نقشه استراتژی بازاریابی جدید برای بازاریابی مستقیم
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|23576||2009||9 صفحه PDF||سفارش دهید||6710 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Knowledge-Based Systems, Volume 22, Issue 5, July 2009, Pages 327–335
Direct marketing is one of the most effective marketing methods with an aim to maximize the customer’s lifetime value. Many cost-sensitive learning methods which identify valuable customers to maximize expected profit have been proposed. However, current cost-sensitive methods for profit maximization do not identify how to control the defection probability while maximizing total profits over the customer’s lifetime. Unfortunately, optimal marketing actions to maximize profits often perform poorly in minimizing the defection probability due to a conflict between these two objectives. In this paper, we propose the sequential decision making method for profit maximization under the given defection probability in direct marketing. We adopt a Reinforcement Learning algorithm to determine the sequential optimal marketing actions. With this finding, we design a marketing strategy map which helps a marketing manager identify sequential optimal campaigns and the shortest paths toward desirable states. Ultimately, this strategy leads to the ideal design for more effective campaigns.
Direct marketing is one of the most effective marketing methods with an aim to maximize the expected profits . A number of cost-sensitive learning methods which focus on predicting profitable customers have been proposed for direct marketing , ,  and . However, a common objective of these methods is to only maximize the short-term profit associated with each marketing campaign. They ignore the interactions among decision outcomes when sequences of marketing decisions are made over time. These independent decision-making strategies cannot guarantee the maximization of total profits generated over a customer’s lifetime because they often inundate profitable customers with frequent marketing campaigns or encourage radical changes in customer behavior . This approach can decrease customer profitability because of the annoyance factor or their budgetary limits per unit time. Some researchers have recognized the importance of sequential decision making to overcome the limitations of isolated decision making. For example, Pednault et al.  and Abe et al.  proposed sequential cost-sensitive learning methods for direct marketing. These sequential cost-sensitive methods, however, fail to consider the cost generated from customer defections. Although a primary objective of direct marketing is to maximize total profit, it is also important to control the probability of customer defection, keeping it under a desirable or acceptable level because the occurrence of a customer defection brings about tangible and intangible loss, (i.e., an increase of acquisition cost of a new customer, loss of word-of-mouth effects, and loss of future cash flows and profits). Since customer switching costs are much lower in e-commerce marketplaces, a company always needs to pay more attention to customer defection. However, current sequential cost-sensitive methods for maximizing profit do not indicate how to control the probability of customer defection while maximizing total profits over the customer’s lifetime. Unfortunately, optimal marketing actions designed to maximize profits often perform poorly in minimizing the probability of customer defection due to a conflict between a profit maximization and defection probability minimization. For example, an optimal marketing action for profit maximization is liable to give up unprofitable customers who are most likely to defect but are profitable from a long-term perspective. In contrast, an optimal marketing action for the minimization of defection probability is apt to unnecessarily sacrifice loyal customers’ profit with excessive marketing cost. To overcome this conflict, we regard the customer defection probability as a constraint and try to control it under the given threshold because, in general, controlling defection probability under the threshold is more cost effective than completely avoiding customer defection with 0%. We also think that most companies have more interest in a strategy which guarantees the maximization of total profits while the defection probability is bounded by a desirable or acceptable level. In this paper, we have developed a sequential decision-making methodology for profit maximization under the given defection probability constraint. For effective sequential learning, we have adopted the Reinforcement Learning algorithm. We have also suggested the concept of a marketing strategy map which visualizes the results of learning such as an optimal marketing action in each state and customer’s behavior dynamics according to suggested marketing actions. This marketing strategy map can help a company identify sequential optimal campaigns and the shortest paths toward desirable states. Ultimately, this strategy leads to the ideal design for more effective campaigns. The rest of this paper is organized in the following manner: In Section 2, a Self-Organizing Map and Reinforcement Learning that are prerequisites for our study are briefly introduced. Section 3 details our method for direct marketing and Section 4 reports experimental results with real-world data sets. Section 5 describes a marketing strategy map and its applications. Finally, Section 6 summarizes our works and contributions.
نتیجه گیری انگلیسی
While direct marketing has garnered a great deal of attention, few studies have addressed the tradeoff between two conflicting objectives such as the profit and defection probability even though these tradeoffs are of great interest to companies. To solve this tradeoff conflict, we have developed a sequential decision-making methodology for profit maximization under the given defection probability constraint. Our method suggests sequential optimal marketing actions for maximizing long-term total profit while controlling the defection probability under the threshold over a customer’s lifetime. In addition, the suggested marketing strategy map clearly shows an optimal action and customers’ behavior dynamics in each state. It also helps a marketing manager identify sequential optimal campaigns and the shortest paths toward desirable states and, ultimately, a design for more effective campaigns. Our experiments demonstrate the feasibility of our proposed method in direct marketing. The proposed method is a practical implementation procedure for direct marketing in telecommunications, online shopping malls, and other highly competitive marketplaces suffering from profit loss and customer defections.