System, program and information processing method for using message of customer

Abstract

To provide a system, program and information processing method for using data related to a customer by easily associating complaints and impressions acquired from the customer with the data.SOLUTION: In a system including an analysis server 1, a customer terminal 4, and a business terminal 5 to uses a message of a customer, the analysis server 1 makes the customer select the necessity / unnecessity of a reply to a message to be transmitted from the customer terminal 4, a vendor replies the message by using the business terminal 5 if the necessity of the reply is selected, while the message transmitted by the customer terminal 4 is stored in a database 12, a summary is created by using artificial intelligence learned by using a learning document on the basis of a plurality of stored messages, and the business terminal 5 displays the summary if the unnecessity of the replay is selected, and the analysis server 1 specifies the customer transmitting a corresponding message and the business terminal 5 displays customer information of the specified customer if a portion of the summary displayed in the business terminal 5 is selected.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a system, a program, and an information processing method for utilizing customer messages. [Background technology]

[0002] BACKGROUND ART Conventionally, companies have been conducting VOC (Voice of Customer) analysis, which involves analyzing complaints and opinions received from customers, with the aim of contributing to improving customer satisfaction (see, for example, Patent Document 1).

[0003] The invention described in Patent Document 1 proposes taking measures to improve business operations by obtaining evaluation scores (e.g., a five-point scale from +2 to -2) and text data (e.g., "good design") from customers, analyzing the text data using predetermined processing, and further weighting the results of the text data analysis based on the evaluation scores. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-109148 Summary of the Invention [Problem to be solved by the invention]

[0005] Many companies hold data on their customers. By linking this data with complaints received from customers, it is possible to identify which complaints come from which customers. However, when a company has a large number of customers, or a large number of complaints and feedback received from customers, linking the complaints with customer data can be extremely cumbersome.

[0006] In view of the above, an object of the present invention is to provide a system for easily associating and utilizing complaints and opinions obtained from customers with data relating to those customers. [Means for solving the problem]

[0007] The above object of the present invention can be achieved by the following means: Note that the parentheses indicate reference symbols of embodiments to be described later, but the present invention is not limited to these.

[0008] The system according to claim 1 comprises: A system for utilizing customer messages, comprising a server (e.g., analysis server 1 shown in FIG. 1), a customer terminal (e.g., customer terminal 4 shown in FIG. 1), and a business terminal (e.g., business terminal 5 shown in FIG. 1), The customer terminal (e.g., customer terminal 4 shown in FIG. 1) prompts the customer to select whether or not to reply to the message that the customer terminal (e.g., customer terminal 4 shown in FIG. 1) is about to send (e.g., step S402 shown in FIG. 3), When a reply is required is selected on the customer terminal (for example, the customer terminal 4 shown in FIG. 1), the agent uses the business terminal (for example, the business terminal 5 shown in FIG. 1) to reply to the message (for example, the message input in the message input section 41a shown in FIG. 5(b)), while When the customer terminal (for example, the customer terminal 4 shown in FIG. 1) selects that no reply is required, the message (for example, the message input in the message input section 42a shown in FIG. 5(c)) sent by the customer terminal (for example, the customer terminal 4 shown in FIG. 1) is stored in a predetermined storage means (for example, the message information tbl_B shown in FIG. 2(b)) (for example, step S101 shown in FIG. 3), Based on the plurality of stored messages (e.g., message information tbl_B shown in FIG. 2(b)), the server (e.g., analysis server 1 shown in FIG. 1) creates a summary (e.g., summary 50c shown in FIG. 6(a)) using artificial intelligence trained using training documents (e.g., step S112 shown in FIG. 4), The business terminal (for example, the business terminal 5 shown in FIG. 1) displays the summary (for example, the summary 50c shown in FIG. 6(a)) (for example, see step S113 shown in FIG. 4), When a portion of the summary (e.g., the portion 51c1 of the summary shown in FIG. 6(b)) displayed on the business terminal (e.g., the business terminal 5 shown in FIG. 1) is selected, the server (e.g., the analysis server 1 shown in FIG. 1) identifies the customer who sent the message corresponding to the portion of the summary (e.g., the portion 51c1 of the summary shown in FIG. 6(b)) (e.g., step S115 shown in FIG. 4), The business terminal (for example, business terminal 5 shown in FIG. 1) displays customer information of the identified customer (for example, customer information 51d shown in FIG. 6(b)) (for example, step S116 shown in FIG. 4).

[0009] The program according to claim 2 comprises: A program for utilizing customer messages, A computer (for example, the analysis server 1 shown in FIG. 1) that is communicably connected to a customer terminal (for example, the customer terminal 4 shown in FIG. 1) and a business terminal (for example, the business terminal 5 shown in FIG. 1) a step in which the customer terminal (e.g., the customer terminal 4 shown in FIG. 1) prompts the customer to select whether or not a reply is required to the message to be sent from the customer terminal (e.g., the customer terminal 4 shown in FIG. 1) (e.g., step S402 shown in FIG. 3); if "return required" is selected on the customer terminal (e.g., the customer terminal 4 shown in FIG. 1), the business terminal (e.g., the business terminal 5 shown in FIG. 1) is used to prompt the agent to reply to the message (e.g., the message input in the message input section 41a shown in FIG. 5(b)); and if "return not required" is selected on the customer terminal (e.g., the customer terminal 4 shown in FIG. 1), the business terminal (e.g., the business terminal 5 shown in FIG. 1) is used to create a summary (e.g., the summary 50c shown in FIG. 6(a)) using artificial intelligence trained using a learning document based on the message (e.g., the message input in the message input section 42a shown in FIG. 5(c)) to be sent from the customer terminal (e.g., the customer terminal 4 shown in FIG. 1) and stored in a predetermined storage means (e.g., the message information tbl_B shown in FIG. 2(b)) (e.g., step S112 shown in FIG. 4); a step of displaying the summary (for example, summary 50c shown in FIG. 6(a)) on the business terminal (for example, business terminal 5 shown in FIG. 1) (for example, step S113 shown in FIG. 4); When a portion of the abstract (e.g., the portion 51c1 of the abstract shown in FIG. 6(b)) displayed on the business terminal (e.g., the business terminal 5 shown in FIG. 1) is selected, a step of identifying the customer who sent the message corresponding to the portion of the abstract (e.g., the portion 51c1 of the abstract shown in FIG. 6(b)) (e.g., step S115 shown in FIG. 4); The method is characterized by executing a step (e.g., step S116 shown in FIG. 4) of displaying customer information of the identified customer (e.g., customer information 51d shown in FIG. 6(b)) on the business terminal (e.g., business terminal 5 shown in FIG. 1).

[0010] The program according to claim 3 is the program according to claim 2, The customer information (e.g., customer information 51d shown in FIG. 6(b)) displayed on the business terminal (e.g., business terminal 5 shown in FIG. 1) is characterized by including information that serves as an index for evaluating the customer loyalty (e.g., "number of visits" included in customer information 51d shown in FIG. 6(b)).

[0011] The program according to claim 4 is the program according to claim 2, The computer (e.g., analysis server 1 shown in FIG. 1) provides a web service that displays analysis results of predetermined numerical data (e.g., analysis results 50b of numerical data shown in FIG. 6(a)) on the business terminal (e.g., business terminal 5 shown in FIG. 1), When a portion of the summary (e.g., portion 51c1 of the summary shown in FIG. 6(b)) displayed on the business terminal (e.g., business terminal 5 shown in FIG. 1) is selected, the method further executes a step (e.g., step S116 shown in FIG. 4) of displaying on the business terminal (e.g., business terminal 5 shown in FIG. 1) the analysis results (e.g., analysis results 51b of the analysis of the numerical data shown in FIG. 6(b)) obtained by analyzing the numerical data regarding the customer who sent the message corresponding to the portion of the summary (e.g., portion 51c1 of the summary shown in FIG. 6(b)).

[0012] The program according to claim 5 is the program according to claim 2, Furthermore, the program executes a step (for example, step S104 shown in FIG. 3) of granting a benefit to a customer when the number of times (for example, message transmission count tbl_Cc shown in FIG. 2(c)) that the same customer has sent the message (for example, a message input in message input section 41a shown in FIG. 5(b) and / or a message input in message input section 42a shown in FIG. 5(c)) reaches a predetermined number, The predetermined number of times or how many times the message needs to be sent to reach the predetermined number of times is not notified to the customer.

[0013] The method according to claim 6, 1. A method for utilizing customer messages, comprising: a step of prompting a customer terminal (e.g., the customer terminal 4 shown in FIG. 1) to select whether or not a reply is required to a message to be sent from the customer terminal (e.g., the customer terminal 4 shown in FIG. 1) (e.g., step S402 shown in FIG. 3); if "return required" is selected on the customer terminal (e.g., the customer terminal 4 shown in FIG. 1), prompting the agent to reply to the message (e.g., a message input in the message input section 41a shown in FIG. 5(b)) using the business terminal (e.g., the business terminal 5 shown in FIG. 1); and, if "return not required" is selected on the customer terminal (e.g., the customer terminal 4 shown in FIG. 1), creating a summary (e.g., the summary 50c shown in FIG. 6(a)) using artificial intelligence based on the message (e.g., the message input in the message input section 42a shown in FIG. 5(c)) to be sent from the customer terminal (e.g., the customer terminal 4 shown in FIG. 1) that is stored in a predetermined storage means (e.g., the message information tbl_B shown in FIG. 2(b)) (e.g., step S112 shown in FIG. 4); a step (for example, step S113 shown in FIG. 4) of displaying the summary (for example, summary 50c shown in FIG. 6(a)) on a business terminal (for example, business terminal 5 shown in FIG. 1); When a portion of the abstract (e.g., the portion 51c1 of the abstract shown in FIG. 6(b)) displayed on the business terminal (e.g., the business terminal 5 shown in FIG. 1) is selected, a step of identifying the customer who sent the message corresponding to the portion of the abstract (e.g., the portion 51c1 of the abstract shown in FIG. 6(b)) (e.g., step S115 shown in FIG. 4); The method is characterized by comprising a step (for example, step S116 shown in FIG. 4) of displaying customer information of the identified customer (for example, customer information 51d shown in FIG. 6(b)) on the business terminal (for example, business terminal 5 shown in FIG. 1). [Effects of the Invention]

[0014] Next, the effects of the present invention will be described with reference to the drawings. Note that the reference symbols in parentheses are those of the embodiments described below, but the present invention is not limited to these.

[0015] According to the inventions of claims 1, 2, and 6, when a server (e.g., the analysis server 1 shown in FIG. 1) displays messages (e.g., message tbl_Bb in message information tbl_B shown in FIG. 2(b)) sent by customers that do not require a reply and that are stored in a predetermined storage means (e.g., message information tbl_B shown in FIG. 2(b)), the server can display a summary (e.g., summary 50c shown in FIG. 6(a)) of the stored messages (e.g., message tbl_Bb in message information tbl_B shown in FIG. 2(b)). When a portion of the summary (e.g., summary 50c shown in FIG. 6(a)) is selected (e.g., portion 51c1 of the summary shown in FIG. 6(b)), customer information (e.g., customer information 51d shown in FIG. 6(b)) of the customer who sent the message corresponding to that portion is displayed on the business terminal (e.g., the business terminal 5 shown in FIG. 1).

[0016] Therefore, the stored messages (for example, message tbl_Bb in message information tbl_B shown in FIG. 2(b)) are not displayed themselves, but rather a summary (for example, summary 50c shown in FIG. 6(a)) is displayed, thereby reducing the time required for the trader to check messages received from customers. Moreover, by selecting a portion of the summary (for example, portion 51c1 of the summary shown in FIG. 6(b)), the customer who sent the corresponding message is identified and displayed, allowing the trader to easily find information about the customer who sent the message related to the summary.

[0017] Therefore, according to the present invention, complaints and opinions obtained from customers can be easily associated with data relating to those customers and used.

[0018] According to the invention of claim 3, by displaying an indicator for evaluating customer loyalty (for example, "number of visits" included in customer information 51d shown in Figure 6(b)), it is easy to emphasize messages from customers who can be evaluated as having high customer loyalty.

[0019] According to the invention of claim 4, the analysis results (e.g., analysis results 51b of the numerical data shown in FIG. 6(b)) of the analysis of the numerical data related to the customers who sent messages related to the selected summary (e.g., part of the summary 51c1 shown in FIG. 6(b)) are displayed, so that it is easy to analyze the customers who sent messages related to the summary.

[0020] According to the invention of claim 5, it is possible to reduce the possibility that a customer will frequently send messages (for example, messages input into the message input section 41a shown in Figure 5(b)) in order to receive special benefits. [Brief explanation of the drawings]

[0021] [Figure 1] 1 is a block diagram illustrating one embodiment of a system for utilizing customer messages according to the present invention. [Figure 2] (a) is a diagram explaining the customer information table stored in the customer server, (b) is a diagram explaining the message information table stored in the analysis server, and (c) is a diagram explaining the customer-user information table stored in the analysis server. [Figure 3] FIG. 10 is a flowchart illustrating an example of use of the system according to the embodiment. [Figure 4] FIG. 10 is a flowchart illustrating an example of use of the system according to the embodiment. [Figure 5] These are diagrams showing application screens displayed on a user terminal, where (a) is a screen asking whether a reply to the message being sent is required, (b) is a screen displayed when a reply to the message is required, and (c) is a screen displayed when a reply to the message is not required. [Figure 6] 10A and 10B are diagrams showing screens displayed on a business terminal, where (a) shows the screen on which the results of an analysis of numerical data and a summary of a message are displayed, and (b) shows the screen when a portion of the message summary is selected. DETAILED DESCRIPTION OF THE INVENTION

[0022] Hereinafter, an embodiment of a system for using customer messages according to the present invention will be specifically described with reference to the drawings. In the following description, when indicating the up, down, left, and right directions, it refers to the up, down, left, and right as viewed from the front shown in the drawing.

[0023] <Overview of the System for Using Customer Messages> In the present embodiment, as shown in FIG. 1, an analysis server 1, an SNS server 2, a customer server 3, a customer terminal 4, and a business terminal 5 are connected via a network N. Hereinafter, each component will be described.

[0024] <Description of the Analysis Server> As shown in FIG. 1, the analysis server 1 is composed of a data processing unit 10 that processes various data, a communication unit 11 that can be connected to the network N by communication means such as wireless LAN, wired LAN, and dial-up, and a database 12. The analysis server 1 provides a so-called BI tool. More specifically, it provides a web service that analyzes numerical data collected by users, visualizes the analysis results by graphing them, and displays them on the business terminal 5 (see FIG. 1) (see FIG. 6).

[0025] By the way, in the database 12 of the analysis server 1 shown in FIG. 1, message information tbl_B shown in FIG. 2(b) and customer-user information tbl_C shown in FIG. 2(c) are stored. The message information tbl_B shown in FIG. 2(b) is composed of a message ID tbl_Ba that uniquely identifies each record in the table, a message tbl_Bb that stores the message received from the customer terminal 4 (see FIG. 1), and a user ID tbl_Bc that indicates the customer who sent the message.

[0026] On the other hand, the customer-user information tbl_C (see FIG. 2(c)) will be described later.

[0027] <Description of the SNS Server> As shown in Fig. 1, the SNS server 2 has the same configuration as the analysis server 1. This SNS server 2 provides so-called social networking services. More specifically, it provides services such as LINE (registered trademark) that enable communication between users or between users and companies.

[0028] The database 22 of the SNS server 2 shown in FIG. 1 stores a table (not shown) that stores user IDs for the SNS server 2 to identify users who use its services.

[0029] <Customer server description> As shown in Figure 1, the customer server 3 has the same configuration as the analysis server 1. The customer server 3 stores in its database 32 a customer information tbl_A (see Figure 2(a)) that stores customer information held by a specific company.

[0030] This customer information tbl_A (see Figure 2(a)) consists of a customer ID tbl_Aa that uniquely identifies each record in the table, the customer's name tbl_Ab, telephone number tbl_Ac, address tbl_Ad, and number of visits tbl_Ae that indicates how many times the customer has visited a commercial facility operated by the company.

[0031] The analysis server 1 (see FIG. 1) stores the customer ID tbl_Aa (see FIG. 2(a)) in association with the user ID of the SNS server 2 in the customer-user information tbl_C shown in FIG. 2(c). More specifically, the customer ID tbl_Aa (see FIG. 2(a)) is stored in the customer ID tbl_Ca (see FIG. 2(c)), and the user ID (not shown) of the SNS server 2 (see FIG. 1) is stored in the user ID tbl_Cb (see FIG. 2(c)).

[0032] Furthermore, the customer-user information tbl_C shown in FIG. 2(c) stores a message sending count tbl_Cc indicating how many times each customer has sent a message (see the message input field 42a in FIG. 5(c)).

[0033] <Customer terminal description> The customer terminal 4 is composed of a smartphone such as an iPhone (registered trademark), a mobile phone, a desktop computer, a notebook computer, etc. Specifically, as shown in Fig. 1, the customer terminal 4 is composed of a CPU 4a, an input unit 4b that can input predetermined data into the customer terminal 4 using a touch panel, a mouse, a keyboard, etc., a ROM 4c that is a writable flash ROM or the like that stores an application (see Fig. 5) for using the social networking service provided by the SNS server 2 (see Fig. 1) and a program such as a browser, a RAM 4d that functions as a work area, buffer memory, etc., a display unit 4e that can display the screen of the application (see Fig. 5) for using the social networking service provided by the SNS server 2 (see Fig. 1), and a communication unit 4f that can connect to the network N by communication means such as wireless LAN.

[0034] <Description of business terminal> As shown in Fig. 1, the business terminal 5 has the same configuration as the customer terminal 4. However, it differs from the customer terminal 4 in that the display unit 5e displays a screen (see Fig. 6) of a web service provided by the analysis server 1 (see Fig. 1).

[0035] Here, we will briefly explain the operation of the system for utilizing customer messages configured as described above. First, a customer of a company uses a customer terminal 4 (see Figure 1) to send a message to the company (see Figure 5(c)) that does not require a reply from an app (see Figure 5) for using the social networking service provided by the SNS server 2. The sent message is transferred from the SNS server 2 (see Figure 1) to the analysis server 1 (see Figure 1) and stored in the analysis server 1 (see Figure 1).

[0036] Meanwhile, a company employee accesses the analysis server 1 (see FIG. 1) using a business terminal 5 (see FIG. 1), and the results of the analysis server 1's (see FIG. 1) analysis of specific numerical data collected by the company are displayed in the browser (see analysis result 50b of the numerical data analysis in FIG. 6(a)). At this time, the analysis server 1 (see FIG. 1) uses artificial intelligence to create a summary of the message sent by the customer and displays the summary together with the results of the analysis of the numerical data (see summary 50c in FIG. 6(a)). When a portion of the displayed summary is selected (see portion of summary 51c1 in FIG. 6(b)), the analysis server 1 (see FIG. 1) identifies the customer who sent the message corresponding to that portion of the summary (see portion of summary 51c1 in FIG. 6(b)), and displays information about that customer on the business terminal 5 (see FIG. 1) (see customer information 51d in FIG. 6(b)).

[0037] The operation of the system for utilizing customer messages in this embodiment can be briefly explained as above.

[0038] <Explanation of the system processing for using customer messages> Next, the processing content of the system for utilizing customer messages shown in this embodiment will be described in detail with reference to the flowcharts shown in Figures 3 and 4. In this explanation, first, with reference to Figure 3, the processing up to the point where the analysis server 1 (see Figure 1) stores a message from the customer terminal 4 (see Figure 1) will be described. Next, with reference to Figure 4, the processing for displaying a summary of the message on the business terminal 5 (see Figure 1) based on the message stored by the analysis server 1 (see Figure 1) will be described.

[0039] <Message storage process> First, the CPU 4a (hereinafter referred to as the customer terminal 4) of the customer terminal 4 (see FIG. 1) performs advance registration with the customer server 3 (see FIG. 1) owned by the company (step S401). More specifically, first, the customer accesses the customer server 3 (see FIG. 1) using a browser or the like pre-installed on the customer terminal 4 (see FIG. 1). Then, the customer enters his / her desired customer ID (e.g., "Taro71"), name (e.g., "Taro Ueda"), telephone number (e.g., "0353214567"), and address (e.g., "Tokyo") into a customer information input form (not shown) displayed in the browser, and taps a "send" button (not shown). As a result, the customer terminal 4 (see FIG. 1) transmits the customer information entered by the customer to the customer server 3 (see FIG. 1) via the network N (see FIG. 1).

[0040] Meanwhile, the data processing unit 30 (hereinafter referred to as the customer server 3) of the customer server 3 (see FIG. 1) stores the customer information received from the customer terminal 4 (see FIG. 1) in the customer information tbl_A (see FIG. 2(a)) of the database 32 (see FIG. 1) of the customer server 3 (step S301). More specifically, as shown in FIG. 2(a), the customer ID (e.g., "Taro71") is stored in the customer ID tbl_Aa, the name (e.g., "Taro Ueda") in the name tbl_Ab, the telephone number (e.g., "0353214567") in the telephone number tbl_Ac, and the address (e.g., "Tokyo") in the address tbl_Ad. When new customer information is registered, "0" is stored in the number of visits tbl_Ae (see FIG. 2(a)).

[0041] Note that advance registration (step S401) and customer information registration (step S301) only need to be performed once, and do not need to be performed again from the next time.

[0042] Next, the customer terminal 4 prompts the customer to select whether or not he or she wishes to receive a reply to the message that the customer is about to send from the customer terminal 4 (step S402). More specifically, after completing advance registration via a browser or the like, the customer then uses the customer terminal 4 to launch an application for using the social networking service provided by the SNS server 2 (see FIG. 1). It is assumed that this customer has already completed user registration for the social networking service provided by the SNS server 2 (see FIG. 1) and has already been assigned a unique user ID (user IDtbl_Cb (see FIG. 2(c))).

[0043] When the app is launched, a list of people with whom messages can be exchanged through the app is displayed (not shown). When the customer selects from the list, for example by tapping, the name of the company that owns the customer server 3 (see FIG. 1) where the customer information was previously registered, the screen shown in FIG. 5(a) is displayed on the display unit 4e of the customer terminal 4. Note that at the time the company name is selected by tapping, only the menu button group 40d is displayed on the screen shown in FIG. 5(a).

[0044] Then, when the customer taps the support button 40d1 of the menu button group 40d shown in Fig. 5(a), a screen is displayed that allows the customer to select whether or not a reply is required for the message that the customer is about to send. More specifically, a confirmation 40a asking "Which would you like?" as shown in Fig. 5(a) is displayed on the display unit 4e of the customer terminal 4, and below that, a "Answered" button 40b and a "No reply" button 40c are displayed.

[0045] Here, if "Reply Required" is selected on the customer terminal 4 (Step S402: Required), the customer terminal 4 sends a message to the SNS server 2 (see FIG. 1) (Step S403). More specifically, when the customer taps the "Reply" button 40b (see FIG. 5(a)) on the screen shown in FIG. 5(a), the customer terminal 4 displays the screen shown in FIG. 5(b) on the display unit 4e of the customer terminal 4. The customer then inputs a message (e.g., "I accidentally used a coupon. Can you do something about it?") into the message input unit 41a, inputs an email address (e.g., "0123@abc.com") into the email address input unit 41b, and taps the "Send" button 41c. This causes the customer terminal 4 to send the message to the SNS server 2 (see FIG. 1) via the communication unit 4f (see FIG. 1).

[0046] Although not shown in Fig. 3, the message sent to the SNS server 2 (see Fig. 1) is distributed to a predetermined server or computer by the data processing unit 20 (hereinafter abbreviated as SNS server 2) of the SNS server 2 (see Fig. 1). Then, the predetermined server or computer displays an incoming message, prompting company employees to reply using the business terminal 5 or other terminals used for business.

[0047] On the other hand, if the customer selects "No reply required" on the customer terminal 4 (step S402: No reply required), the customer terminal 4 also sends a message to the SNS server 2 (see FIG. 1) (step S404). More specifically, when the customer taps the "No reply" button 40c (see FIG. 5(a)) on the screen shown in FIG. 5(a), the customer terminal 4 displays the screen shown in FIG. 5(c) on the display unit 4e of the customer terminal 4. The customer then enters a message (e.g., "I haven't won any prizes, so it's boring.") in the message input unit 42a and taps the "Send" button 42c. This causes the customer terminal 4 to send the message to the SNS server 2 (see FIG. 1) via the communication unit 4f (see FIG. 1). Note that, because a reply is not required this time, the screen shown in FIG. 5(c) does not display an email address input section such as the email address input section 41b shown in FIG. 5(b).

[0048] Next, when the SNS server 2 (see FIG. 1) receives a message for which a reply is not required, it transfers the message to the analysis server 1 (step S201). More specifically, the SNS server 2 (see FIG. 1) adds a user ID (see user IDtbl_Cb in FIG. 2(c)) that the SNS server 2 uses to identify the user using its service to the message (see FIG. 5(c)) that does not require a reply received from the customer terminal 4 (see FIG. 1), and then transmits the message to the analysis server 1 (see FIG. 1).

[0049] Next, the SNS server 2 sends an automatic response to the message for which a reply is not required to the customer terminal 4 (step S202). That is, for example, an automatic response message saying "Your message has been received" is displayed on the display unit 4e (see FIG. 1) of the customer terminal 4 (not shown).

[0050] Next, the data processing unit 10 (hereinafter referred to as analysis server 1) of the analysis server 1 (see FIG. 1) stores the message in the database 12 of the analysis server 1 (see FIG. 1) (step S101). More specifically, as shown in FIG. 2(b), an ID (e.g., "0001") automatically generated by the analysis server 1 (see FIG. 1) is stored in message ID tbl_Ba, a message forwarded and received from the SNS server 2 (see FIG. 1) that does not require a reply (e.g., "It's boring because I haven't won any prizes") is stored in message tbl_Bb, and a user ID (e.g., "u1234") indicating the sender of this message is stored in user ID tbl_Bc.

[0051] Next, the analysis server 1 (see Figure 1) updates the number of times the user has sent a message (step S102). More specifically, it searches the customer-user information tbl_C shown in Figure 2(c) for the user ID (e.g., "u1234") that indicates the sender of the received message, and identifies the record of the sender. It then adds 1 to the number of times the message has been sent tbl_Cc (see Figure 2(c)) of the identified record.

[0052] It is assumed that the customer-user information tbl_C (see FIG. 2(c)) has already been registered at an arbitrary timing. For example, the arbitrary timing may be when the customer registers customer information in the customer server 3 (see FIG. 1). In this case, a section for inputting the customer's user ID for the social networking service provided by the SNS server 2 (see FIG. 1) may be provided in the customer information input form (not shown) displayed in the browser. In this case, the customer server 3 (see FIG. 1) transfers the input user ID and the customer ID tbl_Aa (see FIG. 2(a)) stored in the customer server 3 (see FIG. 1) to the analysis server 1 (see FIG. 1), and the analysis server 1 (see FIG. 1) can register this information in the customer-user information tbl_C (see FIG. 2(c)).

[0053] Next, the analysis server 1 (see FIG. 1) checks whether the number of message transmissions has reached a predetermined number (step S103). More specifically, it checks whether the number of message transmissions tbl_Cc (see FIG. 2(c)), to which 1 has been added, has reached a predetermined number (e.g., "3 times").

[0054] If the predetermined number of times (e.g., "3 times") has been reached, the analysis server 1 (see FIG. 1) transmits an electronic coupon to the customer terminal 4 (see FIG. 1) (step S104). That is, as a bonus for the fact that the number of times of sending messages that do not require a reply has reached the predetermined number (e.g., "3 times"), the analysis server 1 transmits the electronic coupon to the customer terminal 4 (see FIG. 1) via the SNS server 2 (see FIG. 1). In response, the customer terminal 4 (see FIG. 1) displays on the display unit 4e (not shown) that the electronic coupon has been acquired.

[0055] Note that the customer may be notified of the predetermined number of times (e.g., "3 times") or the number of times remaining to reach the predetermined number of messages (e.g., "2 more times") in order to encourage further message sending, but this may not be done at all. Alternatively, the customer may not even be informed that a special benefit will be awarded if the customer sends a predetermined number of messages. In this way, the customer is not given an opportunity to become aware of the special benefit, and the possibility that the customer will send messages frequently in order to receive the special benefit can be reduced.

[0056] In this way, the analysis server 1 (see FIG. 1) stores messages (see FIG. 5(c)) that do not require a reply from the customer in the message information tbl_B (see FIG. 2(b)).

[0057] <Process to display message summary> Next, referring to FIG. 4, a process in which the analysis server 1 (see FIG. 1) displays a message summary on the business terminal 5 (see FIG. 1) will be described. First, a company employee or the like logs in to the analysis server 1 (see FIG. 1) using the business terminal 5 (see FIG. 1) (step S501). More specifically, the company employee or the like accesses the analysis server 1 (see FIG. 1) from a browser pre-installed on the business terminal 5 (see FIG. 1), enters an account name and password on a login screen (not shown) displayed within the browser, and clicks a login button (not shown). As a result, the business terminal 5 (see FIG. 1) sends the entered account name and password to the analysis server 1 (see FIG. 1).

[0058] Next, the analysis server 1 grants permission to log in (step S111). That is, the analysis server 1 (see FIG. 1) grants permission to log in if the account name and password received from the business terminal 5 (see FIG. 1) are correct. If permission to log in is granted, the analysis server 1 (see FIG. 1) displays the BI tool (see FIG. 6(a)) provided by the analysis server 1 (see FIG. 1) in the browser displayed on the display unit 5e of the business terminal 5 (see FIG. 1).

[0059] Next, the business terminal 5 (see FIG. 1) requests the analysis server 1 (see FIG. 1) to display the results of the analysis of the numerical data (step S502). More specifically, when an employee of the company clicks on the operation button 50a of the BI tool (see FIG. 6(a)) displayed on the browser displayed on the display unit 5e of the business terminal 5 (see FIG. 1), the business terminal 5 (see FIG. 1) specifies predetermined numerical data (for example, numerical data on sales by region) and transmits a request for analysis of the specified numerical data to the analysis server 1 (see FIG. 1).

[0060] Next, the analysis server 1 creates a summary from the stored messages (step S112). More specifically, upon receiving the specification of numerical data and a request to display the analysis results from the business terminal 5 (see FIG. 1), the analysis server 1 (see FIG. 1) first analyzes the specified numerical data and generates screen data for displaying the analysis results (see analysis results 50b of the numerical data analyzed in FIG. 6(a)). The analysis server 1 (see FIG. 1) also uses artificial intelligence to generate summaries (see summary 50c in FIG. 6(a)) of multiple messages that do not require a reply and are stored in message information tbl_B (see FIG. 2(b)).

[0061] Here, we will explain how to generate summaries. First, as shown in FIG. 2(b), message information tbl_B stores multiple messages that do not require a reply and were received from the customer terminal 4 (see FIG. 1). Among these, message tbl_Bb with message ID tbl_Ba "0001" is "It's boring because I can't win any prizes," and message tbl_Bb with message ID tbl_Ba "0002" is "I wish it were easier to win prizes." Since both of these messages are about the difficulty of winning prizes, the AI ​​combines these two messages and summarizes them as, for example, "It's hard to win prizes." (See FIG. 6(b)). The analysis server 1 (see FIG. 1) associates this summary with the corresponding message IDs tbl_Ba "0001" and "0002" and stores them in a temporary storage area (not shown).

[0062] Similarly, the artificial intelligence compiles the messages with message IDs tbl_Ba "0003" and "0004" and summarizes them as, for example, "The cleaning is not thorough." (See FIG. 6(b)) The analysis server 1 (See FIG. 1) associates this summary with the corresponding message IDs tbl_Ba "0003" and "0004" and stores them in a temporary storage area (not shown).

[0063] The artificial intelligence that performs summarization is trained in advance using training documents in which, for example, a plurality of sentences are used as input data and summaries of the sentences are used as output data.

[0064] Next, the analysis server 1 displays the summary together with the analysis results of the numerical data (step S113). More specifically, the analysis server 1 (see FIG. 1) displays the analysis results and summary generated as described above on the display unit 5e of the business terminal 5, as shown in FIG. 6(a).

[0065] Next, the business terminal 5 notifies the analysis server 1 (see FIG. 1) of the summary selected by the company employee or the like (step S503). More specifically, as shown in FIG. 6(a), the company employee or the like views the summary generated by the analysis server 1 (see FIG. 1) on the business terminal 5 (see FIG. 1), and selects a part of the summary (for example, "The prizes are hard to win") by clicking or the like using the input unit 5b such as a mouse. The business terminal 5 (see FIG. 1) notifies the analysis server 1 (see FIG. 1) that a part of the summary (for example, "The prizes are hard to win") has been selected.

[0066] Next, the analysis server 1 identifies the customer who sent the message corresponding to the part of the selected summary (step S114). More specifically, it reads out the message IDs tbl_Ba "0001" and "0002" stored in association with the part of the selected summary (e.g., "It's hard to win the prizes.") from a temporary storage area (not shown). Next, the analysis server 1 (see FIG. 1) refers to the message information tbl_B shown in FIG. 2(b) and identifies the user IDs tbl_Bc "u1234" and "u5678" as the senders of the message IDs tbl_Ba "0001" and "0002." Then, the analysis server 1 (see FIG. 1) refers to the customer-user information tbl_C shown in FIG. 2(c) and identifies that the user IDs tbl_Cb "u1234" and "u5678" correspond to the customer IDs tbl_Ca "Taro71" and "HanaU."

[0067] Next, the analysis server 1 (see FIG. 1) requests the customer information of the identified customer from the customer server 3 (see FIG. 1) (step S115). More specifically, the analysis server 1 sends the request for the customer information to the customer server 3 (see FIG. 1) along with the customer IDs tbl_Ca "Taro71" and "HanaU" (see FIG. 2(c)) identified as described above.

[0068] Next, the customer server 3 (see FIG. 1) transmits the customer information to the analysis server 1 (see FIG. 1) (step S211). More specifically, the customer server 3 (see FIG. 1) refers to the customer information tbl_A shown in FIG. 2(a) and identifies records in which the customer ID tbl_Aa is "Taro71" and "HanaU." Then, the customer server 3 (see FIG. 1) transmits the customer information contained in the identified records (for example, the information stored in the customer ID tbl_Aa, name tbl_Ab, address tbl_Ad, and number of visits tbl_Ae shown in FIG. 2(a)) to the analysis server 1 (see FIG. 1).

[0069] Next, the analysis server 1 (see FIG. 1) displays the customer information on the business terminal 5 (see FIG. 1) together with the analysis results of the numerical data related to the customer information received from the customer server 3 (see FIG. 1) (step S116). More specifically, the analysis server 1 (see FIG. 1) first narrows down the numerical data used in the analysis results 50b (see FIG. 6(a)) being displayed on the business terminal 5 to the numerical data related to the customer information received from the customer server 3 (see FIG. 1) (for example, the information stored in the customer ID tbl_Aa, name tbl_Ab, address tbl_Ad, and number of visits tbl_Ae shown in FIG. 2(a)).

[0070] Here is a more detailed example of how to narrow down the numerical data. For example, if the numerical data consists of "customer ID," "visit date and time," and "store identifier," this numerical data is narrowed down to numerical data where "customer ID" is the customer ID (see customer IDtbl_Aa shown in Figure 2(a)) included in the customer information received from the customer server 3 (see Figure 1).

[0071] In this way, the narrowed-down numerical data is analyzed, and screen data is generated for displaying the analysis results (see analysis results 51b of the numerical data analyzed, shown in FIG. 6(b)). Furthermore, the analysis server 1 (see FIG. 1) generates screen data for displaying customer information (see customer information 51d, shown in FIG. 6(b)) received from the customer server 3 (see FIG. 1) on the business terminal 5 (see FIG. 1).

[0072] Then, the analysis server 1 (see Figure 1) sends screen data for displaying the analysis results generated as described above and screen data for displaying customer information to the business terminal 5 (see Figure 1).

[0073] The business terminal 5 (see FIG. 1) receives the screen data from the analysis server 1 (see FIG. 1) and displays, on the display unit 5e, customer information 51d of the customer who sent the message corresponding to the selected summary and analysis results 51b of the numerical data related to that customer, as shown in FIG. 6(b). Note that a portion 51c1 of the summary selected by a company employee or the like is displayed in an emphasized manner.

[0074] While viewing the screen shown in FIG. 6(b), company employees can easily associate a portion of the summary 51c1 with the customer who sent the message related to that portion of the summary. As shown in FIG. 6(b), customer information 51d includes information that serves as an index for evaluating a customer's loyalty (in the illustration, "number of visits"). This makes it possible to confirm whether the message is from a customer who can be evaluated as having high customer loyalty. Other information that serves as an index for customer loyalty includes the number of times the service is used, the annual transaction amount, and the average purchase price.

[0075] According to the present embodiment described above, the analysis server 1 (see FIG. 1) stores messages that customers have designated as not requiring a reply in the message information table tbl_B (see FIG. 2(b)), and then, upon request from a company employee, has the AI ​​generate a summary 50c (see FIG. 6(a)) from the message. This reduces the time required to review messages, even when there are many. Furthermore, when a company employee selects a portion 51c1 of the summary (see FIG. 6(b)), the analysis server 1 identifies the customer who sent the message and displays the customer information 51d (see FIG. 6(b)) of that customer on the business terminal 5 (see FIG. 1), allowing the employee to easily learn about the customer who sent the message.

[0076] As described above, according to this embodiment, messages (complaints and opinions) obtained from customers can be easily associated with data relating to the customers and used.

[0077] <Description of Modifications> It should be noted that the system for utilizing customer messages shown in this embodiment is merely an example, and various modifications and changes are possible within the scope of the gist of the present invention as set forth in the claims.

[0078] For example, in this embodiment, messages are stored in the database 12 of the analysis server 1 (see FIG. 1) as message information tbl_B (see FIG. 2(b)), but they may be stored anywhere. For example, they may be stored in the SNS server 2 (see FIG. 1), or in another server or computer (not shown). Similarly, the customer information tbl_A (see FIG. 2(a)) and customer-user information tbl_C (see FIG. 2(c)) may also be stored anywhere.

[0079] Furthermore, in this embodiment, the SNS server 2 (see FIG. 1) transfers the message to the analysis server 1 (see FIG. 1) and sends an automatic response to the customer terminal 4 (see FIG. 1), but the analysis server 1 (see FIG. 1) may directly receive the message from the customer and send an automatic response to the customer terminal 4 (see FIG. 1). Similarly, the operations performed by the analysis server 1, the SNS server 2, and the customer server 3 (see FIG. 1) may be performed by any server within the scope of the present invention, or may be performed by another server not shown.

[0080] In this embodiment, an example has been shown in which, regardless of the type of message, 1 is always added to the number of message transmissions tbl_Cc (see FIG. 2(c)) of the identified record whenever a message is sent. However, this is not limiting, and for inappropriate messages such as abusive language, 1 may not be added to the number of message transmissions tbl_Cc (see FIG. 2(c)) of the identified record. For example, an artificial intelligence may be trained in advance to learn the content of inappropriate messages. Then, the analysis server 1 (see FIG. 1) uses artificial intelligence to check whether or not any inappropriate messages are included in the messages sent from the SNS server 2, and if any inappropriate messages are included, 1 may not be added to the number of message transmissions tbl_Cc (see FIG. 2(c)) of the identified record. [Explanation of symbols]

[0081] 1 Analysis Server (Server) 12 Database (prescribed storage means) 4. Customer terminal 41a Reply-required message input section (the message you are about to send) 42a No reply required message input section (message you are about to send) 5 Business terminals 50b, 51b Analysis results of numerical data (analysis results of specified numerical data) 50c Summary 51c1 Part of the summary 51d Customer information tbl_Bb messages (multiple stored messages) tbl_Cc Message sent count (number of times a message was sent)

Claims

1. A system for utilizing customer messages, comprising a server, a customer terminal, and a business terminal, The customer terminal allows the customer to select whether or not a reply is required to the message to be sent from the customer terminal, If a reply is required on the customer terminal, the agent uses the business terminal to reply to the message, while When a reply is not required at the customer terminal, the message sent by the customer terminal is stored in a predetermined storage means, The server creates a summary based on the plurality of stored messages using artificial intelligence trained using training documents; the business terminal displays the summary; When a portion of the summary displayed on the business terminal is selected, the server identifies the customer who sent the message corresponding to the portion of the summary; The business terminal displays customer information of the identified customer.

2. A program for utilizing customer messages, A computer communicably connected to the customer terminal and the business terminal, The customer terminal allows the customer to select whether or not a reply is required to the message that is about to be sent from the customer terminal, and if a reply is required is selected on the customer terminal, the business terminal is used to prompt the agent to reply to the message, while if a reply is not required is selected on the customer terminal, a summary is created using artificial intelligence that has been trained using a learning document based on the message sent from the customer terminal, which is stored in a predetermined storage means; displaying the summary on the business terminal; When a portion of the abstract displayed on the business terminal is selected, identifying the customer who sent the message corresponding to the portion of the abstract; and a program for causing the business terminal to display customer information of the identified customer.

3. 3. The program according to claim 2, wherein the customer information displayed on the business terminal includes information that serves as an index for evaluating customer loyalty of the customer.

4. The computer provides a web service that displays analysis results of predetermined numerical data on the business terminal, 3. The program according to claim 2, further comprising a step of displaying, on the business terminal, the analysis results of the analysis of the numerical data relating to the customer who sent the message corresponding to the portion of the summary when the portion of the summary displayed on the business terminal is selected.

5. Furthermore, the program executes a step of granting a benefit to the customer when the number of times the same customer has sent the message reaches a predetermined number, 3. The program according to claim 2, wherein the predetermined number of times or how many times the message must be sent before the predetermined number of times is reached is not notified to the customer.

6. 1. A method for utilizing customer messages, comprising: a step in which the customer terminal prompts the customer to select whether or not a reply is required to the message that the customer is about to send from the customer terminal, and when a reply is required is selected on the customer terminal, the business terminal is used to prompt the agent to reply to the message, while when a reply is not required is selected on the customer terminal, a summary is created using artificial intelligence based on the message sent by the customer terminal, which is stored in a predetermined storage means; displaying the summary on a business terminal; When a portion of the abstract displayed on the business terminal is selected, identifying the customer who sent the message corresponding to the portion of the abstract; and causing the business terminal to display customer information of the identified customer.

Images