing Research.International Journal of Marketing Studies, 8(3), p.58. Mehta, J. (2008).Advertising, marketing and sales management. 1st ed. Jaipur, India: Book Enclave. Merceron, A., Blikstein, P. and Siemens, G. (2015). Learning analytics: From big data to meaningful data.Journal of Learning Analytics, 2(3), pp.4-8. Predictive Analytics: The Core of Data Mining. (2016).International Journal of Science and Research (IJSR), 5(5), pp.2075-2079. Raphel, M., Raphel, N. and Raye, J. (2005).The complete idiot's guide to winning customer loyalty. 1st ed. Indianapolis, Ind.: Alpha Books. Solomon, M. (2012).High-tech, high-touch customer service. 1st ed. New York: American Management Association. Stinson, P. (2008).Sales, marketing, business and finance. 1st ed. New York: Ferguson Pub. Timm, P. (2001).Seven power strategies for building customer loyalty. 1st ed. New York: AMACOM.