Customer analyzer

Abstract

To rationally evaluate a person in charge of a service.SOLUTION: A customer analyzer comprises: an answer acquisition unit that obtains main answers from a user to an inquiry for surveying willingness to purchase a new service from a person currently in charge of providing or mediating a predetermined service, an inquiry for surveying willingness to continue to use the service provided by the person, and an inquiry for surveying willingness to recommend the person in charge to other users, and also obtains sub-answers on preconditions for the main answers; and an achievement calculation unit that calculates an achievement value for the person on the basis of a plurality of main answers related to the person. The achievement calculation unit selects main answers to be used for the calculation of the achievement value on the basis of the sub-answers.SELECTED DRAWING: Figure 14

Description

[Technical field]

[0001] The present invention relates to customer analysis, and more particularly to a customer analysis technique for realizing customer-oriented business operations. [Background technology]

[0002] In recent years, the importance of customer experience has been increasing. Customer experience value is not just the objective and rational material and monetary value of a product or service, but also the subjective and emotional value that appeals to customer emotions and covers the entire customer experience, from pre-purchase promotion to post-purchase support for the product or service.

[0003] Bernd H. Schmitt, a professor at Columbia University, classifies emotional value into five categories: Sense, Feel, Think, Act, and Relate. Customers recognize emotional value, or customer experience value, when their five senses are stimulated by something being delicious or pleasant to the touch (Sense), when their inner emotions are stimulated by something being cool or cute (Feel), when their intellectual desires or self-development are stimulated (Think), when they experience a different lifestyle than they have experienced before (Act), and when they have a sense of belonging or sharing by participating in an activity (Relate).

[0004] Customer experience value is said to have a strong correlation with a company's profitability, and advanced companies are said to be focusing on improving customer experience value (see Patent Document 1). [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2006-268405 A Summary of the Invention [Problem to be solved by the invention]

[0006] However, there is no established method for measuring customer experience value. For example, one proposed method is to ask customers, "Are you satisfied with our company's service?" and have them rate it on a 5-point scale. However, companies that receive high ratings in this way do not necessarily boast high profitability. A more rational method for measuring customer experience value is needed.

[0007] The present invention was completed based on the recognition of the above problems, and its main purpose is to provide a technique for rationally evaluating customer representatives, in particular, when measuring customer experience value. [Means for solving the problem]

[0008] A customer analysis device in one embodiment of the present invention includes a behavioral index calculation unit that obtains answers from a user who has received a service to one or more first type questions defined for investigating behavioral willingness for a specified service, and calculates a behavioral index based on the one or more answers; an emotion index calculation unit that obtains answers from the user who has received the service to a plurality of second type questions defined for investigating customer feelings toward the service, and calculates multiple types of emotion indexes based on the multiple answers; a correlation calculation unit that calculates a correlation coefficient between the behavioral index and each of the multiple types of emotion indexes; and an influence display unit that graphically displays the influence of each of the multiple emotion indexes on the behavioral index based on the calculated correlation coefficients.

[0009] In another aspect of the present invention, a customer analysis device includes an answer acquisition unit that acquires main answers from a user to questions to investigate the willingness to purchase a new service from a current representative who provides or mediates a specified service, questions to investigate the representative's willingness to continue using the service, and questions to investigate the representative's willingness to recommend the service to other users, as well as secondary answers regarding the prerequisites to the main answers, and a performance calculation unit that calculates a performance value of the representative based on the multiple main answers regarding the representative. The result calculation unit selects a main answer to be used for calculating the result value based on the secondary answers. Effect of the Invention

[0010] According to the present invention, it becomes easier to rationally evaluate service personnel. [Brief description of the drawings]

[0011] [Figure 1] FIG. 2 is a schematic diagram for explaining the relationship between a business index, a behavioral index, and an emotion index. [Diagram 2] FIG. 1 is a schematic diagram for explaining a behavior index. [Diagram 3] FIG. 13 is a diagram showing survey results regarding the strength of correlation between recommendation intention and CX and business indicators. [Figure 4] FIG. 13 is a data structure diagram of an emotion index. [Diagram 5] FIG. 2 is a functional block diagram of a customer analysis device. [Figure 6] 13 is a flowchart showing a process of calculating a psychological correlation coefficient. [Figure 7] FIG. 13 is a screen diagram of an expected value screen. [Figure 8] FIG. 13 is a diagram showing an evaluation value screen. [Figure 9] FIG. 13 is a screen diagram of an emotion index screen. [Figure 10] FIG. 13 is a screen diagram of a psychological correlation screen. [Figure 11] FIG. 13 is a screen diagram of a customer psychology analysis screen for an Internet bank. [Figure 12] FIG. 13 is a diagram showing a customer psychology analysis screen for a brick-and-mortar bank. [Figure 13] FIG. 13 is a functional block diagram of a customer analysis device according to a modified example. [Figure 14] An excerpt from the questionnaire regarding recommendation intention is shown below. [Figure 15] An excerpt from the questionnaire regarding intention to continue is shown below. [Figure 16] An excerpt from the questionnaire regarding purchase intentions is shown below. [Figure 17] 13 is a flowchart showing a process of calculating a result value. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] FIG. 1 is a schematic diagram for explaining the relationship between business indicators, behavioral indicators, and emotion indicators. A business index is an index that indicates the profitability of a service provided by a company. Examples of business indexes include sales, operating profit margins, and repeat customer rates. In this embodiment, customer experience value is expressed by two types of indexes: "behavioral index" and "emotional index." Note that in this embodiment, "service" is not limited to intangible services such as meals and accommodation, but also includes the purchase of tangible items such as merchandise.

[0013] Behavioral indicators indicate the strength of behavioral intentions toward a service (e.g., willingness to purchase). Emotional indicators indicate the type of emotions and the extent to which the user who actually received the service was moved. When users can strengthen their positive behavioral intentions toward a service, in other words, when the behavioral indicator is large, it is believed that business indicators showing profitability will also improve. Also, when users have positive emotions toward a service, in other words, when the emotional indicator is large, it is believed that the behavioral indicator will also be high. It is believed that the emotional indicators of users who sense customer experience value will be increased. The psychological model shown in Figure 1 models the correlation between the three elements of emotions encouraging behavior, and behavior leading to profits.

[0014] The customer analysis device in this embodiment analyzes customer sentiment in order to improve profitability (business indicators) and proposes which areas of the service should be improved based on the analysis results.

[0015] In the following, the financial services provided by financial institutions such as banks and securities companies will be described.

[0016] By the way, there is an index for exploring customer experience value called NPS (Net Promoter Score). NPS was proposed in 2003 by Frederick F. Reichheld of Bain & Company, an American consulting firm. NPS first asks users to answer the question, "How likely are you to recommend this company (product, service, brand) to a close friend or colleague?" on a scale of 0 to 10. NPS considers 9 points or more as "Promoters," 8 or 7 points as "Passives," and 6 points or less as "Detractors," and the index value is the percentage of promoters minus the percentage of detractors in the user group surveyed. NPS has been recognized as having a certain degree of effectiveness and is now the de facto standard.

[0017] However, NPS also has its weaknesses. Financial services are particularly weak points for NPS. It has been found that when NPS is used to explore customer experience in financial services, the NPS tends to be very low. In a survey conducted by the applicant, in response to the question, "Do you think that financial institutions are not something you would recommend to others?", approximately 80% of people answered "I think so" or "I think so to some extent" (sample size: 12,612).

[0018] When we asked users who had thoughts that were an obstacle to analyzing customer psychology as described above (hereafter referred to as "noise thinking") why they did so, many of them answered, "(The choice of financial institution) is my own responsibility," "I can't take responsibility if something happens," "People have different expectations and priorities," and "I don't like talking about money with other people."

[0019] NPS is difficult to use in services such as financial services, where the noise thought "this is not something I would recommend to others" is easily introduced. Therefore, in this embodiment, we propose the following behavioral index to compensate for the shortcomings of NPS.

[0020] FIG. 2 is a schematic diagram for explaining the behavior index. In this embodiment, the behavioral index is broken down into three elements: "intention to recommend," "intention to continue," and "intention to purchase." As with NPS, recommendation intention is calculated by asking users to rate the service on a scale of 1 to 10 in response to the question, "Are you likely to recommend the service to a close friend or colleague?". Hereafter, the evaluation value of recommendation intention is called the "recommendation value." A recommendation value is obtained for each user.

[0021] The noise thought in recommendation intention is "(This service) is not something I would recommend to others." As mentioned above, because such noise thoughts exist in recommendation intention, there is a tendency for the recommendation value to be low (decrease bias).

[0022] Continuation intention indicates the user's willingness to continue using the service. Regarding continuation intention, the user is asked to rate the question "How likely is it that you will continue to use our service in the future?" on a scale of 1 to 10. For example, 100% is 10 points, 90% to less than 100% is 9 points, and less than 10% can be 0 points. Hereinafter, the evaluation value of continuation intention is called the "continuation value." A continuation value is obtained for each user.

[0023] Even if a user has noise thoughts about recommendation intention, if they like the service, they are likely to continue using it. Therefore, the continuation value functions as a complement to the decrease bias of the recommendation value.

[0024] On the other hand, the noise thought in the continuation intention is "I don't like the service, but changing the service is bothersome." Because of this kind of noise thought (inertia continuation) in the continuation intention, the continuation value tends to be high (upward bias).

[0025] Purchase intention indicates the user's willingness to use a service for the first time. Regarding purchase intention, the user is asked to rate the question "How likely is it that you will continue to buy, sell, trade, and manage risky products with our company in the future?" on a scale of 10 points. For example, 100% is 10 points, 90% or more but less than 100% is 9 points, and less than 10% can be 0 points. Hereinafter, the evaluation value of purchase intention is called the "purchase value." A purchase value is obtained for each user.

[0026] Even if a user has noise thoughts about their continuance intention, they are unlikely to purchase a new financial product if they do not like the service. Therefore, the purchase value functions as a complement to the upward bias of the continuance value.

[0027] The noise thought in purchase intention is, "There is nothing else I want to buy." Because of this noise thought, the purchase value tends to be low (decrease bias).

[0028] Even if a user has noise thoughts about purchase intentions, if they like a service, they are more likely to recommend it to others. Therefore, the recommendation value functions as a complement to the decrease bias of the purchase value.

[0029] In this embodiment, a recommendation value, a continuation value, and a purchase value are obtained based on the above-mentioned three questions (hereinafter referred to as "first type questions") regarding recommendation intention, continuation intention, and purchase intention, and an average value of the recommendation value, continuation value, and purchase value is calculated as a behavior index. The behavior index is calculated for each user. Hereinafter, the behavior index calculated in this manner is referred to as a "CX index" or simply "CX". The inventor hypothesized that the CX index may be able to more appropriately index customer psychology by offsetting noise thoughts contained in recommendation intention, continuation intention, and purchase intention.

[0030] Figure 3 shows the survey results regarding the strength of correlation between recommendation intent and CX and business metrics. The inventor conducted a survey by dividing the subjects into those who deposit the most of their financial assets in banks (A: bank-preference group) and those who deposit the most of their financial assets in securities companies (B: securities-preference group). Note that the survey excluded those who had noise thoughts regarding recommendation intentions.

[0031] First, we defined five business indicators: (R1) amount of deposits at the target financial institutions, (R2) amount of investment in risky products at the target financial institutions, (R3) wallet share (the ratio of the amount of deposits at the target financial institutions to the customer's total financial assets), (R4) percentage increase or decrease in the amount of deposits at the target financial institutions compared to five years ago, and (R5) percentage increase or decrease in the amount of investment in risky products at the target financial institutions compared to five years ago.

[0032] When correlation coefficients between recommendation values ​​only (NPS: intention to recommend) and business indicators were calculated for 862 bank preferences, the correlation coefficients of NPS with "(R1) deposit amount" and "(R2) investment amount in risky products" were over 0.7, confirming a relatively strong correlation. On the other hand, the correlation coefficients of recommendation values ​​with "(R3) wallet share," "(R4) rate of change in deposit amount," and "(R5) rate of change in investment amount in risky products" were below 0.7, indicating that no very strong correlations could be found.

[0033] Meanwhile, when the correlation coefficient between CX (average of recommendation value, purchase value, and continuation value) and business indicators was calculated for 2,980 bank preference respondents, the correlation coefficient was 0.7 or higher for all four business indicators except for "(R1) deposit amount." Additionally, the correlation coefficient for "(R1) deposit amount" was 0.68, which is slightly lower than the recommended value. As far as bank preference respondents are concerned, it was confirmed that CX (average of the three values) has a stronger positive correlation with various business indicators than the recommendation value alone.

[0034] When we calculated the correlation coefficient between the recommendation value and business indicators for 1,758 securities preferences, the correlation coefficient of NPS with all five business indicators was below 0.5. Therefore, the NPS, which is only the recommendation value, cannot be said to be an appropriate indicator for exploring the customer experience value of securities preferences.

[0035] Meanwhile, when the correlation coefficient between CX (average of three values) and business indicators was calculated for 6,150 securities-preference individuals, the correlation coefficient was 0.7 or higher for all four business indicators except for "(R3) Wallet Share." In addition, the correlation coefficient for "(R4) Wallet Share" was 0.67, which is at least a better result than the recommendation value alone (NPS). For securities-preference individuals, a clear advantage of CX over NPS was confirmed.

[0036] As described above, even though people without noise thinking were surveyed and analyzed, CX has a stronger positive correlation with business indicators overall than NPS. In particular, it was confirmed that CX is effective for securities preferential people. Since the hypothesis that profitability can be improved by improving CX as a behavioral indicator was confirmed, it is necessary to consider what measures should be taken to improve CX as the next step. In this embodiment, as described in relation to FIG. 1, the emotional indicators that drive the behavioral indicators (CX) are defined, and measures to improve the behavioral indicators are considered by analyzing the emotional indicators.

[0037] FIG. 4 is a data structure diagram of the emotion index. The applicant has collected and accumulated tens of thousands of questionnaires from customers who have received financial services. These questionnaires include free comments. The inventor analyzed the vast amount of free comments and summarized the diverse evaluations and requests regarding financial services into three axes: "reliability," "convenience," and "economic rationality."

[0038] Credibility includes the following five types of evaluation points (hereinafter referred to as "emotion points"). (P1) Empathy (does the servicer show empathy for the customer's problems?) (P2) Capability (Does the servicer have the expertise to meet the customer's needs?) (P3) Personality (Do you think the servicer is kind and enthusiastic about their work?) (P4) Effectiveness of risk management (Are risks in financial services being appropriately managed?) (P5) Transparency (are there any lies or secrets, and are honest responses being made?)

[0039] Convenience encompasses the following five emotional points: (P6) Anytime (Can I receive the service at any time?) (P7) Ease (Can you easily get the service you want?) (P8) Speed ​​(Is the service executed promptly?) (P9) Understandability (Are the procedures and steps for receiving the service easy to understand?) (P10) Useful (Was the service useful?)

[0040] Economic rationality includes the following three emotional points: (P11) Costs (are fees and other costs appropriate?) (P12) Profit (Were you able to make a profit through the service?) (P13) Added value (Did you feel the added value?)

[0041] For each of the 13 emotional points listed above, users are asked to input the level of their expectations before receiving the service (hereafter referred to as "expectation value") and their level of satisfaction after actually receiving the service (hereafter referred to as "evaluation value"), each on a scale of 10 points.

[0042] In this embodiment, the emotion index is calculated by subtracting the expected value from the evaluation value. 13 types of emotion indexes are calculated for each user corresponding to the 13 types of emotion points. Next, the correlation coefficient of each of the 13 types of emotion indexes with the behavior index (CX) is calculated (described in detail later).

[0043] In order to analyze the emotion index, various questions (hereinafter referred to as "second-type questions") are prepared for the user in advance. For example, for "(P1) Empathy", a question is prepared: "Do you think that servicers make an effort to understand your needs?". Based on the user's answer to this question (expectation value and evaluation value), the emotion index for "(P1) Empathy" is calculated.

[0044] Other examples include questions such as "Were you able to trade without missing the timing of the buy or sell at the moment" for "(P8) Speed" and "Do you think the fees are low" for "(P11) Cost". One emotion point may be associated with one type 2 question, or multiple type 2 questions may be associated with one emotion point. For example, for a certain emotion point PX, a weighted average expected value and a weighted average evaluated value may be calculated by weighting the answers (expected value and evaluated value) to multiple type 2 questions, and the weighted average evaluated value - the weighted average expected value may be calculated as the emotion index for emotion point PX.

[0045] By calculating the difference between the evaluation value and the expected value as the emotion index, the emotion index indicates the size of the gap between expectations and reality. When the result exceeds expectations, the emotion index takes a positive value. On the other hand, when the result falls short of expectations, the emotion index takes a negative value. Human psychology has a tendency to have a large gap between expectations and reality, leading to people being moved or dissatisfied. The emotion index in this embodiment quantifies the extent to which a customer's emotions have been stirred for multiple emotion points.

[0046] FIG. 5 is a functional block diagram of the customer analysis device 100. As shown in FIG. Each component of the customer analysis device 100 is realized by hardware including computing units such as a CPU (Central Processing Unit) and various coprocessors, storage devices such as memory and storage, and wired or wireless communication lines connecting them, and software stored in the storage devices and supplying processing instructions to the computing units. The computer program may be composed of device drivers, an operating system, various application programs located at higher layers than these, and libraries that provide common functions to these programs. Each block described below represents a functional block, not a hardware-based configuration.

[0047] The customer analysis device 100 includes a user interface processor 110, a communication unit 114, a data processor 112, and a data storage unit . The user interface processing unit 110 accepts operations from the user and is responsible for processing related to the user interface, such as image display and audio output. The communication unit 114 is responsible for communication processing with external devices via the Internet. The data storage unit 116 stores various data. The data processing unit 112 executes various processes based on data acquired by the user interface processing unit 110 and the communication unit 114 and data stored in the data storage unit 116. The data processing unit 112 also functions as an interface between the user interface processing unit 110, the communication unit 114, and the data storage unit 116.

[0048] The user interface processing unit 110 includes an input unit 118 that receives input from a user, and an output unit 120 that outputs various types of information such as images and sounds to the user. The output unit 120 includes an influence display unit 122 and an improvement notification unit 124. The influence display unit 122 displays the magnitude of influence of each of a plurality of emotion indicators on a behavior indicator. Specifically, the influence display unit 122 displays screens (described below) shown in Figs. 7, 8, 9, 10, 11, and 12. The improvement notification unit 124 notifies the analyst of emotion points that are effective in improving the behavior indicator.

[0049] The communication unit 114 includes a transmission unit 132 that transmits data and a reception unit 134 that receives data.

[0050] The data processing unit 112 includes a behavior index calculation unit 126, an emotion index calculation unit 128, and a correlation calculation unit 130. The behavior index calculation unit 126 calculates a behavior index based on the above-mentioned first type of question. Specifically, the behavior index calculation unit 126 collects answer data to the first type of question in the user group to be surveyed, calculates a recommendation value, a continuation value, and a purchase value by the above-mentioned method, and calculates the average value of these three values ​​as the behavior index. The behavior index calculation unit 126 calculates the behavior index for each user. The behavior index calculation unit 126 may calculate a behavior index (average behavior index) for the entire group by calculating the average value of the behavior indexes of the user group to be surveyed.

[0051] The emotion index calculation unit 128 calculates the emotion index based on the above-mentioned second type question. Specifically, the emotion index calculation unit 128 calculates the emotion index for each user on an individual basis for each of the 13 emotion points. The emotion index calculation unit 128 calculates the emotion index for each user. The emotion index calculation unit 128 may calculate the average emotion index of the user group to be surveyed, thereby calculating the emotion index for the entire group for each emotion point.

[0052] The correlation calculation unit 130 calculates the correlation coefficient of the emotion index and the behavior index obtained from multiple users. Specifically, the correlation calculation unit 130 first collects the emotion index (hereinafter, referred to as "emotion index (P1)") of multiple users for "(P1) empathy". Next, the correlation calculation unit 130 calculates the correlation coefficient (hereinafter, referred to as "psychological correlation coefficient") based on the formula of Pearson product-moment correlation coefficient for the set of emotion index (P1) and the set of behavior index of these multiple users. The correlation calculation unit 130 calculates 13 types of psychological correlation coefficients in the same manner for each of the 13 types of emotion indexes.

[0053] FIG. 6 is a flowchart showing the process of calculating the psychological correlation coefficient. First, the receiving unit 134 collects answer data based on the first type question and the second type question from a user who has received financial services. When there is unprocessed answer data (Y in S10), the behavior index calculation unit 126 first calculates a behavior index based on the answer to the first type question included in the new answer data (S12). Next, the emotion index calculation unit 128 calculates 13 types of emotion indexes based on the answers to the second type questions included in the answer data of the same user (S14). One type of behavior index and 13 types of emotion indexes are calculated for each user. The same process is repeated for all answer data.

[0054] After calculating the behavioral index and the emotional index from all the response data (N in S10), the correlation calculation unit 130 calculates a psychological correlation coefficient for each emotional index (S16). The influence display unit 122 graphically displays the influence of the emotional index in a predetermined format based on the calculation result (S18). At this time, the improvement notification unit 124 also displays the improvement points. The improvement notification unit 124 notifies the analyst of the emotional points where the psychological correlation coefficient is equal to or greater than a first threshold value, for example, 0.5, and the emotional index is equal to or less than a second threshold value, for example, 0, as improvement points.

[0055] FIG. 7 is a diagram showing the expected value screen 140. The influence display unit 122 displays the expectation screen 140 according to instructions from the analyst. The vertical axis shows the average expectation (hereinafter referred to as "average expectation"). Graph L1 shows a brick-and-mortar bank, and graph L2 shows an Internet bank. For example, in the Internet bank user group, the average expectation for "(P3) Personality" is "6.44." In contrast, in the brick-and-mortar bank user group, the average expectation for "(P3) Personality" is "8.42." Therefore, it can be seen that users of brick-and-mortar banks have higher expectations for the personality of the servicer than when using an Internet bank.

[0056] Additionally, in the Internet bank user group, the average expectation for "(P11) Cost" is "8.08." This shows that users have an expectation for low costs when using Internet banks. In contrast, in the branch bank user group, the average expectation for "(P11) Cost" is "7.40." This shows that branch bank users have lower expectations regarding costs than Internet bank users.

[0057] In the brick-and-mortar bank user group, the average expectation value for "(P2) Ability" is "7.20." In contrast, in the Internet bank user group, the average expectation value for "(P2) Ability" is "6.43," indicating that expectations are relatively low.

[0058] FIG. 8 is a diagram showing the evaluation value screen 142. The influence display unit 122 displays the evaluation value screen 142 according to instructions from the analyst. The vertical axis indicates the average evaluation value (hereinafter referred to as the "average evaluation value"). In the Internet bank user group (graph L2), the average evaluation value for "(P2) ability" is "5.64", and the average evaluation value for "(P3) personality" is "5.78". In contrast, in the branch bank user group (graph L1), the average evaluation value for "(P2) ability" is "7.00", and the average evaluation value for "(P3) personality" is "8.12".

[0059] FIG. 9 is a screen diagram of the emotion index screen 144. The influence display unit 122 displays the emotion index screen 144 according to instructions from the analyst. The vertical axis indicates the average emotion index (evaluation value-expectation value) of multiple users (hereinafter referred to as the "average emotion index"). As described above, a high emotion index indicates "satisfaction beyond expectations," and a low emotion index indicates "dissatisfaction below expectations."

[0060] In the user group of branch banks (Graph L1), the average emotional index (P4: Effectiveness of risk management) was -1.30. This means that users were highly dissatisfied with the "effectiveness of risk management (P4)" at branch banks. On the other hand, in the user group of branch banks, the average emotional index (P3: Personality) was -0.30. This means that users' expectations for "P3) Personality" at branch banks were largely met.

[0061] In addition, the average emotion index (P8: Speed) for the Internet bank user group (Graph L2) is 0.04. This means that Internet banks are almost able to meet users' expectations regarding "(P8) Speed."

[0062] FIG. 10 is a screen diagram of the psychology correlation screen 146. The influence display unit 122 displays the psychology correlation screen 146 according to instructions from the analyst. The vertical axis indicates the psychology correlation coefficient between the emotion index and the behavior index (CX). In the case of a branch bank (graph L1), the psychology correlation coefficient of "(P3) Personality", one of the emotion points, is "0.95". Therefore, it is considered that giving customers a good impression of the servicer's personality is likely to lead to an improvement in the behavior index. According to the emotion index screen 144 in FIG. 9, the average emotion index (P3: Personality) is "-0.30", which is by no means low, but it can be seen that there is value in further improvement in "(P3) Personality".

[0063] In brick-and-mortar banks, the psychological correlation coefficient for "(P4) Effectiveness of risk management" is -0.45. Therefore, it is thought that improvements in risk management are unlikely to lead to improvements in behavioral indicators. According to the emotion index screen 144 in FIG. 9, the average emotion index (P4: Effectiveness of risk management) is -1.30, which is very low. Although there is a lot of room for improvement in "(P4) Effectiveness of risk management," such improvements are unlikely to lead to improvements in customer experience value.

[0064] The improvement notification unit 124 refers to the emotion index and the psychology correlation coefficient and notifies the analyst of the improvement points. As described above, in this embodiment, the improvement notification unit 124 notifies the analyst of the emotion points where the psychology correlation coefficient is 0.5 or more and the emotion index is 0 or less as improvement points. Therefore, according to the emotion index screen 144 in FIG. 9 and the psychology correlation screen 146 in FIG. 10, the improvement notification unit 124 suggests "(P3) personality", "(P5) transparency", "(P6) anytime", and "(P8) speed" as improvement points for the brick-and-mortar bank (graph L1). In addition, the improvement notification unit 124 suggests "(P3) personality", "(P4) effectiveness of risk management", and "(P7) simplicity" as improvement points for the Internet bank (graph L2).

[0065] FIG. 11 is a screen shot of a customer sentiment analysis screen 148 for Internet banking. The influence display unit 122 displays a customer psychology analysis screen 148 as the analysis result for the Internet bank user group according to instructions from the analyst. The horizontal axis indicates the average emotion index, and the vertical axis indicates the psychology correlation coefficient. The (average) emotion index is higher toward the right, that is, it indicates a higher level of satisfaction with expectations. The psychology correlation coefficient is higher toward the top, that is, it indicates a strong positive correlation between the emotion index and the behavior index.

[0066] According to the customer psychology analysis screen 148, the sentiment index regarding convenience is generally large, and the psychology correlation coefficient is also large. Therefore, it is considered that the profitability can be realized by concentrating management resources on further improving convenience for Internet banks.

[0067] FIG. 12 is a screen view of a customer psychology analysis screen 150 for brick-and-mortar banks. The influence display unit 122 displays a customer psychology analysis screen 150 as the analysis result for the user group of the branch bank according to the instruction from the analyst. The horizontal axis indicates the average emotion index, and the vertical axis indicates the psychology correlation coefficient.

[0068] According to the customer psychology analysis screen 150, some of the emotional indices related to convenience are thought to contribute greatly to improving behavioral indices and, by extension, profitability, but the effect is not as great as when targeting Internet banks. On the other hand, for "(P3) Personality," although a high level of satisfaction is already being provided, it is thought that there is value in working on further improvement. Also, although the emotional index for "(P01) Empathy" is high, it is difficult to link it to improving behavioral indices.

[0069] The customer analysis device 100 has been described above based on the embodiment. According to this embodiment, the positive correlation between profitability and behavioral indicators can be increased by defining behavioral indicators using multiple elements that offset each other's noise thinking. In particular, for "services that are difficult to recommend to others," such as financial services, it was found that CX, a new behavioral indicator, has a higher correlation than NPS.

[0070] In this embodiment, multiple emotion indexes are set as emotional factors that affect the behavior index. By calculating the psychological correlation coefficient as the degree of influence of each emotion index on the behavior index, it is possible to find an emotion point that is effective for improving the behavior index. According to this control method, it is easy to reform organizational operations efficiently and rationally by intensively investing limited management resources in improving appropriate emotion points.

[0071] The present invention is not limited to the above-mentioned embodiment and modified examples, and the components can be modified without departing from the gist of the present invention. Various inventions can be formed by appropriately combining multiple components disclosed in the above-mentioned embodiment and modified examples. In addition, some components can be deleted from all the components shown in the above-mentioned embodiment and modified examples.

[0072] [Variations] In the present embodiment, the average value of the recommendation value, the continuation value, and the purchase value is calculated as the behavior index. As a modified example, the behavior index may be calculated by taking a weighted average of these three values. For example, the weighting coefficients of the three values ​​may be adjusted so that the positive correlation coefficient with the business index indicating profitability is high.

[0073] [Evaluation by staff] As a result of further consideration, the inventor came to the conclusion that a behavioral indicator (CX) that indicates behavioral intentions toward a service would not only be beneficial to companies that provide services, but could also be used to evaluate the work of employees in the field who actually interact with customers and introduce, provide, and broker services, such as customer service representatives and branch managers.

[0074] The strength of recommendation intention, continuation intention, and purchase intention are greatly influenced not only by the attractiveness of the service, but also by the ability of the agent. As mentioned above, these behavioral indicators include noise thoughts, but recommendation intention, continuation intention, and purchase intention can complement each other's noise thoughts. For example, it is thought that it would be possible to evaluate the work of agent A by aggregating the recommendation value, continuation value, and purchase value from customers for agent A and calculating the total or average value.

[0075] However, some customers may give the recommendation value a score of 0 (the lowest score) for the reason that "No matter who the agent is, I have no intention of recommending him / her to others. I cannot take responsibility." Although noise thoughts about recommendation intentions can be supplemented by continuity values, it is unfortunate for the agent himself / herself to receive the lowest rating for such a reason, and there may be cases where it is not appropriate as a business evaluation. In particular, when the number of customers associated with one agent is small, the business evaluation is easily influenced by noise thoughts. Therefore, the inventors thought that improvements were necessary when using behavioral indicators to evaluate the business of agents.

[0076] FIG. 13 is a functional block diagram of a customer analysis device 100 in a modified example. In the customer analysis device 100 in the modified example, the data processing unit 112 further includes a result calculation unit 200. The result calculation unit 200 performs a work evaluation of the person in charge. The customer analysis device 100 acquires answer data on behavioral indicators from the customer using the input unit 118 or the receiving unit 134 as an "answer acquisition unit."

[0077] Figure 14 shows an excerpt from the questionnaire regarding willingness to recommend. The customer inputs a maximum of 10 points for a first type question regarding the agent: "How likely is it that you would recommend your current agent to a close friend or relative?" The input value at this time becomes the recommendation value. The questionnaire may be provided on paper, or the transmission unit 132 of the customer analysis device 100 may provide the questionnaire as a web page to a tablet computer or the like. The questionnaire includes various questions regarding the emotion indexes described above, as well as three types of first type questions regarding the agent (recommend, continue, purchase).

[0078] Hereinafter, the answers to the first type of questions regarding the recommended values, etc. will be referred to as the "main answers." The recommended values, etc. will be calculated based on the main answers.

[0079] In addition to the first type of question, a question that should be a prerequisite for the first type of question is defined: "Please mark this box if the reason you selected the above number is that you would not recommend the person in charge at the bank / securities company to others, not just your current person in charge." Hereafter, the answer to this prerequisite question is called the "secondary answer."

[0080] Figure 15 shows an excerpt from the questionnaire regarding continuance intention. Regarding the rep, the customer enters a score out of 10 in response to the first type question, "How likely is it that you will continue to use... through your current rep?" As with the recommendation intention, the prerequisite question for the continuation intention is also defined as "Please check here if the reason you selected the above number is that you will not be using any rep at the bank / securities company, not just your current rep."

[0081] Figure 16 shows an excerpt from the purchase intention questionnaire. Regarding the rep, the customer enters a score out of 10 for the first type question, "How likely is it that you will purchase financial products through your current rep again in the future?" For purchase intention, as with recommendation intention, a prerequisite question is defined: "If the reason you selected the above number is that you will not purchase financial products from the bank / securities firm, not limited to your current rep, please check this box."

[0082] The performance calculation unit 200 calculates the work evaluation of the person in charge by tallying up the main answers and secondary answers to the first type of questions shown above. At this time, the main answer with a check mark in the secondary answer is excluded from the work evaluation calculation.

[0083] FIG. 17 is a flowchart showing the process of calculating the result value. First, the receiving unit 134 collects answer data based on the first type question, the premise question, and the second type question from a user who has received financial services. When there is unprocessed answer data (Y in S20), the result calculation unit 200 determines whether a check has been made in the secondary answer for any of the recommended value, the continuation value, and the purchase value (S22). When a check has been made (Y in S22), the main answer at that time is excluded from the calculation target for business evaluation (S24). When no check has been made (N in S22), the main answer is included in the calculation target.

[0084] The average value of the recommendation value, continuation value, and purchase value included in the main answer that is the subject of the calculation is calculated as the performance value for each person in charge (S26). The output unit 120 displays the performance value for each person in charge (S28).

[0085] According to the above control method, it is possible to more appropriately evaluate the performance of individual employees by excluding recommendation values ​​that are low due to reasons beyond the individual employee's efforts. If the performance value is calculated by branch office instead of by employee, it can also be applied to the performance evaluation of branch managers.

[0086] As shown in FIG. 14, etc., a part of the premise questions may include a question such as "Please tell us why you selected the above numbers," and the customer may input or write a free comment. In this free comment field, a reason that cannot be attributed to the person in charge, such as "Because I am elderly, I do not intend to purchase any more new products," or "The current person in charge is new, so I do not know well yet," may be written. Questionnaires containing such comments may be excluded from the targets of business evaluation. The analysis of the comments may be judged by a human being, or the result calculation unit 200 may use natural language processing to detect questionnaires containing comments that include any exclusion condition (e.g., "because I am elderly") and exclude them. [Explanation of symbols]

[0087] 100 Customer analysis device, 110 User interface processing unit, 112 Data processing unit, 114 Communication unit, 116 Data storage unit, 118 Input unit, 120 Output unit, 122 Impact display unit, 124 Improvement notification unit, 126 Behavioral index calculation unit, 128 Emotion index calculation unit, 130 Correlation calculation unit, 132 Transmission unit, 134 Reception unit, 140 Expectation value screen, 142 Evaluation value screen, 144 Emotion index screen, 146 Psychological correlation screen, 148 Customer psychology analysis screen, 150 Customer psychology analysis screen, 200 Results calculation unit, L1 graph, L2 graph

Claims

[Claim 1] An answer acquisition unit that acquires a main answer from a user to each of a question for investigating willingness to purchase a new service from a current person who provides or mediates a predetermined service, a question for investigating willingness of said person to continue using said service, and a question for investigating willingness of said person to recommend said service to other users, together with a secondary answer regarding a precondition for the main answer; and a performance calculation unit that calculates a performance value of the person in charge based on a plurality of main answers related to the person in charge, The customer analysis device is characterized in that the result calculation unit selects a main answer to be used for calculating a result value based on the sub-answers.

Images