Method and system for executing nudge marketing using customer value decay function

Abstract

The present disclosure relates to a method and system of executing an efficient marketing strategy targeting customers, and more particularly, a method and system of calculating a customer decay curve using a customer value decay function, and finding a customer range eligible for nudge marketing to maximize marketing effectiveness.

Description

METHOD AND SYSTEM FOR EXECUTING NUDGE MARKETING USING CUSTOMER VALUE DECAY FUNCTION

[0001] The present disclosure relates to a method and system of executing an efficient marketing strategy targeting customers, and more particularly, a method and system of calculating a customer decay curve using a customer value decay function, and finding a customer range eligible for nudge marketing to maximize marketing effectiveness.

[0002] A customer lifetime value (CLV) refers to a present value of a profit flow that consumers are expected to purchase over a predetermined period of time, and is an indicator used by a seller to maximize profitability from a long-term perspective.

[0003] There are various models (linear regression model, RFM model, decision-tree based boosting algorithm, deep learning algorithm, etc.) to predict a CLV, but all conventional models have severe deviations due to data imbalance and have very poor prediction performance for most individual retail purchase patterns, which occur intermittently.

[0004] The present disclosure is proposed in consideration of these problems, and proposed to provide a methodology that can capture a point in time when nudge marketing is possible using a customer value decay function while maintaining the accuracy of data analysis results based on customer groups.

[0005] Unlike existing disclosures that aimed to analyze and predict a customer lifetime value (CLV) based on customer data, an aspect of the present disclosure is to provide a function that can capture a section in which the customer lifetime value decreases or increases.

[0006] Another aspect of the present disclosure is to provide a system that can appropriately determine a point at which nudge marketing can be actively performed using this function, and a range of customers subject to nudge marketing.

[0007] Meanwhile, technical problems of the present disclosure are not limited to the above-mentioned problems, and other technical problems which are not mentioned herein will be clearly understood by those skilled in the art from the description below.

[0008] The present disclosure is contrived to solve the foregoing problems, and a nudge marketing execution method using a customer value decay function according to the present disclosure may include (a) collecting customer data; (b) preprocessing the collected customer data; (c) classifying individual customers or customer groups based on the preprocessed customer data; (d) determining an analysis target from the classified individual customers or customer groups; (e) calculating a customer value decay function for the analysis target; and (f) extracting a customer range subject to nudge marketing by referring to the customer value decay function.

[0009] Furthermore, in the nudge marketing execution method using a customer value decay function, the analysis target may be a specific customer extracted from the customer data, or a specific customer group among a plurality of customer groups classified from the customer data.

[0010] Furthermore, in the nudge marketing execution method using a customer value decay function, the customer value decay function in the step (e) may be a purchase probability decay function of the analysis target.

[0011] Furthermore, in the nudge marketing execution method using a customer value decay function, the customer value decay function in the step (e) may be a purchase amount ratio decay function of the analysis target.

[0012] Furthermore, in the nudge marketing execution method using a customer value decay function, the step (f) may include calculating a churn probability of the analysis target, the churn probability being a value indicating a probability that the analysis target will churn from a purchase activity, wherein the churn probability of the analysis target is a value obtained by subtracting a value at an arbitrary point in time on the normalized customer value decay function from 1.

[0013] Furthermore, in the nudge marketing execution method using a customer value decay function may further include, subsequent to the step (f): (g) executing a marketing strategy targeting customers included within the extracted customer range.

[0014] Furthermore, in the nudge marketing execution method using a customer value decay function, the step (g) may include generating one or more executable marketing strategies targeting customers included in the customer range; and executing the generated marketing strategies to the customers included in the customer range.

[0015] Furthermore, in the nudge marketing execution method using a customer value decay function, the step (g) may include generating two or more marketing strategies targeting customers included in the customer range; and executing a marketing strategy selected from the plurality of marketing strategies targeting customers included in the customer range.

[0016] Meanwhile, a computing device for executing nudge marketing according to another embodiment of the present disclosure may include a central processing unit and a memory, wherein the central processing unit executes instructions for executing a nudge marketing execution method using a customer value decay function stored in the memory, and the nudge marketing execution method using the customer value decay function includes (a) collecting customer data; (b) preprocessing the collected customer data; (c) classifying individual customers or customer groups based on the preprocessed customer data; (d) determining an analysis target from the classified individual customers or customer groups; (e) calculating a customer value decay function for the analysis target; and (f) extracting a customer range subject to nudge marketing with reference to the customer value decay function.

[0017] Furthermore, in the computing device, the customer value decay function in the step (e) may be a purchase probability decay function of the analysis target.

[0018] Furthermore, in the computing device, the customer value decay function in the step (e) may be a purchase amount ratio decay function of the analysis target.

[0019] According to the present disclosure, there is an effect in that a customer lifetime value (CLV) and a customer value decay function can be quantified and utilized over time.

[0020] In addition, according to the present disclosure, an analysis of consumers and a marketing strategy to be followed up may be programmed, thereby having an effect of more easily and effectively implementing the marketing strategy from the perspective of a user (seller, etc.).

[0021] Meanwhile, the effects of the present disclosure may not be limited to the above-mentioned effects, and other technical effects which are not mentioned herein will be clearly understood by those skilled in the art from the description below.

[0022] FIG. 1 is a diagram to conceptually understand a nudge marketing execution method using a customer value decay function according to the present disclosure.

[0023] FIG. 2 shows an example of a customer value decay function.

[0024] FIG. 3 shows respective steps of a nudge marketing execution method according to the present disclosure in an order thereof.

[0025] FIGS. 4 and 5 are tables for understanding an actual process of determining a customer range subject to nudge marketing.

[0026] FIG. 6 shows another implementation example of a customer range determination step.

[0027] FIG. 7 shows a list of data that can be referenced when determining a customer range.

[0028] FIG. 8 shows how a marketing strategy for nudge marketing is executed for a specific customer.

[0029] FIG. 9 shows how one marketing strategy selected from a plurality of previously generated marketing strategies is executed for a specific customer.

[0030] FIG. 10 shows another implementation example of determining a customer range.

[0031] The details of the objects and technical configurations of the present disclosure and operational effects thereof will be more clearly understood from the following detailed description based on the accompanying drawings appended hereto. Hereinafter, embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings.

[0032] Embodiments disclosed herein should not be interpreted as limiting or used to limit the scope of the present disclosure. It is apparent for those skilled in the art that a description including embodiments herein has various applications. Therefore, any embodiments described in the detailed description of the present disclosure are illustrative for better understanding of the present disclosure and are not intended to limit the scope of the present disclosure to the embodiments.

[0033] Functional blocks illustrated in the drawings and described hereunder are only examples of possible implementations. In other implementations, other functional blocks may be used without departing from the concept and scope of the detailed description. Furthermore, one or more functional blocks of the present disclosure are illustrated as separate blocks, but one or more of the functional blocks of the present disclosure may be a combination of various hardware and software elements that execute the same function.

[0034] In addition, an expression that some elements are "included" is an expression of an "open type", and the expression simply denotes that the corresponding elements are present, but should not be construed as excluding additional elements.

[0035] Moreover, in case where it is mentioned that one element is "connected" or "coupled" to the other element, it should be understood that one element may be directly connected to the other element, but another element may be present therebetween.

[0036]

[0037] FIG. 1 is a diagram for schematically understanding a marketing execution process according to the present disclosure, more precisely, a process of executing nudge marketing.

[0038] As can be seen from the drawing, in the present disclosure, the service server 200 may collect customer data that can be collected from a plurality of consumers, and the service server 200 may calculate a customer value decay function for a specific customer or a specific customer group therefrom, and capture a nudge marketing point in time, that is, a nudge marketing point, from the customer value decay function to execute effective nudge marketing for the customer or customer group.

[0039] One of the main features of a nudge marketing execution method and system according to the present disclosure is to calculate a customer value decay function, wherein the customer value decay function denotes a function that quantifies a phenomenon in which when a customer purchases a specific product or service, a probability of the customer purchasing a product again or an expected value of an estimated purchase amount decreases as time passes from the point of purchase.

[0040] The customer value decay function according to the present disclosure may have various features and behaviors for a customer (u; e.g. customer information, a purchased product, a purchase time, an inflow path, etc.), a last purchase point or a reference time (t) when making a subsequent prediction, a prediction period (d; e.g., a period from present to 30 days later) as variables, and the function may be expressed in the form of a non-linear graph with time as the x-axis and customer value as the y-axis. Here, a decreased customer value may be calculated as a numerical value according to a relational expression depending on what value is entered into each variable in the customer value decay function.

[0041] In the customer value decay function according to the present disclosure, in order to offset as much as possible factors that may have a spot effect on the probability of purchase, such as specific days of the week and specific times for each specific customer or specific customer group, randomly and irregularly input values may be statistically processed (e.g., averaging over groups or period / cycle, time / field-windowed averaging, time / field-windowed filtering, etc.), and converted into a meaningful decay function through this process.

[0042] Such a customer value decay function may be used in a variety of ways, and for example, through the customer value decay function, the churn likelihood of a specific customer or a specific customer group may be identified probabilistically. Additionally, when the y-axis value is defined as an estimated sales amount in the customer value decay function and then an integral operation is performed within a predetermined section of the x-axis, a customer lifetime value (CLV) may be calculated. In addition, in case where a decrease in customer value over time can be calculated by using the customer value decay function, a user (seller) may use the function to execute marketing, especially nudge marketing, targeting a specific customer or customer group, thereby using it to prevent the churn of the customer or customer group. In particular, when trying to prevent the churn of a customer or customer group, for example, in case where a maximum customer retention period without a purchase (e.g. 6 months) is defined as a customer churn reference point, and the customer's most recent purchase time is defined as the origin, a tendency for customer value to decrease during a period between the origin (the most recent purchase point) and the customer churn reference point may be quantitatively quantified for each customer or customer group, and from this, a probability of churn for the customer or customer group may be computed or a point in time when nudge marketing can be executed (nudge marketing point) may be calculated, thereby preventing the churn of the customers.

[0043]

[0044] Meanwhile, the customer value decay function may be calculated from two major aspects. FIG. 2 shows an example of a customer value decay function, in which a purchase probability (dashed line) and a purchase amount ratio (solid line) are shown for a specific period. The customer value decay function in the form shown in FIG. 2 may be normalized by setting values at a point in time when the customer last purchased a product or service and a maximum customer retention period to 1 and 0, respectively, and the use of probability operations that are already well-defined academically may be allowed through the normalization transformation. A graph result as shown in FIG. 2 may also be understood as a (residual) customer value, wherein in FIG. 2, the closer it is to the origin (point of last purchase) in time, the higher the value is, and the farther away it is from the origin (the longer a purchase has not been made), the lower the value becomes, and thus as a result, it can be seen that the (residual) customer value decreases and a customer churn probability (customer churn probability = 1 - (customer value decay function value)) increases.

[0045] With reference to FIGS. 1 and 2, a customer value decay function to be covered in this detailed description, and a rough concept of a method of executing marketing using this function have been described.

[0046]

[0047] FIG. 3 shows a step-by-step list in a nudge marketing execution method (hereinafter referred to as a 'nudge marketing execution method') using a customer value decay function according to the present disclosure.

[0048] Prior to giving a full description of FIG. 3, it is understood that the nudge marketing execution method according to the present disclosure may be executed by a computing device 100 equipped with a central processing unit and a memory. The type of such a computing device may include both a portable terminal such as a smartphone, a PDA, a tablet PC, and a terminal fixedly placed at a predetermined position, such as a desktop PC. The central processing unit may also be referred to as a controller, a microcontroller, a microprocessor, a microcomputer, or the like. Furthermore, the central processing unit may be implemented by hardware or firmware, software, or a combination thereof, and configured to include an application specific integrated circuit (ASIC) or a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), or a field programmable gate array (FPGA) when implemented using hardware, and configured with firmware or software to include a module, a procedure, a function or the like that performs the foregoing functions or operations when implemented using firmware or software. In addition, the memory may be implemented as Read Only Memory (ROM), Random Access Memory (RAM), Erasable Programmable Read Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), flash memory, Static RAM (SRAM), a hard disk drive (HDD), a solid state drive (SSD) or the like.

[0049] In some cases, the computing device 100 may be a server, and in this case, the server may be a device that stores and executes a program, that is, a set of instructions, for actually implementing the nudge marketing execution method according to the present disclosure. The type of the server may be at least one server PC managed by a specific user or may be a type of cloud server provided by another company, that is, a type of cloud server that the user can sign up to use.

[0050] Additionally, in some cases, the nudge marketing execution method according to the present disclosure may be executed not on a single computing device 100 but on a computing system consisting of a plurality of computing devices. Within the computing system, a plurality of computing devices required to execute the nudge marketing method may be respectively set to perform different operations.

[0051] As such, the nudge marketing execution method according to the present disclosure may be executed by a computing device or computing devices equipped with a central processing unit and a memory, and the description of FIG. 3 will be continued below.

[0052] Referring to the drawing, the nudge marketing execution method may first include a step S101 of collecting customer data. Customer data has a concept that includes all types of data that can be referred to in generating a customer lifetime value (CLV) or customer value decay function, which may include all types of data that can be actively or passively collected by hardware devices in a process of purchasing, by any customer, a product or service. Customer data may include, for example, inflow analysis data including inflow channel information (channel type or path accessed to purchase a product or service), membership registration information (membership registration status, information entered during membership registration, etc.), customer behavior information (time spent on a page during a purchase, number of clicks, entered search terms, customer location at the time of a purchase, etc.), or the like, customer attribute data including gender, age, grade, or the like, purchase attribute data including first purchase date, last purchase date, purchase frequency, purchase amount, purchased product, or the like. The process of collecting customer data may be carried out in a variety of ways, for example, including methods of receiving some of the above data directly from a user terminal (a smartphone, a PC, etc.) owned by a customer, receiving data using network / offline media from a company that specializes in collecting and processing customer data, or the like.

[0053] Subsequent collecting customer data, a step S103 of preprocessing the collected customer data may be performed. In the preprocessing step, processing required to perform subsequent steps, such as a process of formalizing unstructured customer data, a process of excluding data outside a specified range (a range set by a user) from among data, and a process of clustering data by attribute, may be carried out.

[0054] Subsequent to the preprocessing step, a step S105 of classifying customers based on customer data may be performed. This is a step in which so-called customer segmentation is carried out, which is understood as a process of extracting a plurality of effective factors from previously collected and preprocessed customer data and classifying customers into an arbitrary number of groups by referring to a similarity between these effective factors. Including the classifying of customers in this manner is to more effectively analyze a large amount of customer data while at the same time maintaining the accuracy and stability of the analyzed result value. That is, when a follow-up analysis is carried out only based on customers' individual data, problems such as excessive deviation in analysis results may occur due to a lack of group tendency, and customer classification in this step may collect customer data with similar effective factors to enable classification into a customer group, thereby allowing the general attributes (customer attributes, purchase attributes, etc.) of customers with similar tendencies to be revealed.

[0055] Subsequent to the customer classification step, a step S107 of determining an analysis target may be performed. In order to find a customer subject to marketing, it is first necessary to determine the customer or a customer group that includes the customer as an analysis target. For example, specific customers or customer groups, such as customers who have made purchases in an online store over the past 6 months, male customers between the ages of 25 and 30, and specific VIP customers who have purchased products or services worth more than 10 million won (KRW, South Korean currency) in the past week may be determined as an analysis target. This step may be implemented to allow a user to directly select a specific customer or customer group through an interface provided by the computing device 100 so as to determine an analysis target, or in some cases, may be implemented to allow the computing device 100 to automatically determine an analysis target and then carry out subsequent steps of the nudge marketing execution method for the analysis target, and then automatically determine a new analysis target again so as to carry out the subsequent steps repeatedly.

[0056] Subsequent to determining the analysis target, a step S109 of calculating a customer value decay function for the analysis target may be performed. It has been mentioned that the analysis target may be a specific customer or a specific customer group, and in this step, the customer value decay function may be calculated according to each analysis target.

[0057] For reference, in step S109, in addition to calculating the customer value decay function, other functions that can be obtained from the customer value decay function may be further calculated. Typically, a customer lifetime value (CLV) function f may be calculated from a customer value decay function g using Equation 1 and Equation 2 below. In equations below, u is a variable representing a specific customer or a specific customer group, and T is a variable representing any given period.

[0058] [Equation 1]

[0059]

[0060] [Equation 2]

[0061]

[0062] On another hand, if g in Equation 2 for a 'purchase case' smoothed over T, different from an instantaneous function g indicating 'a purchase event' inside Equation 2 having of value 1 (purchased) or 0 (otherwise), is calculated for a specific customer or customer group u, and f is obtained therefrom, then f may be a function for obtaining a total number of purchases expected within a given period, wherein furthermore, a purchase probability density function P of the customer or customer group may be obtained from Equation 3 below.

[0063] [Equation 3]

[0064]

[0065] On the other hand, if g for a 'purchase amount' is calculated for a specific customer or customer group u, and f is obtained therefrom, then f may be a function for obtaining a total estimated purchase amount within a given period, which may be a function with the same meaning as the customer lifetime value (CLV) function.

[0066] Judging from the above relational expressions, if either the customer value decay function g or the customer lifetime value function f can be calculated, then the remaining functions may be naturally calculated by referring to the above relational expressions. In addition, since conditions defined when calculating the customer value decay function g are generally more meaningful, when preprocessing is performed on given customer data, a specific analysis target (customer, customer group) may be selected to extract and accumulate attributes and features for a given period, then a value thereof will be derived as a result value for the feature, thereby performing learning for calculating the customer value decay function g using the derived result value.

[0067] Subsequent to step S109, the nudge marketing execution method may further include a step S111 of extracting a customer range subject to nudge marketing. This step is basically performed when the previous analysis target was a specific customer group, wherein assuming that one customer group includes a plurality of customers, which may also be understood as a step of determining which range of customers from among the plurality of customers will be targeted for nudge marketing. For reference, when the analysis target is a specific customer, the step may be replaced with a step of determining whether the customer is a target of nudge marketing.

[0068] FIGS. 4 and 5 are tables for understanding an actual process of determining a customer range subject to nudge marketing. Specifically, FIG. 4 is a table that calculates and organizes purchase probabilities of customers A, B, and C by period, and FIG. 5 is a table that calculates and organizes estimated cumulative purchase amounts by period for the same customers. Tables shown in FIGS. 4 and 5 may include values that can be obtained by the customer lifetime value function f derived from the customer value decay function g.

[0069] Referring to FIG. 4, it is expected that customer A's purchase likelihood does not decrease significantly even as time passes, while it is expected that customer B's purchase likelihood decreases significantly after 20 days, and customer C's purchase likelihood drops sharply after 10 days to a point where he or she is hardly expected to make a purchase after 30 days. From the perspective of a user (seller) among customers A, B, and C in a customer group, it may be considered to execute nudge marketing at an appropriate point in time targeting the customer B rather than the customer A, whose purchase likelihood is expected to remain at or higher than a predetermined level even without a nudge, or the customer C who is hardly expected to make a purchase after a predetermined period, and the computing device 100 may be manipulated such that the customer B is included in a customer range targeted for marketing based on purchase likelihood, or the customer B is directly targeted for marketing.

[0070] Meanwhile, referring to a table in FIG. 5 based on estimated cumulative purchase amounts, cumulative purchase amounts for customers A and B are expected to continuously increase as the period passes, while an initial purchase amount for the customer C is expected to no longer increase. From the perspective of a user (seller), it would be preferable to select the customers A and B, whose cumulative purchase amounts are expected to increase, as marketing targets, and therefore, the computing device 100 may be manipulated to determine the customers A and B as a customer range subject to marketing.

[0071] Meanwhile, a customer range subject to marketing may be determined by referring to both the purchase likelihood in FIG. 4 and the cumulative purchase amount in FIG. 5. In other words, when the customer B can be selected as a marketing target in the condition of FIG. 4, and the customers A and B can be selected as a marketing target in the condition of FIG. 5, ultimately, only the customer B who satisfies both the conditions in FIGS. 4 and 5 may be targeted for nudge marketing.

[0072] For reference, a description referring to FIGS. 4 and 5 has been made based on an embodiment in which a user directly determines a customer range subject to marketing, but the customer range subject to marketing may also be implemented to be autonomously set by the computing device 100. The computing device 100 may be designed to autonomously calculate under what condition and to what range it is most preferable to include customers as marketing targets by referring to a nudge marketing effect for each customer range setting case learned by an artificial intelligence algorithm, and to determine on its own a customer range subject to marketing.

[0073] FIG. 6 is a flowchart for explaining a modified embodiment of the customer range extraction step S111 described above. Referring to FIG. 6, a modified embodiment of the customer range extraction step, may include, subsequent to calculating, by the computing device 100, a customer value decay function (S209), a step of calculating a customer churn probability (S2111), and a step of determining a range of customers subject to marketing based on the customer churn probability (S2113).

[0074] In the foregoing description of the customer value decay function, it has been mentioned that a probability of customer churn may be calculated from this function, wherein the step S2111 may include obtaining a customer churn probability from the customer value decay function in this manner. Although not shown separately in the drawing, for the customers A, B, and C belonging to a specific customer group, a probability of customer churn, that is, a likelihood of customers churning out of purchasing activities, may be quantified and presented in a table, wherein for example, if the customer A's churn probability is calculated as 90%, the customer B's churn probability as 25%, and the customer C's churn probability as 30%, then the user (seller) may determine the customer A, who is expected to have the highest churn probability, as a customer range subject to marketing.

[0075] FIG. 7 shows a list of references that can be referred to when determining a customer range subject to nudge marketing. The user (seller) or computing device 100 may determine a customer range subject to marketing by referring to criteria related to purchasing activities, such as customer churn probability, purchase probability, and estimated purchase amount, or determine a customer range subject to marketing by referring to criteria related to individual customer attributes, such as gender, age group, and region of residence. A plurality of the above criteria may be combined to specify a customer range that satisfies all conditions such as, for example, men in their 20s living in the region of Seoul, a customer churn probability within a range of 30-70% within 30 days, and recording a purchase amount of above 300,000 won within the next two months. For reference, it is understood that the references listed in FIG. 7 are illustratively shown to aid understanding of the disclosure, and are not limited thereto.

[0076] In the above, the step S111 of determining a customer range subject to marketing has been described.

[0077]

[0078] Referring again to FIG. 3, subsequent to determining a customer range subject to marketing, nudge marketing targeting customers for whom the range has been determined may be performed (S113). The meaning of nudge marketing being implemented essentially may denote that a specific marketing strategy targeting the customers has been implemented, and in this case, the marketing strategy may include at least an execution period and a nudge type.

[0079] FIG. 8 shows one marketing strategy by way of example, wherein when a marketing strategy for specific nudge marketing is executed for the customer B, the marketing strategy may include (i) an execution period, for example, an execution period indicating the marketing strategy to be executed within 20 to 30 days from the present point in time, and furthermore, may include (ii) a nudge type, for example, a nudge methodology called 'inducing a click' of the customer B. In addition, if necessary, the marketing strategy may further include a more specific marketing strategy description such as, for example, inducing a cursor to move to a specific button on the screen and outputting a special effect animation character to induce a click.

[0080] Meanwhile, the marketing strategy may preferably be generated subsequent to determining a customer range subject to the nudge marketing. The conditions considered in the process of determining a customer range may be referred to so as to generate a marketing strategy through calculating in real time at what point in time nudge marketing can be implemented for the customer range to maximize the marketing effect and which type of nudge can best induce a customer to make a purchase.

[0081] On the other hand, a marketing strategy may be determined by selecting the most appropriate marketing strategy at the present time from a plurality of previously generated marketing strategies. If a first implementation example described above involves generating a marketing strategy in real time after the customer range is determined, then a second implementation example is characterized in that when the customer range is determined, one specific marketing strategy among a plurality of previously generated marketing strategies is selected with reference to the customer range and the conditions considered in determining the customer range. FIG. 9 is a drawing for explaining an implementation example in which any one from among a plurality of marketing strategies is selected, wherein it is shown in the drawing, that when customer D is determined to be a target for nudge marketing, a third marketing strategy out of a total of three marketing strategies is selected to implement nudge marketing targeting the customer D.

[0082] FIG. 10 shows another example of executing nudge marketing, wherein a table in which a probability of a customer churning at the present point in time, a probability of a customer churning when a nudge is provided to the customer according to each marketing strategy, and a probability of a customer who will actually take an action based on a nudge are summarized is shown. The table in the drawing describes a first marketing strategy (click inducement), a second marketing strategy (event participation inducement), and a third marketing strategy (purchase conversion inducement), and in this situation, the user (seller) or the computing device 100 may select the first marketing strategy, which is a marketing strategy that is most likely to induce a customer action while significantly decreasing the probability of customer churn.

[0083] In the above, the steps for executing nudge marketing have been described with reference to FIGS. 8 to 10.

[0084] In the above, a nudge marketing execution method using a customer value decay function according to the present disclosure and a system therefor have been described. Meanwhile, the present disclosure is not limited to the foregoing specific embodiments and application examples, it will be of course understood by those skilled in the art that various modifications may be made without departing from the gist of the present disclosure as defined in the following claims, and it is to be noted that those modifications should not be understood individually from the technical concept and prospect of the present disclosure.

Claims

1.A nudge marketing execution method using a customer value decay function, the method comprising:(a) collecting customer data;(b) preprocessing the collected customer data;(c) classifying individual customers or customer groups based on the preprocessed customer data;(d) determining an analysis target from the classified individual customers or customer groups;(e) calculating a customer value decay function for the analysis target; and(f) extracting a customer range subject to nudge marketing by referring to the customer value decay function.2.The method of claim 1, wherein the analysis target is a specific customer extracted from the customer data, or a specific customer group among a plurality of customer groups classified from the customer data.3.The method of claim 2, wherein the customer value decay function in the step (e) is a purchase probability decay function of the analysis target.4.The method of claim 2, wherein the customer value decay function in the step (e) is a purchase amount ratio decay function of the analysis target.5.The method of claim 2, wherein the step (f) comprises:calculating a churn probability of the analysis target, the churn probability being a value indicating a probability that the analysis target will churn from a purchase activity, andwherein the churn probability of the analysis target is a value obtained by subtracting a value at an arbitrary point in time on the normalized customer value decay function from 1.6.The method of claim 2, further comprising, subsequent to the step (f):(g) executing a marketing strategy targeting customers included within the extracted customer range.7.The method of claim 6, wherein the step (g) comprises:generating one or more executable marketing strategies targeting customers included in the customer range; andexecuting the generated marketing strategies to the customers included in the customer range.8.The method of claim 6, wherein the step (g) comprises:generating two or more marketing strategies targeting customers included in the customer range; andexecuting a marketing strategy selected from the plurality of marketing strategies targeting customers included in the customer range.9.A computing device for executing nudge marketing, the computing device comprising a central processing unit and a memory,wherein the central processing unit executes instructions for executing a nudge marketing execution method using a customer value decay function stored in the memory, andwherein the nudge marketing execution method using the customer value decay function comprises:(a) collecting customer data;(b) preprocessing the collected customer data;(c) classifying individual customers or customer groups based on the preprocessed customer data;(d) determining an analysis target from the classified individual customers or customer groups;(e) calculating a customer value decay function for the analysis target; and(f) extracting a customer range subject to nudge marketing with reference to the customer value decay function.10.The computing device of claim 9, wherein the customer value decay function in the step (e) is a purchase probability decay function of the analysis target.11.The computing device of claim 9, wherein the customer value decay function in the step (e) is a purchase amount ratio decay function of the analysis target.

Images