Information providing apparatus, method and program

Abstract

To provide an information providing apparatus, method and program for visually identifying, for each evaluation attribute, degrees of contribution of evaluation factors with respect to re-purchase intension or recommendation intention of a customer.SOLUTION: An information providing apparatus includes: investigation data acquisition means which acquires evaluation data including evaluation values and evaluation attributes of evaluation items for a target commodity or a service, and evaluation values of multiple evaluation items related to re-visit intension, re-purchase intension, or other person recommendation intension; multivariable analysis means which performs multivariable analysis using, as objective variables, evaluation values of re-visit intention, re-purchase intension, or other person recommendation intention, and using, as explanatory variables, the evaluation values of the multiple evaluation items, to calculate degrees of contribution of the explanatory variables to the objective variables; and analysis result output means which causes a network-connected information terminal to display a result analyzed by the multivariable analysis means. The multivariable analysis means divides the evaluation data into multiple groups based on selected evaluation attributes and calculates degrees of contribution of the explanatory variables in the selected groups with respect to the objective variables.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an information providing device, method, and program for providing information that contributes to a customer's intention to repeat or recommend a store. In the present invention, the concept of "intention to repeat" includes not only the intention to revisit a store, but also the intention to repurchase products and services. Note that the target stores in the present invention include not only physical stores (real stores) but also stores in virtual space. [Background technology]

[0002] Conventionally, by collating the results of monitor surveys and in-store questionnaires, customer evaluations of a target store are analyzed and used to improve the service of the target store. Patent Document 1 discloses a technology that displays a display showing the psychological state of customers toward the target store (for example, whether their satisfaction is high or low) superimposed on map information of the target store. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2013-089035 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, conventional methods including Patent Document 1 simply display the survey results or questionnaire results, and it is not possible to grasp the degree to which each factor contributes to the customer's intention to repeat purchase (or to recommend). In addition, there is a demand for a display means that not only calculates the contribution of evaluation factors with high accuracy, but also allows easy understanding of each evaluation attribute.

[0005] An object of the present invention is to provide an information providing device, method, and program that enable visual understanding of the contribution of evaluation factors to a customer's intention to repeat purchase or recommend, for each evaluation attribute. [Means for solving the problem]

[0006] In the service industry, guidance is given on operational quality improvement measures to improve the re-visit rate (intention to repeat), but the proposals do not correspond to the operational quality of the store or the expectations of the customers, and there is a problem that the improvement effect is limited. The inventor hypothesized that the background to such problems is that the operational quality of the store and the expectations of the customers change depending on the contents handled by the store, the status of the improvement efforts on the store side, the customer segment, etc. In product sales, it is important to accurately grasp the evaluation factors that affect the intention to repeat purchase and use them to improve the product and the creative of promotions in order to improve the re-purchase rate (intention to repeat). Therefore, the inventor invented an information providing device and a program that make it possible to easily grasp effective measures to improve the repeat rate by factor analyzing the evaluation factors expected to be effective in customers' repeat purchases, collecting a statistically significant number of evaluations from repeat customers, and visually displaying the results of the statistical analysis in a diagram.

[0007] The information provision device of the present invention comprises: a survey data acquisition means for acquiring evaluation data including evaluation values ​​of evaluation items and evaluation attributes for a target product or service, the evaluation data including evaluation values ​​of multiple evaluation items related to intention to revisit the store, intention to repurchase, or intention to recommend to others; a multivariate analysis means for performing a multivariate analysis based on the evaluation data using the evaluation value of intention to revisit the store, intention to repurchase, or intention to recommend to others as a dependent variable and the evaluation values ​​of the multiple evaluation items in the evaluation data as a multiple explanatory variables, thereby calculating the contribution of the explanatory variables to the dependent variable; and an analysis result output means for displaying the analysis results by the multivariate analysis means on an information terminal capable of communicating via a network, wherein the multivariate analysis means divides the evaluation data into multiple groups based on a selected evaluation attribute, and calculates the contribution of the explanatory variables in the selected group to the dependent variable. In the above information providing device, the evaluation attributes may include the age or generation, sex, income or earnings, and / or residential area of ​​the evaluator. In the information providing device, the evaluation attribute may include a name of a store that provides a target service. In the above information providing device, the analysis result output means may be configured to display the plurality of explanatory variables on the information terminal using a graphic having a size corresponding to the degree of contribution. In the above information providing device, the analysis result output means may be configured to display the figures in a radial arrangement. The information providing device further comprises a correlation analysis means for calculating a correlation between the plurality of explanatory variables, and the analysis result output means displays on the information terminal a line connecting pairs of the explanatory variables, the line having a thickness and / or a type corresponding to the strength of the correlation. It can be configured as follows. In the above information providing device, the multivariate analysis means can be configured to calculate the contribution of the explanatory variables in the two selected groups to the target variable, and the analysis result output means can be configured to display the analysis results of the two selected groups by the multivariate analysis means side by side.

[0008] The information provision method of the present invention is an information provision method executed by a computer to perform the following steps: acquiring evaluation data including evaluation values ​​of evaluation items and evaluation attributes for a target product or service, the evaluation data including evaluation values ​​of multiple evaluation items related to intention to revisit a store, intention to repurchase, or intention to recommend to others; a multivariate analysis step of calculating the contribution of the explanatory variables to the objective variable by performing multivariate analysis based on the evaluation data using the evaluation values ​​of the multiple evaluation items in the evaluation data as objective variables and the evaluation values ​​of the multiple evaluation items in the evaluation data as multiple explanatory variables; and an analysis result output step of displaying the analysis results in the multivariate analysis step on an information terminal capable of communicating via a network, wherein the multivariate analysis step divides the evaluation data into multiple groups based on a selected evaluation attribute, and calculates the contribution of the explanatory variables in the selected group to the objective variable. The program of the present invention causes a computer to execute the following: survey data acquisition means for acquiring evaluation data including evaluation values ​​of evaluation items and evaluation attributes for a target product or service, the evaluation data including evaluation values ​​of multiple evaluation items related to intention to revisit the store, intention to repurchase, or intention to recommend to others; multivariate analysis means for performing multivariate analysis based on the evaluation data using the evaluation value of intention to revisit the store, intention to repurchase, or intention to recommend to others as a dependent variable and the evaluation values ​​of the multiple evaluation items in the evaluation data as a multiple explanatory variables, thereby calculating the contribution of the explanatory variables to the dependent variable; and analysis result output means for displaying the analysis results by the multivariate analysis means on an information terminal capable of communication via a network, wherein the multivariate analysis means divides the evaluation data into multiple groups based on a selected evaluation attribute, and calculates the contribution of the explanatory variables in the selected group to the dependent variable. Effect of the Invention

[0009] According to the present invention, it is possible to easily grasp the contribution of each evaluation factor to a customer's intention to revisit a store, repurchase, or recommend a product, for each evaluation attribute. [Brief description of the drawings]

[0010] [Figure 1] 1 is a configuration diagram of an information providing system according to a first embodiment. [Diagram 2] FIG. 2 is a diagram illustrating functions of the information providing device according to the first embodiment. [Diagram 3] FIG. 2 is a diagram showing an example of a questionnaire for a monitor survey of the present embodiment. [Figure 4] FIG. 2 is a diagram showing an example of a questionnaire for a questionnaire survey according to the present embodiment. [Diagram 5] FIG. 2 is a diagram illustrating a multivariate analysis means according to the first embodiment. [Figure 6] FIG. 11 is a diagram showing analysis results showing the contribution of each explanatory variable to a revisit intention at a specific restaurant in the first embodiment. [Figure 7] FIG. 11 is a diagram showing the analysis results showing the contribution of each explanatory variable to the intention to revisit a particular beauty salon according to the first embodiment. [Figure 8] FIG. 7 is a diagram showing an analysis result showing the contribution of each explanatory variable to an intention to revisit a restaurant different from that in FIG. 6 according to the first embodiment. [Figure 9] FIG. 10 is a diagram showing the analysis results showing the contribution of each explanatory variable to the intention to revisit a store when only the survey data of people in their twenties is extracted from FIG. 8. [Figure 10] FIG. 10 is a diagram showing the analysis results showing the contribution of each explanatory variable to the intention to revisit a store when only the survey data of people in their 30s is extracted from FIG. 8. [Figure 11] FIG. 9 is a diagram in which the contribution degree of each explanatory variable in FIG. 8 and the correlation coefficient of each explanatory variable are graphically displayed in a radial manner by the analysis result output means according to the first embodiment. [Figure 12] FIG. 10 is a diagram in which the contribution degree of each explanatory variable in FIG. 9 and the correlation coefficient of each explanatory variable are graphically displayed in a radial manner by the analysis result output means according to the first embodiment. [Figure 13] FIG. 11 is a diagram in which the contribution degree of each explanatory variable in FIG. 10 and the correlation coefficient of each explanatory variable are graphically displayed in a radial manner by the analysis result output means according to the first embodiment. [Figure 14]FIG. 11 is a diagram showing analysis results showing the contribution of each explanatory variable to repurchase intention for a specific beverage according to the second embodiment. [Figure 15] FIG. 15 is a diagram in which the contribution degree of each explanatory variable in FIG. 14 and the correlation coefficient of each explanatory variable are graphically displayed in a radial manner by the analysis result output means according to the second embodiment. [Figure 16] FIG. 15 is a diagram showing the analysis results showing the contribution of each explanatory variable (excluding fit) to repurchase intention when only the survey data of men is extracted from FIG. 14. [Figure 17] FIG. 17 is a diagram in which the contribution degree of each explanatory variable in FIG. 16 and the correlation coefficient of each explanatory variable are graphically displayed in a radial manner by the analysis result output means according to the second embodiment. [Figure 18] This figure shows the analysis results indicating the contribution of each explanatory variable (excluding fit) to repurchase intention when only survey data of those under 40 years old is extracted from Figure 14. [Figure 19] FIG. 20 is a diagram in which the contribution degree of each explanatory variable in FIG. 18 and the correlation coefficient of each explanatory variable are graphically displayed in a radial manner by the analysis result output means according to the second embodiment. [Figure 20] This figure shows the analysis results indicating the contribution of each explanatory variable (excluding fit) to repurchase intention when only the survey data of those aged 40 and over is extracted from Figure 14. [Figure 21] FIG. 21 is a diagram in which the contribution of each explanatory variable in FIG. 20 and the correlation coefficient of each explanatory variable are graphically displayed in a radial manner by the analysis result output means according to the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] First Embodiment The information provision system 1 of the first embodiment analyzes evaluation factors in a specific business format, allowing customers to easily understand the evaluation factors that influence the intention to revisit or recommend a store by graphically displaying them, and has the configuration described below.

[0012] Fig. 1 is a configuration diagram of an information providing system according to this embodiment. As shown in Fig. 1, the information providing system 1 according to this embodiment is composed of an information providing device 10 functioning as a server, a management information terminal 20 functioning as a management client, an information terminal 30 used by customers, a survey data collection device 40, and the Internet 50. Note that, in the example shown in Fig. 1, one each of the information providing device 10, the management information terminal 20, and the survey data collection device 40 is illustrated, but this configuration is not limited thereto, and two or more of each may be used.

[0013] The information providing device 10 is, for example, a PC server, and has a calculation unit 11, a storage unit 12 that stores an analysis program 121, and a communication unit 13 that enables communication with the information terminal 30 and the survey data collection device 40.

[0014] The management information terminal 20 is an information terminal for inputting data to the information providing device 10 and the survey data collecting device 40 and for executing the analysis program 121 and the survey data collecting program 141, and is, for example, a personal computer. A user can output various analysis results (for example, Figs. 6-8, 11) on the display device of the management information terminal 20 by executing the analysis program 121 described later from the management information terminal 20. In addition, in order to classify (segment) the survey data into a plurality of groups by evaluation attributes (attributes such as the gender, age, and region of the respondent), the user can output analysis results by evaluation attribute (for example, Figs. 9-10, 12-13) on the display device of the management information terminal 20 by transmitting the evaluation attribute selected from the management information terminal 20 to the analysis program 121 described later.

[0015] The information terminal 30 is an information terminal used by customers and is used to input survey data into the database 122. In the example of Fig. 1, the information terminal 30 is a smartphone, but is not limited to this and can be configured as any information terminal that runs a web browser, such as a personal computer or tablet.

[0016] The survey data collection device 40 is, for example, a PC server, and has a calculation unit 41, a storage unit 42 that stores a database 122 and an input program 123, and a communication unit 13 that enables communication with the information providing device 10 and the information terminal 30. The input program 123 includes a survey data input means that provides the database 122 with a web page that enables survey data to be input from the information terminal 30. In this embodiment, the survey data is acquired by having store customers enter the data using a web browser on the information terminal 30 via the input program 123 in the form of answering a questionnaire, but it is also possible to process the survey data collected from customers using another means, such as paper media, and upload it from the management information terminal 20 to the database 122.

[0017] 2, the calculation unit 11 of the information providing device 10 executes the analysis program 121 stored in the storage unit 12 to realize (1) a survey data acquisition means, (2) a multivariate analysis means, and (3) an analysis result output means. Each of these means will be described below.

[0018] (1) Method of collecting survey data The survey data acquisition means is a means for storing survey data to be used in the multivariate analysis means in the storage unit 12. In this embodiment, a screen for inputting a query for acquiring survey data from the database 122 is provided to the management information terminal 20. The survey data in this embodiment is data on customer satisfaction based on responses from customers of the store, and consists of two types of data: monitor survey data and questionnaire survey data.

[0019] The monitor survey in this embodiment is a survey in which a monitor (surveyor) actually uses the services of a store and answers certain questions based on his or her own experience. Since the surveyor is paid a fee, the monitor survey can have a large number of questions. On the other hand, compared to a questionnaire survey, it may be difficult to perform a reliable multivariate analysis using only the monitor survey, depending on the N number.

[0020] The questionnaire survey in this embodiment is an in-store survey with predetermined evaluation items, the number of questions is less than half that of a monitor survey, and the questions are suitable for multivariate analysis. On the other hand, the number of N can be larger than that of a monitor survey (for example, more than 100 cases can be collected per store per month), making it suitable for statistical analysis. Since the number of N is large, the questionnaire survey makes it possible to grasp evaluation trends by customer segment classified by evaluation attributes such as the respondent's gender, age, income or income, residential area, purpose of use, and amount of use. In this embodiment, it is also possible to use monitor survey data and questionnaire survey data together, making it possible to perform a statistical analysis of the strength of the influence of key evaluation items on overall satisfaction by business type, while also making it possible to grasp issues at a specific business level through the monitor survey.

[0021] The evaluation items to be written on each questionnaire are, for example, items related to QSCA (Quality Service Cleanliness Atmosphere). For target restaurants in the food and beverage category, items such as "customer service," "service speed," "food," "cleanliness," "space and atmosphere," and "cost performance" are examples. For target restaurants in the beauty category, items such as "customer service," "counseling," "skill," "cleanliness," "atmosphere," and "cost performance" are examples. In any business type, questions necessary for calculating a regression with the intention to revisit the store are set. In addition, questions for calculating a regression with the intention to recommend to others, such as whether the customer would like to use the target store to others, may be set. The above-mentioned items are just an example, and there may be several tens of items depending on the type (genre), size, season, etc. of the target store.

[0022] FIG. 3 is a diagram showing an example of a survey form for a monitor survey of this embodiment. The attribute information for the surveyor conducting the monitor survey is entered into each attribute information item. Note that the age, sex, place of residence, and other information of the monitor surveyor is registered in advance, so this information is not written on the survey form. The level 1 questions consist of a recommendation intention and a revisit intention. The level 1 questions are the objective variables of the multiple regression analysis described later. The evaluation of the level 1 questions is, for example, a multi-level evaluation of 0 to 10 points. The level 2 questions are factors that affect the level 1 questions, which are the objective variables. The level 3 questions are explanatory variables when performing multiple regression analysis with the level 2 questions as the dependent variables. Here, the level 2 questions and the level 3 questions are in a relationship of a higher concept and a lower concept. For example, the lower concept of "evaluation of customer service" in the subcategory A of the level 2 questions is "evaluation of greeting" and "evaluation of food serving" in the subcategory A of the level 3 questions. However, level 3 questions are not provided for all the subcategories of the level 2 questions. In addition, an evaluation of fit may be provided as another question equivalent to the level 2 questions. The question about fit is, for example, a multi-level evaluation for a question such as "Do you feel that our restaurant suits your taste?" In this embodiment, a level 3 question explaining the question about fit is not provided, but a level 3 question may be provided.

[0023] FIG. 4 is a diagram showing an example of a questionnaire for the questionnaire survey of this embodiment. The attribute information of the survey respondent is entered into each attribute information item. Since the survey respondent is not a registered respondent like the surveyor in a monitor survey, he or she is also asked to enter evaluation attribute information such as age and gender. The level 1 questions are the same as those in the monitor survey, and consist of items on willingness to recommend and willingness to revisit. In order to reduce the burden on the customers who respond to the questionnaire, the questionnaire survey is set with a small number of questions that serve as observed variables, and in this embodiment, it is composed of level 2 questions and level 2 checks. Here, the level 2 questions and the level 2 checks correspond to each other in terms of subcategories. For example, the level 2 check corresponding to "evaluation of customer service" in subcategory A of the level 2 questions is "dissatisfaction with customer service (multiple choices)". From another perspective, the level 2 check can be explained as a replacement of the level 3 questions, which were conducted in a multi-level evaluation, with a multi-level evaluation. In the questionnaire survey, other questions equivalent to the level 2 questions include evaluations of fit and satisfaction.

[0024] The sense of fit in this embodiment is a variable representing the preference structure of A, such as an individual's preference for a target, that an individual A feels that option b is more preferable than option a, or that utility b is greater than utility a. The comfort or harmony of the space and atmosphere of the survey target is used as an explanatory variable, and is positioned as a variable that includes lifestyle factors and reference group effects. By including the sense of fit that takes into account the tastes of the respondents to the monitor survey and questionnaire survey as an evaluation item, a function that enables the output of the correlation between the sense of fit and clustered customer attributes may be provided in the multivariate analysis means described later.

[0025] (2) Multivariate analysis methods The multivariate analysis means of this embodiment uses multiple regression analysis to analyze how each evaluation item influences the score of recommendation intention and revisit intention (hereinafter referred to as "revisit intention score"). Here, regression analysis is a statistical method that clarifies the relationship between a numerical value indicating a result (objective variable) and a numerical value that is a factor (explanatory variable), and multiple regression analysis is a statistical method that, when explaining an objective variable, quantifies to what extent each explanatory variable influences the result and makes a future prediction. By quantifying the influence of explanatory variables using multiple regression analysis, it becomes easier to plan ways to improve services. It also makes it possible to predict future changes in the objective variable that will accompany expected changes in explanatory variables.

[0026] In this embodiment, evaluation factors that can be improved by efforts are set as explanatory variables, and evaluation factors that are difficult to improve by efforts such as location and appearance are not set as explanatory variables. Specifically, "service," "food," "cleanliness," "space / atmosphere," "cost performance," etc. are set as explanatory variables for restaurants, and monitor surveys and questionnaire surveys are conducted in which the questions are evaluation items related to the explanatory variables and evaluation items for calculating the revisit intention score. FIG. 6 is a diagram showing an example of the results of calculating the contribution of each explanatory variable to the revisit intention score (objective variable) by multiple regression analysis when the target store is a restaurant. Here, VIF (Variance Inflation Factor) is an index for evaluating multicollinearity in multiple regression analysis. Multicollinearity means that there is a high correlation between explanatory variables, and it is considered better to remove the variable from the analysis when the VIF value is large. Here, the coefficient of determination (R 2 ) is an index that shows how closely the predicted values ​​of the dependent variables obtained by multiple regression analysis match the actual values ​​of the dependent variables. In general, the closer the coefficient of determination is to 1, the higher the explanatory power.

[0027] FIG. 5 is a diagram for explaining the multivariate analysis means according to this embodiment. The multivariate analysis means of this embodiment analyzes (a) multiple regression analysis of levels 2 and 3, (b) simple correlation of level 3 evaluation factors, (c) multiple regression analysis of levels 1 and 2, and (d) simple correlation of level 2 evaluation factors. In this embodiment, by performing a hierarchical multiple regression analysis, it is possible to improve the explanatory power of the influence (contribution) of the level 2 evaluation factors on the will to recommend or the will to revisit the store. In other words, when performing a multiple regression analysis with the level 3 evaluation factors as explanatory variables and the evaluation values ​​of the level 1 evaluation items as objective variables, it may not be possible to accurately grasp which evaluation factor is dominant, but this embodiment solves this problem by providing level 2 evaluation factors, which are intermediate concepts.

[0028] The analysis (a) is a first multiple regression analysis in which the level 2 evaluation items are the objective variables (intermediate objective variables) and the level 3 evaluation items are the explanatory variables. This analysis makes it possible to understand the level 3 evaluation items (evaluation factors) that have an influence on the level 2 evaluation items (intermediate objective variables). For example, in Figure 3, it is possible to quantify the influence (degree of contribution) of each of the evaluation of the taste of the food (B-1), the evaluation of the variety of food (B-2), and the evaluation of the amount of food (B-3) on the evaluation of the food (B). Analysis (b) quantifies the correlation between evaluation factors at level 3. For example, in Figure 3, if simple correlation analysis reveals a high correlation between the evaluation of the variety of food (B-2) and the evaluation of the amount of food (B-3), it can be seen that an increase in the evaluation of one will also increase the evaluation of the other. Analysis (c) is a second multiple regression analysis with the level 1 evaluation items as the objective variables and the level 2 evaluation items as the explanatory variables. This analysis makes it possible to understand the level 2 evaluation items (evaluation factors) that have an influence on the level 1 evaluation items. For example, in Figure 3, the influence (contribution) of each of the evaluations of customer service (A), food (B), cleanliness (C), space and atmosphere (D), and cost performance (E) on the intention to recommend or revisit the restaurant can be quantified. Analysis (d) calculates a correlation coefficient that quantifies the correlation between evaluation factors at level 2. For example, in Figure 3, if simple correlation analysis reveals a high correlation between the evaluation of food (B) and the evaluation of cost performance (E), it can be seen that an increase in one evaluation will also increase the other evaluation.

[0029] Figure 6 shows the analysis results showing the contribution of each level 2 explanatory variable (evaluation factor) when the level 1 revisit intention for a specific restaurant is used as the objective variable. In the example shown in Figure 6, it can be seen that "cost performance" has the highest contribution to the intention to revisit, followed by "service." In addition, the VIF of each explanatory variable is not a large value, which shows that no problem of multicollinearity has occurred. Also, in Figure 6, the coefficient of determination R 2 is large at .711, indicating that this is an analysis with high explanatory power.

[0030] In this embodiment, "customer service," "counseling," "skill," "cleanliness," "atmosphere," and "cost performance" were set as explanatory variables (evaluation factors) of level 2 for beauty salons, and a monitor survey was conducted in which questions were asked about the level 2 evaluation factors and level 3 evaluation items, which are subordinate concepts to level 2. FIG. 7 is a diagram showing an example of the results of calculating the contribution of each explanatory variable (evaluation factor) of level 2 to the customer's intention to revisit (objective variable) by multiple regression analysis when the target store is a beauty salon. In the example shown in FIG. 7, it can be seen that "customer service," "skill," and "cost performance" have a high contribution to the intention to revisit. In addition, the VIF of each explanatory variable is not a large value, so it can be seen that there is no problem of multicollinearity. In FIG. 7, the coefficient of determination R 2 Since the mean is .502, it can be said that the analysis has sufficient explanatory power.

[0031] Figure 8 shows the analysis results showing the contribution of each explanatory variable (evaluation factor) when the intention to revisit a restaurant different from that in Figure 6 is used as the objective variable. In the example shown in Figure 8, it can be seen that "cost performance" has the highest contribution to the intention to revisit, followed by "service" and then "space and atmosphere". In addition, the VIF of each explanatory variable is not a large value, which shows that there is no problem with multicollinearity. Also, in Figure 8, the coefficient of determination R 2 Since the difference is .658, it can be seen that the analysis has sufficient explanatory power. Furthermore, in FIG. 8, the difference value between the average score, which is the standard score for the target store and the same business type, is also calculated for each explanatory variable. Here, the standard score can be the average score of the score data in the same business type, or an arbitrary score that the target store aims for. In addition, the difference value between the standard score in a business type that is a higher conceptualization of a store that serves a specific type of food (for example, a restaurant business type) or in other business types that have common ingredients (for example, a yakiniku business type and a teppanyaki business type) may be used. Although not shown, in FIG. 8, the difference value between the average score for the revisit intention score in the same business type is also calculated, and the difference value was -0.1.

[0032] FIG. 9 is a diagram showing the analysis results showing the contribution of each explanatory variable (evaluation factor) to the intention to revisit, which is the objective variable, when only the data of respondents in their 20s are extracted from the survey data related to FIG. 8. In the example shown in FIG. 9, it can be seen that "cost performance" has the highest contribution to the intention to revisit, followed by "food" and then "service", so it can be seen that the explanatory variables with the second highest contribution are different from those of all generations shown in FIG. 8. In addition, as for the evaluation difference from the standard score, the values ​​of all explanatory variables were negative for all generations, but all except for two explanatory variables are higher than the standard score, so it can be seen that the evaluation is high in those in their 20s. Although not shown in FIG. 9, the difference value from the average score of the same business type for the intention to revisit was also calculated, and the difference value was 5.1. In addition, in Figure 9, the VIF of each explanatory variable is not large, so there is no problem with multicollinearity. 2 Since the mean is .717, it can be said that this analysis has high explanatory power.

[0033] FIG. 10 is a diagram showing the analysis results showing the contribution of each explanatory variable (evaluation factor) to the intention to revisit, which is the objective variable, when only the data of respondents in their 30s are extracted from the survey data related to FIG. 8. In the example shown in FIG. 10, it can be seen that "cost performance" has the highest contribution to the intention to revisit, followed by "space and atmosphere", and then "customer service", and the overall tendency is close to the permutation of the contribution of all generations shown in FIG. 8. In addition, the evaluation difference from the standard score for the intention to revisit and each explanatory variable is lower (dark red) than the evaluation difference for all generations shown in FIG. 8, so it can be seen that the evaluation in the 30s is particularly low. Although not shown in FIG. 10, the difference value from the average score for the same business type for the intention to revisit was also calculated, and the difference value was -2.8. In addition, in Figure 10, the VIF of each explanatory variable is not a large value, so there is no problem with multicollinearity. 2 Since the mean is .646, it can be said that the analysis has sufficient explanatory power.

[0034] (3) Analysis result output means The analysis result output means of this embodiment not only outputs the analysis result in a table format as shown in Figs. 6 to 10, 14, 16, 18, and 20, but also graphically displays the degree of contribution of explanatory variables to the customer's intention to revisit, allowing the customer to visually grasp the degree of contribution (see Figs. 11 to 13, 15, 17, 19, and 21). In this embodiment, each explanatory variable is graphically displayed as a circle of a size proportional to the degree of contribution to the revisit intention score. In addition, in this embodiment, when the degree of contribution is equal to or less than a certain level, the circle is drawn with a predetermined minimum radius, and is not made to be smaller than the minimum radius. Note that the size of the circle does not necessarily have to be proportional to the numerical value of the degree of contribution, and it may be highlighted (for example, the top two explanatory variables may be displayed large by multiplying the numerical value of the degree of contribution by a coefficient). In this embodiment, explanatory variables with a higher rating than the average rating of the same business type are displayed in green (diagonal lines), and explanatory variables with a lower rating than the average rating of the same business type are displayed in red (dots). Here, in this embodiment, the greater the difference between the rating of the explanatory variable and the average rating, the darker the color is drawn.

[0035] In addition, in this embodiment, the contribution display line connecting the explanatory variable, which is the intention to revisit, and the explanatory variable is drawn with a thickness and type according to the contribution. That is, the line connecting the explanatory variable and the objective variable with a high contribution to the intention to revisit is displayed thick, the line connecting the explanatory variable and the objective variable with a medium contribution is displayed thin, and the line connecting the explanatory variable with a low contribution is drawn as a dotted line. In this way, by changing the thickness and type of the contribution display line depending on the magnitude of the contribution, it is possible to visually grasp the contribution of each explanatory variable to the objective variable.

[0036] Any rule can be set for the method of displaying the correlation degree display line, and for example, the top 5 pairs that are 0.4 or more after rounding off the least significant digit may be displayed. Furthermore, the correlation degree display line of this embodiment adopts a rule that those that are 0.7 or more after rounding off the least significant digit are displayed as double lines, those that are less than 0.7 to 0.6 or more are displayed as thick solid lines, those that are less than 0.6 to 0.5 or more are displayed as solid lines of normal thickness, those that are less than 0.5 to 0.4 or more are displayed as thin dotted lines, and those that are less than 0.4 are not displayed.

[0037] FIG. 11 is a diagram in which the contribution of each explanatory variable to the intention to revisit the restaurant shown in FIG. 8 and the correlation coefficient of each explanatory variable are graphically displayed by the analysis result output means. In FIG. 11, the diagonal lines indicate the first color, green, and the dots indicate the second color, red. In FIG. 11, "service" and "service speed" are colored in dark red, so it can be seen at a glance that "service" and "service speed" are particularly lower in score than the average score for the same business type.

[0038] In addition, in FIG. 11, each explanatory variable is drawn by a circle of a size according to the numerical value of the contribution degree to the revisit intention score. From the size of the circle in FIG. 11, it can be seen that "cost performance" has the strongest influence on the revisit intention score, followed by "space and atmosphere" and "customer service". In addition, since the scores of "service speed" and "cleanliness" are negative, it can be inferred that normal operation is not being performed. And, considering the small size of the circle for "food" and the situation of other explanatory variables, it is possible to judge from the graphical radial display in FIG. 11 that the evaluation is that the attention is not paid to "food", which is the original value provided, and that the cost performance is the main focus, without examining the numerical value of the contribution degree. In this embodiment, the circles corresponding to each explanatory variable are arranged at the vertices of a regular hexagon, but the drawing method is not limited to this, and the explanatory variables may be arranged at the vertices of any regular polygon according to the number of explanatory variables, or may be arranged at equal intervals on the circumference. Also, the arrangement order of the figures indicating the explanatory variables may be changed by dragging.

[0039] In addition, in Figure 11, it is possible to visually grasp the degree of correlation between explanatory variables from the thickness and type of the correlation lines connecting the explanatory variables. For example, it can be seen that "Cuisine" has a strong influence on "Cost Performance", which is connected with a double correlation line, that "Service" has a large influence on "Cost Performance", which is connected with a thick solid correlation line, and that "Service" also has a significant influence on "Space / Atmosphere" and "Cuisine", which are connected with solid correlation lines. In this way, by graphically displaying the correlation coefficients of each explanatory variable with correlation lines of different thicknesses and types, it is possible to easily grasp that in the restaurant shown in Figure 11, food and "service" are highly correlated with "cost performance", followed by "space / atmosphere" and "cleanliness" and "service". Note that the explanatory variables connected with correlation lines are not limited to the top five pairs, and any number of correlation lines can be displayed.

[0040] Fig. 12 is a diagram in which the contribution of each explanatory variable to the intention to revisit and the correlation coefficient of each explanatory variable are graphically displayed by the analysis result output means using the data of respondents in their 20s related to Fig. 9. The arrangement of the circles showing the explanatory variables in Fig. 12 is the same as that in Fig. 11, so that the difference in the contribution of the explanatory variables between all generations and those in their 20s can be easily compared. Here, Fig. 11 and Fig. 12 may be displayed side by side on the same screen. In Figure 12, the diagonal lines represent the first color, green, and the dots represent the second color, red. In Figure 12, the explanatory variables other than "service speed" and "space / atmosphere" are colored dark green, so you can see at a glance that the explanatory variables other than "service speed" have particularly higher scores than the average score for the same business format.

[0041] From the size of the circles in Figure 12, it can be visually seen that "cost performance" has the strongest influence on the revisit intention score, followed by "customer service" and "service speed." In comparison with Figure 11, it can be seen that the contribution of "space and atmosphere," which had the third highest contribution rate for all generations, was low for those in their 20s. Also, in Figure 12, it is possible to visually grasp the degree of correlation between explanatory variables from the thickness and type of the correlation display lines connecting the explanatory variables. For example, it can be seen that "cost performance" is strongly influenced by "service speed", which is connected with a double line, followed by "food", which is connected with a solid line, and "service" is strongly influenced by "space and atmosphere", which is connected with a double line.

[0042] Fig. 13 is a diagram in which the contribution of each explanatory variable to the intention to revisit and the correlation coefficient of each explanatory variable are graphically displayed by the analysis result output means using the data of respondents in their 30s related to Fig. 10. The arrangement of the circles showing the explanatory variables in Fig. 13 is the same as that in Fig. 11, so that the difference in the contribution of the explanatory variables between all generations and those in their 30s can be easily compared. Here, Fig. 11 or Fig. 12 and Fig. 13 may be displayed side by side on the same screen. In Fig. 13, the diagonal lines indicate the first color, green, and the dots indicate the second color, red. In Fig. 13, "intention to revisit" and all explanatory variables are colored red, so it can be seen at a glance that "intention to revisit" and all explanatory variables have lower scores than the average score.

[0043] From the size of the circles in Figure 13, it can be visually seen that "cost performance" has the strongest influence on the revisit intention score, followed by "customer service" and "space and atmosphere provided." In comparison with Figure 12, it can be seen that the contribution of "space and atmosphere" is higher than that of people in their 20s. Also, in Figure 13, it is possible to visually grasp the degree of correlation between explanatory variables from the thickness and type of the correlation display lines connecting the explanatory variables. For example, it can be seen that "cost performance" is strongly influenced by "food", which is connected with a double line, followed by "service" and "space / atmosphere", which are connected with solid lines.

[0044] In this way, the analysis result output means can provide insight for efficiently carrying out store improvement efforts by graphically visualizing and providing the magnitude of the contribution of each evaluation factor to the revisit intention score, which changes with store improvement efforts. Note that in this embodiment, the explanatory variables are represented by circles whose size is proportional to the contribution, but the figures representing the contribution are not limited to this and any figure can be used, for example, a polygon or an illustration of food or a character. Also, the arrangement order of the figures representing the explanatory variables may be configured to be interchangeable by a drag operation.

[0045] As described above, according to the information provision system 1 of this embodiment, the analysis results obtained by multivariate analysis of dependent variables and explanatory variables (evaluation factors) are graphically visualized and provided, making it possible to instantly grasp the dependent variables for efficiently carrying out improvement efforts. In addition, by providing a graphical visualization of explanatory variables with high correlation coefficients, it becomes possible to easily predict the impact of improving the explanatory variables.

[0046] The information provision system 1 according to this embodiment can also provide a graphical visualization of the results of a multivariate analysis performed using a repurchase intention or a recommendation intention for a specific product as a response variable.

[0047] <Second embodiment> The information provision system 1 according to the second embodiment includes an information provision device 10 functioning as a server, an administration information terminal 20 functioning as an administration client, an information terminal 30 used by customers, a survey data collection device 40, and the Internet 50, and the hardware configuration is the same as that of the first embodiment. The second embodiment differs from the first embodiment in that the multivariate analysis means executed by the information provision device 10 has a function of analyzing, by multiple regression analysis, how each evaluation item affects the repurchase intention score for a specific product.

[0048] FIG. 14 is a diagram showing the analysis results showing the contribution of each level 2 explanatory variable (evaluation factor) when level 1 repurchase intention for a specific beverage (a so-called energy drink) is used as the objective variable. In the example shown in FIG. 14, it can be seen that "fit" has the highest contribution to repurchase intention, followed by "flavor." In addition, the VIF of each explanatory variable is not a large value, which shows that no problem of multicollinearity has occurred. Also, in FIG. 14, the coefficient of determination R 2 Since the mean score is .565, it can be seen that the analysis has sufficient explanatory power. Furthermore, in Figure 14, the difference value between the average score and the same type of beverage is also calculated for each explanatory variable.

[0049] The analysis result output means of this embodiment allows a customer to visually grasp the degree of contribution of explanatory variables to repurchase intention by graphically displaying the degree of contribution of each explanatory variable to repurchase intention for a specific beverage shown in FIG. 14 and the correlation coefficient of each explanatory variable. As in the first embodiment, explanatory variables with a high degree of contribution to repurchase intention are displayed as large circles, and explanatory variables with a low degree of contribution are displayed as small circles. Also, circles corresponding to each explanatory variable are arranged at the vertices of a substantially regular heptagon. In FIG. 15, the diagonal lines indicate the first color, green, and the dots indicate the second color, red.

[0050] In this embodiment, as in the first embodiment, explanatory variables with a higher rating than the standard rating are colored green (diagonal lines), explanatory variables with a lower rating than the standard rating are colored red (dots), and the greater the difference between the rating of the explanatory variable and the standard rating, the darker the color is drawn. In FIG. 15, "Package" is colored dark green, so it is immediately visible that the rating is particularly higher than the standard rating. Also, "Repurchase Intention" and "Effectiveness" are colored red, so it is immediately visible that the ratings are lower than the standard rating and that "Effectiveness" has a negative impact on "Repurchase Intention".

[0051] FIG. 16 is a diagram showing the analysis results of extracting only male respondent data from the survey data related to FIG. 14 and showing the contribution of each explanatory variable (evaluation factor) to the objective variable, repurchase intention. However, in FIG. 16, the analysis is performed by excluding "fit" from the explanatory variables in FIG. 14. Although "fit" is excluded in FIG. 16, it is possible to perform the analysis by excluding one or more explanatory variables other than this (the same applies to FIGS. 18, 20, and 22 described later). In the example shown in FIG. 16, it can be seen that "cost performance" has the highest contribution to repurchase intention, followed by "flavor", and then "brand attitude". It can also be seen that the evaluation difference from the standard score is lower than the standard score for "quantity / ingredients", "package", and "brand attitude".

[0052] Fig. 17 is a diagram in which the contribution of each explanatory variable to repurchase intention and the correlation coefficient of each explanatory variable are graphically displayed by the analysis result output means using the male respondent data related to Fig. 16. In Fig. 17, the number of explanatory variables excluding "cost performance" is six, and each explanatory variable is arranged at the vertex of a regular hexagon. Here, Fig. 17 and Fig. 15 may be displayed side by side on the same screen. In Figure 17, the diagonal lines indicate the first color, green, and the dots indicate the second color, red. In Figure 17, "Amount / Ingredients", "Packaging" and "Brand Attitude" are colored red, so the scores are lower than the average score for the same type of beverage, and the other explanatory variables and objective variables are colored green, so you can immediately see that the scores are higher than the average score for the same type of beverage.

[0053] From the size of the circles in Figure 17, it can be visually seen that "cost performance" has the strongest influence on the repurchase intention score, followed by "flavor" and "brand." In addition, from the thickness and type of the correlation display lines connecting the explanatory variables, it is possible to visually understand the degree of correlation between the explanatory variables. For example, it can be seen that "taste" and "amount / ingredients" connected with dotted lines have an influence on "cost performance," and that "effect / efficacy" connected with a solid line has a strong influence on "amount / ingredients."

[0054] FIG. 18 is a diagram showing the analysis results of extracting only data of respondents under 40 years old from the survey data related to FIG. 14, and showing the contribution of each explanatory variable (evaluation factor) to the objective variable, repurchase intention. However, in FIG. 18, "fit" is excluded from the explanatory variables in FIG. 14 for the analysis. In the example shown in FIG. 18, it can be seen that "taste" has the highest contribution to repurchase intention, followed by "brand attitude". In addition, it can be seen that the evaluation difference from the standard score is higher than the standard score for all explanatory variables except "packaging".

[0055] Fig. 19 is a diagram in which the contribution of each explanatory variable to repurchase intention and the correlation coefficient of each explanatory variable are graphically displayed by the analysis result output means using the data of respondents under 40 years old related to Fig. 18. In Fig. 19, the number of explanatory variables excluding "cost performance" is six, and each explanatory variable is arranged at the vertex of a regular hexagon. Here, Fig. 19 and Fig. 15 may be displayed side by side on the same screen. In Fig. 19, the diagonal lines indicate the first color, green, and the dots indicate the second color, red. In Fig. 19, the explanatory variables and the objective variable other than "packaging" are colored green, so it can be seen at a glance that the score is higher than the average score of the same type of beverage.

[0056] From the size of the circles in Figure 19, it can be visually understood that "Taste" has the strongest influence on the repurchase intention score, followed by "Brand attitude". In addition, from the thickness and type of the correlation display lines connecting the explanatory variables, it is possible to visually understand the degree of correlation between the explanatory variables. For example, it can be seen that "Taste" is influenced by "Quantity / Component", which is connected by a dotted line.

[0057] FIG. 20 is a diagram showing the analysis results of extracting only data of respondents aged 40 or over from the survey data related to FIG. 14, and showing the contribution of each explanatory variable (evaluation factor) to the objective variable, repurchase intention. However, in FIG. 20, "fit" is excluded from the explanatory variables in FIG. 14 for the analysis. In the example shown in FIG. 20, it can be seen that "taste" has the highest contribution to repurchase intention, followed by "brand attitude." In addition, it can be seen that the evaluation difference from the standard score is below the standard score for all explanatory variables except "packaging."

[0058] Fig. 21 is a diagram in which the contribution of each explanatory variable to repurchase intention and the correlation coefficient of each explanatory variable are graphically displayed by the analysis result output means using the data of respondents aged 40 or over related to Fig. 20. In Fig. 21, the number of explanatory variables is six, and each explanatory variable is arranged at the vertex of a regular hexagon. Here, Fig. 21 and Fig. 15 or Fig. 21 and Fig. 19 may be displayed side by side on the same screen. In Figure 21, the diagonal lines indicate the first color, green, and the dots indicate the second color, red. In Figure 21, the explanatory variables and the objective variable other than "packaging" are colored red, so it is possible to immediately see that the scores are lower than the average score for the same type of beverage. In addition, by comparing with Figure 19, it can be seen that the scores for each explanatory variable are all high for those under 40 years old, whereas the scores for each explanatory variable are all low for those over 40 years old.

[0059] From the size of the circles in Figure 21, it can be visually seen that "cost performance" has the strongest influence on the repurchase intention score, followed by "taste". In addition, from the thickness and type of the correlation display lines connecting the explanatory variables, it is possible to visually see the degree of correlation between the explanatory variables. For example, it can be seen that "taste" is strongly influenced by "brand attitude", "amount / ingredients", and "packaging", which are connected by solid lines, followed by "effect / efficacy", which is connected by dotted lines.

[0060] Although the preferred embodiment of the present invention has been described above, the technical scope of the present invention is not limited to the description of the above embodiment. Various modifications and improvements can be made to the above embodiment, and such modifications and improvements are also included in the technical scope of the present invention.

[0061] For example, in the above-described embodiment, a configuration was exemplified in which multiple regression analysis, which is a type of multivariate analysis, was used to determine the contribution of each evaluation factor to the revisit intention score, but the present invention is not limited to this configuration, and principal component analysis, factor analysis, correspondence analysis, multidimensional scaling, cluster analysis, path analysis, analysis of variance, conjoint analysis, regression analysis, logistics regression analysis, etc. may also be used. In addition to the above-described embodiment, the information providing device 10 can also be configured to collect condition information such as customer segments, usage scenes (daily use, anniversary use, etc.), seasons, and weather, and to analyze evaluation factors that contribute to repeat purchase or recommendation intentions by taking these condition information into consideration. Also, by adding graphic attributes and psychographic attributes to the questionnaire, analysis by customer segment is possible. Also, by linking with a system that manages POS (Point of Sales) and purchase history, it is possible to grasp evaluation trends by ordered product. In the above embodiment, the attributes of the survey respondents are used as the evaluation attributes, but the location of the store (store name) may also be included as a target of the evaluation attributes. In this case, the influence (contribution) of each explanatory variable on the objective variable for each store name may be analyzed based on the response results of the same survey, and a comparative analysis may be performed that takes into account location conditions, etc. For example, a comparative analysis may be performed for a golf course in City A and a golf course in City B using the intention to revisit as the objective variable and the course, facilities, food, customer service, atmosphere, and cost performance as explanatory variables, thereby making it possible to more appropriately analyze the influence of each objective variable on the intention to revisit score.

[0062] Furthermore, in the above-mentioned first to third embodiments regarding the service industry, examples have been described using survey data targeting restaurants, but the present invention can also be applied to other industries, such as the accommodation industry, food and beverage service industry, entertainment industry, lifestyle-related service industries such as hairdressing and beauty salons and public baths, education and skills teaching industry, finance industry, insurance industry, retail industry, information and communications industry, goods rental industry, professional service industries (professionals, etc.), advertising industry, real estate industry, and transportation industry. In the above-described embodiment, an example was described in which a hierarchical multiple regression analysis was performed to enhance the explanatory power of the influence (contribution) of evaluation factors on recommendation intention or repeat purchase intention, but the technical concept of the present invention is also applicable to multivariate analysis using non-hierarchical evaluation factors. [Explanation of symbols]

[0063] 1. Information provision system 10...Information providing device 20...Administrative information terminal 30…Information terminal 40...Survey data collection device 50…Internet

Claims

1. A survey data acquisition means for acquiring evaluation data including evaluation values ​​and evaluation attributes of evaluation items for a target product or service, the evaluation data including evaluation values ​​of a plurality of evaluation items related to a willingness to revisit a store, a willingness to repurchase, or a willingness to recommend to others; a multivariate analysis means for performing a multivariate analysis using an evaluation value of a store revisit intention, a repurchase intention, or an intention to recommend to others based on the evaluation data as a response variable and using evaluation values ​​of a plurality of evaluation items in the evaluation data as a plurality of explanatory variables to calculate a contribution of the explanatory variables to the response variable; and an analysis result output means for displaying the analysis result by the multivariate analysis means on an information terminal capable of communicating via a network, The information providing device, characterized in that the multivariate analysis means divides the evaluation data into a plurality of groups based on a selected evaluation attribute, and calculates the contribution of the explanatory variables to the target variable in the selected group.

2. 2. The information providing device according to claim 1, wherein the evaluation attributes include the age or generation, sex, income or revenue, and / or residential area of ​​the evaluator.

3. 3. The information providing device according to claim 1, wherein the evaluation attribute includes a name of a store that provides a target service.

4. 2. The information providing device according to claim 1, wherein said analysis result output means causes said plurality of explanatory variables to be displayed on said information terminal in the form of a graphic having a size corresponding to said contribution degree.

5. 5. The information providing device according to claim 4, wherein said analysis result output means displays said figures in a radial arrangement.

6. Further, a correlation analysis means for calculating a correlation between the plurality of explanatory variables is provided, The information providing device according to claim 4 or 5, characterized in that the analysis result output means displays on the information terminal a line connecting pairs of explanatory variables, the line having a thickness and / or a type corresponding to the strength of the association.

7. the multivariate analysis means calculates the contribution of the explanatory variables to the objective variable in the two selected groups; 2. The information providing apparatus according to claim 1, wherein said analysis result output means displays the analysis results of the two groups selected by said multivariate analysis means side by side.

8. Acquiring evaluation data including evaluation values ​​and evaluation attributes for evaluation items regarding the target product or service, the evaluation data including evaluation values ​​for a plurality of evaluation items related to intention to revisit the store, intention to repurchase, or intention to recommend to others; a multivariate analysis step of calculating the contribution of the explanatory variables to the objective variable by performing a multivariate analysis using the evaluation values ​​of the plurality of evaluation items in the evaluation data as the objective variable, the evaluation value of the intention to revisit the store, the intention to repurchase, or the intention to recommend to others based on the evaluation data; An information providing method, comprising: an analysis result output step of displaying the analysis result in the multivariate analysis step on an information terminal capable of communicating via a network, the information providing method being executed by a computer, An information providing method characterized in that, in the multivariate analysis step, the evaluation data is divided into a plurality of groups based on selected evaluation attributes, and the contribution of the explanatory variables to the target variable in the selected groups is calculated.

9. A survey data acquisition means for acquiring evaluation data including evaluation values ​​and evaluation attributes of evaluation items for a target product or service, the evaluation data including evaluation values ​​of a plurality of evaluation items related to intention to revisit a store, intention to repurchase, or intention to recommend to others, in a computer; a multivariate analysis means for performing a multivariate analysis using an evaluation value of a store revisit intention, a repurchase intention, or an intention to recommend to others based on the evaluation data as a response variable and using evaluation values ​​of a plurality of evaluation items in the evaluation data as a plurality of explanatory variables to calculate a contribution of the explanatory variables to the response variable; and an analysis result output means for displaying the analysis result by the multivariate analysis means on an information terminal capable of communicating via a network, The program, wherein the multivariate analysis means divides the evaluation data into a plurality of groups based on a selected evaluation attribute, and calculates the contribution of the explanatory variable to the objective variable in the selected group.

Images