Proposed solution selection assistance device and proposed solution selection assistance method

Abstract

A proposed solution selection assistance device (100) according to the present invention is characterized by comprising: a problem description passage extraction unit (21) whereby a passage in which a problem is described is extracted from a past proposal on the basis of empirical rules; a problem list generation unit (22) that repeats, for individual solutions, processing in which a problem it is possible to solve by introducing a solution is extracted from the extracted passage using a generation AI, and generates a problem list describing combinations of the solutions and the extracted problems; a solution candidate extraction unit (23) that extracts, from the generated problem list, a solution corresponding to a customer problem actually being faced by a customer; and a proposed solution selection unit (24) that determines a priority order for actually proposing the extracted solutions.

Description

Proposed Solution Selection Support Apparatus and Proposed Solution Selection Support Method

[0001] The present invention relates to a proposed solution selection support apparatus and a proposed solution selection support method.

[0002] Inexperienced sales representatives often do not know which of their company's solutions should be proposed to customers for customer issues obtained through customer interviews. As a result, sales representatives require a lot of time for proposal preparation such as investigating past cases and making inquiries within the company, leading to inefficient sales activities. The solution proposal support system of Patent Document 1 searches for a solution corresponding to a customer issue obtained from a customer interview from an issue list in which solutions and issues that can be solved by the solutions are manually associated with each other.

[0003] Japanese Unexamined Patent Publication No. 2023-135499

[0004] It is expected that the types of solutions will increase in the future. However, in the system described in Patent Document 1, it becomes more difficult to manually create an issue list in a short time. Accordingly, a method of creating an issue list using generative AI (Artificial Intelligence), which is good at document interpretation and summarization, is conceivable.

[0005] However, if information such as past proposals and activity information created by sales representatives in the past is simply used as input information for generative AI to extract an issue list, the amount of information becomes enormous, which requires extraction time and cost, making it impossible to efficiently create an issue list. Furthermore, there are cases where customers do not clearly state their own potential issues to sales representatives. In this case, it is impossible to select a solution that solves the issue. Therefore, even if a new solution is introduced, the issue remains unsolved. Accordingly, an object of the present invention is to accurately extract an issue and a solution for solving the issue in a short time.

[0006] The proposed solution selection support device of the present invention is characterized by comprising: a problem description section extraction unit that extracts sections describing problems from past proposals based on empirical rules; a problem list generation unit that generates a problem list describing combinations of the solution and the extracted problems by repeatedly extracting problems that can be solved by introducing the solution from the extracted sections using generation AI for each solution; a solution candidate extraction unit that extracts solutions corresponding to customer problems actually faced by the customer from the generated problem list; and a proposed solution selection unit that determines the priority of the extracted solutions to be actually proposed. Other means will be described in the embodiments for carrying out the invention.

[0007] According to the present invention, problems and solutions to solve them can be accurately and quickly identified.

[0008] This is a diagram of the proposed solution selection support system. This is a hardware configuration diagram of the proposed solution selection support device. This is a sequence diagram of processing centered on the proposed solution selection support device. This is a diagram showing an example of selection information. This is a diagram showing an example of switching position definition information. This is a diagram showing an example of solution information. This is a diagram showing an example of past proposal information. This is a diagram showing an example of a problem list. This is a flowchart of the problem list generation processing procedure. This is a diagram showing an example of a problem list creation screen. This is a diagram showing an example of solution constraint information. This is a diagram showing an example of a customer problem list (before change). This is a diagram showing an example of customer constraint information. This is a flowchart of the solution candidate extraction processing procedure. This is a diagram showing an example of a customer problem inquiry screen. This is a diagram showing an example of a problem list (after customer problem allocation). This is a diagram showing an example of an additional customer problem inquiry screen. This is a diagram showing an example of a customer problem list (after change). This is a diagram showing an example of a solution implementation customer list. This is a diagram showing an example of evaluation results (initial value). This is a diagram showing an example of evaluation results (after problem addition). This is a flowchart of the proposed solution selection processing procedure. This is a diagram showing an example of a recommended solution confirmation screen. This is a diagram showing an example of a detailed solution selection reason confirmation screen.

[0009] Hereafter, embodiments of the present invention (referred to as "this embodiment") will be described in detail with reference to the figures.

[0010] (Introduction) To better understand the features of this embodiment, we will first describe the overall picture of this embodiment. This embodiment enables sales representatives to extract problems that can be solved by introducing a solution from a vast amount of past proposals they have created in the past, in a short time and at low cost. To this end, this embodiment uses selection information that indicates the parts to be included in the extraction and the parts to be excluded using keywords, and switching position definition information that indicates the boundary between the parts to be included in the extraction and the parts to be excluded using the relative positions before and after. As a result, this embodiment efficiently and automatically extracts only the slide pages of past proposals that contain problems that can be solved by introducing a solution.

[0011] In this embodiment, using the generated AI, a list of issues is generated from the extracted slide pages in this manner, summarizing the correspondence between customer issues and solutions that can solve those issues. This saves time spent searching for solutions to customer issues and reduces search costs through the use of the generated AI.

[0012] In this embodiment, solutions corresponding to customer issues obtained by the sales representative during customer interviews are extracted from the list of issues created in this manner. In this embodiment, from among the multiple solutions listed in the issue list, solutions in which all issues that can be solved match the customer issues, as well as solutions in which only some of the issues that can be solved match the customer issues, are extracted as candidates. Then, this embodiment presents (displays) the extracted solutions and the candidates for additional customer issues to the sales representative.

[0013] This embodiment prioritizes the candidate solutions extracted in this way for actual proposal. This allows the embodiment to compensate for any oversights in the understanding of customer issues by sales representatives and customers, and furthermore, to quickly propose solutions that are suitable for the customer according to the priority.

[0014] Figure 1 is a diagram showing the configuration of the proposed solution selection support system 10. The proposed solution selection support system 10 comprises a proposed solution selection support device 100, a user terminal 300, and a data management device 400. The proposed solution selection support device 100 is an information processing device such as a personal computer or a server computer, and comprises a storage unit 110, an arithmetic unit 120, an input unit 130, and a display unit 140.

[0015] The storage unit 110 stores past proposal information 31, selection information 32, switching position definition information 33, solution information 34, prompt information 35, CRM (Customer Relationship Management) information 36, solution constraint information 37, customer constraint information 38, issue list 39, customer issue list 40, solution implementation customer list 41, and evaluation results 42. Details of each data will be described later. Note that a large number of past proposals 43 may be stored in the storage unit 110, or they may be stored in the user terminal 300 or data management device 400 and then transmitted to the proposed solution selection support device 100 as needed.

[0016] The calculation unit 120 includes a problem description extraction unit 21, a problem list generation unit 22, a solution candidate extraction unit 23, and a proposed solution selection unit 24. The problem list generation unit 22 includes a generation AI 22b. The operation of each of these programs will be described later.

[0017] Network 200 connects the user terminal 300, data management device 400, and proposed solution selection support device 100 in a communication manner. The user terminal 300 is an information processing device such as a personal computer. Sales representatives operate the user terminal 300. The data management device 400 is, for example, a system such as PDM (Product Data Management) or a similar device. The data management device 400 consists of a storage device for storing data (product data) and a computer for sending and receiving data.

[0018] The sales representative registers selection information 32 with the proposed solution selection support device 100 in advance via the user terminal 300. The proposed solution selection support device 100 then acquires past proposal information 31 from the data management device 400 via the input unit 130. The input unit 130 is also a program that controls the input processing to the proposed solution selection support device 100.

[0019] The issue description section extraction unit 21 uses the selection information 32 and past proposal information 31 to extract the necessary parts (issue description sections) from past proposals 43 for issue extraction. The issue list generation unit 22 uses the extracted issue description sections, prompt information 35 and CRM information 36 to cause the generation AI 22b to create an issue list 39, which is a list summarizing the correspondence between issues and solutions that can solve those issues. The prompt information 35 is input (instruction) information for the generation AI (details below).

[0020] Next, upon obtaining customer issues, the sales representative registers the customer issue list 40 and customer constraint information 38 with the proposed solution selection support device 100 via the user terminal 300. The solution candidate extraction unit 23 extracts candidate solutions from the issue list 39 that include the customer issues listed in the customer issue list 40. The proposed solution selection unit 24 calculates the priority order for actually proposing the extracted solution candidates to the customer based on customer satisfaction with the issues, the number of companies that have implemented the solutions, etc., and creates an evaluation result 42. The display unit 140 presents the evaluation result 42 to the sales representative. The display unit 140 is also a program that controls the output processing from the proposed solution selection support device 100.

[0021] The proposed solution selection support device 100 utilizes past proposal information 31 and selection information 32 to extract only the slides containing issues from a vast number of past proposals 43. The proposed solution selection support device 100 then allows the generating AI 22b to refer only to the extracted slides. Through this process, the proposed solution selection support device 100 quickly and inexpensively creates a list of issues 39 in advance, summarizing the issues that can be solved by introducing a solution for each solution.

[0022] The sales representative uses the problem list 39 to extract candidate solutions that can actually solve the problems the customer is facing. At that time, the proposed solution selection support device 100 performs processing such as giving higher priority to solutions that have been implemented by other customers who have the same problems as a customer who previously implemented a solution. Calculating the priority of proposed solutions in this way makes it easier for the sales representative to select a proposed solution.

[0023] Figure 2 is a hardware configuration diagram of the proposed solution selection support device 100. The proposed solution selection support device 100 includes a CPU (Central Processing Unit) 11, RAM (Random Access Memory) 12, ROM (Read Only Memory) 13, auxiliary storage device 14, display device 15, input device 16, media reader 17, and information transmission / reception device 18.

[0024] The CPU 11 is a unit that performs various calculations. The CPU 11 performs various processes by executing a predetermined proposed solution selection support program loaded from the auxiliary storage device 14 into the RAM 12. The proposed solution selection support program is, for example, an application program that can be executed on the OS (Operating System) program, and includes the aforementioned problem description location extraction unit 21, problem list generation unit 22, solution candidate extraction unit 23, and proposed solution selection unit 24. The proposed solution selection support program may be installed in the auxiliary storage device 14 from a portable storage medium via a media reader 17, for example.

[0025] RAM 12 is memory that temporarily stores programs executed by the CPU 11 and data necessary for program execution. ROM 13 is memory that temporarily stores programs necessary for starting the proposed solution selection support device 100. Auxiliary storage device 14 is a non-volatile storage device such as an HDD (Hard Disk Drive) or SSD (Solid State Drive).

[0026] The display device 15 is a device such as a CRT (Cathode Ray Tube) display, LCD (Liquid Crystal Display), or organic EL (Electro-Luminescence) display. The input device 16 is a device such as a keyboard, mouse, or microphone. The media reader 17 is a device that reads information from portable storage media such as CD-ROMs. The information transmission / reception device 18 is a device such as a wired LAN, wireless LAN, dial-up router, or infrared communication device that transmits and receives data with external devices.

[0027] The arithmetic unit 120 in Figure 1 corresponds to the CPU 11 in Figure 2. The storage unit 110 in Figure 1 corresponds to the auxiliary storage device 14 in Figure 2. The input unit 130 in Figure 1 controls the input device 16 in Figure 2. The display unit 140 in Figure 1 controls the display device 15 in Figure 2.

[0028] Figure 3 is a sequence diagram of the process centered on the proposed solution selection support device 100. The outline of each step in Figure 3 will be explained below. These processes will be described again later using the flowcharts in Figures 9, 14, and 22.

[0029] In step S201, the user terminal 300 instructs the proposed solution selection support device 100 to start the issue list generation process, in accordance with the instructions of the sales representative. Furthermore, the user terminal 300 registers selection information 32 with the proposed solution selection support device 100 so that sections of the past proposal document 43 that describe issues that can be solved by introducing a solution can be extracted in a short time. The proposed solution selection support device 100 then receives the start command and the selection information 32. However, if the selection information 32 is already registered in the data management device 400, or if the selection information 32 is not needed because all slide pages of the past proposal document 43 are to be used, then registration of the selection information 32 is not necessary. In addition, the user terminal 300 can also check the selection information 32 already registered in the data management device 400 and change only the sections that need to be added or modified.

[0030] In step S202, the proposed solution selection support device 100 requests past proposal information 31 and CRM information 36 from the data management device 400. The data management device 400 then receives this request. In step S203, the data management device 400 transmits the past proposal information 31 and CRM information 36 to the proposed solution selection support device 100. The proposed solution selection support device 100 then receives this information.

[0031] In step S204, the proposed solution selection support device 100 uses the received past proposal information 31 and CRM information 36, as well as the selection information 32 previously registered by the user terminal 300, to extract only the slide pages from the vast number of past proposals 43 that describe problems that can be solved by introducing the solution. Details of the extraction method will be described later.

[0032] In step S205, the proposed solution selection support device 100 uses the generation AI 22b to repeatedly extract, for each solution, the issues that a key solution can solve, from only the extracted slide pages. Through this process, the proposed solution selection support device 100 generates an issue list 39 that associates the extracted issues with the solutions (details described later). Furthermore, the proposed solution selection support device 100 uses the generation AI 22b to repeatedly extract, for each solution, the issues that a key solution can solve, from publicly available vendor information such as the internet, customer contact information, contract status, sales representative sales activity information, and CRM information 36 that manages customer relationship information. Through this process, the proposed solution selection support device 100 generates an issue list 39 that associates the extracted issues with the solutions (details described later).

[0033] In step S206, the proposed solution selection support device 100 stores the generated issue list 39 in the data management device 400. In step S207, the proposed solution selection support device 100 transmits the generated issue list 39 to the user terminal 300.

[0034] Up to this point, it is essentially a preliminary stage before the sales representative makes contact with a specific customer. From here on, it is essentially the main stage, where the sales representative makes contact with a specific customer. In step S208, the user terminal 300, in response to the sales representative's instructions, commands the proposed solution selection support device 100 to start the solution candidate extraction process. Furthermore, the user terminal 300 registers the specific customer's problem (customer problem) with the proposed solution selection support device 100. The registration screen at this time will be described later. The proposed solution selection support device 100 then receives the activation command and the customer problem.

[0035] In step S209, the proposed solution selection support device 100 requests the problem list 39 and solution constraint information 37 from the data management device 400. The data management device 400 then receives this request. In step S210, the data management device 400 transmits the problem list 39 and solution constraint information 37 to the proposed solution selection support device 100. The proposed solution selection support device 100 then receives this information.

[0036] In step S211, the proposed solution selection support device 100 refers to the problem list 39 and the solution constraint information 37 and extracts candidate solutions from the problem list 39 that can solve the customer problem. In step S212, if there are similar but potential problems that were not explicitly stated as customer problems, the proposed solution selection support device 100 presents (displays) such problems to the user terminal 300 as additional customer problem candidates. The user terminal then receives this feedback.

[0037] In step S213, the user terminal 300 registers the additional customer issues selected by the sales representative from the presented list of additional customer issues as additional customer issues in the proposed solution selection support device 100. The proposed solution selection support device 100 then receives these additional registered issues, registers them to itself, and acquires candidate solutions that can solve them. In step S214, the proposed solution selection support device 100 uses the solution implementation customer list 41, solution constraint information 37, and customer constraint information 38 to determine the priority of the acquired candidate solutions to be proposed. The priority is determined based on the customer satisfaction level of the issues that can be solved, the degree to which they conform to the customer constraint information 38, etc. (details below).

[0038] In step S215, the proposed solution selection support device 100 stores the result of the processing in step S214 as evaluation result 42 in the data management device 400. In step S216, the proposed solution selection support device 100 displays the result of the processing in step S214 as evaluation result 42 on the user terminal 300. The display screen will be described later.

[0039] (Selection Information) Figure 4 shows an example of selection information 32. Selection information 32 is information for quickly extracting only the slide pages that describe "problems that can be solved by introducing the solution" from each of the past proposals 43 that were used when the solution was proposed to other customers in the past. Selection information 32 holds keywords, section numbers, and solution IDs (Identifiers) in relation to each other. Keywords are characteristic words of each past proposal 43 described in the past proposal information 31. Characteristic words are, for example, slide titles. Section numbers are information for linking with the switching start position and selection flag of the switching position definition information 33 in Figure 5. Solution ID is the solution ID of the solution to which the selection information 32 is applied. "ALL" means all solution IDs.

[0040] (Switching Position Definition Information) Figure 5 shows an example of switching position definition information 33. Switching position definition information 33 defines which slides of the past proposal 43 shown in the past proposal information 31 will be basically deleted, or which will be used without deletion, and stores the section number, switching start position, and selection flag in relation to each other. The section number is a number that distinguishes the slide section and is the same as the section number of the selection information 32 in Figure 4. The switching start position indicates which page of the slide the selection will start from. The end position is the switching start position of the next line. The selection flag is either "D" or "A". Of these, "D" means to delete the slide page defined by the section number. "A" means to use the slide page without deleting it.

[0041] (Solution Information) Figure 6 shows an example of solution information 34. Solution information 34 stores the solution ID and solution name in relation to each other. For example, the first row of the record means that the name of the solution with solution ID "001" is "Sales Opportunity Management".

[0042] Hereafter, we will explain how to extract only the slide pages (necessary information) that contain information (locations describing the problems) related to "problems that can be solved by introducing the solution" using the data shown in Figures 4, 5, 6, and 7.

[0043] (Past Proposal Information) Figure 7 is a diagram showing an example of past proposal information 31. Past proposal information 31 maintains the solution ID, past proposal No., slide No., and slide title in an interrelated manner. Slides "1" to "10" correspond to past proposal No. "a001". This indicates that past proposal 43, for which past proposal No. is "a001", consists of 10 slides. Past proposal 43 is a single sales file presented to any customer in the past, and includes at least one combination of problem and solution. The solution for solution ID "001" is the "Sales Opportunity Management" solution (see Figure 6).

[0044] There are the following empirical rules for determining how to extract necessary information. That is, the necessary information to be extracted generally tends to be located at the beginning of a past proposal 43 composed of a plurality of slide pages. However, there are exceptional cases where no necessary information exists even within the pages at the beginning. Conversely, necessary information generally tends not to be located at the end of the past proposal 43. However, there are exceptional cases where necessary information exists even within the pages at the end. Based on experience, slide titles in such exceptional cases are predetermined.

[0045] In order to efficiently extract slides that include problem description sections by utilizing the aforementioned empirical rules, FIG. 4, FIG. 5, and FIG. 6 are used. First, in order to define that the necessary information is from the first page to page P at the beginning of the slides, the switching position definition information 33 in FIG. 5 is used. The fact that the switching start position "0%" is described in the first line of FIG. 5 and the switching start position "70%" is stored in the second line means that the slide section number "1" corresponds to the 0% to 70% range of pages at the beginning of the slides.

[0046] Regarding the selection of slides to be deleted or restored, first, the first line of FIG. 5 means that slides located within the first 70% of the beginning are basically utilized without being deleted. However, if a keyword of section number "1" in FIG. 4 is included as a slide title within the first 70% of the beginning, the slide containing that keyword is deleted as unnecessary. Consequently, even if located at the beginning, sections including "table of contents", "company introduction", or "promotion structure" are deleted as they are not considered necessary information.

[0047] Next, the second line of FIG. 5 means that slides located between 70% and 90% in the middle are basically deleted. However, if a keyword of section number "2" in FIG. 4 is included in a slide title between 70% and 90% in the middle, the slide containing that keyword is restored as necessary information without being deleted. Consequently, even if located in the middle, sections including "case examples", "samples", or "future vision" are restored as necessary information.

[0048] The solution ID "ALL" in FIG. 4 means that selection (deletion, restoration) is performed for all solutions. The solution ID "001" in the fourth row means that selection (utilization, deletion, restoration) is performed only for the solution whose solution ID is "001".

[0049] Finally, the third row of FIG. 5 means that slides positioned after 90% of the end of a slide deck are basically utilized without being deleted. At this time, there are no exceptionally deleted portions even if they are in the end section.

[0050] According to the above rules, when selecting slide titles in FIG. 7, first, slides whose slide numbers are from "1" to "7" are positioned within 70% of the end of the slide deck, so they are utilized without being deleted. However, the slide title "Table of Contents" of the slide whose slide number is "2" matches the selection information 32 in FIG. 4. Therefore, the slide whose slide number is "2", which was once scheduled to be utilized, is determined not to be necessary information and is deleted. The slide whose slide number is "3", which was once scheduled to be utilized, is also deleted for the same reason.

[0051] Subsequently, slides whose slide numbers are "8" and "9" are positioned between 70% and 90% of the middle section of the slide deck, so they are deleted. However, the slide title of the slide whose slide number is "9" is "Cases of other companies". In the fourth row of FIG. 4, the keyword is "case", the section number is "2", and the solution ID is "001". Therefore, the slide whose slide number is "9", which was once deleted, is determined to be necessary information and is restored. Finally, the slide whose slide number is "10" is positioned after 90% of the end of the slide deck, so it is utilized without being deleted.

[0052] As described above, this embodiment efficiently extracts only the seven other slides containing the problem description by deleting the three slides with slide numbers "2", "3", and "8". This process significantly improves the efficiency of the subsequent process using the generation AI. In the example of this embodiment, the sample slides number only 10 pages, but if the number of slides exceeds 100 pages, or even if there are a huge number of such files spanning several years, this process will greatly contribute to reducing processing time and costs.

[0053] The above discussion concerns the matching of slide titles (Figure 7) and keywords (Figure 4). However, slide titles may be replaced with characteristic words (frequently occurring words, etc.) for each page or paragraph. The keywords in Figure 4 are words that are used or not used in conjunction with the topic. The transition start position in Figure 5 may be replaced with a percentage of the number of characters instead of a percentage of the number of pages. In this case, the transition start position can be said to be a relative position before or after which the topic is likely to be included or not included.

[0054] In conclusion, the issue description extraction unit 21 extracts from past proposals 43 sections that contain keywords used with the issue and are in a relative position before or after the section that is likely to contain the issue, and / or deletes from past proposals 43 sections that contain keywords not used with the issue and are in a relative position before or after the section that is likely to not contain the issue.

[0055] (Issue List) Figure 8 shows an example of the Issue List 39. The Issue List 39 holds customer issues, solution IDs, and applicability in an interrelated manner. The customer issue is the content of the customer issue that can be solved by implementing the solution with the solution ID. The solution ID is the solution ID of the solution that can solve the customer issue. In the Applicability column, a check mark is stored if the customer issue matches the customer issue that the customer is actually facing. If they do not match, the column is left blank. The usage of the Issue List 39 will be described later.

[0056] (Procedure for Generating a Problem List) Figure 9 is a flowchart of the procedure for generating a problem list. The problem description extraction unit 21 executes steps S301 and S302 of the problem list generation procedure. The problem list generation unit 22 repeats steps S303 to S305 of the problem list generation procedure for each solution that is the key. For example, the problem list generation unit 22 executes steps S303 to S305 for solution "001", then executes steps S303 to S305 for solution "002", ..., and finally executes step S306. The trigger for starting the problem list generation procedure will be described later.

[0057] In step S301, the issue description extraction unit 21 obtains past proposal information 31, selection information 32, prompt information 35, and CRM information 36 from the data management device 400. In step S302, as described above in the example in Figure 7, the issue description extraction unit 21 extracts the slide pages containing the issue description from the past proposals 43 based on empirical rules.

[0058] In step S303, the issue list generation unit 22 uses the generation AI 22b to extract issues from the issue extraction locations that can be solved by the key solution. At this time, the issue list generation unit 22 inputs prompt information 35 to the generation AI 22b. Prompt information 35 is an instruction (input information) to the generation AI. The prompt information 35 here contains, for example, "Extract the sections from {{issue description sections in past proposals}} that describe issues that can be solved by introducing a solution characterized by XXX, and output each in a bulleted list of 100 characters or less." The section "characterized by XXX" describes the specific role (function) of the key solution (the same applies below). The issue list generation unit 22 can add roles to the prompt information 35, provide examples of content to be extracted or excluded, and instruct the AI ​​to summarize the extracted content.

[0059] Past proposal document 43 is, for example, a multi-page slide. The section describing the problem is a single slide. The section describing the problem includes "the problem that can be solved by the key solution." In other words, the inclusion relationship is Past proposal document > Problem description section > "the problem that can be solved by the key solution." The amount of information in the problem description section is far less than the amount of information in the original Past proposal document 43 from which it is extracted. Therefore, extracting "the problem that can be solved by the key solution" from the problem description section is less burdensome in steps S303 to S305 than directly extracting "the problem that can be solved by the key solution" from Past proposal document 43.

[0060] In step S304, the issue list generation unit 22 uses the generation AI 22b to extract issues that can be solved by the key solution from publicly available information (vendor information) on the Web. At this time, the issue list generation unit 22 inputs prompt information 35 to the generation AI 22b. The prompt information 35 here contains, for example, "Extract 10 issues from {{vendor information}} that can be solved by introducing a solution characterized by XXX, summarize each of them in 50 characters or less, and output them as a bulleted list." The number of issues to extract, the number of characters in the summary, etc., can be changed as needed.

[0061] In step S305, the issue list generation unit 22 uses the generation AI 22b to extract issues from the CRM information 36 that can be solved by the key solution. In particular, the activity information within the CRM information 36 includes information such as "customer problems, issues, and challenges" obtained through interviews with customers. At this time, the issue list generation unit 22 inputs prompt information 35 to the generation AI 22b. The prompt information 35 here contains, for example, "Extract the sections from the {{activity information of the CRM information}} that describe issues that can be solved by introducing a solution characterized by ○○○, summarize each of them in 100 characters or less, and output them as a bulleted list." The number of issues to extract, the number of characters in the summary, etc., can be changed as needed.

[0062] In step S306, the issue list generation unit 22 removes duplicates of the issues extracted in steps S303 to S305 and creates the issue list 39 shown in Figure 8. The solution ID in the issue list 39 contains the ID of the solution used as the key. After that, the issue list generation process procedure is terminated.

[0063] The task list creation screen 51 (Figure 10) may be used as a trigger to start the task list generation process procedure shown in Figure 9. Figure 10 is a diagram showing an example of the task list creation screen 51. The task description location extraction unit 21 displays the task list creation screen 51 on the user terminal 300. The contents displayed in the selection keyword confirmation field and the task list creation solution selection field of the task list creation screen 51 correspond to the selection information 32 and solution information 34, respectively.

[0064] The sales representative can register and delete records in the selection information 32. The sales representative registers the solution ID and solution name in the solution information 34, and further selects the solution for which they want to create the issue list 39. When the sales representative presses the registration completion button, the issue description location extraction unit 21 executes the processes in steps S301 and S302 of Figure 9 with the contents specified in the selection keyword confirmation field. Subsequently, when the sales representative presses the execute button, the issue list generation unit 22 executes the processes in steps S303 to S306 of Figure 9 with the contents specified in the issue list creation solution selection field.

[0065] Next, we will explain how sales representatives select the most appropriate solution for a customer's challenges from among many potential solutions.

[0066] (Solution Constraint Information) Figure 11 shows an example of solution constraint information 37. Solution constraint information 37 holds the solution ID, implementation cost, and implementation period in an interrelated manner. The implementation cost is the amount required to implement the solution. The implementation period is the period required to implement the solution.

[0067] (Customer Issue List (Before Change)) Figure 12 shows an example of the Customer Issue List (Before Change) 40. The Customer Issue List (Before Change) 40 maintains the customer issue number, customer name, customer issue, and whether or not an issue exists, all in a related manner. While the Issue List 39 lists issues that are not limited to a specific customer, the Customer Issue List (Before Change) 40 lists issues specific to a particular customer.

[0068] (Customer Constraints) Figure 13 shows an example of customer constraint information 38. Customer constraint information 38 holds customer name, budget amount, and implementation period in an interrelated manner. The budget amount is the amount that the customer has set aside for the implementation of the solution. The implementation period is the period that the customer has set aside for the implementation of the solution.

[0069] (Solution Candidate Extraction Processing Procedure) Figure 14 is a flowchart of the solution candidate extraction processing procedure. In step S401, the solution candidate extraction unit 23 receives the customer issue list (before modification) 40 and customer constraint information 38 from the sales representative via the user terminal 300. At this time, the solution candidate extraction unit 23 displays the customer issue inquiry screen 61 shown in Figure 15 on the user terminal 300. Although we are in the middle of Figure 14, the explanation will now move to Figure 15.

[0070] Figure 15 shows an example of the customer issue inquiry screen 61. The customer issue inquiry section on the left side of the customer issue inquiry screen 61 is a list extracted from the issue list 39 in Figure 8, containing only customer issues without duplication. The sales representative understands the issues the customer is facing through interviews with the customer, etc. If the customer issue on the screen matches the issue that was interviewed by the customer, they place a check mark in the "Matching" column; otherwise, they leave the "Matching" column blank.

[0071] Furthermore, the sales representative registers the customer's name in the customer name field in the upper right section of the customer issue inquiry screen 61, and enters the budget amount and implementation period that the customer has secured for solution implementation in the customer constraints field in the lower right section. After completing the registration of these three items, the sales representative presses the registration completion button. This creates the customer issue list (before change) 40 in Figure 12 and the customer constraints information 38 in Figure 13. The explanation returns to Figure 14.

[0072] In step S402, the solution candidate extraction unit 23 extracts solutions from the issue list 39 that address the actual customer issues the customer faces. Specifically, the solution candidate extraction unit 23 stores a check mark in the "Applicability" column for records in the issue list 39 that have the same customer issues as the customer issues in the customer issue list (before modification) 40.

[0073] For example, in the customer issue list (before modification) 40 in Figure 12, customer issue No. 1 for customer X is "centralized management of customer information," and a check mark is placed in the issue presence / absence column. In this case, the solution candidate extraction unit 23 stores a check mark in the applicability column for all records in the issue list 39 in Figure 8 whose solution IDs are "001" to "004" and whose customer issue is "centralized management of customer information." The solution candidate extraction unit 23 performs the same process for other customer issues that have check marks in the customer issue list (before modification) 40, and creates the issue list (after customer issue allocation) 39 in Figure 16. We are still in the middle of Figure 14, but the explanation will now move on to Figure 16.

[0074] (Issue List (After Customer Issue Assignment)) Figure 16 shows an example of the issue list (after customer issue assignment) 39. The issue list (after customer issue assignment) 39 has the same structure as the issue list 39 in Figure 8. Compared to Figure 8, in Figure 16, a check mark is stored in the "Yes / No" column. The explanation returns to Figure 14.

[0075] In step S403, the solution candidate extraction unit 23 outputs the additional customer issue inquiry screen 71. Specifically, the solution candidate extraction unit 23 displays the additional customer issue inquiry screen 71 (Figure 17) on the user terminal 300 so that the sales representative can confirm the contents of the issue list (after customer issue allocation) 39 (Figure 16). Although we are still in the middle of Figure 14, the explanation will now move to Figure 17.

[0076] Figure 17 shows an example of the Additional Customer Issue Inquiry Screen 71. In the Additional Customer Issue Inquiry section, the Solution Candidate Extraction Unit 23 presents customer issues ("Centralized Management of Customer Information", "Strengthening Sales Capabilities", and "Data Analysis") and issues other than customer issues ("Visualization of Sales Activities", "Sales Forecasting", and "Inventory Management") in the Customer Issue section. The Solution Candidate Extraction Unit 23 also presents solutions in the Issue Resolution Solution section that can solve at least one customer issue and simultaneously solve at least one issue other than customer issues. Furthermore, in the "Additional Availability" section, the Solution Candidate Extraction Unit 23 displays a black circle for issues already registered as customer issues and a white circle for issues that are not yet registered and are therefore candidates for addition.

[0077] Sales representatives can check the additional customer issue inquiry screen 71 to see if there are any customer issues that were missed during the initial issue entry. If there are any missing customer issues, the sales representative places a black circle in the "Additional Issues" column and then presses the "Add Issue Complete" button. For example, if the customer issues "Sales Forecast" and "Inventory Management" are added, the customer issue list (before change) 40 in Figure 12 will be updated to the customer issue list (after change) 40 in Figure 18.

[0078] (Customer Issue List (Revised)) Figure 18 shows an example of the Customer Issue List (Revised) 40. The Customer Issue List (Revised) 40 has the same structure as the Customer Issue List 40 in Figure 12. Compared to Figure 12, Figure 18 has an added check mark in the Issue Presence / Absence column. The explanation returns to Figure 14.

[0079] Subsequently, the solution candidate extraction unit 23 repeats the process in S402 and updates the issue list (after customer issue allocation) 39 in Figure 16. In this way, the sales representative can identify the solution that should actually be proposed to the customer from among many candidates that can solve the customer's problem. The sales representative can also fill in any missing customer issues and add customer issues as needed. After that, the solution candidate extraction process procedure is terminated.

[0080] Next, we will explain how to present the most appropriate proposed solution to the customer's challenges and constraints from among the extracted solution candidates to the sales representative.

[0081] (Solution Implementation Customer List) Figure 19 shows an example of the Solution Implementation Customer List 41. The Solution Implementation Customer List 41 shows the challenges recognized by customers who have already implemented the solution. In other words, the Solution Implementation Customer List 41 shows which solutions other companies with the same challenges have implemented. The more other companies with the same challenges have implemented a solution, the more likely it is to be a suitable solution for the target customers in this case.

[0082] The solution implementation customer list 41 maintains the customer name, issue group, and solution ID in an interrelated manner. The issue group is a binary number (bit value) assigned digit by digit from the most significant digit, in ascending order of the customer issue number in the customer issue list (after modification) 40. In each bit, "0" indicates no issue, and "1" indicates an issue. The customer issues in Figure 18 are customer issue No. 1, customer issue No. 3, customer issue No. 4, customer issue No. 5, and customer issue No. 6. Therefore, the issue group after filling in the missing customer issues for customer X is "101111".

[0083] (Evaluation Results (Initial Values)) Figure 20 shows an example of the evaluation results (initial values) 42. The evaluation results (initial values) 42 maintains the solution ID, customer satisfaction with the problem, number of implementing companies, difference from budget, difference from requested period, additional issues to be resolved, and priority, all of which are interrelated. Initially, there are many blank fields. Customer satisfaction with the problem is created from the customer issue list (before modification) 40 in Figure 12. Customer satisfaction with the problem is the ratio of the number of issues that each solution can solve (numerator) to the total number of issues that need to be solved (denominator).

[0084] The additional customer issue inquiry screen 71 in Figure 17 shows the following regarding the issue resolution solution "001": • Two check marks are displayed in the issue resolution solution "001" column. • Three black circles are displayed in the "Additional" column. • Only one customer issue, "Centralized management of customer information," has both a check mark and black circles. • In conclusion, the issue resolution solution "001" can solve only one of the three customer issues. • Therefore, the customer issue satisfaction score in the first row of Figure 20 is "1 / 3".

[0085] Similarly, the additional customer issue inquiry screen 71 in Figure 17 shows the following regarding the issue-solving solution "002": • Three check marks are displayed in the issue-solving solution "002" column. • Three black circles are displayed in the "Additional" column. • The customer issues for which both check marks and black circles are displayed are "Centralized management of customer information," "Strengthening sales capabilities," and "Data analysis." • In conclusion, the issue-solving solution "002" can solve all three of the three customer issues. • Therefore, the customer issue satisfaction score in the second row of Figure 20 is "3 / 3."

[0086] Details regarding the number of companies implementing the solution, the difference between the budget and the actual cost, the difference between the requested timeframe and the actual cost, and the additional issues to be addressed and their priorities will be described later.

[0087] (Evaluation Results (After Addition of Issues)) Figure 21 shows an example of the evaluation results (after addition of issues) 42. Specifically, Figure 21 corresponds to an example where, as shown in Figure 18, the number of customer issues (number of check marks indicating the presence or absence of issues) increased to five due to a review of customer issues for customer X. For example, in Figure 20, the customer issue satisfaction for solution ID "003" was "3 / 3", but "sales forecasting" and "inventory management" were added due to a review of customer issues. Of these, "sales forecasting" is considered an issue for solution ID "003". As a result, the customer issue satisfaction for solution ID "003" becomes "4 / 5". Further explanation of Figure 21 will be given later.

[0088] (Proposed Solution Selection Process Procedure) Figure 22 is a flowchart of the proposed solution selection process procedure. The proposed solution selection unit 24 evaluates which solutions can solve the customer's problem and calculates the priority of the proposed solutions. In step S501, the proposed solution selection unit 24 registers an additional customer problem using the additional customer problem inquiry screen 71 in Figure 17. This process is as described above in the screen explanation of Figure 17.

[0089] In step S502, the proposed solution selection unit 24 calculates the number of companies that have implemented a solution within the same issue group. In Figure 18, the customer issues are No. 1, No. 3, No. 4, No. 5, and No. 6. Then, customer X's issue group is "101111". The proposed solution selection unit 24 then extracts the customer names and solution IDs from the solution implementation customer list 41 in Figure 19 whose issue group is the same as customer X's, "101111".

[0090] The proposed solution selection unit 24 aggregates the number of companies that have adopted each solution ID. The aggregated results show that for solution ID "001", there is one company, "Customer 1", for solution ID "002", there is one company, "Customer 2", and for solution ID "003", there are three companies, "Customer 1", "Customer 2", and "Customer 3". The maximum number of companies that have adopted all solutions is "3".

[0091] Therefore, the ratio of companies that have adopted the solution (number of companies / maximum number) is 1 / 3 for solution ID "001", 1 / 3 for solution ID "002", 3 / 3 for solution ID "003", and 0 / 3 for solution ID "004".

[0092] In step S503, the proposed solution selection unit 24 determines whether the solution constraints are within the scope of the customer constraints. Specifically, the proposed solution selection unit 24 compares the customer constraint information 38 in Figure 13 with the solution constraint information 37 in Figure 11 to determine whether the solution constraints satisfy the customer constraints.

[0093] First, regarding the implementation cost, customer X's budget is "10 million yen," while the implementation cost for solution ID "001" is "2 million yen." Therefore, the difference between the budget and the implementation cost, "-8 million yen," obtained by subtracting the budget amount in Figure 13 from the implementation cost in Figure 11, is within the constraint range. The implementation cost for solution ID "002" is "6 million yen," so the difference between that amount and the budget, "-4 million yen," is within the constraint range. The implementation cost for solution ID "003" is "9 million yen," so the difference between that amount and the budget, "-1 million yen," is within the constraint range. The implementation cost for solution ID "004" is "20 million yen," so the difference between that amount and the budget, "+10 million yen," is outside the constraint range.

[0094] Next, regarding the implementation period, the implementation period for customer X is "9 months," while the implementation period for solution ID "001" is "2 months." Therefore, the difference between the required period (calculated by subtracting the implementation period in Figure 13 from the implementation period in Figure 11) and the implemented period ("-7 months") is within the constraint range. The implementation period for solution ID "002" is "3 months," so the difference between that and the required period ("-6 months") is within the constraint range. The implementation period for solution ID "003" is "6 months," so the difference between that and the required period ("-3 months") is within the constraint range. The implementation period for solution ID "004" is "1 year," so the difference between that and the required period ("+3 months") is outside the constraint range.

[0095] In step S504, the proposed solution selection unit 24 uses the calculation results from steps S502 and S503 to determine the priority of the solutions to be proposed from among the multiple solution candidates. In this case, the proposed solution selection unit 24 has determined the priority of the solution candidates as follows.

[0096] (1) Solutions with high customer satisfaction will be ranked higher. (2) If there is no difference in (1), solutions with a large proportion of companies that have adopted the solution will be ranked higher. (3) If there is no difference in (2), solutions with implementation costs and implementation periods within the constraints will be ranked higher. (4) If there is no difference in (3), solutions with the smallest excess of implementation costs and implementation periods will be ranked higher.

[0097] As a result, the proposed solution selection unit 24 selected solution "003," which has the highest priority, as the solution to propose to the customer. In this case, high customer satisfaction with the problem was given a high priority, but other constraints may be given higher priority depending on the situation. The proposed solution selection unit 24 updates the evaluation result (initial value) 42 to the evaluation result (after problem addition) 42 using the processing results of steps S501 to S504. After that, the proposed solution selection processing procedure is terminated.

[0098] Figure 23 shows an example of the recommended solution confirmation screen 81. The proposed solution selection unit 24 displays the recommended solution confirmation screen 81 via the user terminal 300. The recommended solution confirmation screen 81 displays the priority and solution ID of the evaluation result (after adding issues) 42 in Figure 21, which was determined in step S504 of Figure 22. If the sales representative wants to check the details that formed the basis of the priority displayed here, they press the details confirmation button. Then, the proposed solution selection unit 24 displays the solution selection reason details confirmation screen 91 in Figure 24 via the user terminal 300.

[0099] Figure 24 shows an example of the Solution Selection Reason Details Confirmation Screen 91. The Solution Selection Reason Details section at the top of the Solution Selection Reason Details Confirmation Screen 91 displays the details of the solution selection reason. The details of the solution selection reason include, for example, the solution ID, customer satisfaction, number of companies that have implemented the solution, the difference in amount from the budget, the difference from the requested period, additional issues to be resolved, and whether or not there are issues that can be resolved by implementing the solution. The proposed solution selection unit 24 creates the Solution Selection Reason Details section using the evaluation results (after issue addition) 42 in Figure 21 and the customer issue list (after modification) 40 in Figure 18. The proposed solution selection unit 24 then copies a portion of the customer issue list (after modification) 40 in Figure 18 and a portion of the customer constraint information 38 in Figure 13 to the lower part of the Solution Selection Reason Details Confirmation Screen 91.

[0100] As described above, by using information such as activity data contained in past proposals, vendor information, and CRM information, the AI ​​can generate a list of issues that can be solved by implementing a solution. This allows even inexperienced sales representatives to extract candidates for addressing customer issues from a vast number of solutions and present a priority list of proposed solutions. This reduces the time it takes from customer visit to proposal.

[0101] It should be noted that the present invention is not limited to the embodiments described above, and various modifications are included. For example, the embodiments described above are described in detail for the purpose of explaining the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the configurations described. Furthermore, it is possible to replace parts of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add configurations from other embodiments to the configuration of one embodiment. It is possible to add, delete, or replace parts of the configuration of each embodiment with other configurations.

[0102] 10 Proposal Solution Selection Support System 21 Problem Description Section Extraction Unit 22 Problem List Generation Unit 22b Generation AI 23 Solution Candidate Extraction Unit 24 Proposal Solution Selection Unit 31 Past Proposal Information 32 Selection Information 33 Switching Position Definition Information 34 Solution Information 35 Prompt Information 36 CRM Information 37 Solution Constraint Information 38 Customer Constraint Information 39 Problem List 40 Customer Problem List 41 Solution Implementation Customer List 42 Evaluation Results 43 Past Proposals 100 Proposal Solution Selection Support Device

Claims

1. A proposal solution selection support device comprising: a problem description section extraction unit that extracts sections describing problems from past proposals based on empirical rules; a problem list generation unit that generates a problem list describing combinations of the solution and the extracted problems by repeatedly extracting problems that can be solved by introducing the solution from the extracted sections using generation AI for each solution; a solution candidate extraction unit that extracts solutions corresponding to customer problems actually faced by the customer from the generated problem list; and a proposal solution selection unit that determines the priority of the extracted solutions for actual proposals.

2. The proposed solution selection support device according to claim 1, characterized in that the solution candidate extraction unit presents to the customer a solution that can solve the customer problem and other problems simultaneously, and presents to the customer problems other than the customer problem.

3. The proposed solution selection support device according to claim 1, characterized in that the problem list generation unit extracts problems that can be solved by introducing the solution from publicly available information and / or CRM information in addition to the past proposals, using a generation AI.

4. The proposed solution selection support device according to claim 1, characterized in that the issue description section extraction unit extracts from the past proposal document sections that contain keywords used together with the issue and are in relative positions before and after the issue that are likely to contain the issue, and / or deletes from the past proposal document sections that contain keywords not used together with the issue and are in relative positions before and after the issue that are likely to not contain the issue.

5. A proposal solution selection support method characterized by: a problem description section extraction unit of the proposal solution selection support device extracts sections from past proposals based on empirical rules where problems are described; a problem list generation unit of the proposal solution selection support device repeatedly extracts problems that can be solved by introducing the solution from the extracted sections using generation AI for each solution; and generates a problem list that describes combinations of the solution and the extracted problems; a solution candidate extraction unit of the proposal solution selection support device extracts solutions that correspond to customer problems actually faced by the customer from the generated problem list; and a proposal solution selection unit of the proposal solution selection support device determines the priority for actually proposing the extracted solutions.

Images