Customer analysis system and customer analysis method

Abstract

Provided is technology for performing customer analysis that captures changes in values that appeal to customers or in products or the like in a timely manner. This customer analysis system comprises a computer having at least an arithmetic device and a storage device. The storage device stores a customer feature score associating a customer with a score of a feature of the customer, an object-for-sale feature appeal score associating an object-for-sale with a score of the appeal the object-for-sale has to the feature of the customer, and customer feedback about the object-for-sale from the customer. The arithmetic device updates the customer feature score and the object-for-sale feature appeal score by using the customer feedback, and outputs the customer feature for which the score increases and the object-for-sale which appeals to the feature.

Description

Customer analysis system and customer analysis method

[0001] The present invention relates to a customer analysis system and a customer analysis method.

[0002] In recent years, in response to the diversification of end-user needs and corporate services, a fundamental review of traditional business operations and systems has become one of the key factors in becoming a company of choice for customers. As competition between companies intensifies, improving customer experience is necessary to ensure customer loyalty. Here, customer experience refers to the value experienced by customers at all points of contact with a company. In order to improve customer experience, it is necessary to understand customers more deeply, at the level of their values. In today's world, where customer values ​​are rapidly changing, the challenge is to estimate customer values ​​and then provide products that, for example, keep up with those changes in values.

[0003] As an example of a technology for providing products based on customer values, the technology described in Patent Document 1 has been proposed. Patent Document 1 states, "According to the customer analysis system and method of the present invention, it is possible to create a model that accurately integrates a model relating to values ​​and lifestyles in a person's life in general with a model relating to customer needs for products or services, and by using this model, it is possible to deepen ties with customers while providing products and services that suit the customers." and "The present invention is a customer analysis system that analyzes customers using a model, comprising: a first model acquisition means for acquiring a first model that indicates general characteristics of a person, including at least values ​​and lifestyles in a person's life in general; a second model acquisition means for acquiring a second model that indicates customer needs for products or services; and a third model creation means for creating a third model that integrates the first model and the second model, wherein the general characteristics that define the first model are values ​​and lifestyles expressed independently of demographic characteristics indicating at least age and gender, the first model includes a correlation between the general characteristics and the values ​​regarding the product or service, the second model includes a correlation between the customer's values ​​regarding the product or service and the details of the product or service selected by the customer, and the third model creation means creates the third model by integrating the first model and the second model using parts of the first and second models that indicate the same or similar values ​​regarding the product or service.

[0004] Japanese Patent Application Laid-Open No. 2022-150503

[0005] By using a model that can be created using the technology described in Patent Document 1, it is possible to offer products based on customer values. However, it is described that the model is created "based on one or more of the following: questionnaire results, psychological test results, brain response test results, and physiological response test results." As customer values ​​change rapidly, it is difficult to update the model by conducting questionnaires and psychological tests, for example, every week. Therefore, when using the technology described in Patent Document 1, it is not possible to capture changes in customer values ​​in a timely manner.

[0006] Furthermore, the technology described in Patent Document 1 cannot capture changes in the customer's values ​​that the product is trying to appeal to. In other words, when using the technology described in Patent Document 1, for example, if the values ​​appealing to chocolate change from "hedonism" to "achievement," this change cannot be captured.

[0007] The present invention has been made in consideration of the above-mentioned problems, and aims to provide a technology that enables customer analysis that captures changes in the values ​​that customers, products, etc. appeal to in a timely manner.

[0008] A customer analysis system according to the present invention includes a computer having at least an arithmetic unit and a storage device. The storage device stores customer characteristic scores that associate customers with scores of their characteristics, sales object characteristic appeal scores that associate sales objects with scores of the sales objects' appeal to the customer characteristics, and customer feedback from customers regarding the sales objects. The arithmetic unit uses the customer feedback to update the customer characteristic scores and sales object characteristic appeal scores, and outputs customer characteristics whose scores are increasing and sales objects that appeal to those characteristics.

[0009] Other problems and solutions disclosed in the present application will be made clear in the detailed description and drawings.

[0010] According to the present invention, it is possible to carry out customer analysis that captures changes in the values ​​that customers, products, etc., appeal to in a timely manner.

[0011] 1 is a diagram illustrating an example of the configuration of an entire system including a customer analysis system according to an embodiment. FIG. 1 is a diagram illustrating an example of the hardware configuration of a customer analysis AP server. FIG. 2 is a diagram illustrating an example of the hardware configuration of a customer analysis DB server. FIG. 2 is a diagram illustrating an example of the hardware configuration of a client terminal. FIG. 3 is a diagram illustrating an example of the hardware configuration of a customer terminal. FIG. 4 is a diagram illustrating an example of the configuration of a purchase record table. FIG. 5 is a diagram illustrating an example of the table configuration of a product master. FIG. 6 is a diagram illustrating an example of the table configuration of a customer master. FIG. 7 is a diagram illustrating an example of the configuration of a product value appeal score table. FIG. 8 is a diagram illustrating an example of the table configuration of a value master. FIG. 9 is a diagram illustrating an example of the configuration of a product review table. A sequence diagram showing an example of the overall flow of processing executed in the customer analysis system. A flowchart showing an example of the flow of a product value appeal score update process. A flowchart showing an example of the flow of a customer value score update process. A flowchart showing an example of the flow of a customer value change / appealing product presentation process. A diagram illustrating an example of an input screen displayed by a client terminal. A diagram illustrating an example of an output screen displayed by a client terminal. A diagram illustrating another example of output information.

[0012] In the following description, an "interface unit" refers to one or more interface devices. The one or more interfaces may be one or more interface devices of the same type (for example, one or more NICs (Network Interface Cards)) or two or more interface devices of different types (for example, a NIC and an HBA (Host Bus Adapter)).

[0013] In the following description, a "storage unit" refers to one or more memories. At least one memory may be a volatile memory or a non-volatile memory. The storage unit may include one or more PDEVs in addition to one or more memories. A "PDEV" refers to a physical storage device, and may typically be a non-volatile storage device (e.g., an auxiliary storage device). A PDEV may be, for example, an HDD (Hard Disk Drive) or an SSD (Solid State Drive).

[0014] In the following description, a "processor unit" refers to one or more processors. At least one processor is typically a CPU (Central Processing Unit). The processor may include a hardware circuit that performs some or all of the processing.

[0015] In the following description, functions are sometimes described using the term "kkk unit" (excluding the interface unit, storage unit, and processor unit). However, the functions may be implemented by one or more computer programs executed by the processor unit, or by one or more hardware circuits (e.g., a field-programmable gate array (FPGA) or an application-specific integrated circuit (ASIC)). When a function is implemented by a program executed by the processor unit, the specified processing is performed using the storage unit and / or the interface unit, as appropriate, and therefore the function may be considered to be at least a part of the processor unit. Processing described using a function as the subject may be processing performed by the processor unit or a device having the processor unit. A program may be installed from a program source. The program source may be, for example, a program distribution computer or a computer-readable recording medium (e.g., a non-transitory recording medium). The description of each function is merely an example; multiple functions may be combined into one function, or one function may be divided into multiple functions.

[0016] In the following description, information may be described using expressions such as "xxx table," but the information may be expressed in any data structure. In other words, to indicate that the information does not depend on the data structure, an "xxx table" may be referred to as "xxx information." In the following description, the structure of each table is an example, and one table may be divided into two or more tables, or all or part of two or more tables may be one table.

[0017] In the following description, "time" is expressed in units of year, month, day, hour, minute, and second, but the time unit may be coarser or finer than that, or may be a different unit.

[0018] In the following description, a "dataset" refers to data (a logical block of electronic data) consisting of one or more data elements, and may be, for example, any of a record, a file, a key-value pair, and a tuple.

[0019] In the following description, a process may be described using a "program" as the subject, but the process described using a program as the subject may also be a process performed by a processor or a device having that processor. Two or more programs may be realized as one program, or one program may be realized as two or more programs.

[0020] Furthermore, in the following description, a "customer analysis system" may be a system made up of one or more physical computers, or may include a system (e.g., a cloud computing system) implemented on a group of physical computing resources (e.g., a cloud infrastructure). When the customer analysis system "displays" the display information, it may mean that the display information is displayed on a display device possessed by the computer, or that the computer transmits the display information to a display computer (in the latter case, the display information is displayed by the display computer).

[0021] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0022] In the following description, the same or similar components will be designated by common reference numerals, and redundant description may be omitted.

[0023] Furthermore, when there are multiple elements having the same or similar functions, the multiple elements may be described by using the same reference numeral with different subscripts to distinguish between them. On the other hand, when there is no need to distinguish between the multiple elements, the subscripts may be omitted.

[0024] Hereinafter, the "features" in claim 1 will be referred to as "values" and the "sales target" as "products," and examples will be described with reference to the drawings.

[0025] <System Configuration Example> Fig. 1 is a diagram showing an example of the configuration of an entire system including a customer analysis system 100 according to this embodiment. The customer analysis system 100 shown in Fig. 1 is a computer system that, in the retail industry, captures rises and falls in customer values ​​and enables product selection that appeals to the rising values ​​of customers.

[0026] 1, this customer analysis system 100 is configured by the cooperation of a customer analysis AP server 101 and a customer analysis DB server 102, which are communicatively connected to each other via a network 104. Of course, this configuration of the customer analysis system 100 is just one example, and these servers may also be implemented as an integrated server device.

[0027] On the other hand, the client terminal 103 is a terminal used by a user who wants to recognize products that appeal to rising customer values ​​in order to determine product lineups, such as in a retail store's product department, and is capable of communicating with, for example, the customer analysis AP server 101 via the above-mentioned network 104.

[0028] Next, a description will be given of the hardware configuration of each server that constitutes the customer analysis system 100. Fig. 2 is a diagram showing an example of the hardware configuration of the customer analysis AP server 101 according to this embodiment.

[0029] The customer analysis AP server 101 in this embodiment includes a storage device 201 configured as an appropriate non-volatile storage device such as a hard disk drive, a memory 203 configured as a volatile storage device such as a RAM, a CPU (Central Processing Unit) 202 which is a computing device that reads out and executes a program 210 stored in the storage device 201 into the memory 203, and serves as a control unit (not shown) to perform overall control of the device itself and perform various judgments, calculations, and control processing, and a communication device 206 which communicates with a client terminal 103 operated by a user via a network 104. The program 210 in the storage device 201 is a program that implements the functions required for the customer analysis system 100 of this embodiment, namely, a product value appeal score update function, a customer value score update function, and a customer value change / appeal product presentation function.

[0030] 3 is a diagram showing an example of the hardware configuration of the customer analysis DB server 102 according to this embodiment. The customer analysis DB server 102 according to this embodiment includes: a storage device 301 configured as a suitable nonvolatile storage device such as a hard disk drive; a memory 303 configured as a volatile storage device such as a RAM; a CPU 302, which is a computing device that reads a program 310 stored in the storage device 301 into the memory 303 and executes it as a control unit (not shown) to perform overall control of the device itself and perform various determinations, calculations, and control processing; and a communication device 306 that is connected to the network 104 and handles communication processing with other devices. In addition to the program 310, the storage device 301 also includes a purchase history table 311, a product master 312, a customer master 313, a product value appeal score table 314, a customer value score table 315, a value master 316, and a product review table 317.

[0031] Next, an example of the hardware configuration of the client terminal 103 according to this embodiment will be described. FIG. 4 is a diagram showing an example of the hardware configuration of the client terminal 103 according to this embodiment. The client terminal 103 includes a storage device 401 configured as a suitable nonvolatile storage device such as a hard disk drive, a memory 403 configured as a volatile storage device such as RAM, a CPU 402 that is a computing device that reads and executes a program 410 stored in the storage device 401 into the memory 403, and serves as a control unit (not shown) to perform overall control of the device itself and perform various judgments, calculations, and control processes, an input device 404 that accepts key inputs and voice inputs from the user, an output device 405 such as a display that displays processed data, and a communication device 406 that is connected to the network 104 and handles communication processing with the customer analysis AP server 101 and the like. In addition to the program 410, the storage device 401 also stores screen data for an input screen 450 and an output screen 451. The screen data for the input screen 450 and the output screen 451 may be part of the program 410.

[0032] Next, an example of the hardware configuration of the customer terminal 105 according to this embodiment will be described. FIG. 5 is a diagram showing an example of the hardware configuration of the customer terminal 105 according to this embodiment. The customer terminal 105 includes a storage device 501 configured as a suitable nonvolatile storage device such as a hard disk drive, a memory 503 configured as a volatile storage device such as RAM, a CPU 502 that reads a program 510 stored in the storage device 501 into the memory 503 and executes it as a control unit (not shown) to perform overall control of the device itself and perform various judgments, calculations, and control processes, an input device 504 that accepts key inputs and voice inputs from the user, an output device 505 such as a display that displays processed data, and a communication device 506 that connects to the network 104 and handles communication processing with the customer analysis AP server 101. In addition to the program 510, the storage device 501 also includes screen data for an input screen 550. The screen data for the input screen 550 may be part of the program 510.

[0033] (Functions of the System) Next, the functions of the customer analysis system 100 of this embodiment will be described. The functions described below are functions that are implemented, for example, by executing the program 210 provided in the customer analysis AP server 101 in the customer analysis system 100. Of course, other servers may also have similar functions, or the functions may be divided and held between servers to cooperate with each other.

[0034] (Product Master Registration) The customer analysis system 100 of this embodiment has the function of accepting product information in which the user describes the product's features from the client terminal 103 and storing the accepted information as product master 312 in the storage device 301.

[0035] (Customer master registration) The customer analysis system 100 of this embodiment has the function of accepting customer information from the customer terminal 105, which contains customer information from customers who use retail stores, etc., and storing the accepted information in the memory device 301 as a customer master 313.

[0036] (Product review registration) The customer analysis system 100 of this embodiment has the function of accepting product review information written by customers who use retail stores, etc., from the customer terminal 105, and storing the accepted information in the storage device 301 as a product review table 317.

[0037] (Product value appeal score update function) The customer analysis system 100 has the function of calculating a score of the values ​​that a product appeals to using a value master 316 that defines the values ​​used to analyze customer values ​​and the values ​​that a product appeals to, and a product review table 317 that stores information on product reviews by customers who use retail stores, etc., and then taking a weighted average of the calculated score and scores calculated in the same way in the past, and storing the result in the product value appeal score table 314.

[0038] (Customer value score update function) The customer analysis system 100 also has the function of calculating a customer's value score using a purchase history table 311 that stores the customer's product purchase history and a product value appeal score table 314 that stores the scores of the values ​​that the products appeal to, taking a weighted average of the calculated score and scores calculated in the same way in the past, and storing the result in the customer value score table 315.

[0039] (Function to present products that appeal to changing customer values) The customer analysis system 100 also has a function to display on the client terminal 103 a screen that presents changes in customer values ​​and products that appeal to rising customer values, using a product value appeal score table 314 that stores scores of values ​​that products appeal to, and a customer value score table 315 that stores scores of customer values.

[0040] (Data Configuration Example) Next, a configuration example of data held by the server that constitutes the customer analysis system 100 of this embodiment will be described.

[0041] 6 is a diagram showing an example of the configuration of a purchase history table 311 provided in the customer analysis DB server 102. This purchase history table 311 is a table generated by, for example, the customer analysis DB server 102 by storing POS data acquired from the POS systems of each store that sells products at any time, and has the following table items:

[0042] 6, the customer code, product code, date, unit price, number of sales, and sales amount are defined as follows: customer code, product code, date, unit price, number of sales, and sales amount. In the configuration illustrated in FIG. 6, the customer code is defined as "C111111," the product code purchased by the customer with this customer code is defined as "P111111," the purchase date is defined as "20230101," the unit price of this product to this customer is defined as "100," the number of sales is defined as "3," and the sales amount is defined as "300." When the customer analysis AP server 101 issues an information acquisition instruction using the customer value score update function or the customer value change / appeal product presentation function, the customer analysis DB server 102 sends this purchase history table 311 to the customer analysis AP server 101.

[0043] 7 is a diagram showing an example of the table configuration of the product master 312 provided in the customer analysis DB server 102. This product master 312 is a table that the customer analysis DB server 102 generates by storing user input values ​​received from the client terminal 103, for example, and has the following items as table items.

[0044] 7, the product code, product name, product category, and product description are defined as "P111111," the product name "Chocolate A," the product category "Confectionery," and the product description "Sweet and energizing chocolate." When the user inputs the conditions via the input screen on the client terminal 103, the customer analysis DB server 102 sends this product master 312 to the customer analysis AP server 101.

[0045] 8 is a diagram showing an example of the table configuration of the customer master 313 provided in the customer analysis DB server 102. This customer master 313 is a table that the customer analysis DB server 102 generates by storing input values ​​received from the customer terminal 105 by customers who use retail stores, for example, and has the following items as table items.

[0046] That is, the customer code, age group, and gender. In the configuration illustrated in FIG. 8, the customer code is defined as "C111112," the customer's age group as "10," and the gender as "no response." When the customer analysis AP server 101 issues an information acquisition instruction using the customer value score update function, the customer analysis DB server 102 will send this customer master 313 to the customer analysis AP server 101.

[0047] 9 is a diagram showing an example of the configuration of the product value appeal score table 314 provided in the customer analysis DB server 102. This shows the table structure of the product value appeal score table 314. This product value appeal score table 314 is a table that the customer analysis DB server 102 generates by storing the calculation results received from the customer analysis AP server 101, for example, and has the following items as table items.

[0048] That is, the product code, the update date, and the value appeal score. Here, the value appeal score is the score at which a product with that product code appeals to a customer's values, and is expressed as a 10-dimensional vector, with each dimension representing a value and the value representing the product's appeal score for each value. In the configuration illustrated in FIG. 9 , the product code is "P111111," the update date of the value appeal score for that product code is "20230131," and the value appeal score is defined as each dimension being the value 1 appeal score, the value 2 appeal score, the value 3 appeal score, etc., with the values ​​being 0.8, 0.1, 0.0, etc. When the customer analysis AP server 101 issues an information acquisition instruction using the product value appeal score update function, the customer value score update function, or the customer value change / appeal product presentation function, the customer analysis DB server 102 transmits this product value appeal score table 314 to the customer analysis AP server 101.

[0049] 10 is a diagram showing an example of the configuration of a customer value score table 315 provided in the customer analysis DB server 102. This customer value score table 315 is a table that the customer analysis DB server 102 generates by storing calculation results received from the customer analysis AP server 101, for example, and has the following items as table items.

[0050] That is, the values ​​are age, gender, update date, and value score. Here, the value score is the value score of a customer of the current age and gender, and is expressed as a 10-dimensional vector, with each dimension representing a value and each value representing the score of each value held by the customer. In the configuration illustrated in FIG. 10 , the customer's age is "10," the gender is "male," the update date of the value score for the current product code is "20230101," and the value score dimensions are defined as value 1 score, value 2 score, value 3 score, ..., with the values ​​defined as 0.3, 0.1, 0.1, etc. Note that the initial values ​​of the value scores for each age and gender are predetermined by the administrator of the customer analysis DB server 102. When the customer analysis AP server 101 issues an information acquisition instruction using the customer value score update function or the customer value change / promotional product presentation function, the customer analysis DB server 102 transmits this customer value score table 315 to the customer analysis AP server 101.

[0051] 11 is a diagram showing an example of the table configuration of the value master 316 provided in the customer analysis DB server 102. This value master 316 is a table that is stored and generated in the customer analysis DB server 102 by, for example, an administrator of the server, and has the following items as table items.

[0052] In other words, it is a value definition. In the configuration illustrated in FIG. 11, the value is "value 1," and the definition of this value is defined as "personal pleasure and sensual satisfaction," etc. When a user registers a product and inputs conditions via the input screen on the client terminal 103, the customer analysis DB server 102 transmits this value master 316 to the customer analysis AP server 101.

[0053] 12 is a diagram showing an example of the configuration of a product review table 317 provided in the customer analysis DB server 102. This product review table 317 is a table generated by the customer analysis DB server 102 by storing input values ​​received from the customer terminal 105 by customers who use retail stores, and has the following items as table items:

[0054] 12, the product code, customer code, response date, and review text are defined as follows: product code, customer code, response date, and review text. In the configuration illustrated in FIG. 12, the product code is "P111111," the customer code is "C111112," and the response date and review text provided by the customer regarding the product are defined as "20230103," "I was satisfied with the rich flavor," etc. When the customer analysis AP server 101 issues an information acquisition instruction using the product value appeal score update function, the customer analysis DB server 102 will send this product review table 317 to the customer analysis AP server 101.

[0055] <System Operation Example> The actual procedure of the customer analysis method in this embodiment will be described below with reference to the drawings. Note that the various operations corresponding to the customer analysis method described below are realized by programs that are read into each memory and executed by each server constituting the customer analysis system 100. These programs are composed of code for performing the various operations described below.

[0056] (Processing Procedure Example 1) Figure 13 is a sequence diagram showing an example of the overall flow of processing executed in customer analysis system 100 as processing procedure example 1 of the customer analysis method related to this embodiment. Specifically, it is a sequence diagram showing the flow in which customer analysis AP server 101, customer analysis DB server 102, etc., which make up customer analysis system 100, receive various requests from client terminal 103 and customer terminal 105 via input screens, and finally output analysis results to the output screen of client terminal 103.

[0057] (Product Registration Request, Product Registration Processing) First, a predetermined user who manages a retail store or the like sends product information describing the characteristics of the product as information about the products they handle from client terminal 103 to customer analysis AP server 101. Having received the product information from client terminal 103, customer analysis AP server 101 executes a product registration function and first instructs customer analysis DB server 102 to acquire information necessary for product registration processing. On the other hand, customer analysis DB server 102 receives this acquisition instruction and transmits information from value master 316 to customer analysis AP server 101 in response to the request.

[0058] The customer analysis AP server 101 receives the information transmitted in the above steps from the customer analysis DB server 102 and executes product registration processing using the product registration function. The results of the product registration processing are transmitted to the customer analysis DB server 102 and registered in the product master 312 and the product value appeal score table 314.

[0059] Here, the result registered in the product value appeal score table 314 is the initial value of the value appeal score of the product to be registered. This initial value can be calculated from the linguistic similarity between the product description of the product registered in the product master 312 and the definition of each value in the value master 316, and can be calculated by normalizing the numerical sequence of each similarity so that the sum is 1. The linguistic similarity can be calculated, for example, by vectorizing each sentence using an embedding model and calculating the cosine similarity.

[0060] (Customer registration request, customer registration process) First, customer information describing the characteristics of a customer who has purchased a product handled by a customer using a retail store or the like is sent from customer terminal 105 to customer analysis AP server 101. Having received the customer information from customer terminal 105, customer analysis AP server 101 executes customer registration processing using a customer registration function. The results of the customer registration process are sent to customer analysis DB server 102 and are registered in customer master 313.

[0061] (Product review registration request, product review registration process) First, product review information in which a customer using a retail store or the like has responded to a product sold at the store is sent from customer terminal 105 to customer analysis AP server 101. Upon receiving the relevant customer information from customer terminal 105, customer analysis AP server 101 executes product review registration processing using a product review registration function, and the results of the product review registration processing are sent to customer analysis DB server 102 and registered in product review table 317.

[0062] (Instruction to acquire information for product value appeal score update processing, acceptance of instruction to acquire information) The customer analysis AP server 101 executes the product value appeal score update function periodically or when a customer registers a product review, and first issues an instruction to the customer analysis DB server 102 to acquire information necessary for the product value appeal score update processing. On the other hand, the customer analysis DB server 102 receives this acquisition instruction and transmits each piece of information stored in its own storage device 301, namely the product value appeal score table 314, the value master 316, and the product review table 317, to the customer analysis AP server 101. Here, "periodically" may mean, for example, once at the end of each month.

[0063] (Product value appeal score update process) The customer analysis AP server 101 obtains the information transmitted in the above steps from the customer analysis DB server 102 and executes product value appeal score update using the product value appeal score update function. The product value appeal score update result is transmitted to the customer analysis DB server 102 and registered in the product value appeal score table 314.

[0064] (Updating the purchase history table) The purchase history of products and customers registered in the product master 312 and customer master 313 is sent from the POS system of the store in question to the customer analysis DB server 102 and stored in the purchase history table 311.

[0065] (Instruction to acquire information for customer value score update processing, acceptance of instruction to acquire information) The customer analysis AP server 101 executes a customer value score update function periodically or when an operation is performed by a predetermined user who manages a retail store, and first issues an instruction to the customer analysis DB server 102 to acquire information necessary for the customer value score update processing. On the other hand, the customer analysis DB server 102 receives this acquisition instruction and transmits each piece of information stored in its own storage device 301, namely the purchase history table 311, the product value appeal score table 314, and the customer value score table 315, to the customer analysis AP server 101. Here, "periodically" may mean, for example, once every weekend.

[0066] (Customer value score update process) The customer analysis AP server 101 obtains the information transmitted in the above steps from the customer analysis DB server 102 and executes a customer value score update using the customer value score update function. The result of the customer value score update is transmitted to the customer analysis DB server 102 and registered in the customer value score table 315.

[0067] (Acceptance of conditions for changes in customer values ​​and presentation of appealing products) The client terminal 103 accepts condition input from the user on the input screen and transmits the input to the customer analysis AP server 101. As shown in the example of FIG. 17, the conditions accepted on this input screen can include the customer's age, gender, the year and month for comparing value scores, and the number of products to be displayed that appeal to specific values.

[0068] (Instruction to acquire information for customer value change / appeal product presentation processing, and acceptance of information acquisition instruction) The customer analysis AP server 101, which has received the above-mentioned input conditions from the client terminal 103, executes the customer value change / appeal product presentation function, and first issues an instruction to the customer analysis DB server 102 to acquire information necessary for processing customer value change / appeal product presentation. On the other hand, the customer analysis DB server 102 receives this acquisition instruction and transmits each piece of information stored in its own storage device 301, namely the product master 312, product value appeal score table 314, and customer value score table 315, to the customer analysis AP server 101.

[0069] (Execution of customer value change / appeal product presentation process, output of customer value change / appeal product presentation information) The customer analysis AP server 101 obtains the information transmitted in the above steps from the customer analysis DB server 102, and executes customer value change / appeal product presentation using the customer value change / appeal product presentation function. The customer analysis AP server 101 transmits the results of this customer value change / appeal product presentation to the client terminal 103 via an output screen. The client terminal 103 displays the results of the customer value change / appeal product presentation on this output screen.

[0070] (Processing Procedure Example 2) FIG. 14 is a flowchart 1400 showing an example of the flow of a process for updating a product value appeal score (hereinafter referred to as a "product value appeal score update process") as a processing procedure example 2 of the customer analysis method according to this embodiment. The product value appeal score update process is specifically executed as a product value appeal score update function possessed by the customer analysis AP server 101. Here, each step of this product value appeal score update process will be described. Details of the calculation of the product value appeal score will be described later using mathematical formulas.

[0071] (Obtaining Latest Update Date) First, in step S1401, the control unit of the customer analysis AP server 101 executes processing to obtain the latest update date from the update date in the product value appeal score table 314 obtained from the customer analysis DB server 102 and stored in the storage device 201. As a result, the latest update date is obtained from the update date in the product value appeal score table 314 obtained from the customer analysis DB server 102 and stored in the storage device 201. When the control unit of the customer analysis AP server 101 completes the processing in step S1401, the process proceeds to step S1402.

[0072] (Obtaining Product Reviews After the Latest Update Date) Subsequently, in step S1402, the control unit of the customer analysis AP server 101 executes processing to extract all items after the update date obtained in step S1401 from the product review table 317 that has already been obtained from the customer analysis DB server 102 and stored in the storage device 201. As a result, all items after the update date obtained in step S1401 are extracted from the product review table 317 that has already been obtained from the customer analysis DB server 102 and stored in the storage device 201. When the control unit of the customer analysis AP server 101 completes the processing in step S1402, the process proceeds to step S1403.

[0073] (Determining Whether There Are Any Product Reviews Since the Latest Update Date) Subsequently, in step S1403, the control unit of the customer analysis AP server 101 executes processing to determine the length of the product review table 317 extracted in step S1402. If it is determined that the length is 1 or greater, that is, that extracted data exists (step S1403: YES), the process proceeds to step S1404. On the other hand, if it is determined that the length is 0, that is, that extracted data does not exist (step S1403: NO), the product value appeal score update processing shown in flowchart 1400 of FIG. 14 is terminated.

[0074] (Calculation of Similarity Between Product Review and Value) Subsequently, in step S1404, the control unit of the customer analysis AP server 101 executes a process of calculating the similarity between each product review in the product review table 317 extracted in step S1402 and each definition of value in the value master 316 already obtained from the customer analysis DB server 102 and stored in the storage device 201. This similarity can be calculated as linguistic similarity, for example, the cosine similarity of the embedding vectors of each sentence. As a result, the similarity between each product review in the product review table 317 extracted in step S1402 and each definition of value in the value master 316 already obtained from the customer analysis DB server 102 and stored in the storage device 201 is calculated. Upon completing the process in step S1404, the control unit of the customer analysis AP server 101 proceeds to step S1405.

[0075] (Normalizing the calculated similarity for each product code) Next, in step S1405, the control unit of the customer analysis AP server 101 executes a process of averaging the numerical sequence of similarity with each value calculated for each product review, for each product code, with each value as the axis, and normalizing the result so that the sum of the values ​​of all the values ​​becomes 1. As a result, the numerical sequence of similarity with each value calculated for each product review, for each product code, is averaged with each value as the axis, and normalized so that the sum of the values ​​of all the values ​​becomes 1. When the process in step S1405 is completed, the control unit of the customer analysis AP server 101 proceeds to step S1406.

[0076] (Weighted Average of Value Appeal Scores) Subsequently, in step S1406, the control unit of the customer analysis AP server 101 acquires, for each product code, from the product value appeal score table 314 already obtained from the customer analysis DB server 102 and stored in the storage device 201, those whose update dates fall within the target period, and then performs a process of weighting the value calculated in step S1405 and the acquired value. Note that this weighted averaging method will be described later using a formula. As a result, for each product code, from the product value appeal score table 314 already obtained from the customer analysis DB server 102 and stored in the storage device 201, those whose update dates fall within the target period are acquired, and then the value calculated in step S1405 and the acquired value are weighted averaged. Upon completing the process in step S1406, the control unit of the customer analysis AP server 101 proceeds to step S1407.

[0077] (The weighted average value is registered as the current product value appeal score) Subsequently, in step S1407, the control unit of the customer analysis AP server 101 executes processing to register the value appeal score calculated in step S1406 in the product value appeal score table 314. As a result, the value appeal score calculated in step S1406 is registered in the product value appeal score table 314. When the processing in step S1407 is completed, the control unit of the customer analysis AP server 101 ends the product value appeal score update processing shown in flowchart 1400 of FIG.

[0078] (Processing Procedure Example 3) FIG. 15 is a flowchart 1500 showing an example of the flow of processing for updating a customer value score (hereinafter referred to as "customer value score update processing") as processing procedure example 3 of the customer analysis method according to this embodiment. Specifically, the customer value score update processing is executed as a customer value score update function of the customer analysis AP server 101. Here, each step of this customer value score update processing will be described. Details of calculation of the customer value score will be described later using mathematical expressions.

[0079] (Obtaining Latest Update Date) First, in step S1501, the control unit of the customer analysis AP server 101 executes processing to obtain the latest update date from the update date in the customer value score table 315 that has already been obtained from the customer analysis DB server 102 and stored in the storage device 201. As a result, the latest update date is obtained from the update date in the customer value score table 315 that has already been obtained from the customer analysis DB server 102 and stored in the storage device 201. When the control unit of the customer analysis AP server 101 completes the processing in step S1501, the process proceeds to step S1502.

[0080] (Obtaining purchase history since latest update date) Subsequently, in step S1502, the control unit of the customer analysis AP server 101 executes processing to extract all items since the update date obtained in step S1501 from the purchase history table 311 already obtained from the customer analysis DB server 102 and stored in the storage device 201. As a result, all items since the update date obtained in step S1501 are extracted from the purchase history table 311 already obtained from the customer analysis DB server 102 and stored in the storage device 201. When the control unit of the customer analysis AP server 101 completes the processing in step S1502, it proceeds to step S1503.

[0081] (Determining Whether Purchase History Exists Since the Latest Update Date) Subsequently, in step S1503, the control unit of the customer analysis AP server 101 executes processing to determine the length of the purchase history table 311 extracted in step S1502. If it is determined that the length is 1 or greater, that is, that extracted data exists (step S1503: YES), the process proceeds to step S1504. On the other hand, if it is determined that the length is 0, that is, that extracted data does not exist (step S1503: NO), the customer value score update processing shown in flowchart 1500 of FIG. 15 is terminated.

[0082] (Obtaining Value Appeal Score of Product in Purchase History) Subsequently, in step S1504, the control unit of the customer analysis AP server 101 executes a process to obtain the latest value appeal score of the relevant product code from the product value appeal score table 314 already obtained from the customer analysis DB server 102 and stored in the storage device 201, using each product code present in the extracted purchase history table 311 as a key. As a result, the latest value appeal score of the relevant product code is obtained from the product value appeal score table 314 already obtained from the customer analysis DB server 102 and stored in the storage device 201, using each product code present in the extracted purchase history table 311 as a key. When the control unit of the customer analysis AP server 101 completes the process in step S1504, the process proceeds to step S1505.

[0083] (Totaling Sales Amounts by Customer Segment and Product Code) Next, in step S1505, the control unit of customer analysis AP server 101 uses the customer code in the extracted purchase history table 311 as a key to obtain the age and gender values ​​from customer master 313 already obtained from customer analysis DB server 102 and stored in storage device 201, and executes a process to totalize sales amounts by customer segment based on customer age and gender and by product code. As a result, using the customer code in the extracted purchase history table 311 as a key, the age and gender values ​​are obtained from customer master 313 already obtained from customer analysis DB server 102 and stored in storage device 201, and sales amounts are totaled by customer segment based on customer age and gender and by product code. When the control unit of customer analysis AP server 101 completes the process in step S1505, it proceeds to step S1506.

[0084] (Calculating customer values ​​from purchase history) Next, in step S1506, the control unit of the customer analysis AP server 101 multiplies the sales amount for each product code by the value appeal score of the corresponding product code, sums the calculated results for each customer segment, and executes a process of normalizing the results so that the sum for each value becomes 1. As a result, the sales amount for each product code is multiplied by the value appeal score of the corresponding product code, and the calculated results are summed for each customer segment and normalized so that the sum for each value becomes 1. When the control unit of the customer analysis AP server 101 completes the process in step S1506, it proceeds to step S1507.

[0085] (Weighted Average of Customer Value Scores) Subsequently, in step S1507, the control unit of the customer analysis AP server 101 acquires, from the customer value score table 315 already obtained from the customer analysis DB server 102 and stored in the storage device 201, those value scores for each age group and gender whose update date falls within the target period, and executes a process of weighted averaging the value calculated in step S1506 and the acquired value score for the relevant age group and gender. As a result, from the customer value score table 315 already obtained from the customer analysis DB server 102 and stored in the storage device 201, those value scores for each age group and gender whose update date falls within the target period are acquired, and then the value calculated in step S1506 and the acquired value score for the relevant age group and gender are weighted averaged. Upon completing the process in step S1507, the control unit of the customer analysis AP server 101 proceeds to step S1508.

[0086] (Weighted average value is registered as the current customer value score) Subsequently, in step S1508, the control unit of the customer analysis AP server 101 executes processing to register the customer value score calculated in step S1507 in the customer value score table 315. As a result, the customer value score calculated in step S1507 is registered in the customer value score table 315. Upon completing the processing in step S1508, the control unit of the customer analysis AP server 101 ends the customer value score update processing shown in flowchart 1500 of FIG.

[0087] (Processing Procedure Example 4) Fig. 16 is a flowchart 1600 showing an example of the flow of a process for presenting customer value change and appealing products (hereinafter referred to as "customer value change and appealing product presentation process") as processing procedure example 4 of the customer analysis method according to this embodiment. The customer value change and appealing product presentation process is specifically executed as a customer value change and appealing product presentation function possessed by the customer analysis AP server 101. Here, each step of this customer value change and appealing product presentation process will be described.

[0088] (Receiving an Analysis Request) First, in step S1601, the control unit of the customer analysis AP server 101 executes processing to receive from the client terminal 103 an analysis request including the values ​​of the customer's age group, gender, year and month for which the value scores are compared, and the number of products displayed that appeal to specific values, which have been input into the input screen of the client terminal 103. As a result, the analysis request including the values ​​of the customer's age group, gender, year and month for which the value scores are compared, and the number of products displayed that appeal to specific values, which have been input into the input screen of the client terminal 103, is received from the client terminal 103. When the processing in step S1601 is completed, the control unit of the customer analysis AP server 101 proceeds to step S1602.

[0089] (Extracting Customer Value Score) Subsequently, in step S1602, the control unit of the customer analysis AP server 101 executes processing to extract the update date and customer value score of the customer value score table 315 using the age and gender of the customer indicated in the analysis request as keys. As a result, the update date and customer value score of the customer value score table 315 are extracted using the age and gender of the customer indicated in the analysis request as keys. When the control unit of the customer analysis AP server 101 completes the processing in step S1602, it proceeds to step S1603.

[0090] (Customer Value Score Aggregation) Furthermore, in step S1603, the control unit of the customer analysis AP server 101 executes a process of dividing the extracted customer value score table 315 by each year and month of the update date described above, averaging the customer value scores in the customer value score table belonging to each year and month, and setting this value as the customer value score for the current year and month. As a result, the extracted customer value score table 315 is divided by each year and month of the update date described above, averaging the customer value scores in the customer value score table belonging to each year and month, and setting this value as the customer value score for the current year and month. When the control unit of the customer analysis AP server 101 completes the process in step S1603, the process proceeds to step S1604.

[0091] (Extraction of Noteworthy Customer Values) Next, in step S1604, the control unit of the customer analysis AP server 101 executes a process to acquire, from the customer value scores for each month tallied in the above steps, the customer value score values ​​for the "from" and "to" months for which values ​​are compared, as indicated by the analysis request, and the customer value score values ​​for the month corresponding to the time of the analysis request, and to acquire the values ​​for which the values ​​for the "from" to "to" periods have increased the most and decreased the most, and the values ​​for the month corresponding to the time of the analysis request, respectively. As a result, from the customer value scores for each month tallied in the above steps, the customer value score values ​​for the "from" and "to" months for which values ​​are compared, as indicated by the analysis request, and the customer value score values ​​for the month corresponding to the time of the analysis request, and to acquire the values ​​for which the values ​​for the "from" to "to" periods have increased the most and decreased the most, and the values ​​for the month corresponding to the time of the analysis request, respectively. When the control unit of the customer analysis AP server 101 completes the process in step S1604, the process proceeds to step S1605.

[0092] (Obtaining the Latest Value Appeal Score for Each Product) Subsequently, in step S1605, the control unit of the customer analysis AP server 101 executes processing to obtain the value appeal score for each product code with the latest update date from the product value appeal score table 314. As a result, the value appeal score for each product code with the latest update date is obtained from the product value appeal score table 314. When the processing in step S1605 is completed, the control unit of the customer analysis AP server 101 proceeds to step S1606.

[0093] (Acquisition of product names that appeal to noteworthy customer values) Subsequently, in step S1606, the control unit of the customer analysis AP server 101 executes a process of acquiring, for each value extracted in step S1604, the product codes with the highest value appeal scores acquired in step S1605, in descending order, for the number of displayed products indicated by the analysis request, and acquiring the product names of the acquired product codes from the product master 312. As a result, for each value extracted in step S1604, the product codes with the highest value appeal scores acquired in step S1605 are acquired, in descending order, for the number of displayed products indicated by the analysis request, and the product names of the acquired product codes are acquired from the product master 312. When the process in step S1606 is completed, the control unit of the customer analysis AP server 101 proceeds to step S1607.

[0094] (Generation of Screen Data) Next, in step S1607, the control unit of the customer analysis AP server 101 executes processing to generate screen data for an output screen, as illustrated in FIG. 18 . In this example, the changes in customer values ​​are displayed using the age and gender of the customer who was the subject of the customer analysis as keys, and the values ​​that were most "on an upward trend" and "on a downward trend" in the comparison years and months of "from when" and "to when" that were the subject of the customer analysis, the values ​​that were maximum and minimum "matching the current situation" and "not matching the current situation" in the year and month at the time of the analysis request, and the products that most appeal to each value are displayed for the number of display products indicated in the analysis request. This generates screen data for the output screen, as illustrated in FIG. 18 . Next, the control unit of the customer analysis AP server 101 executes processing to transmit the generated screen data to the client terminal 103 and display it on the output screen. This causes the generated screen data to be transmitted to the client terminal 103 and displayed on the output screen of the client terminal 103. A user viewing the output screen on the client terminal 103 can, for example, at a glance recognize products that appeal to values ​​that are on the rise or fall from the screen contents of Fig. 18, and can select products that appeal to the rising customer values ​​and increase sales. When the control unit of the customer analysis AP server 101 completes the processing in step S1607, it ends the customer value change / appeal product presentation processing shown in flowchart 1600 of Fig. 16.

[0095] 17 is a diagram specifically illustrating an example of an input screen displayed by the client terminal 103. The input screen in FIG. 17 allows input of conditions for the customer value change / promotion product presentation process. The conditions can include items such as customer segment (age, gender), comparison date (from and to), and number of displayed products.

[0096] 18 is a diagram specifically illustrating an example of an output screen displayed by the client terminal 103. In FIG. 18, the ten human values ​​classified by Shalom H. Schwartz are used as examples of values. The output screen of FIG. 18 displays the time series transition of the customer's value scores for each year and month (changes in the customer's values), as well as values ​​that are rising or falling in the comparison year and month, values ​​that are in line with or incompatible with the current situation at the time of the analysis request, and products that appeal to each value, for the number of displayed products.

[0097] FIG. 19 is a diagram showing another example of output information displayed on the output screen. Note that in FIG. 19, the ten human values ​​classified by Shalom H. Schwartz are used as examples of values. The customer analysis AP server 101 can also generate a line graph of changes in values ​​by age group as the output screen, as shown in FIG. 19. The line graph shown in FIG. 19 indicates that, for example, between 2018 and 2023, the values ​​of hedonism, universalism, and philanthropy have particularly increased, while the values ​​of tradition and power have particularly decreased, among teenagers. This data can be used as a reference for considering future store initiatives to appeal to teenagers.

[0098] Next, the calculation of the value appeal score of a product will be described.

[0099] The value appeal score of a product is calculated, for example, using the following (Equation 1), (Equation 2), (Equation 3), and (Equation 4).

[0100]

[0101]

[0102]

[0103]

[0104] The right-hand term in (Equation 1), Values, is a list in which each dimension is a definition sentence for each value, and review is the nth review sentence for product i at time t. Similarity is a function that calculates the linguistic similarity between Values ​​and review, and the linguistic similarity can be calculated by vectorizing each dimension of Values ​​and the review sentence using an embedding model and then calculating their cosine similarity. The result of this Similarity calculation is added up for the number N of review sentences for product i, and the result becomes the left-hand term, P' feedback.

[0105] In (Equation 2), P′ feedback is normalized so that the sum of the values ​​is 1. This vector can be interpreted as the value appeal score at time t calculated from customer product reviews (feedback) for product i.

[0106] P on the right hand side of (Equation 3) is the value appeal score of product i at time a, and τ is the number of days from time a to time t. K is any positive value, and T is the length of the period covered by past product value appeal scores used to calculate the product value appeal score. The right hand side of (Equation 3) allows the product value appeal score P from time t-T to time t-1 and the product value appeal score P feedback calculated by (Equation 2) to be added together, with the closer the score calculation time is to the current time, the heavier the weighting.

[0107] Equation 4 normalizes the vector calculated by equation 3 so that the sum becomes 1.

[0108] In this way, the product value appeal score P of the product i at time t can be calculated.

[0109] Next, calculation of a customer's value score will be described.

[0110] The customer's value score is calculated, for example, using the following (Equation 5), (Equation 6), (Equation 7), and (Equation 8).

[0111]

[0112]

[0113]

[0114]

[0115] Sales, the right-hand term in (Equation 5), is the total sales amount for customer segment j's purchase of product i at time t. Customer segment j refers to a segment classified by customer age and gender. P is the value appeal score for product i at the time closest to time t, and the product of Sales and P is multiplied and added up for the M products purchased by customer segment j to obtain C' feedback, the left-hand term.

[0116] In (Equation 6), C′ feedback is normalized so that the sum of the values ​​is 1. This vector can be interpreted as the value score of customer segment j calculated based on the customer's purchasing history (feedback).

[0117] C on the right-hand term of (Equation 7) is the value score of customer segment j at time a, and τ is the number of days from time a to time t. K is any positive value, and T is the length of the period covered by past customer value scores used to calculate the customer value score. Using the right-hand term of (Equation 7), the customer's value score C from time t-T to time t-1 and the value score C feedback calculated by (Equation 6) can be added together, with the closer the score calculation time is to the current time, the heavier the weighting.

[0118] Equation (8) normalizes the vector calculated by equation (7) so that the sum of the values ​​becomes 1.

[0119] In this way, the value score C of customer segment j at time t can be calculated.

[0120] The above-described embodiment of the present invention can be summarized as follows.

[0121] (1) As described above, the customer analysis system 100 according to this embodiment includes a customer analysis AP server 101, which is a computer having at least an arithmetic device (CPU 302) and a storage device 301. The storage device 301 stores customer feature scores that associate customers with scores of their features, sales object feature appeal scores that associate sales objects with scores of the sales objects' appeal to the customer features, and customer feedback from customers regarding the sales objects. The arithmetic device (CPU 302) uses the customer feedback to update the customer feature scores and sales object feature appeal scores, and outputs customer features whose scores are increasing and sales objects that appeal to those features.

[0122] This makes it possible to select sales targets that appeal to the characteristics of customers whose scores are increasing and increase sales.

[0123] (2) Furthermore, the customer feedback includes the customer's purchasing history for the sales object, and the calculation device (CPU 302) updates the customer characteristic score using a value obtained by multiplying the customer's purchasing history for the sales object by the sales object characteristic appeal score. In this way, it is possible to capture changes in customer characteristics in a timely manner using indirect but frequent feedback from the customer regarding the sales object.

[0124] (3) In addition, the customer feedback includes customer reviews of the sales object, and the storage device 301 further stores definitions of customer characteristics. The calculation device (CPU 302) associates a characteristic score with the review based on the customer review of the sales object and the definition of the customer's characteristic, and uses the characteristic score to update the customer characteristic score and the sales object characteristic appeal score. In this way, it is possible to more accurately capture changes in the characteristics appealing to the sales object and the customer characteristics using infrequent but direct feedback from customers about the sales object.

[0125] (4) The storage device 301 further stores information on customer attributes and sales targets, and the calculation device (CPU 302) calculates initial values ​​for the customer feature score and the sales target feature appeal score based on the customer attribute information, the sales target information, and the definition of the customer's features. This allows the initial values ​​to be set more accurately than if they were set randomly.

[0126] (5) As another example, the customer characteristics are the customer's values, i.e., the values ​​of a person. For this reason, the 10 human values ​​classified by Shalom H. Schwartz can be used as the customer characteristics.

[0127] The best mode for carrying out the present invention has been specifically described above, but the present invention is not limited to this and can be modified in various ways without departing from the spirit of the present invention.

[0128] For example, in the above embodiment, the index value was calculated based on the ten human values ​​classified by Shalom H. Schwartz, but any index can be used as long as it can define the characteristics of a person.

[0129] According to this embodiment, it is possible to ascertain whether a customer's values ​​are rising or falling, and to select products that appeal to the rising values ​​of customers and increase sales. Furthermore, in product demand forecasting for a retail store, for example, by analyzing customers and products at the value level beyond the unit of product or product category, it becomes possible to make product demand forecasts that cannot be made on a product or product category basis.

[0130] The present invention is not limited to the above-described embodiment, and can be implemented using any components without departing from the spirit of the present invention.

[0131] The above-described embodiments, examples, and modifications are merely examples, and the present invention is not limited to these details as long as the features of the invention are not impaired. Furthermore, although various embodiments, examples, and modifications have been described above, the present invention is not limited to these details. Other aspects that can be considered within the scope of the technical idea of ​​the present invention are also included within the scope of the present invention.

[0132] In the above figures, the control lines and information lines shown are those that are considered necessary for explanation, and do not necessarily show all the control lines and information lines that are actually implemented. For example, it can be considered that almost all components are actually connected to each other.

[0133] Furthermore, the layout of each functional unit of the customer analysis system 100 described above is merely an example. The layout of each functional unit can be changed to an optimal layout in terms of the performance, processing efficiency, communication efficiency, etc. of the hardware and software included in the customer analysis system 100.

[0134] 100 Customer analysis system 101 Customer analysis AP server 102 Customer analysis DB server 103 Client terminal (other terminal) 104 Network 201 Storage device 202 CPU 203 Memory 206 Communication device 210 Program 301 Storage device 302 CPU 303 Memory 306 Communication device 310 Program 311 Purchase record table 312 Product master 313 Customer master 314 Product value appeal table 315 Customer value table 316 Value master 317 Product review table 401 Storage device 402 CPU 403 Memory 404 Input device 405 Output device 406 Communication device 410 Program 450 Input screen 451 Output screen 501 Storage device 502 CPU 503 Memory 504 Input device 505 Output device 506 Communication device 510 Program 550 Input screen

Claims

1. A customer analysis system comprising a computer having at least an arithmetic device and a storage device, wherein the storage device stores customer feature scores that associate customers with scores of their features, sales object feature appeal scores that associate sales objects with scores of how the sales objects appeal to the customer's features, and customer feedback from the customers regarding the sales objects, and the arithmetic device uses the customer feedback to update the customer feature scores and the sales object feature appeal scores, and outputs customer features whose scores have increased and the sales objects that appeal to those features.

2. The customer analysis system of claim 1, wherein the customer feedback includes the customer's purchasing history for the sales object, and the computing device updates the customer feature score using a value obtained by multiplying the customer's purchasing history for the sales object by the sales object feature appeal score.

3. The customer analysis system of claim 1, wherein the customer feedback includes customer reviews of the sales object, the storage device further stores definitions of the customer's characteristics, and the computing device associates a characteristic score with the review based on the customer's review of the sales object and the definition of the customer's characteristics, and uses the score to update the customer characteristic score and the sales object characteristic appeal score.

4. The customer analysis system described in claim 3, wherein the storage device further stores information on customer attributes and sales targets, and the calculation device calculates initial values ​​of a customer feature score and a sales target feature appeal score based on the information on customer attributes and sales targets and a definition of the customer's characteristics.

5. The customer analysis system according to claim 1, wherein the customer characteristics are the customer's values.

6. A customer analysis method performed using a computer having at least an arithmetic device and a storage device, wherein the storage device stores customer feature scores that associate customers with scores of their features, sales object feature appeal scores that associate sales objects with scores of how the sales objects appeal to the customer's features, and customer feedback from the customers regarding the sales objects, and the arithmetic device uses the customer feedback to update the customer feature scores and the sales object feature appeal scores, and outputs the customer features whose scores have increased and the sales objects that appeal to those features.

7. A computer program having at least an arithmetic device and a storage device, wherein the storage device stores: customer feature scores that associate customers with scores of their features; sales object feature appeal scores that associate sales objects with scores that the sales objects appeal to the customer's features; and customer feedback from the customers regarding the sales objects; the computer program causes the arithmetic device to execute a process of updating the customer feature scores and the sales object feature appeal scores using the customer feedback, and outputting the customer features whose scores have increased and the sales objects that appeal to those features.

Images