On the customer lifetime value: a mathematical perspective

Accurately predicting Customer Lifetime Value (CLV) is paramount in optimizing customer relationship management. This study introduces a novel deep learning approach, employing a neural network model to forecast CLV for a leading international e-commerce retailer. This investigation delves into the interplay of key performance indicators—Net Promoter Score (NPS), Average Transaction Value (ATV), and Customer Effort Score (CES)—and their collective impact on CLV. The research utilized a dataset of 15,000 customer profiles and crafted a sequential neural network with dense layers optimized through hyperparameter tuning and regularization to thwart overfitting. The model's efficacy was assessed on a 10 % test set, revealing its adeptness at capturing intricate, nonlinear predictor-customer lifetime value (CLV) relationships. Consistency in training and test performance metrics underscored the model's generalizability, while high R-squared and explained variance scores confirmed the predictive strength of the chosen factors. This research seeks to provide a solution for building an artificial intelligence model with an artificial neural network algorithm to predict customer lifetime value. By incorporating this principle, the study aims to leverage the power of neural networks to forecast CLV, enabling retailers to make informed decisions and optimize customer relationships accurately. This study finds that Net Promoter Score (NPS) and Customer Effort Score (CES) have a powerful impact on the neural network model's ability to predict Customer Lifetime Value (CLV) accurately. On the other hand, while Average Transaction Value (ATV) exhibits the least impact, it still significantly contributes to the accuracy of Customer Lifetime Value (CLV) predictions. These results underscore the importance of incorporating customer feedback metrics like NPS and CES when building predictive models for CLV estimation. Integrating critical customer metrics into neural networks gives retailers enhanced insights, enabling precise customer segmentation, resource allocation, and strategic growth. The study paves the way for future research to refine further CLV prediction, including dataset expansion, customer profile enrichment, and prescriptive marketing optimization.

In Chapter 6, first we discuss customer relationship marketing (CRM)-based analytical models, both for purchase and non-purchase customer behavior. Various models are discussed including traditional metrics such as recency–frequency–monetary value (RFM) model, share of wallet (SOW) model, and past customer value (PCV) model, as well as contemporary models based on customer lifetime value (CLV) concepts. CLV is an advanced forward-looking metric and several variations of the family of CLV models are briefly discussed, including model formulation, estimation, and key merits and demerits. Next, we provide models for customer engagement including customer referral value (CRV), customer influence value (CIV), and customer knowledge value (CKV). Finally, we discuss strategies for maximizing CLV, segmentation strategies, extending CLV to customer engagement value, and linking customer equity to shareholder value. At the end of the chapter, we provide key takeaways and conclude with discussion questions and HBS and Ivey cases. But first, to give a flavor of CRM analytical models, we provide some real-life vignettes.

The paper focuses on lifetime value assessment and its implementation and application in business processes. The lifetime value is closely connected to customer relationship management. The paper presents results of three consecutive researches devoted to issues of customer relationship management. The first two from 2008 and 2010 were conducted as quantitative ones; the one from 2009 had qualitative nature. The respondents were representatives of particular companies. The means for data collection was provided by ReLa system. We will focus on individual attributes of lifetime value of a customer, and relate them to approaches of authors mentioned in introduction. Based on the qualitative research data, the paper focuses on individual customer lifetime value parameters. These parameters include: the cost to the customer relationship acquisition and maintenance, profit generated from a particular customer, customer awareness value, the level of preparedness to adopt new products, the value of references and customer loyalty level. For each of these parameters, the paper provides specific recommendations. Moreover, it is possible to learn about the nature of these parameter assessments in the Czech environment.

This paper presents a direction of development of 'the customer value' concept relying on the results of previous research of the models for the calculation of customer lifetime value (CLV), presented in scientific journals. Evolutionary path of this concept has been observed in the context of relationship marketing and customer relationship management. The aim of this paper is to present the achievements in the development of this concept since 1980 to date and to encourage our country's scientific and professional community to further research the models of calculation of customer lifetime value, as well as management of companies to apply these models to calculate CLV and CE (Customer Equity) of their customers.

Accurate customer LifeTime Value (LTV) predictions are crucial for customer relationship management, especially in Supply Chain Platforms (SCP), which involve effectively managing the service resources in business decision-making. Previous LTV prediction methods usually rely on ample historical customer data, which is not available in the early stages of a customer's lifecycle. It makes the modeling of the historical customer data a difficult task due to the data sparsity. Besides, the long-tail distribution of customer LTV also brings new challenges to the prediction of LTV. To tackle the above issues, we propose CDLtvS, a novel Cross Domain method for customer Lifetime value prediction in SCP. It leverages rich cross-domain information from upstream platforms to enhance LTV predictions in downstream platforms. Firstly, CDLtvS pre-trains the customer representations by an LTV modeling framework named LtvS in source and target domains separately. Specifically, LtvS incorporates the Expert Mask Network (ExMN), which not only effectively models the long-tail distribution of LTV in single-domain but also resolves cross-domain learning model bias resulting from this distribution. Then, the various-level alignment mechanism is introduced to keep the consistency of knowledge transferring from source to target domains on both sparse and non-sparse data. Comprehensive experiments on real-world data from JD, one of the world's largest supply chain platforms, demonstrate that CDLtvS achieves a normalized mean average error of 0.3378 in LTV prediction, outperforming 16.3% to the baseline. Additionally, the improvements of ≥2.3% across various data sparsity levels (0% -- 80%) provide valuable insights into cross-domain LTV modeling.

A Note on Willingness to Spend and Customer Lifetime Value for Firms with Limited Capacity

Customer Lifetime Value (LTV) prediction is a critical task in online games, as it helps with the formulation of refined game operation strategies, resource allocation, and personalized recommendation. Accurate LTV values enable to identify and target high-value users, enhance user retention and further long-term revenue growth. However, LTV prediction in online games faces unique challenges. Most in-app purchases (IAP) games have various fixed recharge levels, and users with different payment preferences show distinct LTV distributions. Moreover, existing methods fail to capture the multi-modal distribution of LTV values, and suffer from bias when predicting LTV values of games that users have not registered, i.e., new users. To address these challenges, we propose HiLTV, a novel hierarchical framework for LTV prediction in online games. We devise hierarchical modules to align with real-world user recharge behaviors, and a Zero-Inflated Mixture-of-Logistic (ZIMoL) loss instead of a unimodal distribution loss is adopted to better model various segments of users. We also introduce a calibration module that enables more robust predictions for new users. Both offline evaluation on real-world industrial datasets over state-of-the-art baselines and online A/B test from a leading game platform demonstrate the superior performance of our method.

Measuring and managing customer relationship risk in business markets

Most segmentation analyses use descriptive variables to group customers into homogenous segments in order to propose appropriate marketing actions and to optimize firms resources allocation. However, descriptive variables are usually fixed in time and lack actionability and responsiveness power. Some studies suggested that value based segmentation is the most significant from the standpoint of marketing activities. The customer lifetime value (CLV) metric, which aims to predict the future value of each customer, is often recommended as an interesting feature to segment customers. However, segmentation based on the two CLV components, number of transactions and lifetime, helps to better explain the customer behavior and to propose more effective marketing actions. In this work, we propose a Multicriterion segmentation approach based both on descriptive variables and on CLV components. The Multicriterion problem is solved using genetic algorithms by generating a set of Pareto-optimal solutions. The empirical analysis shows the ability of the proposed approach to characterize customer segments and to propose appropriate marketing actions.

The authors evaluate the usefulness of customer lifetime value (CLV) as a metric for customer selection and marketing resource allocation by developing a dynamic framework that enables managers to maintain or improve customer relationships proactively through marketing contacts across various channels and to maximize CLV simultaneously. The authors show that marketing contacts across various channels influence CLV nonlinearly. Customers who are selected on the basis of their lifetime value provide higher profits in future periods than do customers selected on the basis of several other customer-based metrics. The analyses suggest that there is potential for improved profits when managers design resource allocation rules that maximize CLV. Managers can use the authors’ framework to allocate marketing resources efficiently across customers and channels of communication.

The rapid growing of information technology (IT) motivates and makes competitive advantages in health care industry. Nowadays, many hospitals try to build a successful customer relationship management (CRM) to recognize target and potential patients, increase patient loyalty and satisfaction and finally maximize their profitability. Many hospitals have large data warehouses containing customer demographic and transactions information. Data mining techniques can be used to analyze this data and discover hidden knowledge of customers. This research develops an extended RFM model, namely RFML (added parameter: Length) based on health care services for a public sector hospital in Iran with the idea that there is contrast between patient and customer loyalty, to estimate customer life time value (CLV) for each patient. We used Two-step and K-means algorithms as clustering methods and Decision tree (CHAID) as classification technique to segment the patients to find out target, potential and loyal customers in order to implement strengthen CRM. Two approaches are used for classification: first, the result of clustering is considered as Decision attribute in classification process and second, the result of segmentation based on CLV value of patients (estimated by RFML) is considered as Decision attribute. Finally the results of CHAID algorithm show the significant hidden rules and identify existing patterns of hospital consumers.

This paper will address a topic common among customer relationship management (CRM) practitioners — the use of lifetime value (LTV) models, especially for customer marketing segmentation. It will draw comparisons between LTV and other common marketing/segmentation metrics and how these are used. It will also discuss some of the challenges facing the analyst who wishes to use LTV and how these might be overcome. Finally, it will illustrate how LTV-based segmentation can be incorporated into a comprehensive model that maximises yield from a finite marketing budget.

As there are a lot of booming online retailers in e-commerce industry in the Internet age, the need of maintaining competitive advantages has become to pay attention to customer relationship management (CRM). To build a successful CRM strategy, it is needed to know individual customer class which can be calculated from Customer Lifetime Value (CLV): the monetary value of customers purchased from the business during their lifetime. CLV modelling allows us to identify customer's predicted business value. It provides the retailers for effectively allocating the resource in their business. This predictive model has been taken on the global Super Store Retail dataset with almost ten thousand transactions. Our model will predict the customers' class of the next year based on their CLV that will help the online retailer to decide which customer should be invested to get long term CRM. Random Forest (RF) algorithm is utilized to train our model and Random Search tuning is conducted to get the best predictive accuracy. The experimental analysis is performed to compare with AdaBoost algorithm on the same dataset.

The objective of this paper is to gain insight into the use of the Lifetime Value (LTV) concept when focusing on relationships with customers.This requires a discussion of both the status quo of LTV and customer-based marketing. As a result of these discussions we conclude that proper use of LTV to support building and maintaining relationships with customers requires a further refinement of the LTV concept. We identify five aspects where current use of LTV is inappropriate for customer-based marketing. Furthermore, we define a model for computing LTV which meets the requirements stipulated by relational strategies. Based on literature study and empirical research we explore the consequences of the improved LTV model for marketing management decisions. We also describe the interaction data that should be available in the customer database.

This exercise is one in a series intended to help students learn how to perform financial calculations in marketing contexts.Maru Keitou, a decorated former collegiate softball player with a PhD from Oxford University, ran Maru Batting Center in the Roppongi district of Tokyo's Minato ward. She had a deep knowledge of the game and of her customers, but she lacked a marketing background. She had recently signed up for a hosted customer relationship management service that would allow her to track the cost of acquiring and serving each of her four main customer segments. Using this data, she could determine which segments to target in the upcoming year.The exercise describes the use of calculations of customer acquisition cost, retention rates, and customer lifetime value in picking between market segments and various options for activities to acquire customers.Maru Keitou, a decorated former collegiate softball player with a PhD from Oxford University, ran Maru Batting Center in the Roppongi district of Tokyo's Minato ward. She had a deep knowledge of the game and of her customers, but she lacked a marketing background. She had recently signed up for a hosted customer relationship management service that would allow her to track the cost of acquiring and serving each of her four main customer segments. Using this data, she could determine which segments to target in the upcoming year.The exercise describes the use of calculations of customer acquisition cost, retention rates, and customer lifetime value in picking between market segments and various options for activities to acquire customers.After completing the exercise, students should be able to: Calculate customer acquisition cost Determine customer break-even Calculate and explain customer lifetime value