Advice generating system, advice generating method and advice generating program

Abstract

To provide a technique for sending a proper advice to a sales representative who conducts on-line business negotiations with a client.SOLUTION: An advice generating server 4 is communicable with an on-line negotiation server 3. The on-line negotiation server 3 conducts on-line negotiations between a sales representative's terminal 1 and a clinet terminal 2. The advice generating server 4 has a control unit 10 and a storage unit 40. The storage unit 40 stores action norm data 44 related to action norms that the sales representative should follow and advice keyword data 48 related to a corresponding advice keyword if action norms are not followed. The control unit 10 has: a conversation acquisition unit 14 which acquires a conversation of the on-line negotiations and determines conversation details; an action norm acquisition unit 18 which acquires action norm data; an action norm conformity determination unit 22 which determines whether the conversation details follow the action norms; and an advice keyword extraction unit 24 which extracts a corresponding advice keyword when the action norms are not followed.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a technique for providing advice to sales representatives in online business negotiations. [Background technology]

[0002] In sales work, supervisors do not provide sufficient training and guidance to their subordinates due to reasons such as "not having enough time" and "variability in ability." Accompanying sales and sales activities, which are particularly important among the training and guidance, are not often carried out.

[0003] In addition, the company is slow to standardize sales theories and know-how. As a result, each employee is stuck learning their own tricks through trial and error in sales activities, which means they are unable to improve their skills or produce the results they want.

[0004] In sales activities, video call systems have been developed that allow for contactless business negotiations with customers (for example, Patent Document 1), and their use has taken root due to their high convenience.

[0005] In these online business meetings, you can share your PC screen and show documents and videos while explaining the product, providing an experience that is just as good as a real business meeting with a customer. In addition, the contents of the business meeting can be obtained as digital data. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 2023-079677 Summary of the Invention [Problem to be solved by the invention]

[0007] Online business negotiations also present challenges that need to be resolved. For example, if the other party's response during a negotiation is slow, the sales representative must immediately correct course and reinforce the content, otherwise the next meeting or the deal will not be successful.

[0008] However, during business negotiations, it is difficult to have the time to objectively look at your own customer service, and you are unable to recognize whether it is good or bad.

[0009] However, a manager who is not present at the business meeting cannot give advice to his subordinates on how to interact with the customer.

[0010] Naturally, such advice is based on the content of conversations during business negotiations, but it is also based on scientific criteria. Numerical results are brought about by causes, which are actions. Numbers roughly correspond to sales performance. Actions include approaches to and interactions with customers, and are expressed as negotiation techniques and sales talk used in interviews. Desirable and unsatisfactory numbers are inevitable, and are a cause and effect relationship with actions. The applicant has extracted the actions that are valued by high-performing individuals and standardized them as norms to use as criteria for evaluation.

[0011] In view of the above-mentioned problems, the present invention provides appropriate advice to sales representatives according to the progress and circumstances of online interviews. [Means for solving the problem]

[0012] The present invention, which solves the above-mentioned problems, is an advice generation system. The advice generation system is implemented in an online business negotiation system or can communicate with the online business negotiation system. The online business negotiation system conducts online business negotiations between a sales representative terminal and a customer terminal. The advice generation system has a control unit and a memory unit. The memory unit stores code of conduct data relating to a code of conduct that sales representatives must follow and advice keyword data relating to advice keywords that correspond to deviations from the code of conduct. The control unit has a conversation acquisition unit that acquires conversations from the online business negotiations and determines the content of the conversation, a code of conduct acquisition unit that acquires the code of conduct data, a code of conduct compliance determination unit that determines whether the content of the conversation complies with the code of conduct, and an advice keyword extraction unit that extracts advice keywords that correspond to deviations from the code of conduct.

[0013] This allows appropriate advice to be given to sales representatives during online business negotiations.

[0014] In the above invention, preferably, the conversation acquisition unit determines, as the conversation content, at least the "topic," "customer reaction," and "salesperson's remarks," whether the "topic" falls into one of the topic categories including product functions, usage effects, and product prices, whether the "customer reaction" falls into one of the topic categories including product functions, usage effects, and product prices, whether the "customer reaction" falls into one of the topic categories including explanations, questions, and suggestions, and whether the "salesperson's remarks" fall into one of the remark categories including explanations, questions, and suggestions.

[0015] This allows you to judge the content of the conversation and compare it to a code of conduct.

[0016] In the above invention, preferably, the code of conduct data is provided for each negotiation progress step, the control unit has a progress step input unit for inputting negotiation progress steps, and the code of conduct acquisition unit acquires code of conduct data corresponding to the negotiation progress steps.

[0017] This allows appropriate codes of conduct to be set for each step in the sales process.

[0018] In the above invention, preferably, the control unit has a progress step confirmation unit that confirms the negotiation progress step based on the judgment of the code of conduct compliance determination unit, the progress step input unit inputs the next negotiation progress step if the code of conduct is complied with, and the code of conduct acquisition unit acquires code of conduct data corresponding to the next negotiation progress step. Sales representatives can increase the likelihood of closing a deal by being aware of the steps in the negotiation process and taking each one carefully.

[0019] In the above invention, preferably, the memory unit stores sales ability level data relating to the sales ability level of each sales representative, the control unit has a sales ability level acquisition unit that acquires the sales ability level data of the sales representative, and the code of conduct compliance determination unit determines the degree of compliance with the code of conduct taking into account the sales ability level.

[0020] As a general rule, we advise you to comply with the Code of Conduct, but as an exception we may make adjustments based on your sales ability level.

[0021] In the above invention, preferably, the memory unit stores characteristic data of each sales representative and / or characteristic data of each customer, and the control unit has a sales activity characteristic determination unit that determines sales activity characteristics based on the characteristic data of each sales representative and / or the characteristic data of each customer, and an auxiliary word addition unit that adds auxiliary words that supplement the advice keywords based on the sales activity characteristics.

[0022] In principle, advice is given to comply with the Code of Conduct, but as an exception, adjustments can be made taking into account the characteristics of business activities.

[0023] In the above invention, preferably, the advice generation system is implemented with a sentence generation system or can communicate with the sentence generation system, and the control unit has a sentence creation instruction unit that instructs the sentence generation system to create a sentence based on the advice keyword.

[0024] The sentence creation instructions input to the sentence generation system are based on advice keywords stored in advance. As a result, even if the processing of the sentence generation system is in a black box state, advice sentences are output that are consistent within a certain range.

[0025] In the above invention, preferably, the advice generation system is implemented with a sentence generation system or can communicate with the sentence generation system, and the control unit has a sentence creation instruction unit that instructs the sentence generation system to create a concise sentence and a detailed sentence based on the advice keyword.

[0026] Concise advice provides a brief outline of areas for improvement in interview approaches and interactions, while detailed advice provides specific areas for improvement in detail.

[0027] In the above invention, preferably, the conversation acquisition unit acquires a conversation at predetermined time intervals, and the advice keyword extraction unit extracts an advice keyword for the conversation.

[0028] This allows appropriate advice to be given to the sales representative as needed.

[0029] The present invention, which solves the above-mentioned problem, is an advice generation method. Online business negotiations are conducted between a sales representative terminal and a customer terminal. An information processing device is used to send advice to the sales representative during the online business negotiations. Code of conduct data relating to the code of conduct that the sales representative must follow and advice keyword data relating to advice keywords corresponding to deviations from the code of conduct are stored, a conversation during the online business negotiations is acquired, the content of the conversation is determined, the code of conduct data is acquired, it is determined whether the content of the conversation conforms to the code of conduct, advice keywords corresponding to deviations from the code of conduct are extracted from the advice keyword data, and advice based on the advice keywords is output.

[0030] This allows appropriate advice to be given to sales representatives during online business negotiations.

[0031] The present invention, which solves the above-mentioned problems, is an advice generation program that causes an information processing device to execute various processes. This makes it possible to send advice to a sales representative during online business negotiations between a sales representative terminal and a customer terminal. The various processes include a process of acquiring a conversation during the online business negotiation and determining the content of the conversation, a process of storing code of conduct data related to a code of conduct that the sales representative must follow and acquiring the code of conduct data, a process of determining whether the content of the conversation conforms to the code of conduct, a process of storing advice keyword data related to advice keywords corresponding to deviations from the code of conduct and extracting advice keywords corresponding to deviations from the advice keyword data, and a process of outputting advice based on the advice keywords.

[0032] This allows appropriate advice to be given to sales representatives during online business negotiations. [Effects of the Invention]

[0033] The advice generation system of the present invention can be applied to existing business negotiation systems.

[0034] The advice generating system of the present invention can provide appropriate advice to salespeople. [Brief explanation of the drawings]

[0035] [Figure 1] A diagram showing an overview of the entire system including the system of the present invention. [Figure 2] Sequence diagram for the entire system [Figure 3] Hardware configuration of the advice generation server [Figure 4] Functional block diagram of the advice generation server [Figure 5] Control flow diagram of the advice generation server [Figure 6] Image of the relationship between negotiation progress steps and the Code of Conduct [Figure 7] An illustration of the code of conduct for each step of the negotiation process [Figure 8] Image of the relationship between sales ability levels and code of conduct [Figure 9] Comparison of sales representative screen and customer screen [Figure 10] Basic functional block diagram of the advice generation server DETAILED DESCRIPTION OF THE INVENTION

[0036] ~Summary~ Figure 1 is a schematic diagram of the entire system including the system of the present invention. The overall configuration will be explained.

[0037] The system of the present invention is an advice generation system centered around an advice generation server 4. The advice generation server 4 is implemented in an online business negotiation system or can communicate with an online business negotiation system. The advice generation server 4 also implements a sentence generation system or can communicate with a sentence generation system.

[0038] The online business negotiation system is composed of a sales representative terminal 1, a customer terminal 2, and an online business negotiation server 3. The sales representative terminal 1, customer terminal 2, and online business negotiation server 3 are connected via a network and are capable of communication. The sales representative terminal 1 is a terminal used by a sales representative, and the customer terminal 2 is a terminal used by a customer.

[0039] The online business negotiation server 3 mediates the sending and receiving of conversation data between the salesperson terminal 1 and the customer terminal 2. Furthermore, the online business negotiation server 3 provides various functions necessary for business negotiations, such as screen sharing and file transfer.

[0040] The text generation system is a system centered around a text generation server 5. The text generation system uses an artificial intelligence algorithm to input text and output new text based on the content of the input text.

[0041] Furthermore, text generation systems can use artificial intelligence algorithms to understand sentence structure, grammar, and meaning.

[0042] Figure 2 is a sequence diagram of the entire system. The operation will now be described.

[0043] In the online business negotiation system, conversation data is transmitted and received bidirectionally between a salesperson terminal 1 and a customer terminal 2 via an online business negotiation server 3. In this way, business negotiations are conducted between the salesperson and the customer.

[0044] The advice generation server 4 receives conversation data between the sales representative and the customer from the online negotiation server 3, and sends an instruction to the sentence generation server 5 to analyze the conversation data.

[0045] The text generation server 5 processes the content of the conversation data based on the analysis instructions, and also performs proofreading and summarization.

[0046] The advice generation server 4 receives the conversation content processed by the sentence generation server 5, performs the present control (described in detail later), and transmits a sentence creation instruction to the sentence generation server 5.

[0047] The advice generation server 4 receives the advice text created by the text generation server 5 and transmits it to the online negotiation server 3.

[0048] The online negotiation server 3 sends the advice text to the salesperson terminal 1. It does not send it to the customer terminal 2.

[0049] The advice text is displayed on the salesperson terminal 1. The advice text is not displayed on the customer terminal 2.

[0050] ~Advice Generation Overview~ 3 is a hardware configuration diagram of the advice generating server 100. The advice generating server 100 is an information processing device.

[0051] The information processing device 100 includes a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, a non-volatile memory 104, and an I / O 105, and each component is connected to each other via a bus 106.

[0052] The CPU 101 executes each function based on a program. The ROM 102 stores the program. The RAM 103 is used as a temporary work area for the CPU 101. The non-volatile memory 104 is an example of a storage device that maintains stored information even if the power supply is cut off. For example, it is a semiconductor memory or a hard disk. The non-volatile memory 104 does not need to be built-in, and may be removable and portable. The I / O 105 is an input / output interface.

[0053] To the I / O 105, for example, a communication unit 107, an input unit 108, a display unit 109, etc. are connected.

[0054] The communication unit 107 is connected to a communication line and has a communication protocol for communicating data with the outside. The input unit 108 is a touch panel, keyboard, mouse, etc., and is used to input operation instructions, etc. The display unit 109 is, for example, a liquid crystal display, an organic EL display, a projector, etc., and displays the results of information processing, etc., which may be displayed on another terminal screen via the communication line, or may be output by a printer. Note that the input unit 108 and the display unit 109 are not essential for the control of the present invention.

[0055] Fig. 4 is a functional block diagram of the advice generation server. The advice generation server 4 is composed of a control unit 10 and a storage unit 40. A CPU 101 executes each function of the control unit 10 via a program. The processing content and processing procedure of each function will be described later using a flowchart (Fig. 5). The storage unit 40 is, for example, a non-volatile memory 104.

[0056] The control unit 10 and the storage unit 40 may be the same device or different devices. Some functions may be located in different devices and integrated via a network.

[0057] The control unit 10 has a progress step input unit 12, a conversation acquisition unit 14, a progress step confirmation unit 16, a code of conduct acquisition unit 18, a sales ability level acquisition unit 20, a code of conduct compliance determination unit 22, an advice keyword extraction unit 24, a sales activity characteristic determination unit 26, an auxiliary word addition unit 28, a sentence creation instruction unit 30, and an advice sentence input / output unit 32.

[0058] The storage unit 40 stores analysis instructions 42 , code of conduct data 44 , sales ability level data 46 , advice keyword data 48 , salesperson characteristic data 50 , and customer characteristic data 52 .

[0059] 5 is a control flow diagram of the advice generating server. The processing contents of each function of the control unit 10 will be explained along the control steps.

[0060] When online negotiation begins on the online negotiation server 3, control by the control unit 10 begins, and when online negotiation ends on the online negotiation server 3, control by the control unit 10 ends. During the online negotiation, the following control loop (step S12 → step S32 → step S12) is repeated.

[0061] In this embodiment, the control unit is set to, for example, 6 minutes, and the overlap length between the control is set to 1 minute. That is, the following control loop (step S12 → step S32 → step S12) is repeated every 5 minutes.

[0062] The progress step input unit 12 inputs negotiation progress steps (step S12). The system of the present application can be used not only for BtoB (business-to-business transactions) but also for BtoC (business-to-consumer transactions) for relatively expensive products such as housing and life insurance. It is mainly assumed that negotiations will be conducted multiple times.

[0063] At the first business negotiation, the progress step input unit 12 inputs the business negotiation progress step 1. At the start of the next business negotiation, the business negotiation progress step K at the end of the previous business negotiation is input. When online business negotiations and real business negotiations are used together, the sales representative determines the business negotiation progress step at the end of the real business negotiation and inputs the business negotiation progress step via the sales representative terminal 1.

[0064] The conversation acquisition unit 14 acquires the conversation of the online business negotiation and determines the content of the conversation (step S14).

[0065] The existing online business negotiation server 3 has the function of converting voice data into text data. The conversation acquisition unit 14 acquires conversation text data between a sales representative and a customer. Furthermore, the conversation acquisition unit 14 outputs the conversation text data and analysis instructions 42 to the text generation server 5. The analysis instructions 42 are stored in the memory unit 40.

[0066] Analysis Instruction 42 states that the "topic" of the conversation in question should be determined to fall into one of the following topic categories, including at least product features, usage effects, product price, as well as casual conversation, customer information, budget amount, and decision-making authority. Customer information refers to information related to the customer's external and internal environments, with the external environment including the actual state and trends of the industry, market, and competition, and the internal environment including the current state and goals of management, operations, business, and products.

[0067] Furthermore, it is written to determine whether the "topic" in question is a topic related to the sales side, a topic related to the customer side, or a topic that is difficult to classify.

[0068] The analysis instructions 42 state that the degree of positivity corresponding to the "topic" of the customer conversation, that is, the customer reaction, should be determined.

[0069] For small talk, the scale was set as "interested (positive) ⇔ bored (negative)", for product functions as "satisfied (positive) ⇔ dissatisfied (negative)", for the effectiveness of use as "clear (positive) ⇔ unclear (negative)", for product price as "cheap (positive) ⇔ expensive (negative)", for customer information as "sufficient understanding (positive) ⇔ insufficient understanding (negative)", for budget amount as "within budget (positive) ⇔ above budget (negative)", and for decision-making authority as "yes (positive) ⇔ no (negative)", with the highest degree of affirmation being 1 and the lowest degree of affirmation (highest degree of negation) being 0.

[0070] The analysis instructions 42 state that the salesperson's utterances should be judged to fall into one of the following utterance categories, including at least explanation, question, proposal, recommendation, request, reply, negotiation, negotiation, consideration, refusal, and acceptance.

[0071] The existing text generation server 5 not only generates text but also has the function of understanding the content of the text. The text generation server 5 appropriately proofreads and summarizes the conversation between the sales representative and the customer, and responds to the analysis instruction 42.

[0072] The conversation acquisition unit 14 determines the content of the conversation based on the response to the analysis instruction 42 .

[0073] The code of conduct acquisition unit 18 acquires code of conduct data corresponding to the current negotiation progress step (step S16). The code of conduct is the approach and interaction that a sales representative should follow. Code of conduct data 44 is stored in the memory unit 40 for each negotiation progress step.

[0074] Figures 6 and 7 are conceptual diagrams showing the correspondence between negotiation progress steps and codes of conduct.

[0075] While actual sales activities are continuous, the steps employees should take and the actions they should take from start to finish are defined as standards. This makes it clear how far a deal has progressed and how it will move forward from here. As a result, it becomes easier to give accurate advice to sales representatives.

[0076] The advice generation system of the present invention sets a code of conduct for each step in the progress of a sales negotiation. For example, in the first half of a sales activity, it is important to clarify the problem, while in the second half, it is important to concretize the solution.

[0077] In the example shown, the sales process is divided into 10 steps, but it should be divided into at least two parts: the first half and the second half, or three parts: the beginning, middle, and end.

[0078] The Code of Conduct includes, for example, "speaker ratio," "topic ratio," "degree of customer response," and "ratio of sales representative remarks." The correspondence between the sales negotiation progress steps and the Code of Conduct is described in detail below.

[0079] The sales ability level acquisition unit 20 acquires sales ability level data of sales representatives. The sales ability level data 46 is stored in the storage unit 40.

[0080] Figure 8 shows the relationship between sales ability levels and codes of conduct. Codes of conduct are actions and interactions that salespeople must follow, but the strictness with which the code of conduct is required varies depending on the sales level of each salesperson. For example, at a high sales level, it is acceptable to deviate slightly from the code of conduct.

[0081] The code of conduct acquisition unit 18 and the sales ability level acquisition unit 20 adjust the content of the code of conduct taking into consideration the sales level of the salesperson (step S18).

[0082] The code of conduct conformance determination unit 22 determines whether the conversation content determined by the conversation acquisition unit 14 conforms to the code of conduct (step S20).

[0083] Note that the conversation content determined by the conversation acquisition unit 14 and the code of conduct acquired by the code of conduct acquisition unit 18 are acquired before the judgment by the code of conduct conformance determination unit 22. In other words, the order of the conversation acquisition (step S14) and code of conduct acquisition (step S16) processes may be changed as appropriate.

[0084] Furthermore, unless the negotiation progress step is updated, the same code of conduct as the previous control unit is used, which means that obtaining the code of conduct (step S16) can be omitted.

[0085] The advice keyword extraction unit 24 extracts advice keywords corresponding to the degree of deviation from the code of conduct (step S22). The advice keyword data 48 is stored in the storage unit 40 so as to correspond to the code of conduct.

[0086] If the conversation content complies with the code of conduct, positive advice keywords are extracted.

[0087] The sales activity characteristic determination unit 26 determines the sales activity characteristic in consideration of the characteristic data of each salesperson and / or the characteristic data of each customer (step S24). The salesperson characteristic data 50 and the customer characteristic data 52 are stored in the storage unit 40.

[0088] The auxiliary word adding unit 28 adds auxiliary words that supplement the advice keywords extracted by the advice keyword extracting unit 24, taking into consideration the sales activity characteristics determined by the sales activity characteristic determining unit 26 (step S26).

[0089] The expression of advice text can be optimized by using not only advice keywords but also auxiliary words that take into account the characteristics of sales activities.

[0090] The writing instruction unit 30 creates writing instructions based on the advice keyword and the auxiliary word, and outputs the writing instructions to the writing generation server 5 (step S28).

[0091] At this time, an instruction to create a concise sentence and an instruction to create a detailed sentence are issued. The instruction to create a concise sentence may be issued for each control loop (step S12 to step S32), and the instruction to create a detailed sentence may be issued every two control loops.

[0092] The existing sentence generation server 5 has a function of generating sentences while understanding the content, and generates concise advice 62 and detailed advice 64 based on the sentence creation instructions.

[0093] The advice sentence input / output unit 32 inputs the concise advice 62 and detailed advice 64 from the sentence generation server 5 and outputs them to the online negotiation server 3 (step S30).

[0094] The progress step confirmation unit 16 confirms whether the current negotiation progress steps are being fully followed based on the determination of the code of conduct compliance determination unit 22 (step S32).

[0095] If the code of conduct conformance determination unit 22 determines that the conversation content does not conform to the code of conduct, it outputs the current negotiation progress step K to the progress step input unit 12. If the code of conduct conformance determination unit 22 continues to determine that the conversation content conforms to the code of conduct (for example, conformance three times in a row), it outputs negotiation progress step K+1 to the progress step input unit 12. At this time, the advice keyword may include an indication that the conversation has progressed to the next negotiation progress step.

[0096] If the code of conduct conformance determination unit 22 determines that the conversation content significantly does not conform to the code of conduct, it outputs a negotiation progress step K-1 to the progress step input unit 12.

[0097] ~Specific examples~ A specific example of conversation acquisition (step S14) will be described.

[0098] The advice generation server 4 acquires the sales representative conversation, "The cost of using this system will be 20,000 yen per user per month. I understand that your company has 15 sales staff, so that will come to 300,000 yen," and the customer conversation, "I think we'll start with three people," and outputs these to the sentence generation server 5 along with instructions for analysis.

[0099] The analysis instructions are written in terms of "topics," "customer reactions," and "sales representative comments."

[0100] The text generation server 5 analyzes that "the topic was mainly about the product price," "the impression is that the customer's reaction is somewhat high (the AI ​​judged the price for all 15 people to be somewhat high, and the price for 3 people to be reasonable)," and "the sales representative's remarks were an explanation of the price."

[0101] The advice generation server 4 repeats the above input and output for each control unit to determine the conversation content. Specifically, it acquires the "speaker ratio," "topic ratio," "degree of customer response," and "sales representative comment content ratio" for that control unit.

[0102] The speaker ratio is the ratio of speech by the sales representative to the entire conversation. It may be based on the cumulative speaking time or the cumulative number of characters.

[0103] Topic ratio is the ratio of a topic to the overall conversation. Continuing on from the example conversation above, if the conversation was about budget limits and cost-effectiveness (the magnitude of synergy gained from everyone using the product), the topic ratios would tend to be expressed as follows: small talk 0%, product features 5%, usage benefits 20%, product price 40%, customer information 20%, budget amount 10%, decision-making authority 5%, and other 0%.

[0104] Additionally, topics can be classified into sales and customer topics, and the ratio of sales topics to the overall conversation can be calculated and used as a separate indicator. Sales topics include product features and product prices, while customer topics include the benefits of use, customer information, budget amounts, and decision-making authority. There are also topics, such as casual conversation, that are difficult to classify as sales or customer topics.

[0105] The customer response level is a numerical representation of the customer's reaction to the topic. In the example conversation above, a negative reaction, such as when the product price is slightly high, will result in a value lower than 0.5. In addition, the average and standard deviation of the customer response level for the control unit are also calculated.

[0106] The salesperson utterance content ratio is the ratio of explanations, questions, proposals, recommendations, requests, replies, negotiations, negotiations, consideration, refusals, acceptances, and other comments made by the salesperson. In the above conversation example, where the salesperson explains the price, explains synergies, and proposes that everyone use the product, the salesperson utterance content ratio tends to be 70% explanations, 5% questions, 25% proposals, 0% recommendations, 0% requests, 0% replies, 0% negotiations, 0% negotiations, considerations, 0% refusals, 0% acceptances, and 0% other comments.

[0107] A specific example of acquiring a code of conduct (step S16) will be described.

[0108] The code of conduct is a set of actions and interactions that salespeople must follow, and is established for each step in the sales negotiation process (see Figure 6). For example, it lists the ratio of people who speak, the ratio of topics covered, the degree of customer reaction, and the ratio of what the salespeople say (see Figure 7).

[0109] In the progress of business negotiations, the focus is initially on clarifying the issues, and gradually on concretizing solutions.

[0110] For example, in sales negotiation step 4, "Research," you clarify the customer's issues by eliciting what the customer has to say. In this case, it is preferable to have a low ratio of sales representatives talking and a low ratio of sales topics. Also, in order to extract more and deeper information from the customer, it is preferable to have a high ratio of "questions" among the sales representative's comments.

[0111] However, when the customer's desire to purchase is not yet high, for example, when the salesperson's comments consist of few questions and many explanations, i.e., when the salesperson is constantly bombarding the customer with explanations of the product's features, the customer will become annoyed and try to end the negotiation.

[0112] In negotiation step 7, "Proposal," the salesperson proposes a product that will help solve the customer's problems. As a result, the salesperson's speaking ratio will naturally be higher compared to negotiation step 4, "Research." In addition, the ratio of "explanation" and "recommendation" in the salesperson's speech ratio will naturally be higher.

[0113] A specific example of adjusting the code of conduct based on the sales ability level (step S18) will be described.

[0114] The code of conduct is the approach and interaction that all salespeople must follow. As long as they adhere to the code of conduct, they can incorporate their own unique style into actual sales negotiations. However, the more specific the code of conduct, the more difficult it becomes to follow the advice and demonstrate originality. This tendency is particularly strong among salespeople with a lot of experience and a proven track record.

[0115] Therefore, the degree of compliance with the code of conduct is adjusted taking into account the sales level.

[0116] In the image diagram shown in Figure 8, salespeople at sales level 1 are required to comply with the code of conduct 100%, while salespeople at sales level 10 are deemed to be in compliance with the code of conduct if they comply with 75% of the code of conduct.

[0117] Another example of adjusting the code of conduct is to limit the rate of small talk in topics discussed by sales representatives at sales level 5 to 20% or less, the rate of small talk in topics discussed by sales representatives at sales level 1 to 10% or less, and the rate of small talk in topics discussed by sales representatives at sales level 10 to 30% or less. Generally, the small talk of high-performing sales representatives leads to relationship building and information gathering, but the small talk of low-performing sales representatives is incoherent and actually slows down the progress of sales negotiations.

[0118] Therefore, even when engaging in casual conversation, the advice displayed on the sales representative terminal 1 differs depending on the sales ability level. A sales representative with sales level 5 (which is the standard value) will be shown "Let's start with casual conversation using topics that will easily lead to a business negotiation," a sales representative with sales level 1 will be shown "Let's limit casual conversation to greetings and then get to the main topic," and a sales representative with sales level 7 will be shown "Let's uncover their needs while incorporating casual conversation about business."

[0119] A specific example of the code of conduct conformance determination (step S20) and advice keyword extraction (step S22) will be described below. For example, negotiation progress step 4 is "research."

[0120] The conversation content is as follows: speaker ratio 70%, topic ratio 60% (details: product features 50%...), customer response rate 0.3, sales representative remarks ratio (explanation 50%, questions 30%...).

[0121] The code of conduct for sales negotiation progress step 4 is: speaker ratio 50% or less, topic ratio 40% or less, customer response rate 0.4 or more, sales representative remarks content ratio (explanation 30% or less, questions 50% or more...).

[0122] By comparing the speaker ratios, it can be seen that, for example, salespeople speak too much and customers speak too little, which does not comply with the code of conduct.

[0123] Comparing the topic ratios reveals that there are more sales-related topics, especially those about product features, and fewer customer-related topics. This does not comply with the Code of Conduct.

[0124] A comparison of customer response rates reveals that the customer response rate is low, which does not comply with the code of conduct.

[0125] Comparing the ratio of sales representatives' comments reveals that they provide a lot of explanations and few questions. This does not comply with the Code of Conduct.

[0126] The comparison amount indicates the degree to which the conversation content does not conform to the code of conduct. If the code of conduct is not conformed, advice keywords suggesting the direction of improvement are extracted. For example, "reduce the amount of sales talk," "reduce discussion of product features," "increase discussion of customer-related topics," "the customer's response is poor, so we will try to correct course immediately," "reduce explanations," "increase questions," etc.

[0127] A specific example of the sentence creation instruction (step S28) and advice sentence input / output (step S30) will be described.

[0128] Based on these advice keywords and comparison amounts, document creation instructions to be input to the document creation server 5 are created.

[0129] For example, salesperson terminal 1 will display advice such as, "If you force or bombard the customer with product information without really getting to know them, you will only leave a bad impression and be rejected before you can even get to closing the deal. Find out how and why the customer became interested, and then add a phrase such as, 'We would like to work with you to find the product that is best suited to your company.' If you can agree on the purpose of the interview early on, it will be easier to structure your conversation."

[0130] The analysis instructions 42 and advice keyword data 48 input by the text generation server 5 are stored in advance. As a result, even if the internal processing of the text generation server 5 is in a black box state, the text generation server 5 outputs advice texts that are consistent within a certain range. Therefore, appropriate advice can be sent to the sales representative.

[0131] A specific example of the business activity characteristic determination (step S24) and the addition of auxiliary words (step S26) will be described.

[0132] Auxiliary words that take into consideration sales activity characteristics that combine salesperson characteristics and customer characteristics can also be input to the sentence generation server 5. Note that advice can be sent to the salesperson even without inputting auxiliary words.

[0133] For the salesperson characteristics, examples of salesperson failures are recorded. For example, technical salespeople tend to be one-sided in explaining product features and have difficulty drawing out the customer's thoughts and requests.

[0134] Customer characteristics include memorizing typical personality types categorized by temperament, including character and personality. For example, are they a "logical" customer who values ​​logic, a "realistic" customer who wants to know measures and prices, a "sensory" customer who relies on intuition and subjectivity, an "emotional" customer who is easily moved by requests, a "scientific" customer who pursues rationality and numbers, a "planning" customer who values ​​the overall structure and procedures, or a "holistic" customer who looks at things from a high perspective or a broad viewpoint?

[0135] Add auxiliary words that take into account the characteristics of sales activities. For example, add "reduce technical explanations using technical terms" and "increase questions that dig deeper into information gathering."

[0136] Without auxiliary words, the advice displayed on the salesperson's terminal 1 becomes, "If you rush to gather information, it becomes akin to 'interrogation,' which is painful for the customer. Stop negotiating and adopt a consultative attitude to open up and make the customer think, 'Maybe I should talk.' The job of sales is not to talk, but to get the other person to talk. It starts with finding out the customer's 'problems' and uncovering the seeds of a deal. Sales talk is all about 'questions,' so rethink your thinking." However, if auxiliary words are added, the advice becomes, "If you don't know the other person, you can't be of any help to them. However, it's difficult to extract information that can be used in sales activities. Most so-called customer understanding is nothing more than a 'product check.' If you limit yourself to a product check, asking about the availability and status of a product, you can only make 'recommendations' that suit your own company's convenience. By going deeper into customer understanding to grasp the customer's current situation and goals related to their management and business, you can make customer-oriented 'proposals.' Deeper information gathering is essential."

[0137] By adding auxiliary words, advice text can be tailored to the characteristics of each sales activity.

[0138] Figure 9 shows a comparison between the salesperson screen and the customer screen. The advice text is displayed only on the salesperson screen, not on the customer screen.

[0139] The concise advice 62 is chosen based on advice keywords and is concise and clear. In addition, an icon corresponding to the advice (for example, a smiling or crying face, or warm or cool colors that represent emotions) is added. The advice is updated for each control unit.

[0140] At 9:25, concise advice 62 was displayed, saying, "Keep the technical explanation to a minimum and ask about the other person's interests," but the situation did not improve, and at 9:30, concise advice 62 was displayed, saying, "You're talking too much. Ask about the circumstances and how you became interested in the product."

[0141] Salespeople can see the concise advice 62 to understand the status of the negotiations and immediately understand what needs to be improved if there are any problems.

[0142] The salesperson can review the history of the concise advice 62 via a scroll bar displayed beside the concise advice 62 .

[0143] The detailed advice 64 is composed of multiple sentences based on advice keywords and taking into account the context before and after. For example, it is updated every two control units. In other words, the update frequency of the detailed advice 64 is lower than the update frequency of the concise advice 62.

[0144] Salespeople may not understand the intention of the salesperson by simply looking at the concise advice 62, and may not be able to make appropriate course corrections.

[0145] Therefore, by looking at the detailed advice 64, you can get an overview of the progress and outcome of the interview, which will make it easier to consider future policies and structures.

[0146] The salesperson can review the history of detailed advice via a scroll bar displayed next to the detailed advice 64.

[0147] A specific example of the progress step confirmation (step S32) will be described.

[0148] In real-life sales activities, there are no clear divisions, but we believe that by being aware of the steps in the progress of sales negotiations and taking each one carefully, the likelihood of closing the deal will increase.

[0149] If the code of conduct continues to be in compliance, it is automatically determined that the sales representative is ready to proceed to the next sales negotiation progress step, and a message is displayed indicating that the sales representative has progressed to the next sales negotiation progress step. This allows sales representatives to structure their interviews while being aware of which sales negotiation progress step they have reached in the overall process. In other words, it is easy to see the overall picture leading up to the closing of a deal.

[0150] In this application, it is assumed that sales negotiations will be conducted over multiple sessions. At the end of each session, a progress history is stored for each negotiation progress step. By checking the progress history, sales representatives can reconsider the policy and structure of the next negotiation while being aware of which negotiation progress step they have reached in the overall process. In other words, it is easy to see the overall picture leading up to the conclusion of a deal.

[0151] ~Summary of effects~ The advice generation system of the present invention is connected to and capable of communicating with the online business negotiation server 3. The online business negotiation system is already in operation and can be easily applied.

[0152] The advice generation system of the present invention is connected to a sentence generation server 5 and is capable of communication. The sentence generation system is already in operation and can be easily applied. The sentence generation system inputs advice keywords and outputs appropriate advice sentences. This allows sales representatives to obtain appropriate advice.

[0153] The analysis instructions 42 and advice keyword data 48 input by the text generation server 5 are stored in advance. As a result, even if the internal processing of the text generation server 5 is in a black box state, the text generation server 5 outputs advice texts that are consistent within a certain range. Therefore, appropriate advice can be sent to the sales representative.

[0154] In principle, the advice generation system of the present invention provides advice that conforms to the code of conduct data 44. This prevents advice from being off the mark in any situation.

[0155] The code of conduct data 44 is provided for each step in the progress of the sales negotiation, which allows appropriate advice to be given according to the progress of the sales negotiation in the sales process.

[0156] In the present advice generation system, the sales representative is aware of the steps in the progress of the sales negotiation and carefully follows each step, thereby increasing the likelihood of closing the deal.

[0157] In addition to the concise advice 62 accompanied by an icon, detailed advice 64 is displayed on the sales representative terminal 1. The sales representative can immediately correct their course with the concise advice 62. Furthermore, the sales representative can see the path to closing the deal with the detailed advice 64.

[0158] ~Variations~ The advice generating system of the present invention is not limited to the above embodiment, and various modifications are possible within the scope of the technical concept of the present invention.

[0159] The text generation system has seen remarkable technological innovation, and various usage methods have been proposed. In the above embodiment, the text generation server 5 is used via communication to generate the concise advice 62 and the detailed advice 64.

[0160] On the other hand, only the detailed advice 64 may be generated using the sentence generation server 5. The concise advice 62 does not have to be displayed, or the advice generation server 4 may generate the concise advice 62.

[0161] The advice generation server 4 may generate the concise advice 62 and display only the concise advice 62. In that case, the detailed advice 64 does not need to be displayed. If the detailed advice 64 is not to be displayed, the sentence generation server 5 does not need to be used.

[0162] As described above, the configuration relating to advice output can be modified in various ways.

[0163] In the above embodiment, the analysis of the conversation content is performed using the sentence generation server 5, but it may be performed as a component of the advice generation server 4 without using the sentence generation server 5.

[0164] In principle, the advice generation system of the present invention provides advice that conforms to the code of conduct data 44. As an exceptional configuration, the degree of compliance with the code of conduct is adjusted taking into account the sales ability level, or the characteristics of sales activities are taken into account. In modified examples, the exceptional configuration may be adopted as appropriate. Note that advice can be sent to sales representatives even without the exceptional configuration.

[0165] In the above embodiment, the code of conduct data 44 is provided for each negotiation progress step, but it may be provided as a single code of conduct.

[0166] 10 is a basic functional block diagram of the advice generation server. Only the minimum necessary configuration is shown as a basic example. The advice generation server 4 is composed of a control unit 10 and a storage unit 40.

[0167] The control unit 10 has a conversation acquisition unit 14, a code of conduct acquisition unit 18, a code of conduct conformance determination unit 22, an advice keyword extraction unit 24, and an advice sentence input / output unit 32. The storage unit 40 stores code of conduct data 44 and advice keyword data 48. The processing content of each function is the same as in the above embodiment.

[0168] The advice generation server 4 receives conversations from the online negotiation server 3 and outputs advice to the online negotiation server 3. [Explanation of symbols]

[0169] 1 Sales representative terminal 2. Customer terminal 3 Online business negotiation server 4 Advice Generation Server 5 Text generation server 10 Control Unit 12 Progress step input section 14 Conversation Acquisition Unit 16 Progress Step Confirmation Section 18 Code of Conduct Acquisition Department 20 Sales Skills Level Acquisition Department 22 Code of Conduct Compliance Judgment Department 24 Advice keyword extraction section 26 Sales Activity Characteristics Determination Department 28 Auxiliary Word Addition Section 30 Text Creation Instructions 32 Advice text input / output section 40 Storage section 42 Analysis instructions 44 Code of Conduct Data 46 Sales force level data 48 Advice Keyword Data 50 Salesperson Characteristics Data 52 Customer characteristic data 62 Concise Advice 64 Detailed Advice 100 Information processing device 101 CPU 102 ROM 103 RAM 104 Non-volatile memory 105 Input / Output Interface Bus 106 107 Communication Unit 108 input units 109 Display Unit

Claims

1. An advice generation system that is implemented in an online business negotiation system that conducts online business negotiations between a sales representative terminal and a customer terminal, or that can communicate with the online business negotiation system, the advice generation system includes a control unit and a storage unit; The storage unit stores code of conduct data relating to a code of conduct that sales representatives should follow and advice keyword data relating to advice keywords corresponding to deviations from the code of conduct, The control unit a conversation acquisition unit that acquires conversations during the online business negotiations and determines the content of the conversations; a code of conduct acquisition unit that acquires the code of conduct data; a behavioral code compliance determination unit that determines whether the conversation content complies with the behavioral code; an advice keyword extraction unit that extracts advice keywords corresponding to deviations from the code of conduct; have An advice generation system comprising:

2. The conversation acquisition unit acquires at least the "topic," "customer's reaction," and "salesperson's remarks" as conversation contents. Regarding the "topic," determine whether it falls into one of the topic categories, including product function, usage effect, and product price; Regarding the "customer reaction," determine the degree of positivity corresponding to the "topic," Regarding the "salesperson's comments," determine whether they fall into one of the following categories: explanation, question, or proposal.

2. The advice generating system of claim 1.

3. The code of conduct data is provided for each negotiation progress step, the control unit has a progress step input unit for inputting negotiation progress steps, The code of conduct acquisition unit acquires code of conduct data corresponding to the negotiation progress step.

2. The advice generating system of claim 1.

4. the control unit has a progress step confirmation unit that confirms the negotiation progress step based on the determination of the code of conduct conformance determination unit, the progress step input unit inputs the next negotiation progress step if the code of conduct is followed; The code of conduct acquisition unit acquires code of conduct data corresponding to the next negotiation progress step.

4. The advice generating system of claim 3.

5. the storage unit stores sales ability level data relating to the sales ability level of each sales representative; the control unit has a sales ability level acquisition unit that acquires sales ability level data of sales representatives, The code of conduct compliance determination unit determines the degree of compliance with the code of conduct based on the sales ability level.

2. The advice generating system of claim 1.

6. the storage unit stores characteristic data of each sales representative and / or characteristic data of each customer; The control unit a sales activity characteristic determination unit that determines sales activity characteristics based on the characteristic data of each salesperson and / or the characteristic data of each customer; an auxiliary word adding unit that adds an auxiliary word that complements the advice keyword based on the sales activity characteristics; 2. The advice generating system of claim 1, further comprising:

7. a text generation system or capable of communicating with said text generation system; The control unit a writing instruction unit that instructs the writing system to write a sentence based on the advice keyword; 2. The advice generating system of claim 1, further comprising:

8. a text generation system or capable of communicating with said text generation system; The control unit a sentence creation instruction unit that instructs the sentence creation system to create a concise sentence and a detailed sentence based on the advice keyword; 2. The advice generating system of claim 1, further comprising:

9. the conversation acquisition unit acquires conversations at predetermined time intervals, The advice keyword extraction unit extracts advice keywords for the conversation.

2. The advice generating system of claim 1.

10. A method for sending advice to a sales representative in an online business negotiation conducted between a sales representative terminal and a customer terminal using an information processing device, comprising: Code of conduct data relating to the code of conduct that salespeople should follow and advice keyword data relating to advice keywords that should be used when the salespeople deviate from the code of conduct are stored; Acquire the conversation of the online business negotiation and determine the content of the conversation; Acquire the code of conduct data; determining whether the content of the conversation complies with the code of conduct; extracting, from the advice keyword data, advice keywords corresponding to deviations from the code of conduct; Outputting advice based on the advice keywords 10. A method for generating advice, comprising:

11. In online business negotiations between sales representative terminals and customer terminals, in order to send advice to sales representatives, A process of acquiring a conversation of the online business negotiation and determining the content of the conversation; A process of storing code of conduct data relating to a code of conduct that sales representatives should follow and acquiring the code of conduct data; determining whether the conversation content complies with the code of conduct; a process of storing advice keyword data relating to advice keywords corresponding to deviations from the code of conduct, and extracting advice keywords corresponding to deviations from the code of conduct from the advice keyword data; A process for outputting advice based on the advice keyword and causing the information processing device to execute the 10. An advice generation program comprising:

Images