JP2002297875A - Customer relation management method, system and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a customer relation management method, a system and a program for standardizing in bound so as to be able to be used in marketing and recording the same. SOLUTION: In this customer relation management method, according to the consumer's psychology corresponding to the inputted inbound which is a consumer's psychology to the commodity or service changing step by step until trading, the inbound is converted into a psychology code and recorded while relating to the customer. A plurality of inbounds are previously recorded taking inbound as standard inbound in a table in relation to a psychology code, and among the standard inbounds, the same or similar inbound as the received inbound is called on a screen and selected by an operator of the call center, thereby inputting the inbound to a computer, and converting the received inbound to the psychology code related to the selected inbound and recording the same.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to a customer relationship management (CRM) using a computer.
nagement) methods, systems and programs.

[0002]

2. Description of the Related Art In the competition between companies, it has become difficult to differentiate in terms of price and product itself. Nowadays, companies are selected by consumers depending on how they treat consumers or the quality of services provided. ing. This makes it difficult for companies to survive market competition with conventional mass marketing-oriented strategies. Therefore, CRM is attracting attention as a management concept for acquiring new customers and building long-term trust with existing customers. CR
M is a management concept that seeks to increase profits by improving the relationship between companies and consumers. More specifically, CRM is an analysis of consumer behavior to read consumer needs, and based on the results of analysis, formulate and implement appropriate marketing for consumers to achieve consumer satisfaction. To improve individual consumers. The analysis of consumer behavior requires the collection and analysis of enormous amounts of data, but as a result of the increase in storage capacity, processing speed, and development of communication technology, C
RM has become a reality, and in the United States, success stories of CRM such as supermarket chains are continually emerging.

[0003] The expected outcome of CRM is to analyze consumers' attributes and behaviors, segment consumers, and change the emphasis on each segment to maximize profits from appropriate costs. Efficiency. Also,
CRM aims to conduct effective one-to-one marketing, such as selling goods and related services to consumers based on a wide range of data on consumers, and selling proposals for consumables at the appropriate time. I do. one·
Two-to-one marketing is a way to provide customers with information tailored to their preferences based on individual consumer analysis, or to make individual proposals based on past purchases and transaction histories. It's marketing to get loyalty. Proper one-to-one marketing for specific segments can reduce costs and increase profits.

[0004] The key to the success of CRM is the construction of a system that makes it possible to utilize enormous data on consumers collected daily at the point of contact between consumers and companies, such as call centers, sales agents and homepages, as management resources. is there. Specifically, a mechanism that determines what kind of information is collected from many points of contact between companies and consumers, and how to process the collected information into information that can be used for management Is key to the success of CRM. At present, each company is trying to transform a call center, which has been a conventional cost center, into a profit center by utilizing non-standard inbounds collected at a call center or the like as management information.

[0005]

The "consumer's voice" heard by a call center operator or the like contains various information useful for management. In particular, since "consumer's voice" is often sent from consumers before taking specific actions such as purchasing a product, it can be said that a lot of information useful for planning marketing is included. Therefore, in the CRM, it is desirable to accurately grasp not only data collected as a result of specific actions such as sales data but also such data of “consumer voice” without omission. For this reason, call centers extract keywords from such “consumer voices” and record them as character data, or record “consumer voices” themselves as voice data.
In addition, while recording the voice of the consumer as character data or voice data, the voice of the consumer, "complaint", "request",
In some cases, these symbols are recorded after being converted into "opinions".

[0006] However, even if an inbound sent to a call center or the like is converted into a symbol and analyzed, it is difficult to find information that is truly useful for the entire management from the information in which the inbound is converted. This is because the consumer's psychology that attempts to transmit “complaints”, “requests”, “opinions”, etc. to the operator is almost irrelevant to the purchasing behavior, and thus the consumer's psychology transmitted to the operator via inbound is assumed. Is converted to symbols such as "complaints", "requests", "opinions", etc., but specific consumer behavior such as purchasing cannot be analyzed from the information obtained by summing up these symbols. is there. Since the amount of information that humans can read and listen to within a limited time is limited and human judgment is subjective, character data or voice data that records "consumer's voice" almost as it is However, it is impossible to quickly and accurately capture market trends and consumer demands. After all, conventional computer-based CRM only assisted in marketing planning based only on data obtained from specific consumer behaviors such as visits to stores, product purchases, contracts, and cancellations. They have not yet used inbound information, which is information at a stage where they have not yet taken concrete actions, for marketing planning.

[0007] In view of the fact that the "consumer voice" includes information before concrete actions are put into practice, it is important to analyze inbound in CRM and formulate a company-wide sales strategy. It can be said that the superiority of data collection, such as what kind of symbols a call center operator hears and converts into a consumer voice, greatly determines whether an effective sales strategy can be formulated. At the stage of collecting inbound, building a mechanism to collect inbound by converting it to a symbol that can be used for management transforms the call center, which is a cost center, into a profit center.

It is an object of the present invention to provide a customer relationship management method, system, and program for standardizing and recording inbound so that it can be used for marketing. An object of the present invention is to provide a customer relationship management method, system, and program for converting inbound business resources.

[0009]

SUMMARY OF THE INVENTION The present invention is intended to find out the consumer sentiment for goods or services that change step by step up to a transaction from behind the inbound, convert the consumer sentiment into symbols, and analyze it by a computer. The above-mentioned object is achieved by standardizing and recording.

[0010] According to the inventions according to Claims 1, 11 and 20, the consumer sentiment for goods or services that change step by step until the transaction is reached, and the inbound is set in accordance with the consumer sentiment corresponding to the input inbound. The first psychological code is converted into a predetermined first psychological code, and the first psychological code is recorded in association with the consumer. That is, the consumer psychology related to the transaction behavior is divided into a plurality of stages in advance, a predetermined first psychological code is associated with each stage, and the inbound and the first psychological code are associated via the consumer psychology. deep.
The association between inbound and consumer psychology, specifically, the association between inbound and first psychological code is essentially subjective, but according to the present invention, using a computer, the inbound and first psychological code Can be quickly and consistently matched without omission.

For example, as shown in FIG. 1, a plurality of inbounds are recorded in advance in a table in association with a psychological code as standard inbounds, and an inbound of the standard inbounds that is the same as or similar to the incoming inbound is displayed on the screen by the call center operator. By inputting the inbound to the computer by calling and selecting the inbound, the incoming inbound is converted to the psychological code associated with the selected inbound and recorded, thereby making the incoming inbound and the first psychological code consistent. Can be quickly associated without any omission. For example, as shown in FIG. 2, a plurality of inbounds associated with the psychological code are recorded in advance as a standard inbound in a table, and the standard inbound is searched by text mining based on the input inbound, and the input inbound is converted. By specifying the first mental code to be performed, the input inbound and the first mental code can be quickly and consistently associated without any omission.

As explained by AIDMA's law, which is a hypothesis of trading behavior proposed by, for example, Roland Hall, a consumer purchases, contracts, cancels, etc. after undergoing several gradual changes in mental state. Put your trading actions into action. Such a gradual change in mental state occurs due to various factors existing in the market. Therefore, managers and the like must formulate marketing that creates factors that cause such gradual changes in the state of mind of consumers in the market. For example, if a business owner succeeds in attracting the consumer's attention to a new product through a TV advertisement or the like, but finds that it is not interested in the product, it will be interested in the product You can plan marketing to create factors in the market. Specifically, for example, by reducing the frequency of TV advertisements and enhancing the product description function in call centers and websites, consumers 'attention is changed to interest and consumers' mental state is advanced toward actual transaction behavior of product purchase You can plan marketing to make it work. As described above, if the consumer's psychology linked to the transaction behavior can be grasped, it becomes possible to plan accurate marketing.

Therefore, when the inbound is converted into the first psychological code by the computer and the first psychological code is recorded in association with the customer as in the present invention, the consumer psychology of the goods or services is at any stage before the transaction. Can be accurately, promptly and completely understood by a computer. According to the present invention, it is possible to obtain an accurate one-to-
One marketing can be planned, accurate segment marketing can be planned by grasping each segment, and accurate mass marketing can be planned by grasping the whole.

[0014] As in the inventions according to the second, twelfth and twenty-first aspects, it is possible to classify consumer sentiment for a product or service into, for example, five-stage divisions and convert inbound into five first mental codes. Specifically, for example, as in the invention according to claim 3, the consumer psychology of a product or service can be divided into a caution stage, an interest stage, a decision stage, a memory stage, and an action stage in chronological order. More specifically, for example, as in the invention according to Claims 4, 13 and 22, inbound relating to advertising, inbound relating to goods or services provided within a predetermined period, inbound relating to a method of trading goods or services, or inbound of goods or services. Each of the inbound regarding the method of use and the inbound regarding the transaction procedure or selection of the product or service can be converted into different first psychological codes. In grasping the consumer sentiment via inbound, it is important to provide in advance a standard that can uniquely identify the first psychological code from the content of inbound. Then, by using a computer, inbound and consumer sentiment can be quickly and consistently and consistently based on such criteria.

The invention according to claims 5, 14 and 23 is intended to convey questions, complaints, requests or opinions in addition to consumer sentiment for goods or services that change step by step until a transaction. Can be analyzed by computer via inbound. According to the invention according to claims 5, 14 and 23, among the consumer psychology to transmit a question, a complaint, a request or an opinion, the inbound is set to a predetermined second psychology according to the consumer psychology corresponding to the input inbound. The second psychological code is converted into a code and recorded in association with the customer. Specifically, for example, it is possible to define the correspondence between the inbound content and the second psychological code as in the invention according to claims 6, 15 and 24. By converting and accumulating the consumer sentiment for transmitting a question, a complaint, a request or an opinion to the second psychological code, it is possible to weight the degree of the consumer's request for the business with respect to specific items. Therefore, marketing can be planned according to the degree of demand.

As a method for quickly and consistently associating the inbound and the consumer sentiment, that is, the psychological code, consistently, for example, as in the inventions according to claims 7, 16 and 25, the first psychological code and / or the The first psychological code and / or the second psychological code to which the input inbound should be converted may be specified by analyzing the inbound and the input inbound recorded in association with the two psychological codes by text mining. still,
As in the inventions according to claims 8, 17 and 26, the inbound obtained as the call voice data may be converted into character data by voice recognition using a computer and then input.

As a method of quickly and consistently associating the inbound and the consumer psychology, that is, the psychological code, without fail, for example, as in the invention according to claims 9, 18 and 27, the incorporation and the psychological code are recorded in association with the psychological code. The inbound may be input by selecting the same or similar inbound as the incoming inbound from the incoming inbounds, and the incoming inbound may be converted into a psychological code associated with the selected inbound.
Incidentally, as in the invention according to claims 10, 19 and 28, in addition to the psychological code, information for specifying when the inbound is input may be recorded in association with the customer.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment showing an embodiment of the present invention will be described below with reference to the drawings. FIG. 3 is a functional block diagram showing a basic configuration of the customer relationship management system according to one embodiment of the present invention. Input means 1, processing means 2,
The recording means 3, the analysis means 4, and the output means 5 are one OS
Can be realized as functional units connected by a bus in one computer system controlled by a single computer system. Each of a plurality of computer systems controlled by one OS and connected to each other by a communication line to operate in cooperation with each other It can also be realized as.

The input means 1 constitutes the input means and the conversion means described in the claims, and is a component for inputting inbound. The input unit 1 standardizes and inputs an inbound signal to be received as any of voice, character, and image information by telephone, electronic mail, facsimile, or the like. The incoming inbound can be standardized and input by inputting which one of the standard inbounds assumed in advance is similar. Alternatively, the inbound data may be converted into character data and input. For example, voice data and image data can be input as character data by inbound by telephone by automatically recognizing the voice of the call and converting it into character data, and by inbound by facsimile by converting it to character data by OCR. By converting the data into character data and inputting the converted data, the data can be standardized using text mining. As a result, it is possible to standardize the inbound without the operator's subjectivity.

The processing means 2 constitutes the processing means and the mining means described in the claims, and is a component for converting symbols by converting inbound into psychological codes. Specifically, for example, as shown in FIG. 1, the input means 1 is used to query a table storing the standard inbound and psychological codes in association with each other, and a standard inbound similar to the incoming inbound is entered by the operator. To select. And the processing means 2
Convert the incoming inbound to the psychological code corresponding to the selected inbound. FIG. 4 shows a specific example of a table in which standard inbound and psychological codes are stored in advance. The contents of this table will be described later in detail. In addition, by analyzing the inbound character data input as it is without standardization by text mining, the inbound input as it is without standardization is converted into a psychological code as shown in FIG. 2 and standardized. Can be. For example, a large amount of inbound character data and psychological codes that have been received in the past are associated with each other and stored in advance in a recording medium that constitutes the second recording means described in the claims, and the content of the inbound that has been received this time Compare the entire inbound with the past inbound by text mining, identify the past inbound with the most similar content to the inbound this time, and enter the psychological code associated with the identified past inbound this time It can be specified as a psychological code that converts the inbound that has been made. As a tool for text mining, ConceptBase, etc. of Just System Inc. is known.

The recording means 3 constitutes the first recording means described in the claims, and is a component for recording the psychological code in association with other data relating to the customer. The recording means 3 manages, for example, a table shown in FIG. 5 and records the psychological code and the customer in association with each other. The customer to which the psychological code is to be associated, that is, the consumer who is the source of the inbound, is specified as follows, for example. For inbound by telephone or facsimile incoming to the call center, the caller's telephone number is specified by CTI (Computer Telephony Integration), and customer attribute information is searched using the telephone number as a search key to inquire the caller's customer. For an inbound incoming call from a consumer accessing a company's homepage, the consumer searches for customer attribute information using the customer code entered in the text box of the homepage as a search key to inquire about the originating customer. The psychological code and other data relating to the customer are associated, for example, via the customer code as shown in FIG.

The analysis means 4 is a component for analyzing various data relating to business such as sales data for each customer, sales data for each product, and psychological codes by data mining to find information useful for planning marketing. . The analysis means 4 is operated, for example, by a company manager. The analysis method performed by the analysis means 4 includes CRM
There are those that analyze data by statistical methods, those that use discrimination methods that use neural networks, those that use data mining techniques that use decision trees, and so on. May be adopted. Also, a plurality of methods may be used independently or in combination in order to improve the accuracy of output evaluation through analysis from different angles. As an analysis tool, for example, SAS
Etc. are known. The output means 5 is a component for outputting the regularity found by the analysis means 4 to a display or a printer.

FIG. 6 is a block diagram specifically illustrating hardware constituting a customer relationship management system according to an embodiment of the present invention. In the following, in a customer relationship management system assuming a business support system of a mobile phone service company, we analyze inbound data together with business data such as monthly call charge that seems to affect transaction behavior and based on the obtained results An embodiment of the present invention will be described as a technique for planning marketing. However, the present invention can be applied to a customer relationship management system of any company that provides products and services to consumers, such as a credit guard service company, a bank, a travel agency, and a department store, regardless of a mobile phone service company.

The database servers 21, 22, and 23 are database management systems constituting a so-called backbone system, such as customer attribute information such as telephone numbers, contract information such as rate plans, and billing information such as monthly call charges. Manage mission-critical data. The information stored in these database servers is all kinds of information accumulated through business, for example, customer names, telephone numbers, ages and genders, sales data for each product or service, monthly billing charges, etc. obtain. Such information is used for analyzing inbound that is converted into a psychological code and recorded, for example, for cluster analysis of customers and inbound analysis for a specific cluster.

The intermediate system 17 constituting the processing means 2 and the recording means 3 constitutes a so-called information system, and is used by managers such as managers and call center operators to easily extract information for CRM. It is desirable to configure as a possible data mart or data warehouse. The CRM information recording unit 18 extracts, from the database servers 21, 22, and 23, data to be referred to when the manager analyzes the psychological code and when the call center operator inquires the originating customer among the core data, and centralizes the data. Record what you do.

FIG. 7 shows an example of data stored in the CRM information recording section 18. The customer code is a unique identifier assigned to each customer, the rate plan is a field indicating the type of call rate contract that the customer has contracted, the usage period is a field indicating the number of months elapsed since the contract, and the customer rank is the Increase customer loyalty according to payment amount
Fields indicated by stages, churn risk is the amount paid for call charges,
This is a deviation value indicating the possibility of cancellation calculated by analysis of a psychological code or the like. The customer response history recording unit 19 records the incoming time, customer code, inbound code, and the like in one record in association with each incoming inbound call.

FIG. 5 shows an example of a table stored in the customer response history recording section 19. In the example shown in FIG. 5, an incoming call code is provided for each incoming call, and the same incoming code is applied to a plurality of inbounds obtained by one incoming call. This is because a single telephone response between the operator and the customer may extend the conversation content to a content corresponding to a plurality of standard inbounds. Another reason is that a user who has logged on to the support page of the homepage may be referred to a plurality of standard inbounds before logging out. The standard inbound is, for example, a method of extracting, for example, several hundred types of contents that can be generalized from inbounds accumulated in the past, and expressing those contents in standard terms.
The default values of the AIDMA code as the first psychological code and the category code as the second psychological code are automatically inputted by referring to the table shown in FIG. In some cases, in the example shown in FIG. 5, an inbound code and another field are defined and recorded. The inbound management unit 20 forms a third recording unit described in the claims, and manages, for example, the table shown in FIG. The inbound code is a unique identifier assigned to each of the standard inbounds. The contents of the table managed by the inbound management unit 20 are updated at any time by an administrator or the like, and when an inbound incoming call that has never occurred before is expected due to a start of a new service or the like, a standard inbound is added to the table in advance. It is desirable to keep. It is also desirable to change the psychological code according to changes in the market. For example, when the category code of "Opinion" is given in the standard inbound service for a specific service, but the incoming call starts to increase, the default value of the category code assigned to the inbound is changed from "Opinion" to "Request". It is desirable to change. .

The AIDMA code and the category code correspond to the first psychological code and the second psychological code, respectively, and are given to each standard inbound according to the contents of the standard inbound. All standard inbound AIDM
There is no need to provide an A code. For example, there is no need to assign an AIDMA code to inbound traffic that is unrelated to trading activity. Further, the AIDMA code and the category code recorded in the table may be set as default psychological codes, and an arbitrary psychological code may be added according to the judgment of the call center operator, for example. For example, the inbound code 0001 contains the AID "Interest"
When an MA code is assigned by default and the operator is convinced that the psychology behind the inbound is not "interest" but "action" for the inbound that is standardized and recorded in the inbound code 0001 through telephone response, A configuration may be adopted in which “interest”, which is the default of the AIDMA code, is changed and recording of the inbound code 0001 and the AIDMA code “action” for the incoming call is allowed.

(AIDMA Code) The AIDMA code is used to describe the consumer sentiment for goods or services that change step by step until the transaction, such as “Attention” and “Interes”.
t (interest), "Decision", "Memory" and "Action"
This is for associating the consumer psychology behind the inbound with the inbound, converting the inbound into the consumer psychology and recording it. In the communication between a company and a consumer in a mobile phone service business, a consumer's decision regarding transaction behavior is often made at an earlier stage than in other businesses.
The code division is defined based on the behavior model in which the division of “sire (desire)” is replaced with “Decision”.
Note that there are various methods for dividing the consumer sentiment into stages or how to draw each stage, and a typical one is known according to the law of AIDMA. AIDMA's law states that the psychological process by which marketing communication can change a consumer's purchasing behavior is "Attention", "Interest",
"Desire", "Memory" and "Action"
(Behavior) ”, and considers that consumers act based on the psychology of each of these stages. Others
AIDAS's law that considers consumer psychology after a transaction is also known. AIDAS's rules are "Attention", "Interest", "Desire",
5 of “Action” and “Satisfaction”
Classify consumer sentiment into stages. As in the case of AIDAS's law, the psychological code system may be formulated by including the consumer psychology after the transaction in the psychological category. The transaction behavior relating to a certain product or service is executed as a result of the mental state changing stepwise with the passage of time due to some external factor from a state where the product or the like is unknown to the customer. The present invention seeks to realize a customer relationship management system that can assist in effective marketing planning by grasping the change in psychological state as movement of the entire market. Whether to classify consumer sentiment on the time axis based on the model is arbitrary. Therefore, the inbound may be classified using an appropriate method according to the content of the inbound among the methods described below, and the methods may be combined and classified using, for example, a method similar to a decision tree. Hereinafter, a specific method for associating each AIDMA code with inbound will be described using an example.

(1) A method of associating an AIDMA code based on the degree of concreteness of a consumer's image of a transaction behaviour: Inbound advertising is classified into Attention. Segment inbound for new products and services into Interest. Classify inbound purchase method into Decision. Classify inbound usage as Memory. Inbound of contract and selection is classified into Action. (Example) The inbound "Please tell me specifically about the rate plan set from next month that can reduce the call charge by 10%" is classified as Interest. The inbound "How do I download ringtone data?"
Classify into. The inbound action "How do I get to the agency xx from xx station?"
(Behavior).

(2) Method of associating AIDMA code based on knowledge of goods and services If the inbound can be assumed to have been issued by the customer at the stage of storing the name of the goods, services or companies, the inbound is considered as Attention. (Caution) If the inbound content can be guessed to have been issued by the customer at the stage of understanding the specific characteristics of the product or service, the inbound is classified as Interest. It is issued by a customer to compare a product or service whose outline has been understood with another product or service to determine whether there is a benefit (use value) to the consumer or to fulfill a certain purpose of life. When it can be guessed that the content is inbound, that is, when the content of the inbound is related to the detailed content of the goods and services, the inbound is classified into Decision. If you can guess that the customer is inbound when trying to reconfirm what is known to the customer,
(Memory). If the inbound is about to initiate a trading action or can be inferred to have been issued when it has already begun, the inbound is acti
Classify on (action). (Example) From the beginning of the year, the inbound “The price plan of XX will be lower than the price plan of △△?” Is classified as Decision.

(3) AID based on the number of inbound
When defining an MA code, the first inbound to get from a consumer with no transaction history is At
It is classified as tention. The second inquiry from a consumer with no transaction history is classified as Interest. If the inbound is repeatedly issued, that is, if the same or similar inbound obtained from the same person has been recorded in the past, the second and subsequent inbounds are classified as Memory.

(4) Other Techniques When an unexpected inbound is obtained as a response to the advertisement, more specifically, a standard inbound similar to the incoming inbound is received by the inbound management unit 20.
If not recorded in Attentio
Classify into n (caution). If there is an inquiry about the availability of a specific service that has begun to be offered in the mobile phone service industry, it will be classified as Attention. Categorize inbound interest in goods and services before or within three months of market launch. If the inbound is presumed to be a reaction to the advertising medium passively stored by the consumer, it is classified as Attention.
If the inbound is presumed to be a reaction to advertising media that is actively stored by consumers over a relatively long period of time, such as ads using print such as newspapers, magazines, and homepages
Classify into Interest. If it is an inbound that is presumed to be a response to in-store advertising such as POP, it is classified into Action.

(Category code) The category code expresses the consumer sentiment that the consumer intends to transmit to the company via inbound in four categories of questions, complaints, requests or opinions, and the consumer sentiment behind the inbound and the inbound. , And convert the inbound into consumer sentiment for recording. Inbound questions and confirmations regarding products and services are classified as “questions”.
Inbound complaints, complaints and doubts about all internal and external operations are classified as "complaints." Inbounds that have not yet been fulfilled and contain requests that can be handled by the company are classified as “requests”. Inbounds that have not yet been fulfilled and contain requirements that the industry must respond to are classified as "opinions."

The CTI system 16 is a system that, when a telephone call is received from a customer at a call center, reversely detects the source telephone number and notifies the intermediate system 17 of the source telephone number. When the CTI system 16, the intermediate system 17, and the client computer 11 installed in the call center cooperate to receive a telephone call from a consumer in the call center, information about the customer is called from the CRM information recording unit 18 and installed in the call center. It is displayed on the display of the client computer 11. From the information displayed on the display of the client computer 11, the operator of the call center can know the name, age, contract information, etc. of the consumer of the caller.

The client computers 11 and 12 constituting the input means 1 display an HTML document and execute a Java applet by a browser, for example, and the WWW server 1
The data managed by the intermediate system 17 is searched, updated, added, analyzed, and used by the activation of the CGI (Common Gateway Interface) in 3. The client computer 11 is a WWW client that supports the work of a call center operator. The client computer 12 is a WWW client for a consumer to access a support page of a company's homepage or the like and directly access the intermediate system 17.

The call center operator and the consumer use the client computer 1
It is desirable to provide the WWW server 13 so that the intermediate system 17 can be accessed only by the browser without being mounted on the WWW server 1 or 12. The WWW server 13 transmits an HTML document to the client computers 11 and 12 by HTTP in response to a request from the client computers 11 and 12. The WWW server 13 has, for example, Java.
The application developed in the above is executed, an applet is downloaded to the client computers 11 and 12, and the client computer 11, 12 and the intermediate system 17 are configured to be able to cooperate with each other by executing the applet. It is desirable to connect an ORB (Object Request Broker) as an interface between the two. Also,
In response to a request from the client computers 11, 12, W
The CGI is started in the WWW server 13 to access the intermediate system 17, and the obtained data is stored in the WWW server 13.
May be attached to the HTML document and returned to the client computers 11 and 12.

The analysis unit 4 and the analysis system 14 as the output unit 5 are provided with a CRM information recording unit 18 and a customer response history recording unit 1 by data mining using, for example, SAS.
Based on the data managed by 9, it analyzes the regularity hidden in customer behavior and outputs useful information for marketing planning to a display or printer.

Hereinafter, a method of converting an inbound into a psychological code and recording it using the customer information management system shown in FIG. 6 will be described. The conversion from the inbound to the psychological code is performed by any of the following methods. The first method is a method in which an operator or a consumer inputs an inbound code using an input device such as a keyboard of the client computers 11 and 12 and simultaneously converts the inbound into a psychological code. The second method is to convert the inbound into a psychological code by text mining. In the third method, an operator directly inputs a psychological code using an input device such as a keyboard of the client computer 11.

Here, the first method will be described. In the first method, an outbound inquiry screen for inbound is displayed on the client computers 11 and 12 using an HTML document, and an HTML message describing the outbound is displayed.
When the WWW server 13 receives a transmission request from the client computers 11 and 12 for the L document,
The psychological code associated with the HTML document in which the outbound is described is recorded in the customer response history recording unit 19. From WWW server 13 to client computer 1
FIG. 8 shows an outbound inquiry document 30 for inbound of the HTML documents transmitted to the browsers 1 and 12 and displayed on the browser. The screen shown in FIG. 8 is another HTM
It is composed of a drill-down section 31, a FAQ section 33 and a topic section 34 in which a character string in which link information to an L document is embedded is arranged, and a text box 32 for inputting a keyword for searching for an answer.

In the drill-down section 31, the FAQ section 33 and the topic section 34, the operator or the consumer selects a character string in which the URL (Uniform Resource Locator) of the linked document is embedded, and the operator listens to the standardized inbound. An HTML document describing an inbound or inbound similar to the content that the consumer wants to ask is reached. FIG. 9 schematically shows a state of a link from the drill-down section 31 to the target HTML document 38. In the example shown in FIG. 9, by hierarchically linking a document displaying a large inbound section to a document displaying a small section, the operator gradually grasps the content of the question or accesses the corporate homepage. The user digs into the contents of the question while searching for multiple HTML documents 30, 35, 36
The standard inbound that is most similar to the content of the question can reach the HTML document 37 in which one or more are described. H with standard inbound
The TML document 37 is provided with a link to an HTML document 38 in which outbound is described.
When the HTML document 38 describing the desired outbound is specified, that is, when the link to the HTML document 38 describing the desired outbound is selected by the operator or the consumer, the intermediate system 17 executes the CGI.
Alternatively, the mental code is specified by referring to the table shown in FIG. 4 of the inbound management unit 20 in cooperation with the Java applet. When the psychological code is specified for the incoming inbound call, the psychological code is recorded in the customer response history recording unit 19, so that the inbound is converted to a psychological code and recorded in association with the inbound source customer. . At this time, the psychological code and the incoming time are added as a new record to the table shown in FIG. 5 together with the customer code. As described above, since the inbound is automatically converted into the psychological code using the computer system, the inbound is quickly and consistently converted into the psychological code without omission.

Next, the second method will be described. The text box 32 receives an input of an arbitrary character string as inbound. Client computer 11, 1
2. The WWW server 13 and the intermediate system 17 are CGI,
Cooperate using a Java applet or the like, perform text mining on an input character string and a pre-recorded standard inbound, specify one or more sets of standard inbounds whose contents are similar to the input character string and their URLs, A WWW server 1 that generates an HTML document in which a list of links to the HTML document in which the standard inbound is described is embedded
3 to the WWW clients 11 and 12. This HTML document is, for example, an HTML document 37 shown in FIG. When a link to an HTML document in which a target standard inbound or outbound is described is selected from the link list displayed on the display of the client computers 11 and 12, the intermediate system 17 changes the table of the inbound management unit 20 as described above. Identify the psychological code by referring to it. Here, a plurality of standard inbounds are extracted without extracting one standard inbound by text mining, and an operator or a consumer is allowed to select a target standard inbound from the plurality of standard inbounds. The standard inbound and the corresponding outbound are extracted, and the procedure for making the operator or the consumer select the standard inbound may be omitted.

In the second method, the inbound is standardized and input as described above, and at the same time, the inbound is converted to a psychological code. In addition, the inbound is once standardized as it is as character data, voice data or image data. Instead, the data may be recorded as raw data, the audio data and the image data may be converted into character data, and then the raw data may be standardized by text mining based on inbound data accumulated in the past. That is, as shown in FIG.
By extracting one or more standard inbounds, the inbounds may be converted into a standard inbound code and a psychological code and recorded in the customer response history recording unit 19.

As described above, the method of converting the inbound into the psychological code and recording it using the customer information management system according to the present embodiment has been described. Next, an example of an analysis method using a psychological code recorded using the customer information management system according to the present embodiment will be described.

The purpose of the analysis using the customer information management system according to the present embodiment is roughly the following three points. (1) To grasp the current situation of customers or consumers (markets) More specifically, to grasp the market segments that have emerged from past corporate activities and the true outcomes that have emerged. (2) Understand the marketing factors that have sent out the current state of the market. More specifically, it is to identify the process and the structure of the factors that have produced results, and measure the result creation effect (ROI). (3) Understand the marketing factors for realizing future business objectives More specifically, predict the outcome of the current market communication formation, identify the most effective market approach procedure, and identify the most effective market approach. Is to identify a combination of strategic marketing factors.

In the analysis for achieving the first object, cluster analysis, discriminant analysis, factor analysis, principal component analysis, and the like are mainly used for the correlation between the inbound code, the psychological code, and the transaction data. Here, the transaction data refers to various information that varies depending on the specific consumption behavior of the customer, such as sales for each product or service, monthly billing amount, and the like. In the analysis for achieving the second object, regression analysis, analysis based on the first rule of quantification, pattern analysis, and the like are mainly used for the correlation between inbound codes, customer attributes, and transaction data. Pattern analysis is based on decision trees, neural networks, and the like. Here, the customer attributes are, for example, age, gender, occupation, and the like. In the analysis for achieving the third object, regression analysis, analysis based on the first rule of quantification, pattern analysis, and the like are mainly used for the correlation between psychological codes, customer attributes, and transaction data.

According to the purposes (1) to (3), the analysis system 14 is operated using these various analysis methods to identify segments and factors, and to analyze a psychological code for these segments and factors. As a result, managers and the like can immediately obtain the latest information useful for marketing planning.

The information obtained by analyzing the customer information including the AIDMA code is different from the information obtained by analyzing the customer information not including the AIDMA code in the following point. In other words, the former indicates at which stage the consumer is present before reaching a specific transaction behavior, while the latter does not. This will be specifically described with reference to FIG. FIG. 10A is a graph in which the inbound AIDMA code for a specific segment is plotted on the time axis for each customer, and FIG. 10B is a graph summarizing the AIDMA code of the customer of this segment for each time section. This is the table shown. The time division is, for example, a division of time by day, week, and month. By recording the incoming time together with the inbound time, the totalization for each time section becomes possible. "At" shown in the graph is "attention",
“I” indicates “interest”, “D” indicates “decision”, “M” indicates “memory”, and “Ac” indicates that there is an inbound incoming call that is converted into a psychological code of “behavior”. From these graphs and tables, we can see the hidden market movements in this segment. Hidden movements are changes in the psychology of the consumer's trading behavior, and are movements in the market that are not represented as specific trading behaviors such as contracts and cancellations. By considering the psychological changes related to the consumer's transaction behavior in conjunction with the outcome of the transaction behavior, managers can plan marketing that leads consumers to favorable transaction behavior or avoids undesirable transaction behavior. The preferable transaction behavior here is, for example, a contract change to a rate plan with a high rate of return, an increase in call time, and the like. Unwanted trading behavior is, for example, cancellation. By analyzing the psychological code for various segments, information useful for management can be obtained. The segmentation can be performed taking into account any marketing related objects such as products, elapsed time after contract, advertising type, means of communication, category code, and the like.

For example, if it can be read from the graph and table of FIG. 10 that the consumer's psychology is focused on attention and interest in a specific segment and the number of decisions is smaller than those in the segment.
It is possible to plan specific marketing measures such as enhancing information provision on purchasing methods for the segment and enhancing information provision on detailed parts of products or services.

[Brief description of the drawings]

FIG. 1 is a schematic diagram for explaining inbound standardization according to an embodiment of the present invention.

FIG. 2 is a schematic diagram for describing inbound standardization according to an embodiment of the present invention.

FIG. 3 is a block diagram showing a customer relationship management system according to one embodiment of the present invention.

FIG. 4 is a schematic diagram showing a table managed by an inbound management unit according to one embodiment of the present invention.

FIG. 5 is a schematic diagram showing a table managed by a customer response history recording unit according to one embodiment of the present invention.

FIG. 6 is a block diagram showing a customer relationship management system according to one embodiment of the present invention.

FIG. 7 is a schematic diagram showing a table managed by a CRM information recording unit according to one embodiment of the present invention.

FIG. 8 is a schematic diagram for explaining a customer relationship management method according to an embodiment of the present invention.

FIG. 9 is a schematic diagram for explaining a customer relationship management method according to an embodiment of the present invention.

FIG. 10 is a schematic diagram for explaining a customer relationship management method according to one embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Input means 2 Processing means 3 Recording means 4 Analysis means 5 Output means 11 Client computer 12 Client computer 13 Server 14 Analysis system 16 CTI system 17 Intermediate system 18 Information recording part 19 Customer correspondence history recording part 20 Inbound management part 21, 22, 23 Database Server

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) H04M 11/00 302 G10L 3/00 551A

Claims (28)

[Claims] 1. An input step of inputting an inbound to a computer, and using the computer, according to a consumer sentiment corresponding to the input inbound, the consumer sentiment for goods or services that change step by step until a transaction is made. Converting the inbound into a predetermined first psychological code and recording the first psychological code in association with the consumer. 2. The consumer sentiment for goods or services is divided into five stages in chronological order, and in the recording step, the inbound corresponding to the consumer sentiment for each stage for goods or services is defined by five first mental codes. 2. The customer relationship management method according to claim 1, wherein the conversion is performed to any one of them. 3. The consumer psychology of a product or service is divided into an attention stage, an interest stage, a decision stage, a memory stage, and an action stage in chronological order, and inbounds emitted at each stage are respectively converted into different first psychological codes. 3. The customer relationship management method according to claim 2, wherein the conversion is performed. 4. In the recording step, inbound related to advertising, inbound related to goods or services provided within a predetermined period, inbound related to a method of trading goods or services, inbound related to use of goods or services, and goods or services. 4. The customer relationship management method according to claim 3, wherein the inbound related to the transaction procedure or the selection is converted into first psychological codes different from each other. 5. In the recording step, the inbound is converted to a predetermined second psychological code according to the consumer psychology corresponding to the input inbound, which is a consumer psychology for transmitting a question, a complaint, a request or an opinion. The customer relationship management method according to any one of claims 1 to 4, wherein the conversion is performed and the second psychological code is recorded in association with the customer. 6. In the recording step, inbound of a question about a product or service, inbound based on negative psychology to a company, inbound of a request for a matter that has not yet been realized and for which an effort to realize is to be considered, and 6. The customer relationship management method according to claim 5, wherein inbound requests for items that have not yet been realized and for which no effort is being made are converted into different second psychological codes. 7. In the recording step, the input inbound is converted by analyzing the inbound and the input inbound recorded in association with the first mental code and / or the second mental code by text mining. The customer relationship management method according to claim 1, wherein the first psychological code and / or the second psychological code to be specified are specified. 8. The method according to claim 1, wherein in the inputting step, the inbound call voice data is recognized using a computer, converted into character data, and the inbound is input. Customer relationship management method. 9. In the inputting step, an inbound is input by selecting an inbound that is the same as or similar to the incoming inbound, which is an inbound recorded in association with the psychological code, and the incoming call is received in the recording step. Inbound,
The customer relationship management method according to claim 1, wherein the conversion is performed to a psychological code associated with the selected inbound. 10. The recording method according to claim 1, wherein in the recording step, information for specifying when the inbound is input is recorded in association with the customer using a computer. Customer relationship management method. 11. An input means for inputting an inbound, and a consumer sentiment for a product or service that changes step by step until a transaction, wherein the inbound is determined in accordance with a consumer sentiment corresponding to the input inbound. A customer relationship management system, comprising: processing means for converting into one psychological code; and first recording means for recording the first psychological code in association with a consumer. 12. The processing means according to claim 5, wherein the inbound corresponding to the consumer sentiment of the goods or the service at each stage is determined.
The customer relationship management system according to claim 11, wherein the conversion is made to one of the first psychological codes. 13. The processing means includes: inbound related to advertising, inbound related to goods or services provided within a predetermined period, inbound related to a method of trading goods or services, inbound related to use of goods or services, and goods or services. 2. The method according to claim 1, wherein the inbound related to the transaction procedure or the selection is converted into first psychological codes different from each other.
2. The customer relationship management system according to item 2. 14. The processing means according to claim 1, wherein said inbound is converted into a predetermined second psychological code in accordance with consumer sentiment corresponding to the input inbound. 14. The method according to claim 11, wherein the recording unit records the second psychological code in association with a customer.
Customer relationship management system described in. 15. The processing means includes: inbound of a question about a product or service, inbound based on negative psychology to a company, inbound of a request for a matter that has not yet been realized and for which an effort to realize is to be considered, and 15. The customer relationship management system according to claim 14, wherein inbound requests for requests that have not yet been realized and for which no effort is being made are converted into different second psychological codes. 16. An inbound and a second recording means for recording the first mental code and / or the second mental code associated with the inbound, and the inbound recorded on the input inbound and the second recording means. Mining means for analyzing the input by inbound text to identify the first psychological code and / or the second psychological code to convert the input inbound. The customer relationship management system according to claim 1. 17. The customer relationship according to claim 11, further comprising conversion means for recognizing inbound call voice data, converting the data into character data, and outputting the character data to the input means. Management system. 18. The apparatus according to claim 18, further comprising third recording means for recording an inbound and a first mental code and / or a second mental code associated with the inbound, wherein the input means is the same as or similar to the incoming inbound. The inbound is input by selecting an inbound from the third recording means, and the processing means converts the incoming inbound into a psychological code associated with the selected inbound. A customer relationship management system according to any one of the preceding claims. 19. The processing means outputs information for specifying when an inbound is input, and the recording means records information for specifying when an inbound is input in association with a customer. Item 19. The customer relationship management system according to any one of Items 11 to 18. 20. An input procedure for inputting an inbound, and a consumer sentiment for a product or service that changes step by step until a transaction, wherein the inbound is set to a predetermined first time in accordance with the consumer sentiment corresponding to the input inbound. A customer relationship management program characterized by causing a computer to execute a processing procedure of converting the first psychological code into a psychological code, and a recording procedure of recording the first psychological code in association with a consumer. 21. The customer relationship management according to claim 20, wherein in the processing procedure, inbound corresponding to consumer sentiment at each stage of the product or service is converted into one of five first psychological codes. program. 22. In the processing procedure, inbound for an advertisement, inbound for a product or service provided within a predetermined period, inbound for a transaction method of the product or service, inbound for a method of using the product or service, and 22. The customer relationship management program according to claim 21, wherein the inbound service transaction procedure or selection is converted into mutually different first psychological codes. 23. In the processing procedure, a question, a complaint,
Convert the inbound to a predetermined second psychological code according to the consumer psychology corresponding to the inbound that is the consumer psychology to transmit the request or opinion, in the recording procedure, in the recording procedure, the second psychological code 22. Recording according to a consumer.
Or the customer relationship management program according to 22. 24. In the processing procedure, inbound of a question about a product or service, inbound based on negative psychology to a company, inbound of a request for a matter that has not yet been realized and for which an effort to realize is to be considered, and 24. The customer relationship management program according to claim 23, wherein inbound requests for items that have not yet been realized and for which no effort is being made are converted into mutually different second psychological codes. 25. A first psychological code to convert an input inbound by analyzing the inbound and the input inbound recorded in association with the first psychological code and / or the second psychological code by text mining. The customer relationship management program according to any one of claims 20 to 24, wherein the program causes a computer to execute a mining procedure for specifying a second psychological code. 26. The computer-readable recording medium according to claim 20, wherein a computer performs a conversion procedure of recognizing inbound call voice data, converting the data into character data, and outputting the character data to the input means. Customer relationship management program. 27. In the input procedure, an inbound is input by selecting an inbound which is the same as or similar to the incoming inbound and which is recorded in association with the psychological code, and in the recording procedure, The customer relationship management program according to any one of claims 20 to 24, wherein the inbound is converted into a psychological code associated with the selected inbound. 28. The method according to claim 20, wherein in the processing procedure, a date when the inbound is input is output, and in the recording procedure, information for specifying a time when the inbound is input is recorded in association with the customer. 28. The customer relationship management program according to any one of Items 27 to 27.