JP6906810B2 - Sales support equipment, programs, and sales support methods - Google Patents

Description

The present invention relates to a technique for supporting the business of a business entity such as a financial institution.

The technology that supports the sales of the business entity is known. For example, Patent Document 1 describes that, regarding sales support in a financial institution, when a search condition is input, a customer is extracted, a map of the location of the customer is output, and correction of the location is accepted. Patent Document 2 describes that a customer attribute including a history of explaining important matters and a daily business report are associated and stored in a business support system of a financial institution.

Patent No. 5497852 Japanese Patent Application Laid-Open No. 2008-181553

However, the technologies of Patent Document 1 and Patent Document 2 can comprehensively analyze accumulated data (customer data, sales entity data, product data, result data, etc.) related to customer sales activities to support sales, etc. There wasn't.

On the other hand, the present invention provides a technique capable of comprehensively analyzing accumulated data related to customer sales activities to support sales and the like.

The present invention relates to (a) customer data including at least one of customer trends and customer attributes for each of a plurality of customers in an entity having at least one product, and (b) for each of the plurality of business entities in the entity. , Business entity data including at least one of the business entity trends and business entity attributes, (c) product data including at least one of the product trends and product attributes for each of the at least one product, and (d) the business entity. In the result of correlation analysis of the customer data, the business entity data, the product data, and the result data, the receiving means for accepting the designation of the objective variable from the result data including at least the transaction history of the plurality of customers. A specific means for identifying a target person from the plurality of business entities based on a specific explanatory variable having a relatively high correlation with the designated objective variable, and the designated objective variable and the designated objective variable for the target person. Provided is a sales support device having an output means for outputting a proposal according to a combination with the specific explanatory variable.

At least a part of the customer data, the business entity data, the product data, and the result data is used as candidate data for explanatory variables, and at least a part of the customer data, the business entity data, the product data, and the result data. May be used as candidate data for the objective variable, and an analytical means for performing the correlation analysis on the explanatory variable and the objective variable may be provided.

The analysis means may perform the correlation analysis using a trained model in which the candidate data of the explanatory variable is given to the input layer and the candidate data of the objective variable is given to the output layer as teacher data and machine learning is performed. good.

The specific explanatory variable is a variable included in the customer data, and the specific means targets a business entity corresponding to a customer whose value of the specific explanatory variable is better than a predetermined criterion among the plurality of business entities. It may be specified as a person.

The proposal may include information that identifies the good customer.

The specific explanatory variable is a variable included in the business entity data, and the specific means identifies a business entity whose value of the specific explanatory variable is worse than a predetermined criterion as the target person among the plurality of business entities. You may.

The proposal may include information for improving the value of the particular explanatory variable.

The specific explanatory variable is a variable included in the product data, and the specific means targets a business entity corresponding to a product in which the value of the specific explanatory variable is better than a predetermined reference among the plurality of business entities. It may be specified as a person.

The proposal may include information identifying the good product.

The specific means may specify the target person from among the plurality of business entities whose designated objective variable values satisfy a predetermined condition.

Even if the objective variable is a variable related to the record of purchasing a specific product among a plurality of the products, and the predetermined condition is a business entity corresponding to a customer who has not purchased the specific product. good.

The objective variable is a variable related to the actual sales of the product, and the predetermined condition may be a condition that the actual performance is worse than the standard in relation to another business entity.

The sales entity may include a salesperson, and the sales entity data may indicate the trends and attributes of the salesperson.

The sales entity data may include a daily report of the sales staff.

The sales entity data may include a telephone call history between the salesperson and the customer.

The sales entity data may include a history of location information of the sales staff.

The sales entity data may include information regarding personnel evaluation of the sales staff.

The sales entity data may include information indicating at least one of the qualifications and skills possessed by the sales staff.

The business entity may be a financial institution, and the customer data may include transaction history and the balance of assets under custody at the financial institution.

Further, the present invention includes, (a) customer data including at least one of customer trends and customer attributes for each of a plurality of customers in an entity having at least one product, and (b) a plurality of customers in the entity. For each of the business entities, the business entity data including at least one of the business entity trends and the business entity attributes, (c) for each of the at least one product, the product data including at least one of the product trends and the product attributes, and (d). ) Correlation analysis of the step of accepting the designation of the objective variable from the result data including at least the transaction history of the plurality of customers in the business entity, the customer data, the business entity data, the product data, and the result data. Based on the specific explanatory variable having a relatively high correlation with the designated objective variable in the result of the above, the step of identifying the target person from the plurality of business entities and the designated person for the target person. Provided is a program for executing a step of outputting a proposal according to a combination of an objective variable and the specific explanatory variable.

Further, the present invention comprises (a) customer data including at least one of customer trends and customer attributes for each of a plurality of customers in an entity having at least one product, and (b) a plurality of business entities in the entity. For each, the business entity data including at least one of the business entity trends and the business entity attributes, (c) for each of the at least one product, the product data including at least one of the product trends and the product attributes, and (d) the sales. In the step of accepting the designation of the objective variable from the result data including at least the transaction history of the plurality of customers in the body, and the result of the correlation analysis of the customer data, the business entity data, the product data, and the result data. A step of identifying a target person from the plurality of business entities based on a specific explanatory variable having a relatively high correlation with the designated objective variable, and the designated objective variable and the designated objective variable for the target person. Provided is a sales support method having a step of outputting a proposal according to a combination with the specific explanatory variable.

According to the present invention, it is possible to comprehensively analyze the accumulated data related to customer sales activities and support sales and the like.

The figure which shows the outline of the sales support system 1 which concerns on one Embodiment. The figure which illustrates the functional structure of the sales support system 1. The figure which illustrates the detail of the analysis means 112. It is a figure which shows the relationship between various data and a correlation analysis part. The figure which illustrates the hardware configuration of the server apparatus 10. The figure which illustrates the hardware configuration of the client apparatus 20. The flowchart which illustrates the operation which concerns on one Embodiment of the sales support system 1. The flowchart which illustrates the operation which concerns on one Embodiment of the sales support system 1. The figure which illustrates the analysis in the correlation analysis unit 1122. The figure which illustrates the output screen of a proposal. The figure which illustrates the output screen of a proposal.

1. 1. Configuration FIG. 1 is a diagram showing an outline of a sales support system 1 according to an embodiment. The sales support system 1 is a system for providing information for supporting sales to a business entity in the business entity. Specifically, the sales support system 1 outputs a proposal for improving the results of the business entity, more specifically, a proposal for improving the sales activity of the business entity to the sales entity. For example, a salesperson (an example of a business entity) of a financial institution such as a bank (an example of a business entity. Here, a business entity means an entity that conducts business activities, specifically, a company, a company, a partnership, an organization, etc.) Here, a business entity is an element that constitutes a business entity, and specifically, a business entity (for example, a department, a branch office, or an individual salesperson of a company) is used to solicit financial products from a customer. It is a support system. That is, the user of the sales support system 1 is, for example, a salesperson of a financial institution and a manager who manages these salespersons. In this example, the sales support system 1 has a server device 10 and a client device 20. Of these, the client device 20 is used by the user. Although only a single client device 20 is shown in FIG. 1 for simplification of the drawing, the sales support system 1 may include a plurality of client devices 20. The server device 10 is connected to the client device 20 used by the user via the network 3.

FIG. 2 is a diagram illustrating the functional configuration of the sales support system 1. The sales support system 1 includes a receiving means 131, a specific means 132, a storage means 105, an analysis means 112, an output means 121, a communication means 201, an input means 202, a display means 203, a control means 204, and a storage means 205. Among these functional elements, the receiving means 131, the specifying means 132, the storage means 105, the analysis means 112, and the output means 121 are mounted on the server device 10. The communication means 201, the input means 202, the display means 203, the control means 204, and the storage means 205 are mounted on the client device 20.

The receiving means 131 accepts the designation of the objective variable from the data group including (a) customer data, (b) business entity data, (c) product data, and (d) result data. Hereinafter, among the plurality of candidates for the objective variable, the variable designated by the receiving means 131 is referred to as a “designated objective variable”.

Customer data refers to data that includes at least one of customer trends and customer attributes. The customer trend refers to the trend of the customer related to the business of the financial institution that is the user of the sales support system 1. For example, the transaction history (for example, the customer, Mr. Taro Yamada purchased the product A on XX, △ Includes information on items such as purchasing product B on the △ day of the month, and depositing ◇◇ yen into the account on the xth month. The customer attribute refers to the attribute of the customer, and includes, for example, information on the age, gender, occupation, place of residence, family structure, annual income, and the like of the customer.

Business entity data refers to data that includes at least one of salesperson data and department / branch data. Salesperson data refers to data that includes at least one of salesperson trends and salesperson attributes. The salesperson trend refers to the trend of the salesperson regarding the business, for example, the daily report of the salesperson, the call history by telephone with the customer (for example, the salesperson, Ichiro Suzuki calls Mr. Q on △ month △ day, etc.), Contact history with customers (for example, salesperson Ichiro Suzuki visited Mr. P's house on October 0), number of times the salesperson visited the customer during a specific period (for example, yearly, monthly, weekly, etc.) (hereinafter, simply Includes at least one of (referred to as "visits") and the history of salesperson location information. A sales employee is an employee who engages in sales positions in a business entity. The salesperson attribute refers to the attribute of the salesperson regarding the business, and includes, for example, information on personnel evaluation of the salesperson, qualification, skill, position, evaluation, work history, age, gender, and the like. The department / branch data refers to data including at least one of the department / branch trends and the department / branch attributes. A department branch is a department or branch as an organization that constitutes a financial institution. The department branch trend refers to the trend of the department branch related to business, and includes, for example, a set of sales staff trends in the department branch. The department branch attribute refers to the attribute of the department branch related to business, and includes, for example, the number of sales staff in the department branch, age distribution, evaluation distribution, location, and the like.

Of the salesperson data, for example, at least a part of the data input by the salesperson as a daily report is used for the salesperson trend. That is, the salesperson trend data is input or added by the salesperson every day. The salesperson attribute is input or changed, for example, when the attribute of the salesperson is changed. Specifically, the attributes of a salesperson are changed when a salesperson acquires a certain qualification or acquires a new skill. The salesperson attribute includes information that identifies the department or branch to which the salesperson belongs. Department branch data is input or generated when the department or branch is created. Of the department branch data, the department branch trends can be obtained from the set of sales staff trends of the sales staff belonging to the department branch. The department / branch attribute is input or changed when the attribute of the department / branch changes. Specifically, when a branch moves, the department / branch attributes are changed.

Product data refers to data that includes at least one of product trends and product attributes. The product trend refers to a trend of a product related to the business of a financial institution that is a user of the sales support system 1. For example, a transaction history (for example, customer Y purchases product A on XX, customer Z is Δ). (Purchase product B on the △ day of the month, etc.) is included. The product attribute refers to the attribute of the product related to the business, and includes, for example, the characteristics of the product, the currency, the risk, and the investment period. The term "commodity" here is a concept that includes services, that is, services.

The result data refers to data showing the results of the business of the financial institution, and refers to data including at least one of the customer status, the sales staff status, the department / branch status, and the product status, for example. The customer status refers to the status of customers related to the business of financial institutions, such as contract volume, profit / loss, turnover rate, transaction volume ranking (or transition), transaction amount up to now (hereinafter referred to as "transaction amount"), and current status. Includes at least one of the number of transactions up to (hereinafter referred to as "number of transactions"), transaction history, amount of assets under custody (or financial assets held), and potential assets. The assets under custody refer to the assets or the like that the financial institution or the like that is the user of the sales support system 1 has received the deposit from the customer. Potential assets refer to assets held by customers other than assets under custody. The salesperson status refers to the status of salespeople related to the business of a financial institution, and includes, for example, at least one of sales ranking (or transition) and operating results. The operating results refer to the sales amount of goods, the amount of assets under custody, the amount of income from commission income, the pass / fail result of the qualification test, the contact behavior history, the coverage rate, income, the number of customers, and the like. The department branch status refers to the status of department branches related to the business of a financial institution, and includes, for example, at least one of sales ranking (or transition), sales performance, contact behavior history, coverage rate, income, and number of customers. The product status refers to the status of products related to the business of a financial institution, and includes, for example, at least one of sales amount ranking (or transition), sales, amount of assets under custody, customer attributes held, and profit or loss.

In this example, the receiving means 131 includes an input means 101, an input means 102, an input means 103, and an input means 104. The input means 101 (an example of the first input means) accepts the input of customer data. The input means 102 (an example of the second input means) accepts the input of the business entity data. The input means 103 (an example of the third input means) accepts input of product data. The input means 104 (an example of the fourth input means) accepts the input of the result data.

The objective variable is a dependent variable in a mathematical model, and its value is determined according to the value of the independent variable in the mathematical model. In the present embodiment, the mathematical model includes customer data, business entity data, product data, and a model showing the correlation between a part of the result data and another part. The objective variable is a variable to be focused on when generating a proposal to a business entity. The independent variable in this mathematical model is called an explanatory variable.

The storage means 105 stores various data and programs. For example, the storage means 105 stores data input via the input means 101, the input means 102, the input means 103, and the input means 104 in the database. The customer data, the business entity data, the product data, and the result data are each input with a predetermined event as a trigger, for example, when each data is updated as a trigger. As an example, customer data is input when a new customer is registered. Customer data, business entity data, product data, and result data are input independently. That is, these four types of data do not have to be input at the same time. Further, at least one of these four types of data may be input from another system (for example, a core banking system or a CRM (Customer Relationship Management) system).

The analysis means 112 uses at least a part of customer data, business entity data, product data, and result data as candidate data for explanatory variables, and at least a part of these as candidate data for objective variables, and uses these candidate data as a mathematical model. (Or machine learning model) is applied to perform correlation analysis. This mathematical model includes at least one of a classification model and a regression model. The analysis means 112 performs correlation analysis (for example, regression analysis) on all items (or variables) that can be explanatory variables and all items (or variables) that can be objective variables, and extracts candidates for explanatory variables. At this time, the analysis means 112 performs a correlation analysis (for example, regression analysis) between all the candidates of the explanatory variable and all the candidates of the explanatory variable other than itself, and correlates the items that are the explanatory variables of one objective variable. Select the explanatory variable so that there is no. In one example, customer data, business entity data, and product data are candidate data for explanatory variables, and result data are candidate data for objective variables. Correlation analysis refers to an analysis of the correlation between a candidate for an explanatory variable and a candidate for an objective variable. In one example, the analysis means 112 performs the correlation analysis using a trained model in which the candidate data of the explanatory variable is given to the input layer and the candidate data of the objective variable is given to the output layer as teacher data and machine learning is performed. In one example, the learning means 111 gives teacher data to the mathematical model and causes the analysis means 112 to learn. At a predetermined timing, the learning means 111 gives new data such as customer data, sales entity data, and product data to the input layer, and the corresponding result data to the output layer as teacher data, respectively, to the analysis means 112. Let them learn.

FIG. 3 is a diagram illustrating the details of the analysis means 112. The analysis means 112 includes a hypothesis database 1121, a correlation analysis unit 1122, a support information database 1123, and a support information generation unit 1124.

Hypothesis database 1121 records hypothesis data. Hypothesis data refers to data that records hypotheses related to business behavior, and more specifically, hypotheses related to the correlation between explanatory variables and objective variables. In one example, the hypothetical data includes data that "a salesperson who has many contacts with customers has good results." In this example, the "number of contacts with the customer" is the explanatory variable, and the "salesperson's performance" is the objective variable. In another example, the hypothetical data includes data that "a customer who purchases product A has an annual income of 10 million yen or more." In this example, "customer's annual income" is the explanatory variable, and "purchase history of product A" is the objective variable. The hypothesis data is input by, for example, a user of the sales support system 1.

FIG. 4 is a diagram showing the relationship between various data and the correlation analysis unit. The correlation analysis unit 1122 learns using the given teacher data and analyzes the correlation between the candidate of the explanatory variable and the candidate of the objective variable.

The support information database 1123 is a database that records information for sales support, and records, for example, templates or drafts of various proposals. The support information generation unit 1124 selects an appropriate proposal template or draft from the information recorded in the support information database 1123 using the analysis result by the correlation analysis unit 1122, and processes it.

The identification means 132 identifies a target person from a plurality of business entities based on a specific explanatory variable. The specific explanatory variable is a variable included in the candidate data of the explanatory variable whose correlation with the designated objective variable satisfies a predetermined condition (for example, the correlation is relatively high) in the result of the correlation analysis.

The output means 121 outputs a proposal according to the combination of the designated objective variable and the specific explanatory variable to the target person. This proposal is a proposal to improve the sales activities of the target person.

The client device 20 is a device (user terminal) that provides a UI to the user. The communication means 201 communicates with the server device 10. The data communicated with the server device 10 includes data indicating information input in the client device 20 or data indicating an instruction from the user. The input means 202 receives input of new data (including at least one of customer data, business entity data, product data, and result data) from the user. The display means 203 displays information to the user. In this example, the information displayed by the display means 203 includes information on the business entity or the business activity of the business entity based on the analysis result transmitted from the server device 10. Information about a sales activity includes, for example, a proposal to a salesperson or a proposal to at least one of the salesperson's manager and the sales entity management department. The sales entity management department is a department that supervises the sales policies, etc. of all sales entities. The storage means 205 stores various data and programs. The control means 204 controls other functional elements.

FIG. 5 is a diagram illustrating a hardware configuration of the server device 10. The server device 10 is a computer device having a CPU (Central Processing Unit) 151, a memory 152, a storage 153, and a communication IF (Interface) 154. The CPU 151 is a control device that executes a program, performs various operations, and controls other hardware elements in the server device 10. The memory 152 is a main storage device that functions as a work area when the CPU 151 executes a program. The storage 153 is a non-volatile auxiliary storage device that records various programs and data. The communication IF 154 is a communication device that communicates with another device according to a predetermined communication standard (for example, Ethernet (registered trademark)).

In this example, the storage 153 records a program (hereinafter, referred to as “server program”) for causing the computer device to function as the server device 10 in the sales support system 1. When the CPU 151 executes the server program, the function of FIG. 2 is implemented in the computer device. At least one of the memory 152 and the storage 153 is an example of the storage means 105 in a state where the CPU 151 is executing the server program. Further, the CPU 151 is an example of an input means 101, an input means 102, an input means 103, an input means 104, a learning means 111, an analysis means 112, and an output means 121.

FIG. 6 is a diagram illustrating a hardware configuration of the client device 20. The client device 20 is a computer device having a CPU 251 and a memory 252, a storage 253, a communication IF 254, a display device 255, and an input device 256, and is, for example, a personal computer, a smartphone, or a tablet terminal.

In this example, the storage 253 stores a program (hereinafter, referred to as “client program”) for causing the computer device to function as the client device 20 in the sales support system 1. When the CPU 251 executes the client program, the function of FIG. 2 is implemented in the computer device. The communication IF 254 is an example of the communication means 201 while the CPU 251 is executing the client program. The CPU 251 is an example of the input means 202 and the control means 204. The display device 255 is an example of the display means 203. At least one of the memory 252 and the storage 253 is an example of the storage means 205.

In this example, the client program is, for example, a general-purpose web browser, and the client device 20 accesses the server device 10 via the web browser. The server device 10 is equipped with a so-called web application function, and the function described below is provided via a web browser.

2. Operation FIGS. 7A and 7B are flowcharts illustrating an operation according to an embodiment of the sales support system 1. The operation of the sales support system 1 is divided into, for example, a preparation stage (FIG. 7A) and a usage stage (FIG. 7B). The preparatory stage includes a process of adjusting the parameters of the mathematical model, for example, a process of giving teacher data to the mathematical model and performing machine learning. The usage stage includes a process using this mathematical model, for example, a process of using a trained model to output a proposal according to a combination with a designated objective variable and a specific explanatory variable.

The flow shown in FIG. 7A is started, for example, when a predetermined event occurs. The specific event referred to here is, for example, an event in which a user logs in to a sales support system. The "user" referred to here is a user in the preparation stage, which may be different from the user in the usage stage described below. In one example, a user in the preparatory stage is an employee of the management department of a financial institution that is a business entity, and a user in the usage stage is a salesperson of a financial institution or its boss. In the following, functional elements such as the input means 101 will be described as the main body of processing, in which hardware elements such as the CPU 151 executing a program such as a server program cooperate with other hardware elements. It means to execute the process.

In step S1, the input means 101, the input means 102, the input means 103, and the input means 104 each receive input of various data. The data input via the input means 101, the input means 102, the input means 103, and the input means 104 is recorded in the database in the storage means 105. Various data are input independently. For example, at one timing, only result data is input, and at another timing, only customer data is input. In one example, various data are input when the data is updated. For example, customer data is input when a new customer is registered.

In step S2, the learning means 111 gives teacher data to the mathematical model in the analysis means 112 and causes machine learning. In this example, since the data input via the input means 101, the input means 102, the input means 103, and the input means 104 includes category data (that is, non-numerical data, for example, the occupation of the customer), all the data are classified into categories. It is converted into data and this category data is given to the mathematical model as teacher data. In one example, the learning means 111 uses all variables included in customer data, business entity data, product data, and result data as candidates for explanatory variables and objective variables, and performs machine learning for all combinations of explanatory variables and objective variables. conduct. The learning in step S2 is executed triggered by a predetermined event. Events that trigger learning include, for example, that new data has been input, that unlearned data has been accumulated up to a predetermined amount, that a predetermined period has passed since the previous learning, or that a user has sent an event. I received an explicit instruction to start learning.

The learning means 111 may narrow down the candidates for the explanatory variables or the candidates for the objective variable by referring to other information. In one example, the learning means 111 narrows down the candidates for the explanatory variables and the objective variables by referring to the hypotheses recorded in the hypothesis database 1121. Specifically, the learning means 111 may perform machine learning using only the variables recorded as the candidates for the explanatory variables and the candidates for the objective variables in the hypothesis database 1121 as the candidates for the explanatory variables and the objective variables, respectively. The correlation analysis unit 1122 may modify the hypothesis recorded in the hypothesis database 1121 or write a new hypothesis in the hypothesis database 1121. The hypothesis is modified or written, for example, according to the user's instruction.

FIG. 8 is a diagram illustrating analysis by the correlation analysis unit 1122. In this example, the correlation analysis unit 1122 performs the analysis according to the hypothesis recorded in the hypothesis database 1121. In this example, the hypothesis database 1121 records the following four hypotheses (a) to (d).
(A) The more frequent visits to customers, the greater the number of contracts concluded.
(B) Customers with higher financial assets are more likely to sell products.
(C) Salespeople who hold FP (Financial Planner) qualifications have a higher investment trust sales amount than salespeople who do not have FP qualifications.
(D) The amount of product sales varies depending on the occupation of the customer.
An example of analysis using these hypotheses will be described below.

In hypothesis a, "the number of visits to customers" is the explanatory variable, and "the number of contracts concluded" is the objective variable. Since all of these are numerical data, regression analysis is used as the analysis method. Regression analysis determines the parameters in the regression model. The process up to the point of determining the parameters is the preparatory stage. If a regression model with this parameter set at the stage of use is used, the number of contracts can be predicted from the number of visits, and conversely, the number of visits required to obtain the number of contracts can be predicted from the number of contracts.

In hypothesis b, "amount of financial assets held" is an explanatory variable, and "presence or absence of product sales" is an objective variable. “Amount of financial assets held” is numerical data, while “whether or not products are sold” is category data. Therefore, discriminant analysis is used as the analysis method. Discriminant analysis determines the parameters in the discriminant model. Using a discrimination model with this parameter set, it is possible to predict whether or not a product will be sold (that is, a prospect) from the amount of financial assets held, or conversely, how much financial assets held will be required to become "with product sales". Can be done.

In hypothesis c, "whether or not a salesperson has FP qualification" is an explanatory variable, and "investment trust sales amount" is an objective variable. "Whether or not a salesperson has FP qualification" is category data, and "Investment trust sales amount" is numerical data. Therefore, quantification type 1 is used as the analysis method. The parameters in the quantification class 1 model are determined by the quantification class 1 analysis. If you use the quantification type 1 model with this parameter set, you can predict the investment trust sales amount from the presence or absence of the FP qualification of the salesperson, and conversely, whether you need the FP qualification to obtain a certain amount of investment trust sales amount. Can be predicted.

In hypothesis d, "customer occupation" is an explanatory variable, and "presence or absence of product sales" is an objective variable. Both "customer occupation" and "presence or absence of product sales" are category data. Therefore, quantification type 2 is used as the analysis method. The parameters in the quantification type 2 model are determined by the quantification type 2 analysis. By using the quantification type 2 model in which this parameter is set, it is possible to predict the presence or absence (that is, the prospect) of product sales from the job type of the customer.

Although the hypotheses a to d have been described as an example in which the explanatory variable and the objective variable are singular, at least one of the explanatory variable and the objective variable may be plural.

See FIG. 7A again. In step S3, the correlation analysis unit 1122 outputs the trained model. Specifically, the correlation analysis unit 1122 stores the parameters of the model obtained in step S2 in the storage means 105. This is the end of the preparation stage. The processing in the preparation stage may be performed again at the timing after the processing in the usage stage described below is started.

FIG. 7B is a flowchart illustrating the processing of the sales support system 1 at the use stage. The process of FIG. 7B is started, for example, when the user logs in to the sales support system 1.

In step S4, the receiving means 131 accepts the designation of the objective variable. The receiving means 131 displays a screen for receiving the designation of the objective variable in the client device 20. This screen contains, for example, a list of objective variables. The user specifies the objective variable from this list. The receiving means 131 accepts the designation of the objective variable by the user. Instead of accepting the designation of the objective variable from the user each time the process is performed, the accepting means 131 may accept the designation of the objective variable from the administrator (or an employee of the management department) in advance. Alternatively, the sales support system 1 may automatically specify the objective variable according to a predetermined algorithm regardless of the user's instruction. In this case, the sales support system 1 designates the objective variable according to the priority specified by the user or determined according to a predetermined algorithm, for example.

In step S5, the identification means 132 identifies a target person from a plurality of business entities. For example, the receiving means 131 displays a screen for receiving the identification of the target person in the client device 20. This screen includes, for example, a list of target person candidates, that is, business entities. The user specifies the target person from this list. In one example, the target person is the user himself. For example, this corresponds to the case where an individual salesperson accesses the sales support system 1 and receives advice. Alternatively, a person different from the user may be the target person. For example, a boss who manages a plurality of sales staff accesses the sales support system 1 and receives advice from the sales staff of his subordinates. The receiving means 131 accepts the designation of the target person by the user. The specifying means 132 identifies the target person according to the information received by the receiving means 131. Instead of accepting the identification of the target person from the user each time the process is performed, the receiving means 131 may accept the identification of the target person from the administrator (or an employee of the management department) in advance. Alternatively, the sales support system 1 may automatically identify the target person according to a predetermined algorithm regardless of the user's instruction. In this case, the sales support system 1 identifies the target person according to the priority specified by the user or determined according to a predetermined algorithm, for example.

In step S6, the support information generation unit 1124 generates a proposal according to the combination of the designated objective variable and the specific explanatory variable. The details are as follows. The support information generation unit 1124 selects an appropriate template or draft from the information recorded in the support information database 1123 based on the analysis result by the correlation analysis unit 1122, processes it, and processes the target items and recommendations. Generate sales behavior support data for classification. The target matter means a matter including any one of the input data, that is, at least one of a customer, a salesperson, a department branch, and a product. The classification of recommendations refers to the classification of the wording of proposals to the target person, for example, (1) customer product trend analysis / support (information on products that should be recommended to the target customer), (2) customer contact status analysis.・ Support (proposals to sales staff or sales staff managers and sales entity management departments), (3) Visit contact efficiency / recommendations (information indicating the efficiency of sales staff's sales activities), ( 4 ) Customer development promotion / recommendations (Information about customers who should recommend the target product), (5) Proposal of wording on compliance (calling attention to sales activities of sales staff), and (6) Degree of correlation / influence between skills, etc. and achievements (sales staff) Information on qualifications or skills recommended for acquisition). Sales behavior support data refers to information related to recommendations selected and processed based on the analysis results. Some specific examples of the recommendations output by the support information generation unit 1124 are shown below. The following specific examples are different from the above hypotheses a to d.

(1) In customer product trend analysis / support, the correlation analysis unit 1122 analyzes the correlation between the combination of product data and customer data and the result data at the preparation stage. The support information generation unit 1124 provides support using this analysis result at the stage of use. For example, the correlation analysis unit 1122 performs correlation analysis using customer trends and customer attributes as explanatory variables and the sales of a specific product (for example, the product with the highest sales amount in the product status) as the objective variable. This analysis is a quantification type 1 analysis illustrated in Hypothesis c. This analysis reveals customer trends and customer attributes (here referred to as "specific trends / attributes") that are highly correlated with sales of this product. Based on the analysis result, the support information generation unit 1124 extracts the customers in charge of the target person who have a specific trend / attribute but have not yet purchased this product as potential customers. The support information generation unit 1124 selects an appropriate proposal template or draft, processes it, and generates specific sales behavior support data. For example, the sales behavior support data is "(Salesperson) Mr. X, recently, product A, which has the highest sales in all stores, is purchased by a customer who is older than XX, occupation △△, and product B. (Sales staff) Among Mr. X's customers, why don't you make a proposal to Mr. P, Mr. Q, and Mr. R? "

(2) In customer contact status analysis / support, the correlation analysis unit 1122 analyzes the correlation between the salesperson data (more specifically, the salesperson trend) and the result data at the preparation stage. The support information generation unit 1124 provides support using this analysis result at the stage of use. For example, the correlation analysis unit 1122 performs correlation analysis using the salesperson trend as an explanatory variable and the operating performance (specifically, for example, the amount of assets under custody) as an objective variable. This analysis is a quantification type 1 analysis illustrated in Hypothesis c. This analysis reveals salesperson trends (herein referred to as "specific trends") that are highly correlated with sales performance. The support information generation unit 1124 generates a warning regarding a specific trend based on the analysis result. For example, when the correlation analysis unit 1122 clarifies by analysis that the number of monthly contacts with the customer is a specific trend, the number of monthly contacts with the customer for the target person (an example of an index of the specific trend) is further determined. When it is found that the comparison with other sales staff (for example, the minimum number of monthly contacts of the sales staff with the top 10% of sales results) is inferior, the support information generation unit 1124 will be based on the analysis result. Then, select an appropriate proposal template or draft, process it, and generate specific sales behavior support data. For example, the sales behavior support data generates sales behavior support data for calling attention to increase the number of monthly contacts to the average number of sales staff with the top 10% of sales results. Specifically, the sales behavior support data shows that "(Salesperson) X's monthly contact with customers is □□ times, which is △△% lower than the average of ○○ times for everyone with the highest sales performance. In order to improve business performance, please make at least XX monthly contacts. " Operating results refer to the amount of assets under custody, the amount of commission income, etc.

(3) Visit contact efficiency ・ In the proposal, the correlation analysis unit 1122 analyzes the correlation between the department branch data and the result data at the preparatory stage. The support information generation unit 1124 provides support using this analysis result at the stage of use. For example, the correlation analysis unit 1122 performs correlation analysis using department branch trends and department branch attributes as explanatory variables and operating results (specifically, for example, the amount of assets under custody) as objective variables. This analysis is a quantification type 1 analysis illustrated in Hypothesis c. This analysis reveals department branch trends and department branch attributes (here referred to as "specific trends / attributes") that are highly correlated with operating results (ie, the amount of assets under custody). There may be a single department / branch included in the “group of departments / branches”. The support information generation unit 1124 generates a warning regarding a specific trend based on the analysis result. For example, when the correlation analysis unit 1122 clarifies by analysis that the number of contacts with the customer is a specific trend / attribute, the contact rate with the customer (specific trend / specific trend /) for the specific department / branch that is the target person. When it is found that the attribute index) is inferior in comparison with other departments (for example, the minimum value of contact rate in the departments with the top 10% of sales results), the support information generation department 1124 Based on the analysis result, an appropriate proposal template or draft is selected and processed to generate specific sales behavior support data. For example, the sales behavior support data generates sales behavior support data for reconsidering the customers who come into contact with each other or for calling attention to increase the number of contacts with customers with a high transaction amount. For example, the sales behavior support data shows that "(branch manager) Y's branch has a low contact rate with major customers, which is △△% compared to the average of XX% of other branches. We recommend that you reconsider the customers you contact because there are concerns about achievement. Or, "(Branch manager) Y's branch has an average contact rate with other major customers. It is low at △△% compared to%. If we continue to do so, we are concerned about achieving our sales department goals, so please increase the number of visits to major customer C and major customer D. " A major customer is a customer who has a high transaction amount to date. The contact rate with a major customer is a value calculated by [the number of contacts with a major customer / the number of contacts with all customers]. Sales department targets include an increase in the amount of assets under custody and an increase in the amount of income from commission income.

( 4 ) In customer development promotion / proposal, the correlation analysis unit 1122 analyzes the correlation between customer data and result data at the preparatory stage. The support information generation unit 1124 provides support using this analysis result at the stage of use. For example, the correlation analysis unit 1122 performs correlation analysis using customer trends and customer attributes as explanatory variables and the rate of increase in assets under custody as objective variables. This analysis is a quantification type 1 analysis illustrated in Hypothesis c. This analysis reveals customer trends and customer attributes (herein referred to as "specific trends / attributes") that are highly correlated with the rate of increase in assets under custody. Based on the analysis result, the support information generation unit 1124 extracts customers who have specific trends / attributes but have few assets under custody as potential customers. The support information generation unit 1124 extracts potential customers from the customers in charge of the sales staff who is the target person. The support information generation unit 1124 selects an appropriate proposal template or draft, processes it, and generates specific sales behavior support data. For example, sales behavior support data includes information that identifies customers with specific trends / attributes. Specifically, the sales behavior support data states, "Among the customers of (salesperson) X, Mr. P, Mr. Q, and Mr. R have a lot of room for development in view of the scale of these potential assets. Therefore, please actively contact these customers and propose inheritance measures and financial products. "

(5) Compliance wording In the proposal, the correlation analysis unit 1122 learns past (compliance) problem cases as teacher data in the preparatory stage, and combines it with customer data, sales entity data, product data, and result data. Analyze the correlation. The support information generation unit 1124 provides support using this analysis result at the stage of use. For example, the correlation analysis unit 1122 performs correlation analysis using customer data, business entity data, product data, and the number of transactions as explanatory variables, and the presence or absence of compliance problems as objective variables. This analysis is a quantification type 2 analysis illustrated in Hypothesis d. This analysis reveals the correlation between customer data, sales entity data, product data, and the number of transactions and the presence or absence of compliance issues. For example, if data of a fictitious transaction case is input as a past problem case, and as a result of learning this, there is a customer who has a large number of transactions or a transaction amount even though the number of contacts is small, there is a possibility of a fictitious transaction. When this is clarified by the analysis of the correlation analysis unit 1122, and further, when the correlation analysis unit 1122 finds the customer data that meets this condition, the support information generation unit 1124 determines the target person, for example, this customer. Generate sales behavior support data to propose the possibility of problem cases to the manager (or higher rank) who manages the sales staff in charge. For example, in the sales behavior support data, "(branch manager) Y and the person in charge (salesperson) X rarely contact customers, but there are customers who make frequent transactions. Now that we have displayed the list, please investigate the investment objectives, investment decisions, or profits and losses of these clients. "

(6) Regarding the degree of correlation / influence between skills and the like and the degree of influence, the correlation analysis unit 1122 analyzes the correlation between the salesperson data (more specifically, the salesperson attribute) and the result data at the preparation stage. The support information generation unit 1124 provides support using this analysis result at the stage of use. For example, the correlation analysis unit 1122 performs correlation analysis using the qualifications and skills of sales staff as explanatory variables and the operating results (for example, the amount of assets under custody) as objective variables. This analysis is a quantification type 1 analysis illustrated in Hypothesis d. By this analysis, the qualifications and skills of sales staff are analyzed to clarify the sales staff attributes (here referred to as "specific attributes") that have a high correlation with sales performance. Then, when the sales performance of the target person is lower than the representative value (for example, average value) of the sales performance of another salesperson (for example, another salesperson in the same position), the specific salesperson assigns a specific attribute. If not, the support information generation unit 1124 generates sales behavior support data for making recommendations regarding specific attributes based on the analysis result. For example, sales support data includes content that recommends taking a course to acquire a specific qualification or a specific skill. Specifically, the sales support data says, "There seems to be room for improvement in XX skills for (salesperson) X's sales activities. In order to improve sales performance, we are aiming to acquire △△ qualifications. What if? We also recommend taking a course on XX skills. "

The analysis performed by the correlation analysis unit 1122 and the support information generated by the support information generation unit 1124 are not limited to the above examples (1) to (6). For example, the correlation analysis unit 1122 may analyze the hypothesis a, and the support information generation unit 1124 may generate support information according to the analysis result. For example, when the analysis reveals that the correlation between the number of visits to customers and the number of contracts is higher than the standard, the target salesperson is another salesperson (for example, the top 10% of sales results). When the number of visits to customers is small compared to (salesperson), the sales support data generated by the support information generation unit 1124 is, for example, "(Salesperson) Mr. X, it seems that the number of visits to customers is small. Please actively visit customers with the goal of ○ times. " Alternatively, the correlation analysis unit 1122 may analyze the hypothesis b, and the support information generation unit 1124 may generate support information according to the analysis result. For example, when the analysis reveals that the correlation between the sales performance of a certain product P (that is, whether or not it is sold) and the amount of financial assets held by the customer is higher than the standard, among the customers of the sales staff who are the target customers. In addition, if there is a customer who has not purchased product P yet and the amount of financial assets held is higher than the standard, the sales support data generated by the support information generation unit 1124 is "(Salesperson) Mr. X, your customer. Please recommend product P to Mr. P, Mr. Q, and Mr. R. Product P is currently sold to this customer base. " The standard of the amount of financial assets held is, for example, a value at which the purchase probability of the product P exceeds the standard value.

The correlation analysis unit 1122 may perform all of the plurality of analyzes including the above example, or may perform only a part of the plurality of analyzes.

See FIG. 7B again. In step S7, the output means 121 outputs the display data for displaying the screen for displaying the information regarding the business behavior support data received from the support information generation unit 1124 to the target client device 20. The target client device 20 is a client device that has accessed the server device 10. The client device 20 transmits a recommendation output request to the server device 10. This output request contains information that identifies the subject matter and the classification of the recommendations. For example, the client device 20 to which the salesperson is logged in sends an output request of a proposal for the customer product trend analysis for the salesperson. Which classification of recommendations is requested is specified, for example, by the user of the client device 20. Alternatively, the software of the client device 20 may select the recommended classification and notify the user of this. Alternatively, the server device 10 may select the recommended classification and send it to the client device 20.

Alternatively, the server device 10 may voluntarily transmit display data for displaying the proposal to the client device 20 triggered by a predetermined event. The event that triggers the transmission of display data is, for example, an event in which the server device 10 receives a login request from the client device 20. Alternatively, the event that triggers this is an event in which a predetermined time has elapsed since the display data of the proposal was output last time.

9 and 10 are diagrams illustrating an output screen of the proposal. This screen is an example of the screen output in (1) Customer product trend analysis / support. In FIG. 9 of this example, the proposal 701 states that "Salesperson X, recently, product A, which has the highest sales in all stores, is purchased by a customer who is older than 〇〇, occupation △△, and product B. In Fig. 10 of this example, the proposal 702 represents the proposal "Why don't you propose to Mr. Taro Yamada and Mr. Ichiro Suzuki among the customers of Mr. X, the salesperson?" It is a UI object.

3. 3. Modifications The present invention is not limited to the above-described embodiment, and various modifications can be performed. Hereinafter, some modification examples will be described. Of the following modifications, two or more may be used in combination.

The business entity that is the target of support in the sales support system 1 is not limited to financial institutions. For example, the sales support system 1 may be applied to an entity other than a financial institution, such as automobile sales and cosmetics sales.

Correlation analysis unit 1122 is not limited to those using AI or machine learning techniques. Correlation analysis unit 1122 may be any one that analyzes the correlation between customer data, sales entity data (salesperson data and department branch data), product data, and result data, and is given, for example. These correlations may be analyzed based on the correlation model. This correlation model may be incorporated into the program or may be input as data later.

The division of functions between the server device 10 and the client device 20 is not limited to that illustrated in the embodiment. For example, a dedicated application program may be installed in the client device 20, and the functions such as the learning means 111, the analysis means 112, and the output means 121 may be implemented in the client device 20 by this application program. According to this example, it is possible to support the sales staff even in a stand-alone environment in which the server device 10 cannot be accessed.

The flow (order of processing), UI screen, and the like described in the embodiments are merely examples, and the present invention is not limited thereto. For example, in the embodiment, an example in which the target person is specified after specifying the objective variable has been described, but the order may be changed. That is, the objective variable may be specified after the target person is specified. Further, in the embodiment, the example in which the proposal is generated after the objective variable is specified and the target person is specified has been described, but the proposal may be generated first. In this case, the support information generation unit 1124 first generates proposals for all possible combinations of the objective variable and the target person. After the proposal is generated, the objective variable is specified and the target audience is specified. The support information generation unit 1124 selects a proposal corresponding to the specified objective variable and the specified target person from the plurality of proposals that have already been generated.

The hardware configurations of the server device 10 and the client device 20 are merely examples. The server device 10 and the client device 20 may have any hardware configuration as long as the required functions can be implemented.

The program executed by the processor such as CPU 151 or CPU 251 may be provided in a state of being recorded on a non-temporary recording medium such as a CD-ROM (Compact Disk Read Only Memory).

1 ... Sales support system, 3 ... Network, 10 ... Server device, 20 ... Client device, 101 ... Input means, 102 ... Input means, 103 ... Input means, 104 ... Input means, 105 ... Storage means, 111 ... Learning means, 112 ... Analytical means, 121 ... Output means, 131 ... Accepting means, 132 ... Specific means, 151 ... CPU, 152 ... Memory, 153 ... Storage, 201 ... Communication means, 202 ... Input means, 203 ... Display means, 204 ... Control Means, 205 ... Storage means, 251 ... CPU, 252 ... Memory, 253 ... Storage, 255 ... Display device, 256 ... Input device, 701 ... Recommendation, 702 ... Recommendation, 1121 ... Hypothesis database, 1122 ... Correlation analysis unit, 1123 ... Support information database, 1124 ... Support information generation unit

Claims (21)

(A) Customer data including at least one of customer trends and customer attributes for each of a plurality of customers in a business entity having at least one product, and (b) business entity trends for each of the plurality of business entities in the business entity. And business entity data including at least one of the business entity attributes, (c) product data including at least one of the product trends and product attributes for each of the at least one product, and (d) the plurality of customers in the business entity. in the result of at least the transaction history data, a reception unit for accepting a designation of the target variable,
Before SL customer data, the operating body data, the product data, and the results correlated object variable and that the specified in the results of correlation analysis of data based on the relatively high specific explanatory variables, the plurality of operating A specific means of identifying the target person from within the body ,
A storage means for storing a database in which proposal templates corresponding to combinations of objective variables and explanatory variables are recorded, and
An output means for outputting a proposal based on a template according to a combination of the specified objective variable and the specific explanatory variable extracted from the database to the target person.
Have,
The specific explanatory variable is a variable included in at least one of the customer data, the business entity data, the product data, and the result data.
The specific means is a sales support device that identifies as a target person a business entity corresponding to a customer whose value of the specific explanatory variable satisfies a predetermined relationship with respect to a predetermined criterion among the plurality of business entities. At least a part of the customer data, the business entity data, the product data, and the result data is used as candidate data for explanatory variables, and at least a part of the customer data, the business entity data, the product data, and the result data. The sales support device according to claim 1, wherein the data is candidate data for the objective variable, and the explanatory variable and the analysis means for performing the correlation analysis on the objective variable are provided. The analysis means performs the correlation analysis using a trained model in which the candidate data of the explanatory variable is given to the input layer and the candidate data of the objective variable is given to the output layer as teacher data and machine learning is performed. The sales support device according to 2. The specific explanatory variable is a variable included in the customer data, and the specific explanatory variable is a variable included in the customer data.
The specifying unit, among the plurality of operating bodies, in any one of claims 1 to 3 values of the specific explanatory variable is specified as the subject operating body corresponding to good customer than a predetermined reference Described sales support device. The sales support device according to claim 4 , wherein the proposal includes information for identifying the good customer. The specific explanatory variable is a variable included in the business entity data, and is
The specifying unit, among the plurality of operating bodies, sales support according to the specific explanatory variable value predetermined reference bad operating member than to any one of claims 1 to 3 identified to as the subject Device. The proposal contains information for improving the value of the particular explanatory variable.
The sales support device according to claim 5. The specific explanatory variable is a variable included in the product data, and is
The specific means identifies as the target person a business entity corresponding to a product in which the value of the specific explanatory variable is better than a predetermined standard among the plurality of business entities.
The sales support device according to any one of claims 1 to 3. The proposal contains information that identifies the good product.
The sales support device according to claim 8. The specific means identifies the target person from among the plurality of business entities whose designated objective variable values satisfy a predetermined condition.
The sales support device according to any one of claims 1 to 9. The objective variable is a variable related to the record of purchasing a specific product among the plurality of the products.
The predetermined condition is a condition that the business entity corresponds to a customer who has not purchased the specific product.
The sales support device according to claim 10. The objective variable is a variable related to the performance of selling the product.
The predetermined condition is a condition that the performance is worse than the standard in relation to other business entities.
The sales support device according to claim 10. The sales entity includes a salesperson
The sales support device according to any one of claims 1 to 12 , wherein the sales entity data indicates trends and attributes of the sales staff. The sales support device according to claim 13 , wherein the sales entity data includes a daily report of the sales staff. The sales support device according to claim 13 or 14 , wherein the sales entity data includes a telephone call history between the sales staff and a customer. The sales support device according to any one of claims 13 to 15 , wherein the sales entity data includes a history of location information of the sales staff. The sales support device according to any one of claims 13 to 16 , wherein the sales entity data includes information on personnel evaluation of the sales staff. The sales support device according to claim 17 , wherein the sales entity data includes information indicating at least one of the qualifications and skills possessed by the sales staff. The entity is a financial institution
The sales support device according to any one of claims 1 to 18 , wherein the customer data includes a transaction history at the financial institution and a balance of assets under custody. On the computer
(A) Customer data including at least one of customer trends and customer attributes for each of a plurality of customers in a business entity having at least one product, and (b) business entity trends for each of the plurality of business entities in the business entity. And business entity data including at least one of the business entity attributes, (c) product data including at least one of the product trends and product attributes for each of the at least one product, and (d) the plurality of customers in the business entity. from among the result data including at least the transaction history, and the step of accepting a specification of the objective variable,
Before SL customer data, the operating body data, the product data, and the results correlated object variable and that the specified in the results of correlation analysis of data based on the relatively high specific explanatory variables, the plurality of operating Steps to identify the target person from inside the body ,
Based on the template corresponding to the combination of the specified objective variable and the specific explanatory variable extracted from the database in which the template of the proposal corresponding to the combination of the objective variable and the explanatory variable is recorded for the subject. It is a program to execute the step to output the proposal .
The specific explanatory variable is a variable included in at least one of the customer data, the business entity data, the product data, and the result data.
In the specifying step, among the plurality of business entities, the business entity corresponding to the customer whose value of the specific explanatory variable satisfies a predetermined relationship with respect to a predetermined criterion is specified as the target person.
program. The computer system (a) for each of the plurality of customers in an entity having at least one product, customer data including at least one of customer trends and customer attributes, and (b) for each of the plurality of business entities in the entity. , Business entity data including at least one of the business entity trends and the business entity attributes, (c) product data including at least one of the product trends and the product attributes for each of the at least one product, and (d) in the business entity. A step of accepting the designation of the objective variable from the result data including at least the transaction histories of the plurality of customers .
Said computer system, before Symbol customer data, the operating body data, the product data, and the results correlated object variable and that the specified in the results of correlation analysis of data based on the relatively high specific explanatory variables , The step of identifying the target person from the plurality of business entities ,
The computer system is used for the combination of the specified objective variable and the specific explanatory variable extracted from the database in which the template of the proposal corresponding to the combination of the objective variable and the explanatory variable is recorded for the subject. Steps to output suggestions based on the corresponding template
Have,
The specific explanatory variable is a variable included in at least one of the customer data, the business entity data, the product data, and the result data.
In the specifying step, among the plurality of business entities, the business entity corresponding to the customer whose value of the specific explanatory variable satisfies a predetermined relationship with respect to a predetermined criterion is specified as the target person.
Sales support method.

Images