Information processing systems, information processing methods, and programs

Abstract

In addition to behavioral logs and customer attribute information related to prospective customers, the system extracts features that indicate the nuances of the prospective customer's response from voice data during phone calls or recorded data of online business meetings. Based on these features, it estimates the prospective customer's needs and generates recommended communication information tailored to that prospective customer. [Solution] An information processing system having at least one control unit, the control unit acquires at least one of the following: behavioral logs and customer attribute information relating to prospective customers; voice data from phone calls and recorded data of online business negotiations; nuance features indicating the nuances of the prospective customer's response from the voice data or recorded data; estimates the prospective customer's needs based on the behavioral logs, customer attribute information, and nuance features; classifies prospective customers into multiple segments based on the estimated needs; analyzes the difference between the group of customers where a success event occurred and the group of customers where a success event did not occur for each segment; and generates recommended communication information to lead prospective customers to a success event based on the results of the difference analysis.

Description

Technical Field

[0001] The present invention relates to an information processing system, an information processing method, and a program.

Background Art

[0002] In recent years, technologies have been proposed for analyzing videos or voices related to sales activities or business negotiations and outputting evaluation information or analysis result information based on the videos or voices.

[0003] For example, a technology has been proposed in which a business video is input, assessment information is estimated based on key actions and evaluation criteria indicating important actions for each phase of the video, and analysis result information is output.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the conventional technologies as described above, although it is possible to evaluate videos or voices related to sales activities or business negotiations and output analysis result information, in addition to the action logs and customer attribute information regarding prospective customers, feature amounts indicating the nuances of the reactions of prospective customers are extracted from voice data during a phone call or recording data of an online business negotiation, and it has not been sufficient to use the feature amounts for estimating the needs of prospective customers.

[0006] Therefore, it has been difficult to appropriately derive recommended communication information according to the prospective customers in consideration of the differences in interests or reactions for each prospective customer. Therefore, the present invention aims to provide an information processing system, information processing method, and program that can extract feature quantities indicating the nuances of a prospective customer's response from voice data during phone calls or recorded data of online business negotiations, in addition to behavioral logs and customer attribute information relating to prospective customers, estimate the prospective customer's needs based on these feature quantities, and generate recommended communication information. [Means for solving the problem]

[0007] (1) An information processing system, Having at least one control unit, The control unit, In addition to behavioral logs and customer attribute information regarding prospective customers, at least one of the following will be acquired: voice data from phone calls and recorded data from online business meetings. Further information on the contents of the sales materials will be obtained. From the aforementioned audio data or video data, nuance features indicating the nuances of the prospective customer's response are extracted. Based on the aforementioned behavioral logs, customer attribute information, and nuance features, the needs of the prospective customer are estimated. Based on the estimated needs, the prospective customers are classified into multiple segments, The web activity, email responses, and form completion history included in the aforementioned activity log are integrated with the nuance features to identify the missing information for the prospective customer. Based on the correspondence between the content information of the sales materials and the nuance features, effective appeals to the prospective customer are determined. For each of the aforementioned segments, the difference between the customer group where the outcome event occurred and the customer group where the outcome event did not occur is analyzed. Difference Based on the analysis results, To supplement the aforementioned missing information and to include the aforementioned effective appeal content, To generate recommended communication information to lead the aforementioned prospective customers to a success event, Information processing system. (2) In the information processing system described in (1), The aforementioned nuance features include at least one of the following: speech rate, intonation, silence time, frequency of interjections, negative expressions, positive expressions, frequency of questions, and response intensity to specific topics. Information processing system. (3) In the information processing system according to (1), the control unit extracts video feature amounts indicating expression changes, gaze changes, or nodding motions from the recorded video data, and uses the video feature amounts as the nuance feature amounts. Information processing system. (4) In the information processing system according to (1), the recommended communication information includes at least one of an email text, a talk script, a winning point, content to be presented, and the next contact timing. Information processing system. (5) In the information processing system according to (1), the control unit visualizes and outputs the correspondence relationship among the needs, the segments, and the nuance feature amounts. Information processing system. (6) In the information processing system according to (1), the control unit extracts the nuance feature amounts based on text information obtained by performing speech recognition processing on the voice data. Information processing system. (7) In the information processing system according to (1), the control unit outputs explanation information indicating the reason for generating the estimated needs or the recommended communication information. Information processing system. (8) An information processing method executed by an information processing system, including a step of acquiring at least one of voice data during a call and recorded video data of an online negotiation in addition to the behavior log and customer attribute information regarding a prospective customer; The process of obtaining information about the contents of sales materials, a step of extracting a nuance feature amount indicating the nuance of the reaction of the prospective customer from the voice data or the recorded video data. estimating the needs of the prospective customers based on the action log, the customer attribute information, and the nuance feature amount; classifying the prospective customers into a plurality of segments based on the estimated needs; A step of integrating the web behavior, email responses, and form completion history included in the aforementioned behavior log with the nuance features to identify information missing from the prospective customer, A step of determining effective appeals to the prospective customer based on the correspondence between the content information of the sales materials and the nuance features, for each of the segments, analyzing the difference between the customer group in which the achievement event is established and the customer group in which the achievement event is not established; Difference based on the analysis result; To supplement the aforementioned missing information and to include the aforementioned effective appeal content, generating recommended communication information for leading the prospective customers to the achievement event; An information processing method including the above. (9) A program for causing a computer to function as the information processing system according to any one of (1) to (23). (1) to (7) any one of (23).

Effect of the Invention

[0008] According to the present invention, in addition to the action log and customer attribute information regarding the prospective customers, a feature amount indicating the nuance of the reaction of the prospective customers is extracted from the voice data at the time of a call or the recorded data of an online negotiation, and the needs of the prospective customers can be estimated based on the feature amount. As a result, recommended communication information corresponding to the prospective customers can be generated in consideration of the differences in interests or reactions of each prospective customer, and the appropriateness of sales activities or negotiation responses can be improved.

Brief Description of the Drawings

[0009] [Figure 1] It is a diagram showing an example of the system configuration of the information processing system. [Figure 2] It is a diagram showing an example of the hardware configuration of the information processing device. [Figure 3] It is a diagram showing an example of the hardware configuration of the terminal device. [Figure 4]This is a block diagram showing an example of the functional configuration of an information processing device. [Figure 5] This is a flowchart illustrating an example of the recommended communication information generation process. [Modes for carrying out the invention]

[0010] Embodiments of the present invention will be described below with reference to the drawings. The various features shown in the embodiments below (including modified examples; the same applies hereinafter) can be combined with each other.

[0011] <Embodiment 1> 1. System Configuration Figure 1 shows an example of the system configuration of the information processing system 1000 according to this embodiment. As shown in Figure 1, the information processing system 1000 includes, as a system configuration, an information processing device 100, a terminal device 110, an external marketing system 120, and a conversation analysis system 130.

[0012] The information processing device 100, terminal device 110, external marketing system 120, and conversation analysis system 130 are connected to each other via a network 150. The network 150 includes a WAN (Wide Area Network), a LAN (Local Area Network), and the Internet, or any combination thereof. The network 150 includes a wired network and a wireless network, or both.

[0013] The information processing system 1000 extracts nuanced features indicating the nuances of a prospective customer's response from audio data during phone calls or recorded data of online business negotiations, in addition to behavioral logs and customer attribute information related to prospective customers. Based on these nuanced features, it estimates the prospective customer's needs and generates recommended communication information that corresponds to those needs. This enables support for sales activities or business negotiations that take into account differences in the interests or responses of each prospective customer.

[0014] The information processing device 100 is a device that provides the main functions of the information processing system 1000 and corresponds to a device having a "control unit" in this disclosure. The information processing device 100 acquires behavioral logs and customer attribute information about prospective customers from the external marketing system 120, and also acquires voice data or video data from the conversation analysis system 130, extracts nuance features based on the various data, and performs needs estimation, segment classification, difference analysis, and generation of recommended communication information.

[0015] Terminal device 110 is a device used by users of the information processing system 1000. Users of the information processing system 1000 include, for example, inside sales representatives, sales representatives, negotiation representatives, sales planning representatives, or sales managers. Through terminal device 110, users perform tasks such as selecting prospective customers, confirming estimated needs, confirming segments, confirming differential analysis results, confirming recommended communication information, and viewing explanatory information.

[0016] The external marketing system 120 is an external system that manages data for each prospective customer, and is, for example, a marketing automation system, a customer relationship management system, a sales support system, or a sales opportunity management system. The external marketing system 120 has an external database that holds behavioral logs and customer attribute information related to prospective customers. The external database stores information such as web page browsing history, email delivery history, email open history, link click history, document download history, form submission history, online event participation history, traffic source information, and customer attribute information such as contact person's name, department, industry, job title, region, and company size.

[0017] The external marketing system 120 provides the behavioral log and customer attribute information in response to a request from the information processing device 100. Furthermore, the external marketing system 120 may also provide historical information indicating whether or not an outcome event occurred, past call history, negotiation results, appointment acquisition history, email reply history, conversation memos, identification information of sales materials used, etc.

[0018] The conversation analysis system 130 is a system that stores voice data from phone calls or recorded data from online business negotiations and provides functions to perform speech recognition processing, video analysis processing, or conversation analysis processing as needed. The conversation analysis system 130 may include a speech recognition model, a nuance extraction model, and a video analysis model.

[0019] The information processing device 100 may acquire nuance features or video features by obtaining voice data or video data corresponding to the determined prospective customer from the conversation analysis system 130, or by sending an analysis request to the conversation analysis system 130. The conversation analysis system 130 may be a separate system from the information processing device 100, or it may be incorporated inside the information processing device 100.

[0020] Here, nuance features are features that indicate the nuances of a prospective customer's response, and may include at least one of the following: speech rate, intonation, silence time, frequency of interjections, negative expressions, positive expressions, question frequency, and response intensity to a specific topic. Nuance features may be composed of acoustic features, text features, video features, or a combination thereof.

[0021] Here, an outcome event is an event that indicates that a certain level of progress has been achieved in sales or negotiation activities with a prospective customer. Examples of outcome events may include, for example, the establishment of a conversation, agreement to send materials, agreement to re-contact, appointment setting, demonstration, request for quotation, conversion into a business deal, or order placement. An outcome event is not limited to one, but may consist of multiple stages defined by the system operator.

[0022] Here, recommended communication information refers to information about communications recommended to lead prospective customers to a success event, and may include at least one of the following: email drafts, talk scripts, key points, content to present, timing of the next contact, recommended appeals, recommended explanation order, etc.

[0023] Here, the information processing system described in the claims may consist of multiple devices or of one device. If the information processing system described in the claims consists of one device, an example of such device is an information processing device 100. If the information processing system described in the claims consists of multiple devices, an example of multiple devices is a configuration that includes at least some of the information processing device 100, a terminal device 110, an external marketing system 120, and a conversation analysis system 130.

[0024] In Figure 1, for simplicity, one terminal device 110, one external marketing system 120, and one conversation analysis system 130 are shown, but the number of each is not limited to one. For example, multiple terminal devices 110 may be connected to the information processing device 100, data may be acquired from multiple types of external marketing systems 120, or multiple conversation analysis systems 130 may be used for different purposes.

[0025] Furthermore, the information processing system 1000 may have different configurations for the system that acquires behavioral logs and customer attribute information, the system that acquires audio data or video data, and the system that acquires the history of achievement events. For example, customer attribute information may be acquired from a customer management system, behavioral logs from a marketing automation system, audio data from a call management system, video data from an online meeting management system, and the history of achievement events from a sales support system.

[0026] 2. Hardware Configuration (1) Hardware configuration of the information processing device 100 Figure 2 shows an example of the hardware configuration of the information processing device 100. As shown in Figure 2, the information processing device 100 includes a control unit 210, a storage unit 220, a communication unit 230, and an internal bus 240 as its hardware configuration. The control unit 210, the storage unit 220, and the communication unit 230 are electrically connected via the internal bus 240.

[0027] The control unit 210 includes at least one of the following: CPU (Central Processing Unit), MPU (Micro Processing Unit), SoC (System on Chip), FPGA (Field Programmable Gate Array), ASIC (Application Specific Integrated Circuit), GPU (Graphics Processing Unit), or a combination of at least two of these. It controls the entire information processing device 100 and executes various processes according to information input from other devices or information stored in the storage unit 220.

[0028] The storage unit 220 includes at least one of the following: HDD (Hard Disk Drive), ROM (Read Only Memory), RAM (Random Access Memory), SSD (Solid State Drive), or a combination of at least two of these, and stores a program and data used by the control unit 210 when executing processing based on the program.

[0029] Examples of data used by the control unit 210 when executing processing based on a program include identification information to identify prospective customers, behavioral logs obtained from the external marketing system 120, customer attribute information, voice data, video data, text information obtained by speech recognition, nuance features, video features, estimated needs for each prospective customer, segment information, differential feature information, missing information, recommended communication information, explanatory information, content information of sales materials, and various thresholds or weight coefficients.

[0030] The storage unit 220 is an example of a computer-readable non-temporary recording medium (storage medium) on which a program is recorded. In this specification, the data used by the control unit 210 when executing processing based on the program is described as being stored in the storage unit 220, but it may also be stored in the storage unit of another device that can communicate with the information processing device 100. In other words, the data may be stored in the storage unit of any device as long as the control unit 210 can access and / or retrieve it.

[0031] The control unit 210 executes processing based on the program stored in the memory unit 220, thereby realizing the functional configuration of the information processing device 100 shown in Figure 4, which will be described later, and the processing of the flowchart shown in Figure 5, which will be described later.

[0032] The communication unit 230 is a communication interface that connects the information processing device 100 to the network 150 and mediates communication with other devices. The communication unit 230 can be implemented, for example, by a network interface card, LAN adapter, optical transceiver, etc. that realize wired communication, and / or by wireless communication modules such as Wi-Fi, Bluetooth®, 5G, LTE, NFC, etc. that realize wireless communication, and circuits that control them.

[0033] Based on instructions from the control unit 210, the communication unit 230 performs data transmission and reception processing over the network 150 in accordance with a predetermined communication protocol, such as TCP / IP, UDP, HTTP, or HTTPS.

[0034] Furthermore, the hardware configurations of the control unit 210, storage unit 220, and communication unit 230 are not limited to one. For example, the information processing device 100 may include multiple CPUs, multiple storage devices, or multiple communication interfaces.

[0035] (2) Hardware configuration of terminal device 110 Figure 3 shows an example of the hardware configuration of the terminal device 110. As shown in Figure 3, the terminal device 110 includes, as a hardware configuration, a control unit 310, a storage unit 320, an input / output unit 330, a communication unit 340, and an internal bus 350. The control unit 310, the storage unit 320, the input / output unit 330, and the communication unit 340 are electrically connected via the internal bus 350.

[0036] The control unit 310 includes at least one of the following: CPU, MPU, SoC, FPGA, ASIC, GPU, etc., or at least two combinations thereof, and controls the entire terminal device 110, as well as executing various processes in response to information input from the input / output unit 330 or other devices.

[0037] The storage unit 320 includes at least one of the following: HDD, ROM, RAM, SSD, etc., or at least two combinations thereof, and stores the program and data used by the control unit 310 when it executes processing based on the program. The storage unit 320 is an example of a computer-readable non-temporary recording medium on which the program is recorded.

[0038] In this specification, the data used by the control unit 310 when executing processing based on a program is described as being stored in the storage unit 320, but it may also be stored in the storage unit of another device that can communicate with the terminal device 110.

[0039] The control unit 310 executes processing based on the program stored in the storage unit 320, thereby realizing display processing, input reception processing, viewing target switching processing, visualization display processing, and screen transition processing in the terminal device 110.

[0040] The input / output unit 330 is a device that inputs information to the terminal device 110 in response to user operations, and also outputs information to the user. The input / output unit 330 receives operation inputs made by the user and transfers these operation inputs as command signals to the control unit 310 via the internal bus 350. The input / output unit 330 may also output information as a GUI screen that can be operated by the user. Examples of the input / output unit 330 include a touch panel, a combination of a display device and an input device, a keyboard, a mouse, etc.

[0041] The communication unit 340 is a communication interface that connects the terminal device 110 to the network 150 and mediates communication with other devices. The communication unit 340 can be implemented, for example, by a network interface card, LAN adapter, optical transceiver, etc. that realize wired communication, and / or by wireless communication modules such as Wi-Fi, Bluetooth®, 5G, LTE, NFC, etc. that realize wireless communication, and circuits that control them.

[0042] Based on instructions from the control unit 310, the communication unit 340 performs data transmission and reception processing via the network 150 in accordance with a predetermined communication protocol.

[0043] In this specification, examples of terminal devices 110 include smartphones, tablet devices, and notebook personal computers, but terminal devices 110 are not limited to these. (3) Hardware configuration of the external marketing system 120 and the conversation analysis system 130

[0044] Since the hardware configuration of the external marketing system 120 and the conversation analysis system 130 is the same as that of the information processing device 100 described above, a detailed explanation will be omitted.

[0045] Furthermore, the control unit of the conversation analysis system 130 may be equipped with a computing device such as a GPU, TPU, or NPU in addition to, or instead of, a CPU, in order to perform inference processing using the speech recognition model, nuance extraction model, and video analysis model at high speed.

[0046] 3. Functional Configuration Figure 4 is a block diagram showing an example of the functional configuration of the information processing device 100. As shown in Figure 4, the information processing device 100 includes, for example, an acquisition unit 401, a nuance feature extraction unit 402, a needs estimation unit 403, a segment classification unit 404, a difference analysis unit 405, a missing information identification unit 406, an appeal content determination unit 407, a recommendation information generation unit 408, an explanatory information generation unit 409, a visualization output unit 410, an external linkage unit 411, and a storage control unit 412. These functions may be realized by the control unit 210 executing a program stored in the storage unit 220.

[0047] The acquisition unit 401 acquires at least one of the following: voice data from phone calls and recorded data from online business meetings, in addition to behavioral logs and customer attribute information related to prospective customers. Behavioral logs may include, for example, web page browsing history, email response history, document download history, form completion history, event participation history, etc. Customer attribute information may include, for example, affiliation, industry, job title, company size, region, etc.

[0048] The acquisition unit 401 may acquire audio data such as call recordings, phone call recordings, and audio tracks from online meetings. It may also acquire video data such as recordings of online business meetings and video recordings of web conferences.

[0049] The acquisition unit 401 may further acquire content information of sales materials. Content information of sales materials may include text, image descriptions, chapter titles, topic classifications, or content identification information contained in proposal materials, product description materials, comparison materials, price description materials, case study materials, demo materials, etc.

[0050] The nuance feature extraction unit 402 extracts nuance features from audio data or video data that indicate the nuances of the prospective customer's response. The nuance features may include at least one of the following: speech rate, intonation, silence time, frequency of interjections, negative expressions, positive expressions, frequency of questions, and response intensity to specific topics.

[0051] The nuance feature extraction unit 402 may extract text features such as negative expressions, positive expressions, question frequency, and response intensity to specific topics based on text information obtained by speech recognition processing of the audio data. For example, it may calculate the frequency of negative words such as "expensive," "difficult," and "not needed now," positive words such as "interested," "positive," and "want to know more," or question expressions.

[0052] Furthermore, the nuance feature extraction unit 402 may extract acoustic features such as speech rate, intonation, silence time, interruption frequency, and response delay time based on the speech waveform. For example, a long silence time immediately following a particular explanation may be treated as a feature indicating insufficient understanding or a cautious response, and an increase in intonation or speech rate may be treated as a feature indicating increased interest.

[0053] Furthermore, the nuance feature extraction unit 402 may extract video features from the recorded data that indicate changes in facial expression, changes in gaze, or nodding movements, and use these video features as nuance features. For example, an increase in the number of nods during the presentation of a particular topic may be treated as a tendency toward a positive response, and an increase in gaze deviation may be treated as a feature indicating a decrease in interest or difficulty in understanding.

[0054] The needs estimation unit 403 estimates the needs of prospective customers based on behavioral logs, customer attribute information, and nuance features. Needs may include information that indicates issues of interest to prospective customers, implementation objectives, comparison points, concerns, or desired values. For example, they may include cost-effectiveness needs, ease of implementation needs, comparison and evaluation needs, case study confirmation needs, security confirmation needs, and support system confirmation needs.

[0055] For example, if a prospective customer is viewing comparison materials, frequently asks questions about competitive comparisons during the call, and shows a strong reaction when price is explained, the needs estimation unit 403 may estimate the need for comparison and the need for cost-effectiveness to be high.

[0056] The segment classification unit 404 classifies prospective customers into multiple segments based on their estimated needs. Segments may include, for example, a cost-effectiveness-focused segment, a comparison-focused segment, a case study-focused segment, an ease-of-implementation-focused segment, a careful consideration segment, and so on.

[0057] The segment classification unit 404 may classify based on behavioral logs, customer attribute information, and nuance features in addition to the needs estimation results. For example, even among prospective customers with the same comparative consideration needs, customers with many positive expressions may be classified as the active consideration group, while customers with many periods of silence or negative expressions may be classified as the cautious consideration group.

[0058] The difference analysis unit 405 analyzes the difference between the customer group in which the outcome event occurred and the customer group in which the outcome event did not occur for each segment. The difference analysis unit 405 may extract differences between the two groups in terms of behavioral logs, customer attribute information, nuance features, needs estimation results, etc.

[0059] The difference analysis unit 405 may generate difference feature information used to generate recommended communication information based on the results of the difference analysis. For example, in a specific segment, the unit may extract differences such as a higher frequency of nodding when case studies are presented and a higher rate of viewing comparative materials in the group where the success event was achieved, while longer periods of silence during price explanations are found in the group where the success event was not achieved.

[0060] The missing information identification unit 406 integrates web behavior, email responses, and form completion history included in the behavior log with nuance features to identify information that is missing for the prospective customer.

[0061] Missing information refers to information that the prospective customer has not yet acquired or understood in order to reach a success event, and may include, for example, a lack of understanding regarding comparative criteria, case studies, cost-effectiveness, implementation procedures, security measures, ease of operation, or support systems.

[0062] For example, if a prospective customer has not yet viewed comparative materials and frequently asks questions when comparative topics are presented, or if there are long periods of silence when price explanations are given, the missing information identification unit 406 may identify missing information regarding comparative decision-making materials or cost-effectiveness.

[0063] The appeal content determination unit 407 may determine effective appeal content for prospective customers based on the correspondence between the content information of the sales materials and the nuance features. For example, if positive expressions or nodding actions increase during the presentation of case study slides, the appeal to case study may be determined as effective appeal content, and if the silence time increases when the price list is presented, the appeal to cost-effectiveness may be determined as effective appeal content.

[0064] The recommendation information generation unit 408 generates recommendation communication information to lead prospective customers to a success event, based on the differential analysis results, estimated needs, missing information, and, if necessary, the content of the appeal.

[0065] Recommended communication information may include, for example, at least one of the following: email template, talk script, key points, content to present, timing of the next contact, recommended order of appeals, or recommended order of explanations.

[0066] The explanatory information generation unit 409 generates explanatory information that indicates the reason for generating estimated needs or recommended communication information. The explanatory information may include, for example, referenced behavior logs, customer attribute information, nuance features, needs estimation results, differential feature information, missing information, and reasons for determining the content of the appeal.

[0067] The visualization output unit 410 visualizes and outputs the correspondence between needs, segments, and nuance features. For example, a heatmap may be displayed for each prospective customer, with nuance features on the horizontal axis and needs categories on the vertical axis, or the feature distribution for each segment may be displayed as a graph.

[0068] The external linkage unit 411 performs linkage processing with the external marketing system 120 or the conversation analysis system 130. The external linkage unit 411 may send and receive information acquisition requests from the acquisition unit 401, analysis requests from the nuance feature extraction unit 402, or content acquisition requests from the recommendation information generation unit 408, etc., in accordance with a predetermined communication protocol and API specifications.

[0069] The storage control unit 412 stores the following in the storage unit 220: behavioral logs acquired by the acquisition unit 401, customer attribute information, voice data, video data, nuance features extracted by the nuance feature extraction unit 402, estimated needs by the needs estimation unit 403, classification results by the segment classification unit 404, difference feature information by the difference analysis unit 405, missing information identified by the missing information identification unit 406, appeal content determined by the appeal content determination unit 407, recommended communication information by the recommendation information generation unit 408, and explanatory information by the explanation information generation unit 409, etc., and reads them out as needed and provides them to each functional unit.

[0070] Note that in Figure 4, for the sake of explanation, each function is shown as a separate function block, but these functions do not need to be physically separate and may be implemented by one program module or multiple program modules.

[0071] 4. Information Processing (1) Overview of the process The acquisition unit 401 acquires behavioral logs and customer attribute information related to prospective customers, as well as at least one of the following: voice data from phone calls and recorded data from online business negotiations. The nuance feature extraction unit 402 extracts nuance features that indicate the nuances of the prospective customer's responses from the voice data or the recorded data. The needs estimation unit 403 estimates the prospective customer's needs based on the behavioral logs, customer attribute information, and nuance features. The segment classification unit 404 classifies the prospective customers into multiple segments based on the estimated needs. The difference analysis unit 405 analyzes the difference between the group of customers who achieved a success event and the group of customers who did not achieve a success event for each segment. The recommendation information generation unit 408 generates recommendation communication information to lead the prospective customers to a success event, based at least on the difference analysis results, estimated needs, and segment information.

[0072] Specifically, the acquisition unit 401 acquires, for each prospective customer, behavioral logs such as web page browsing history, email open history, document download history, and form completion history, as well as customer attribute information such as industry, job title, and company size, and telephone call recording data or online business meeting recording data. The nuance feature extraction unit 402 extracts nuance features such as speech rate, silence time, affirmative expressions, negative expressions, question frequency, and facial expression changes from this conversation data.

[0073] The needs estimation unit 403 integrates the behavioral logs, customer attribute information, and nuance features to estimate the needs of prospective customers. The segment classification unit 404 classifies prospective customers into multiple segments based on the estimated needs. The difference analysis unit 405 analyzes the difference between the group where the outcome event occurred and the group where it did not for each segment, and extracts features that are likely to contribute to the occurrence of the outcome event or information categories that are likely to be lacking.

[0074] The missing information identification unit 406 may identify information that is missing for the prospective customer by integrating web behavior, email responses, and form completion history included in the behavior log with nuance features. The appeal content determination unit 407 may determine effective appeal content for the prospective customer based on the correspondence between the content information of the sales materials and the nuance features. The recommendation information generation unit 408 generates recommended communication information such as email drafts, talk scripts, key points, content to be presented, or next contact timing, based on the differential analysis results, estimated needs, segment information, missing information, and, if necessary, the appeal content.

[0075] The explanatory information generation unit 409 generates explanatory information that shows the reason for generating estimated needs or recommended communication information. The visualization output unit 410 visualizes and outputs the correspondence between needs, segments, and nuance features.

[0076] This process allows for the generation of personalized recommendation communication information, taking into account the differences in each prospective customer's interests and responses. As a result, sales activities and negotiation responses can be optimized.

[0077] (2) Details of the process Next, the details of the processing performed by the information processing device 100 will be described. Figure 5 is a flowchart showing an example of the recommended communication information generation process in this embodiment. This process is realized by the control unit 210 of the information processing device 100 executing a program stored in the storage unit 220. In the process shown in Figure 5, the following steps are executed sequentially: acquisition process (step S501), nuance feature extraction process (step S502), needs estimation process (step S503), segment classification process (step S504), difference analysis process (step S505), recommended communication information generation process (step S506), explanatory information generation process (step S507), and visualization output process (step S508).

[0078] First, the acquisition unit 401 acquires behavioral logs and customer attribute information related to prospective customers, as well as at least one of the following: voice data from phone calls and recorded data from online business negotiations (step S501). The acquisition unit 401 may, for example, specify a customer ID, conversation session ID, or deal ID that identifies a prospective customer, call the APIs of the external marketing system 120 and the conversation analysis system 130, and acquire various data associated with that prospective customer.

[0079] Next, the nuance feature extraction unit 402 extracts nuance features from the audio data or video recording data that indicate the nuances of the prospective customer's reaction (step S502).

[0080] For example, the nuance feature extraction unit 402 may extract speech rate, intonation, silence time, frequency of interjections, etc., based on the speech waveform. It may also extract negative expressions, positive expressions, question frequency, or response intensity to a specific topic based on the text information obtained by speech recognition processing. Furthermore, if video data is available, video features indicating changes in facial expression, changes in gaze, or nodding movements may be extracted and used as nuance features.

[0081] Next, the needs estimation unit 403 estimates the needs of prospective customers based on behavioral logs, customer attribute information, and nuance features (step S503).

[0082] For example, if a prospective customer has viewed the pricing information page multiple times, and has a high frequency of questions about pricing in the audio data, and has a long attention time when the topic is presented in the video data, the needs estimation unit 403 may estimate a high cost-effectiveness need or price comparison need.

[0083] Furthermore, if a prospective customer has not viewed case study materials and their affirmative responses or nods increase during the explanation of a case study, the needs estimation unit 403 may estimate a high need to review case studies.

[0084] Next, the segment classification unit 404 classifies prospective customers into multiple segments based on their estimated needs (step S504). For example, prospective customers may be classified into segments such as cost-effectiveness-focused, comparison-focused, case study-focused, and careful consideration segments.

[0085] Next, the difference analysis unit 405 analyzes the difference between the group of customers for whom the outcome event occurred and the group of customers for whom the outcome event did not occur for each segment (step S505).

[0086] The difference analysis unit 405, for example, refers to past customer data and calculates the frequency, mean, or distribution of features derived from behavioral logs, customer attribute information, nuance features, needs estimation results, etc., for the group where the outcome event was successful and the group where it was not, and evaluates the differences therebetween. For example, it may extract positive expressions that are frequently observed in the successful group, short silence times in the successful group, high response intensity to implementation examples in the successful group, and content viewing history that is more common in the successful group.

[0087] The difference analysis unit 405 may generate difference feature information for use by the recommendation information generation unit 408 based on the difference analysis results. The difference feature information may include, for example, a feature identifier, a difference score, statistics for groups that meet the criteria, statistics for groups that do not meet the criteria, related topic identification information, and related content identification information.

[0088] Next, the recommendation information generation unit 408 generates recommendation communication information to lead prospective customers to a success event based on the difference analysis results (step S506).

[0089] The recommendation information generation unit 408 may, for example, integrate estimated needs, missing information, appeal content, differential feature information, and sales material content information to generate at least one of an email draft, a talk script, key selling points, content to be presented, and the timing of the next contact.

[0090] For example, for prospective customers who belong to a comparison-focused segment, lack sufficient information for comparison, and frequently ask questions about comparison topics, it may be recommended to use talk scripts that emphasize comparative advantages, comparison table content, and timing for re-contact within a short period.

[0091] Furthermore, for prospective customers who belong to the cautious consideration segment, exhibit long periods of silence when discussing price, and frequently use negative language, it may be recommended to include supplementary explanations of cost-effectiveness in the email, provide case study content, and suggest a more staggered timing for the next contact.

[0092] Next, the explanatory information generation unit 409 generates explanatory information indicating the reason for generating the estimated needs or recommended communication information (step S507). The explanatory information may include, for example, the referenced behavior log, nuance features, differential feature information, estimated needs, missing information, and the reason for determining the content of the appeal.

[0093] Next, the visualization output unit 410 visualizes and outputs the correspondence between needs, segments, and nuance features (step S508). The visualization output unit 410 may, for example, display the prospective customer's attribute information, estimated needs, segments, main nuance features, difference analysis results, recommended communication information, and explanatory information on the same screen on the terminal device 110.

[0094] In this way, recommended communication information can be generated that takes into account the differences in interest or response of each prospective customer, and can be used in sales activities or negotiations.

[0095] As an example of the difference analysis processing performed by the difference analysis unit 405, the difference analysis unit 405 may calculate the difference features by following the procedure below.

[0096] First, the difference analysis unit 405 obtains label information indicating whether or not a success event occurred for past prospective customer data, and based on this label information, divides the prospective customer group into two groups: a group where the success event occurred and a group where the success event did not occur.

[0097] Next, the differential analysis unit 405 generates multiple features for each prospective customer based on the behavioral log, customer attribute information, and nuance features. These features may include, for example, the following values: (a) Number of web page views within a specified period (b) Number of views of specific category pages (c) Number of emails opened or open rate (d) Number of document downloads (e) Number of form submissions (f) Speech rate (g) Change in intonation (h) silence time (i) Frequency of nodding in agreement (j)Negative expression frequency (k) Affirmative expression frequency (l) Frequency of questions (m) Intensity of response to a specific topic (n) Amount of facial expression change (o) Change in gaze (p) Frequency of nodding (q) Attribute features such as company size code, industry code, and job title code

[0098] The difference analysis unit 405 calculates statistical values ​​for each of the aforementioned features for both the group where the outcome event occurred and the group where the outcome event did not occur.

[0099] Next, the difference analysis unit 405 calculates a difference index between the group where the outcome event occurred and the group where the outcome event did not occur. As the difference index, for example, the difference in mean values, the difference in ratios, the odds ratio, the correlation coefficient, the information gain, or the statistical test value may be used.

[0100] The difference analysis unit 405 may calculate a difference score for each feature based on the calculated difference index and select the top feature with the highest difference score as the difference feature.

[0101] Furthermore, the difference analysis unit 405 may refer to the classification results by the segment classification unit 404 and perform the difference analysis process for each segment.

[0102] As an example of the needs estimation process performed by the needs estimation unit 403, the needs estimation unit 403 may estimate the needs of prospective customers by following the procedure below.

[0103] First, the needs estimation unit 403 reads behavioral logs, customer attribute information, and nuance features for the target prospective customer.

[0104] Next, the needs estimation unit 403 refers to the correspondence between each predefined needs category and the relevant features. For example, for cost-effectiveness needs, the number of times the price page is viewed, the frequency of questions asked when price is discussed, the frequency of negative expressions, etc., may be associated, while for the need to check case studies, the presence or absence of viewing case study pages, the frequency of positive expressions when case studies are discussed, the frequency of nodding, etc., may be associated.

[0105] Next, the needs estimation unit 403 may calculate a needs fit score for each needs category. The needs fit score may be calculated, for example, as a weighted sum of behavioral log fit, attribute fit, nuance fit, etc.

[0106] The needs estimation unit 403 may estimate the need category with the highest score as the primary need based on the calculated needs suitability score, or it may estimate a composite need by combining multiple need categories that are above a predetermined threshold.

[0107] As an example of the missing information identification process by the missing information identification unit 406, the missing information identification unit 406 may identify the information missing for a prospective customer by following the procedure below.

[0108] First, the missing information identification unit 406 reads the web activity, email responses, and form completion history included in the activity log.

[0109] Next, the missing information identification unit 406 extracts from the behavior log content categories that have not been accessed, content categories that have received little response, or topic categories that may be insufficiently explained.

[0110] Furthermore, the missing information identification unit 406 refers to nuance features and extracts categories indicating a lack of understanding or the presence of concerns based on factors such as increased silence time, increased question frequency, and increased negative expressions when a specific topic is presented.

[0111] The missing information identification unit 406 may determine the missing information by integrating the candidate missing categories derived from the behavior log and the candidate missing categories derived from the nuance features.

[0112] For example, if comparative materials have not been reviewed and the frequency of questions increases when competitive comparison topics are presented, then "lack of comparative information" can be identified as missing information.

[0113] Furthermore, if there is no history of viewing case studies and there are many questions regarding implementation results, "insufficient understanding of case studies" can be identified as missing information.

[0114] As an example of the appeal content determination process by the appeal content determination unit 407, the appeal content determination unit 407 may determine effective appeal content based on the correspondence between the content information of the sales materials and the nuance features.

[0115] For example, if increased attention and question frequency are observed when a price comparison slide is presented in sales materials, then the price comparison appeal may be determined to be an effective selling point.

[0116] Furthermore, if the frequency of positive expressions and nods increases when slides showcasing case studies are presented, the case study appeal may be deemed an effective selling point.

[0117] The recommendation information generation unit 408 generates recommendation communication information for prospective customers based on estimated needs, missing information, differential feature information, and effective appeal content.

[0118] The recommendation information generation unit 408 may, for example, generate a follow-up email that includes estimated needs and content to supplement missing information in the email draft.

[0119] Furthermore, the talk script should be structured to prioritize topics that received a high response, and the key points for success may include topics to avoid, things to confirm, recommended content, and the timing for the next contact.

[0120] Furthermore, the content to be presented may be selected from among case studies, comparison materials, pricing documents, feature descriptions, or security descriptions.

[0121] The timing of the next contact may be determined based on factors such as the timing of email opening, the time elapsed since the end of the conversation, the trend of changes in nuance features, or the trends of past successful event groups.

[0122] Through the above processing, it is possible to generate recommended communication information tailored to the prospective customer by comprehensively using the prospective customer's behavior logs, customer attribute information, and the nuances of their responses included in the conversation.

[0123] The acquisition unit 401 acquires behavioral logs and customer attribute information related to prospective customers, as well as at least one of voice data from phone calls and recorded data from online business negotiations. The nuance feature extraction unit 402 extracts nuance features that indicate the nuances of the prospective customer's response from the voice data or the recorded data. The needs estimation unit 403 estimates the needs of prospective customers based on the behavioral logs, customer attribute information, and nuance features. The segment classification unit 404 classifies prospective customers into multiple segments based on the estimated needs. The difference analysis unit 405 analyzes the difference between the customer group where an outcome event occurred and the customer group where an outcome event did not occur for each segment. The recommendation information generation unit 408 can generate recommendation communication information to lead prospective customers to an outcome event based on the difference analysis results. This allows for the appropriate deriving of recommended communication information tailored to each prospective customer, taking into account differences in their interests or responses, thereby improving the effectiveness of sales activities and negotiations.

[0124] The nuance feature extraction unit 402 can use at least one of the following as nuance features: speech rate, intonation, silence time, frequency of interjections, negative expressions, positive expressions, question frequency, and response intensity to specific topics. This makes it possible to grasp the nuances of a prospective customer's response from multiple aspects of the conversation, thereby improving the accuracy of needs estimation and the generation of recommended communication information.

[0125] The nuance feature extraction unit 402 extracts video features from the recorded data that indicate changes in facial expression, changes in gaze, or nodding movements, and these video features can be used as nuance features. This makes it possible to capture the reaction tendencies of prospective customers that are difficult to grasp from audio or text information alone, and makes it easier to estimate the prospective customer's level of interest or understanding more accurately.

[0126] The missing information identification unit 406 integrates web behavior, email responses, and form completion history included in the behavior log with nuance features to identify information that is missing from the prospective customer. This allows for the simultaneous identification of information that the prospective customer has not yet encountered and information that was misunderstood or raised concerns during the conversation, enabling a more appropriate identification of information that needs to be supplemented for the prospective customer.

[0127] The recommendation information generation unit 408 can generate at least one of the following as recommended communication information: an email draft, a talk script, key points, content to present, and the timing of the next contact. This makes it possible to present support information tailored to multiple contact methods or situations in sales activities or business negotiations, facilitating continuous support for prospective customers.

[0128] The visualization output unit 410 can visualize and output the correspondence between needs, segments, and nuance features. This allows users to visually understand the relationship between the estimated needs of prospective customers and the response features that underlie them, making it easier for them to decide whether to use the recommended communication information.

[0129] The acquisition unit 401 further acquires content information of the sales materials, and the appeal content determination unit 407 can determine effective appeal content for prospective customers based on the correspondence between the content information of the sales materials and the nuance features. This makes it possible to identify topics or content that prospective customers are likely to respond to, and to select more appropriate explanations or presentation content for prospective customers.

[0130] The nuance feature extraction unit 402 can extract nuance features based on text information obtained through speech recognition processing of the audio data. This allows for the extraction of negative expressions, positive expressions, question frequency, or response intensity to specific topics on a text basis, and enables the estimation of semantic response tendencies contained in the audio content.

[0131] The explanatory information generation unit 409 can output explanatory information that shows the reason for generating estimated needs or recommended communication information. This allows users to understand what behavioral logs, customer attribute information, nuance features, differential features, or missing information were used to derive the recommendation results, thereby increasing their acceptance of the recommendation results.

[0132] (Variation 1) In this embodiment, an example using at least one of audio data or video data has been described, but it is not limited to this. It may also be an embodiment using only audio data, only video data, or a combination of both.

[0133] This modified example allows for flexible system configuration depending on the available data environment.

[0134] (Modification 2) In this embodiment, nuance features such as speech rate, intonation, and silence time are given as examples, but are not limited to these. For example, delay time to response, frequency of repeated utterances, emotion polarity score, keyword repetition frequency, and laughter frequency may also be included.

[0135] This modified version allows for a more multifaceted understanding of the nuances in a prospective customer's response.

[0136] (Variation 3) In this embodiment, we have described an example of generating email templates, talk scripts, and key points as recommended communication information, but we are not limited to this. For example, we may also generate chat reply templates, proposed agendas for online meetings, demo presentation order, or explanation scenarios for the next business meeting.

[0137] This modified version can be applied to multiple communication channels.

[0138] (Modification 4) In this embodiment, an example was described in which effective appeal content is determined by referring to the content information of sales materials, but this is not limited to this. Content information from product web pages, help articles, white papers, case study videos, etc., may also be used.

[0139] This variation allows for an expansion of the content options presented to potential customers.

[0140] (Variation 5) In this embodiment, an example of performing difference analysis for each segment has been described, but it is not limited to this. The difference analysis unit 405 may analyze the difference between the group where the outcome event occurred and the group where it did not occur, based on industry, job title, company size, source of inflow, or region.

[0141] This modified version allows for more precise differential analysis based on customer attributes.

[0142] (Experimental variation 6) In this embodiment, an example of outputting explanatory information has been described, but it is not limited to this. The explanatory information generation unit 409 may generate explanatory information that includes numerical information such as a difference score, a needs suitability score, a deficiency score, or a recommendation reason score.

[0143] This modification allows for a more detailed understanding of the rationale behind the generation of recommended communication information.

[0144] (Example 7) In this embodiment, an example of generating a single recommended communication piece has been described, but the recommendation information generation unit 408 may generate multiple candidate recommended communication pieces. The terminal device 110 may display the candidates in a manner that can be selected by the user.

[0145] This modified version allows sales representatives to select the most suitable candidate according to their actual sales strategy.

[0146] (Variation 8) In this embodiment, examples using a speech recognition model, a nuance extraction model, or a video analysis model have been described, but the embodiment is not limited to these. Various feature extraction processes may be performed using rule-based methods, statistical models, machine learning models, or generative AI models, etc.

[0147] This modified example allows for the selection of an appropriate analysis method depending on the system environment.

[0148] (Extreme variation 9) In this embodiment, an example of generating recommended communication information after a conversation has been described, but the embodiment is not limited to this. The conversation may also be analyzed in real time, and recommended content or responses may be presented during the conversation.

[0149] This variation makes it easier for sales representatives to respond appropriately during conversations.

[0150] (Variation 10) In this embodiment, examples have been described that involve voice data from phone calls or recorded data from online business negotiations, but the embodiment is not limited to these. Recorded or video data from face-to-face business negotiations, voice recordings accompanying chat negotiation logs, etc., may also be used.

[0151] This modified version can be applied to multiple sales contact situations.

[0152] Although embodiments have been described above, these are presented as examples and are not intended to limit the scope of the invention. Novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. The embodiments are included in the scope and spirit of the invention, as well as in the claims of the invention and its equivalents. [Explanation of symbols]

[0153] 100: Information Processing Device 110: Terminal device 120: External Marketing System 130: Conversation Analysis System : Speech recognition model Nuance extraction model : Video analysis model 150: Network 210: Control Unit 220: Storage section 230: Communications Department 1000: Information Processing System

Claims

1. An information processing system, Having at least one control unit, The control unit, In addition to behavioral logs and customer attribute information regarding prospective customers, at least one of the following will be acquired: voice data from phone calls and recorded data from online business meetings. Further information on the contents of the sales materials will be obtained. From the aforementioned audio data or video data, nuance features indicating the nuances of the prospective customer's response are extracted. Based on the aforementioned behavioral logs, customer attribute information, and nuance features, the needs of the prospective customer are estimated. Based on the estimated needs, the prospective customers are classified into multiple segments, The web behavior, email responses, and form completion history included in the aforementioned behavior log are integrated with the nuance features to identify the information missing from the prospective customer. Based on the correspondence between the content information of the sales materials and the nuance features, effective appeals to the prospective customer are determined. For each segment, the difference between the customer group where the outcome event occurred and the customer group where the outcome event did not occur is analyzed, and based on the results of the difference analysis, recommended communication information is generated to lead the prospective customers to the outcome event, which includes supplementing the missing information and the effective appeal content. Information processing system.

2. In the information processing system described in claim 1, The aforementioned nuance features include at least one of the following: speech rate, intonation, silence time, frequency of interjections, negative expressions, positive expressions, frequency of questions, and response intensity to specific topics. Information processing system.

3. In the information processing system described in claim 1, The control unit extracts video features from the recorded data that indicate changes in facial expression, changes in gaze, or nodding movements. The aforementioned video features are used as the aforementioned nuance features. Information processing system.

4. In the information processing system described in claim 1, The aforementioned recommended communication information includes at least one of the following: email draft, talk script, key points, content to present, and timing of the next contact. Information processing system.

5. In the information processing system described in claim 1, The control unit visualizes and outputs the correspondence between the needs, the segment, and the nuance feature quantity. Information processing system.

6. In the information processing system described in claim 1, The control unit extracts the nuance features based on the text information obtained by speech recognition processing of the audio data. Information processing system.

7. In the information processing system described in claim 1, The control unit outputs explanatory information indicating the reason for generating the estimated needs or the recommended communication information. Information processing system.

8. An information processing method performed by an information processing system, The process includes acquiring at least one of the following: behavioral logs and customer attribute information related to prospective customers, as well as voice data from phone calls and recorded data from online business meetings. The process of obtaining information about the contents of sales materials, A step of extracting nuance features that indicate the nuances of the prospective customer's reaction from the aforementioned audio data or video data, A step of estimating the needs of the prospective customer based on the behavioral log, the customer attribute information, and the nuance features, A step of classifying the prospective customers into multiple segments based on the estimated needs, The process involves integrating the web behavior, email responses, and form completion history included in the behavior log with the nuance features to identify information missing from the prospective customer. A step of determining effective appeals to the prospective customer based on the correspondence between the content information of the sales materials and the nuance features, For each segment, the process involves analyzing the difference between the customer group where the outcome event occurred and the customer group where the outcome event did not occur, supplementing the missing information based on the results of the difference analysis, and generating recommended communication information that includes the effective appeal content to lead the prospective customers to the outcome event. Information processing methods including

9. Computers, A program for functioning as an information processing system according to any one of claims 1 to 7.

Images