Прогнозирование оттока клиентов на основе паттернов изменения их поведения

PT Kimia Farma Trading and Distribution (KFTD) is a company engaged in the distribution and trading services of Indonesian health products, on a national scale. In 2022, the company aims to increase sales to be awarded as one of the top 3 national pharmaceutical product distributors by 2024. Their current strategy is to provide customers with delayed payment permission and integrated complaint services. However, the offers and services are the same for all customers which does not consider customer track record hence it is not cost-effective. One way to increase sales is by enhancing customer satisfaction and loyalty by implementing Customer Relationship Management (CRM) strategies. Several strategies can be carried out, namely analysis of associations related to pharmaceutical products, and analysis of customer segmentation and clustering of products. The method used in this study was the K-means clustering algorithm combined with the Recency Frequency Monetary (RFM) model. Experiments showed that the optimal clustering results are 4 therefore they are categorized into 4 customer segments, namely Superstar, Golden, Typical, and Occasional Customers.

The glowworm swarm optimisation (GSO) algorithm is a novel algorithm with the simultaneous computation of multiple optima of multimodal functions. Data-clustering techniques are classification algorithms that have a wide range of applications. Since K-means algorithm is easy to fall into the local optimum by the selection of initial clustering centre, GSO algorithm is used for the telecom customers clustering. The customer consumption data is extracted by means of the recency frequency monetary (RFM) model and the standardised data is clustered automatically using the GSO algorithm's synchronous optimisation ability. In the clustering optimisation algorithm, adaptive step size is used instead of the original fixed step size to avoid local optimisation of the algorithm and obtain higher accuracy. Compared with K-means clustering algorithm, GSO approach can automatically generate the number of clusters and use RFM model to reduce effectively the size of the data processing. The results of the experiments demonstrate that the GSO-based clustering technique is a promising technique for the data clustering problems.

As obesity has become a worldwide problem, a number of health programs have been designed to encourage participants to maintain a healthier lifestyle. The stakeholders often desire to know how effective the programs are and how to target the right participants. Motivated by a real-life health program conducted by an Australian supermarket chain, we propose a novel method to track customer behavior changes induced by the program and investigate the program’s effect on different segments of customers, split according to demographic factors like age and gender. The method: (1) derives customer preferences from the transaction data, (2) captures the customer behavior changes via a temporal model, (3) analyzes the program effectiveness on different customer segments, and (4) evaluates the program influence using a one-year data set obtained from a major Australian supermarket. Our results indicate that while overall the program had positive effect in encouraging customers to buy healthy food, its impact varied for the different customer segments. These results can inform the design of personalized health programs that target specific customers in the future and benefit more people. Our method can also be applied to other programs that use transaction data and customer profiles.

This research investigates how policy changes affect customer attitudes and behaviors, as well as firms' service operations outcomes, in the context of social media customer service. It aims to fill the gap in knowledge regarding the interaction between customers and social media customer service agents and its effect on companies’ customer service operation performance. The first essay examines how social media customer service agents’ personalized profiles can influence customer emotions, complaints, and satisfaction. The findings suggest that humanized customer service leads to an increase in positive sentiment in public tweets, fewer complaints, and a higher probability of customer satisfaction. The study highlights the importance of human warmth in customer service in eliciting positive customer attitudes and satisfaction on social media. The second essay explores the impact of the COVID-19 pandemic on changes in customer behavior in the context of social media customer service. The findings show that during the lockdowns and physical distancing, customers expressed more positive emotions and were more likely to express satisfaction after interacting with customer service agents, but female customers expressed a lower level of satisfaction after interacting with the agents. The findings of these studies contribute to our understanding of the impact of humanized customer service and the COVID-19 pandemic on customer behavior and satisfaction and have implications for companies seeking to provide a more positive and personalized customer experience on social media platforms.

As a result of the COVID-19 pandemic, business and consumer behavior towards online consumption and digital payments has changed. This trend, although significant, may be observed, to a varying extent, among customers across the European Union. This study aims to investigate if, and to what extent, the COVID-19 pandemic has had a powerful impact on online consumer behavior in EU (European Union) countries. The empirical research using a soft modeling method supplements theoretical deliberations based on critical literature review. A conceptual model was adopted. Three first-order latent variables were selected for the analysis: “online customer behavior”, “online activity” and “willingness to spend online”, to which all analyzed indicators (14) were classified. The analysis of the loadings of latent variables for 2019 and 2020 allowed the impact of the pandemic on customer behavior to be observed, although this was not the same across all EU countries. Clustering performed with the hit map enabled the identification of four groups of countries. Significant changes in behavior were observed in countries such as Poland, Ireland, Romania, Hungary, Slovenia, Spain, and Finland. The results of the classification based on the values of three latent variables indicate that only the composition of one group (Bulgaria and Romania) remained unchanged during the study period. Research discussion was presented, and further fields of study were identified.

The omnichannel business has becomes a hot topic due to the fast development on e-commerce and the customers' acquaintance with multichannel shopping mode. Various business organizations have started to work on omnichannel business issue in order to satisfy the new trend of customer demand and tend to devote their efforts to both online and offline business. Thus, there is no doubt that understanding the shopping behavior for online customers is vital for the omnichannel business. The RFM (recency, frequency, monetary) model and the k-means clustering method are commonly used to extract customers' information and segment customers, respectively. To extend the RFM model, we divide the total frequency and monetary information into weekly level data, and as a consequence, the number of variables corresponding to one customer increases significantly, leading to the problem of high-dimensional analysis. To address this issue, in this paper we extend the regularized k-means clustering method with L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> -norm for independent case to the clustering method with elastic net penalty with a focus on correlated variables. Our simulation results show that the proposed method performs better than the standard k-means method by providing lower error rates and can select variables simultaneously under 4 different scenarios. A real example of an online retailer is presented to illustrate the use of the proposed method and highlight its high potential in clustering high-dimensional applications. In particular, the number of variables is reduced from 108 to 98 without any loss on clustering accuracy.

Usual methods of segmentation have been designed, relying solely on the components of Recency (R), Frequency (F) and Monetary (M) in which customers’ behavior changes over time are not considered. Accordingly, in order to achieve a desired segmentation method, this study aims to apply a set of statistical calculations, such as line slope and the derivative with respect to time and data mining methods such as K-means and Self-Organizing Maps (SOM) to define new parameters for studying the changes trending of customer purchasing behavior. The results show that considering the slope of the line of customer behavior changes (R, F, and M) and the higher value for recent behaviors of customers compared to that of their past behavior in customer segmentation would thereby increase the accuracy of predicting the future behavior and cause the customers of each section to become more homogeneous. Based on the suggested method, customers are categorized into four segments: best, spender, repeater and missed ones each of them are divided into two subcategories of ascending and descending segments, which leads to better and more accurate understanding of customers in different segments according to how of their purchasing behavior change. Finally, the characteristics of each segments and sub-segments are described and appropriate strategies are provided for managing its customers.

The profound impact of the coronavirus pandemic on global tourism activity and the hospitality industry has rendered statistical approaches on tourism-demand forecasting obsolete. Furthermore, literature review shows the absence of studies on the supply chain in the HoReCa (hotel, restaurant, catering) sector from a sustainability perspective that also addresses economic and social aspects, and not only environmental ones. In this context, the objective of this article is to carry out a prospective analysis on how the changes in the behaviour of consumers during the pandemic and the uncertainties regarding the exit from the health emergency can give rise to social trends with a high impact on the HoReCa sector in the coming years and, specifically, how they will affect the HoReCa supply chain. In the absence of investigations due to the proximity of what has happened, public sources and reports of international relevance have been identified and analysed from the future studies and strategic and competitive intelligence disciplines. The HoReCa sector in Spain has been chosen as field of observation. This analysis draws the future of the HoReCa sector, describes the changes in customer behaviour regarding food and beverages, explains the changes in distribution chains, and reflects on the impact of potential scenarios on the sector. The confluence of all these changes and trends can even configure a new supply chain in the hospitality sector with the emergence of new actors and the increase of access routes to a new final customer for whom security prevails in all its dimensions: physical, emotional, economic, and digital.

In this study, the authors assess the effects of a data breach announcement (DBA) by a multichannel retailer on customer behavior. They exploit a natural experiment and use individual customer transaction data from the retailer to conduct a detailed and systematic empirical examination of the effects of a DBA on customer spending and channel migration behavior. To identify the effects, the authors compare the change in customer behavior before and after the DBA between a treatment group (customers whose information is breached) and a control group (customers whose information is not breached) using the difference-in-differences modeling framework. They find that although the data breach results in a significant decrease in customer spending, customers of the firm migrate from the breached to the unbreached channels of the retailer. The findings further reflect that customers with a higher retailer patronage are more forgiving because the negative effects of the DBA are lower for customers with a higher level of patronage. The authors propose and empirically test for the role of customer data vulnerability as the behavioral mechanism that drives customer behavior subsequent to a DBA. The authors offer prescriptions for managers on how to engage with customers following DBAs.

Hospitality services rank among significant realisation factors of tourism. Due to its increasing importance and high potential, tourism falls into significant elements effecting a possible development of rural areas, especially in the area of maintaining and supporting small and medium businesses. This paper deals with the issues of customer behaviour in the Czech Republic with the emphasis on hospitality services. The main purpose was to identify and analyse development changes in customer behaviour in the area of these services. The pursuit of revealing the causes of such established changes and finding possible links and connections between the individual factors was an inseparable part of it. Furthermore, we compare the materially-technical base and the existing accommodation services in the period after 1989 with the contemporary situation and we also outline new trends in customer behaviour in the Czech Republic. To reach these set targets, a secondary date analysis was conducted and after that, a primary research in the form of questionnaire survey among consumers was carried out. During the statistical evaluation of the research, the frequency of the occurrence of each question was evaluated (e.g. Information on accommodation facilities is usually found on the Internet for 64% of respondents.), then dependences among verbal features were measured and the expressed hypotheses were tested (e.g. The assumption: less than 30% people are looking for accommodation in the category from 201 CZK to 300 CZK was not rejected and so we can say that approximately 30% people find their accommodation in the category 201–300 CZK). Important results of the conducted surveys with the relevant discussions are a part of this paper.

Trading failure is the main reason for a dispute of C2C e-commerce. So predict the behavior of transactions can assist buyers and sellers negotiated transactions, helps to reduce transaction disputes. Separate the success and failure purchase record, then establish decision-making model through the C5.0 decision tree and RFM(Recency, Frequency, Monetary) model on consumer purchase behavior data, quantify the importance of the decision variables, the demonstration experiment shows the prediction accuracy is more than 80%.

In today's highly competitive landscape, businesses across various sectors, including corporations, retail, and banking, are striving to attract and retain customers. Customer segmentation has emerged as a vital strategy in this endeavor, enabling organizations to identify and cater to distinct customer groups. This study leverages the RFM (Recency, Frequency, Monetary) model to segment customers based on their purchasing behavior. Initially, an RFM analysis is conducted to quantify customer value, which is then further analyzed using the k-means clustering algorithm to identify distinct customer segments. By understanding and targeting these segments, businesses can enhance customer relationships, improve marketing strategies, and foster organizational resilience. This approach not only aids in acquiring new customers but also in retaining existing ones, ultimately contributing to sustained competitive advantage.

This study explores customer behavior towards secondary banking products, including credit cards, insurance, and wealth management services, among clients of commercial banks. It identifies key factors influencing adoption decisions, such as product awareness, perceived value, and trust in the bank, while also examining barriers to uptake. The findings provide insights for banks to enhance cross-selling strategies, improve customer satisfaction, and increase retention through tailored product offerings. This case study investigates the impact of credit cards, insurance, and wealth management products on customer retention within commercial banks. Through a comprehensive analysis of customer behavior and preferences, the study identifies key factors that influence retention rates. This descriptive study, aimed to analyze customer behavior towards secondary banking products such as credit cards, insurance, and wealth management services. Utilizing a simple random sampling method, a sample of 330 customers was surveyed using a questionnaire as the primary data collection tool. The findings revealed that these secondary products have a minimal impact on customer retention, indicating the need for banks to develop more attractive schemes to enhance customer loyalty and retention. Furthermore, effective wealth management strategies tailored to individual customer needs contribute significantly to long-term retention. The study concludes by highlighting the importance of personalized financial services in fostering strong relationships and reducing customer attrition in the competitive banking sector.

Due to high competition in today's business and the need for satisfactory communication with customers, companies understand the inevitable necessity to focus not only on preventing customer churn but also on predicting their needs and providing the best services for them. The purpose of this article is to predict future services needed by wireless users, with data mining techniques. For this purpose, the database of customers of an ISP in Shiraz, which logs the customer usage of wireless internet connections, is utilized. Since internet service has three main factors to define (Time, Speed, Traffics) we predict each separately. First, future service demand is predicted by implementing a simple Recency, Frequency, Monetary (RFM) as a basic model. Other factors such as duration from first use, slope of customer's usage curve, percentage of activation, Bytes In, Bytes Out and the number of retries to establish a connection and also customer lifetime value are considered and added to RFM model. Then each one of R, F, M criteria is alternately omitted and the result is evaluated. Assessment is done through analysis node which determines the accuracy of evaluated data among partitioned data. The result shows that CART and C5.0 are the best algorithms to predict future services in this case. As for the features, depending upon output of each features, duration and transfer Bytes are the most important after RFM. An ISP may use the model discussed in this article to meet customers' demands and ensure their loyalty and satisfaction.

An approach for sales forecasting