Purchase behavior analyzer, purchase behavior analysis system, purchase behavior analysis method, and computer program

Abstract

To provide information from new viewpoints which can contribute to developing of marketing strategy.SOLUTION: A generation unit generates purchase state information including information on a total purchase amount which is a total amount of purchase prices in a subject for each analysis unit period, by using purchase history information including the subject which purchases merchandise, the timing of purchasing the merchandise, and a purchase price of the merchandise. An identification unit identifies the merchandise or merchandise classification which is a factor of fluctuation of the purchase state in the subject in which the purchase state has fluctuated as a fluctuation factor, by using the purchase history information, and the purchase state information. An output unit outputs fluctuation factor information showing the identified fluctuation factor.SELECTED DRAWING: Figure 20

Description

Technical Field

[0001] The present disclosure relates to a purchase behavior analysis device, a purchase behavior analysis system, a purchase behavior analysis method, and a computer program for analyzing customer purchase behavior.

Background Art

[0002] Various marketing strategies have been established to increase the sales of retailers. For example, Patent Document 1 (Japanese Patent Application Laid-Open No. 2017-220155) discloses the following marketing strategy. That is, in the marketing strategy disclosed in Patent Document 1, customers are regularly ranked using data on the purchase amount, purchase frequency, and last purchase date of customers obtained from an e-commerce site, and promotional measures such as sending emails for promotion according to the rank are implemented.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the marketing strategy as described above, promotional measures using customer rank information are implemented. Such customer rank information is considered useful information when formulating a marketing strategy, but when considering the development of a marketing strategy, information from a new perspective is desired. Therefore, the main object of the present disclosure is to provide a purchase behavior analysis device, a purchase behavior analysis system, a purchase behavior analysis method, and a computer program that can provide information from a new perspective that can contribute to the formulation of a marketing strategy.

Means for Solving the Problems

[0005] In order to achieve the above object, in one aspect, the purchase behavior analysis apparatus in the present disclosure uses purchase history information including the entity that purchased the product, the timing of the purchase of the product, and the purchase amount of the product, and generates purchase situation information including information on the total purchase amount, which is the total amount of the purchase amount of the entity for each analysis unit period, by a generation unit; a specifying unit that specifies, as a change factor, a product or product classification that is a factor in the change in the purchase situation of an entity whose purchase situation has changed, using the purchase history information and the purchase situation information; an output unit that outputs change factor information representing the specified change factor and includes.

[0006] In addition, in one aspect, the purchase behavior analysis system in the present disclosure uses purchase history information including the entity that purchased the product, the timing of the purchase of the product, and the purchase amount of the product, and generates purchase situation information including information on the total purchase amount, which is the total amount of the purchase amount of the entity for each analysis unit period, by a generation means; a specifying means that specifies, as a change factor, a product or product classification that is a factor in the change in the purchase situation of an entity whose purchase situation has changed, using the purchase history information and the purchase situation information; an output means that outputs change factor information representing the specified change factor and includes.

[0007] Furthermore, in one aspect, the purchase behavior analysis method in the present disclosure uses a computer to generate purchase situation information including information on the total purchase amount, which is the total amount of the purchase amount of the entity for each analysis unit period, using purchase history information including the entity that purchased the product, the timing of the purchase of the product, and the purchase amount of the product; specify, as a change factor, a product or product classification that is a factor in the change in the purchase situation of an entity whose purchase situation has changed, using the purchase history information and the purchase situation information; output change factor information representing the specified change factor.

[0008] Furthermore, as one aspect of the computer program in the present disclosure, processing for generating purchase situation information including information on the total purchase amount, which is the total amount of purchase amounts for each analysis unit period for the subject, using purchase history information including the subject who purchased the product, the timing of the purchase of the product, and the purchase amount of the product; processing for specifying, as a change factor, a product or product classification that is a factor in the change in the purchase situation for a subject whose purchase situation has changed, using the purchase history information and the purchase situation information; processing for outputting change factor information representing the specified change factor; are executed by a computer.

Effect of the Invention

[0009] According to the present disclosure, it is possible to provide information from a new perspective that can contribute to the formulation of a marketing strategy.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Embodiments for Carrying Out the Invention

[0011] Hereinafter, embodiments according to the present disclosure will be described with reference to the drawings.

[0012] ≪First Embodiment≫ The purchasing behavior analysis apparatus according to the first embodiment of the present disclosure is a computer device (for example, a server), and has a function of analyzing the purchasing behavior of customers and providing information based on the analysis results to users. As shown in FIG. 1, the purchasing behavior analysis apparatus 1 of the first embodiment is connected to a terminal device 20. The terminal device 20 is an information device (computer device) and has a communication function, an information reception function, and a display control function. The communication function is a function of communicating information with a communication device (device with a communication function) that is directly or indirectly connected via an information communication network. The information reception function is a function of receiving information input using an input device such as a keyboard, a mouse, or a touch panel. The display control function is a function of controlling the display device 21 to display information on the screen. The type of the terminal device 20 here is not limited as long as it is an information device having a communication function, an information reception function, and a display control function, but specific examples include a personal computer, a smartphone, a tablet, and the like. When the terminal device 20 is a personal computer, the display device 21 may be external, but when the terminal device 20 is a smartphone or a tablet, the display device 21 is integrated with the terminal device 20.

[0013] The terminal device 20 uses the communication function, the information reception function, and the display control function to input information to the purchasing behavior analysis apparatus 1 or display the information output from the purchasing behavior analysis apparatus 1 on the display device 21. That is, the user of the purchasing behavior analysis apparatus 1 inputs information to the purchasing behavior analysis apparatus 1 or obtains the information output from the purchasing behavior analysis apparatus 1 from the screen display of the display device 21 by the terminal device 20. Note that the user of the purchasing behavior analysis apparatus 1 may be, for example, a person who formulates a marketing strategy, or may be a person who provides the analysis results of the purchasing behavior analysis apparatus 1 to the person who formulates the marketing strategy.

[0014] In the example of FIG. 1, there is one terminal device 20 connected to the purchasing behavior analysis apparatus 1, but there may be a plurality of terminal devices 20 connected to the purchasing behavior analysis apparatus 1, and the number of connected terminal devices 20 is not limited.

[0015] The purchase behavior analysis device 1 is also connected to a database 50. The database 50 is a storage medium (storage device) that stores data (information). Here, the data (information) used by the purchase behavior analysis device 1 in processing and the data (information) generated by the processing of the purchase behavior analysis device 1 are stored in the database 50. For example, purchase history information is stored in the database 50 as information used in the processing of the purchase behavior analysis device 1.

[0016] The purchase history information is information in which the entity that purchased the product, the timing of purchasing the product, and the purchase amount of the product are associated. The product here is something that is bought and sold, including not only articles such as food and clothing, but also services such as logistics services and consultation services, and information (for example, information provided via the Internet) and rights (for example, the right to receive a service) that are bought and sold.

[0017] An example of purchase history information is shown in FIG. 2. In the example shown in FIG. 2, the purchase history information includes a customer ID (identification), purchase product information, purchase amount information, purchase timing information, and purchase store information. The customer ID is information that represents the customer who is the entity that purchased the product, and is information such as an identification number assigned to each customer to identify the customer. In the example of FIG. 2, the customer ID is given as an example of the customer's identification information. However, for example, the customer's identification information may be any information that can identify the customer, and may be information other than the customer ID, such as a name.

[0018] The purchased product information is information representing the products purchased by the customer. The purchased product information may be, for example, information representing the name of the product or information representing the category (classification) of the product. The purchase amount information is information representing the purchase amount of the product represented by the purchased product information. The purchase timing information is information representing the timing when the customer purchased the product represented by the purchased product information, and is, for example, information representing the purchase date. The purchase store information is information representing the store where the product represented by the purchased product information was purchased. The purchase store information may be, for example, the name of the purchase store or the identification number (store number) as long as the purchase store can be identified, and is not limited here. Also, it is assumed that the purchase store includes an EC (Electronic Commerce) site on the Internet.

[0019] In the database 50, furthermore, customer attribute information is stored as information used for the processing of the purchase behavior analysis device 1, for example. The customer attribute information is information representing the attributes of the customer. In the example of FIG. 3, it includes attribute information representing the name, age, gender, and address of the customer, and these attribute information are associated with the customer ID. Note that the customer attribute information may include, for example, customer attribute information appropriately determined in consideration of the marketing strategy, and is not limited to the example of FIG. 3.

[0020] The method of acquiring the customer's purchase history information and customer attribute information as described above, and the control procedure for storing the acquired information in the database 50 are not limited here, so the description thereof is omitted.

[0021] In the database 50, furthermore, customer rank history information is stored. The customer rank history information is information representing the rank history for each customer as shown in the conceptual diagram of FIG. 4, for example. The rank here is information representing the evaluation of the customer determined (judged) from the purchase behavior. Here, the method of determining the rank is not limited as long as it is a method using information related to the purchase behavior. As an example, there is a method of ranking the customer using a rank determination criterion given in advance.

[0022] For example, as a rank determination criterion, a rank determination criterion as shown in FIG. 5 is given. In the example of FIG. 5, the rank determination criterion is a criterion for ranking a customer's rank into one of ranks A to F using the number of times of purchasing a product (purchase frequency) and the total amount of the purchase amount of the product (total purchase amount) for each preset rank determination period (for example, in units of three months). In the example of FIG. 5, the ranks increase in the order of F, E, D, C, B, A. Here, the customer ranking is performed, for example, for a store of interest (hereinafter also referred to as the store of interest) for formulating a marketing strategy. The store of interest here may be one store, or for example, when the same operator operates a plurality of stores of the same type, these plurality of stores may be collectively regarded as one store of interest.

[0023] The purchase behavior analysis device 1 calculates, for example, for each customer, information on the purchase frequency, which is the number of times of purchasing a product at the store of interest during the rank determination period, and the total purchase amount, which is the total amount of the purchase amount, from the purchase history information. Further, the purchase behavior analysis device 1 uses the purchase frequency and the total purchase amount calculated for each customer and the rank determination criterion to determine a rank for each customer, and adds information representing the determined rank to the customer rank history information in the database 50. For example, it is assumed that for a certain customer, it is calculated that the purchase frequency at the store of interest during the rank determination period is 3 times and the total purchase amount is 22,000 yen. In this case, based on the rank determination criterion of FIG. 5, this customer is determined to have a rank of "E". Note that there are various methods for ranking customers, and it is not limited to the ranking method described above, nor is the information used for ranking limited to the above example. For example, there is also a method of ranking only based on the information of the total purchase amount.

[0024] As described above, various information related to the processing of the purchase behavior analysis device 1 is stored in the database 50.

[0025] As shown in FIG. 1, the purchase behavior analysis device 1 includes an arithmetic unit 10 and a storage device 30. The storage device 30 includes a storage medium for storing data and a computer program (hereinafter also simply referred to as a program) 31. There are multiple types of storage devices, such as magnetic disk devices and semiconductor memory elements. Furthermore, there are multiple types of semiconductor memory elements, such as RAM (Random Access Memory) and ROM (Read Only Memory). There are many types in total. A computer device is equipped with multiple storage devices with different uses. Here, the type and number are not limited, and the description thereof is omitted. Also, here, the multiple storage devices provided in the computer device are collectively represented as the storage device 30 without distinction.

[0026] The arithmetic unit 10 is composed of a processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit). By reading and executing the program 31 stored in the storage device 30, the arithmetic unit 10 can have various functions based on the program 31. The purchase behavior analysis device 1 includes a generation unit 13, a specification unit 15, and an output unit 17 as functional units related to purchase behavior analysis.

[0027] The generation unit 13 generates purchase situation information for each customer using the purchase history information stored in the database 50. The purchase situation information is information including the information on the total purchase amount for each predetermined analysis unit period. Here, as an example of the analysis unit period, the above-mentioned rank determination period (for example, in units of three months) can be cited. In this case, as a specific example of the purchase situation information, information including the history information of the total purchase amount for each rank determination period, such as the total purchase amount for the three months from January to March (rank determination period) is ○○○ yen, and the total purchase amount for the three months from April to June is □◇◎〇 yen, can be cited.

[0028] Here, the generation unit 13 performs the ranking as described above for each customer based on the purchasing behavior during the rank determination period (analysis unit period). The generation unit 13 also includes information on the rank for each customer during the rank determination period (analysis unit period) determined by the ranking process in the purchase status information.

[0029] Furthermore, depending on the ranking method, the generation unit 13 may also include information other than the total purchase amount used for ranking (for example, information on the number of purchases during the rank determination period (analysis unit period)) in the purchase status information as shown in Figure 6.

[0030] An example of the timing at which the generation unit 13 generates the above-described purchase status information is when it receives a command to start analysis from the terminal device 20 connected to the purchasing behavior analysis device 1. After the generation unit 13 generates the purchase status information, the generation unit 13 may update the purchase status information, for example, every time an analysis unit period elapses.

[0031] The identification unit 15 uses the purchase history information and the purchase status information to identify, as a factor of change, a product or product category that is a factor in the change in the purchase status of a customer whose purchase status has changed. To give a specific example, the identification unit 15 refers to rank information included in the purchase status information and groups customers at the store of interest according to the trend in rank change. Examples of groups here include a group of customers whose rank rose from B to A during the analysis unit period (rank determination period) and a group of customers whose rank fell from C to D. Another example of a group may be a group of customers whose rank rose from D to B and a group of customers whose rank was D in the previous ranking but not B in the current ranking. In other words, groups may be formed by dividing customers into groups with the same rank in the previous ranking but a focused rank (the rank of interest) and other ranks in the current ranking.

[0032] The specific unit 15 identifies the factors of variation in the purchase situation during the analysis unit period for the group as described above, taking the product or product classification that is the factor of variation in the purchase situation as the factor of variation, thereby identifying the factors of variation during the analysis unit period for the customers included in the group. Whether the specific unit 15 identifies a product or a product classification as the factor of variation is set by, for example, the device designer in consideration of the requests of the user of the purchase behavior analysis device 1 (the person who receives information based on the analysis results from the purchase behavior analysis device 1). Note that, as described above, what the specific unit 15 identifies as the factor of variation, whether it is a product or a product classification, is appropriately set and not limited. However, in the following description, to avoid complexity in the explanation, it is assumed that the product classification is identified. Also, in the following description, the product classification that is the factor of variation in the change in the total purchase amount of the customer is also referred to as the factor of variation. Furthermore, it is not always the case that one factor of variation is identified as the factor of variation during the analysis unit period, and there may be cases where a plurality of factors of variation are identified.

[0033] As an example of a method by which the specific unit 15 identifies the factor of variation, a method using AI (Artificial Intelligence) technology can be mentioned. In this method, a factor-of-variation identification model is used. As an example of the factor-of-variation identification model, it is a model that takes as input the purchase history information of the customers included in the group as described above and the attribute information of those customers, and outputs information on the factor of variation.

[0034] The variable factor identification model is generated by machine learning using the customer's purchase history information as learning data. For example, in order to select the learning data for the variable factor identification model, a learning data selection screen as shown in FIGS. 7 and 8 is displayed on the display device. In the examples of FIGS. 7 and 8, six ranks are set in the order of F, E, D, C, B, and A, with the rank increasing as described above. In the example of FIG. 7, classification items (for example, classification items such as the classification of rising from rank B to rank A (hereinafter also referred to as rank change classification items)) for classifying the variation trend of the rank are listed in characters. In the example of FIG. 8, the rank change classification items are represented using symbols (arrows). For example, the upward arrow AA shown in FIG. 8 represents a classification item of rising from rank C to rank A. Also, the downward arrow represents a classification item with a decreased rank.

[0035] The plurality of purchase history information as learning data candidates are each classified into any one of the rank change classification items as described above using the rank determination period and the rank history shown in the purchase situation information of the corresponding customer.

[0036] On the learning data selection screen as shown in FIGS. 7 and 8, a display column T for the number of selected data is further displayed. The display column T for the number of selected data is a screen display part representing the number of positive example data and negative example data selected as learning data respectively. Here, the positive example data is the purchase history information of the subject showing the variation in the purchase situation that is the focus (for example, the variation such as from rank C to rank D). The negative example data is the purchase history information of the subject whose purchase situation variation is otherwise (for example, a subject that is not the variation such as from rank C to rank D that is the focus).

[0037] For example, assume that using the learning data selection screen, a system designer or the like selects (designates) the rank variation classification item (variation of the rank to be focused on) of the learning data candidates to be used for learning the variation factor identification model. As a result, the respective numbers of positive example data and negative example data to be displayed in the display column T for the number of selected data are calculated and the calculated results are displayed in the display column T. In the example of FIG. 7, the selected rank variation classification item has a check in the check column. Also, in the example of FIG. 8, the arrow representing the selected rank variation classification item is displayed with, for example, its color or filling pattern changed like an arrow surrounded by a dotted frame.

[0038] The specifying unit 15 uses the variation factor identification model as described above to specify the variation factors in the analysis unit period for the group as described above, thereby specifying the variation factors in the analysis unit period of the customers included in the group.

[0039] Furthermore, the specifying unit 15 may execute the following specifying process. The specifying process is a process of specifying, for each of a plurality of divided periods set in the analysis unit period, a product or product classification presumed to lead to the purchase behavior of the customer as a purchase factor. The plurality of divided periods set in the analysis unit period may be periods obtained by equally dividing the analysis unit period, or may be a plurality of divided periods appropriately designated according to matters to be analyzed by a system designer or the like. Here, the method of setting the divided periods is not limited. Although the method of setting the divided periods is not limited in this way, as a specific example, one week centered on the day one week back from the end point of the analysis unit period is set as one divided period, and furthermore, one week centered on the day two weeks back is set as one divided period, and so on, the divided periods are set.

[0040] The specifying unit 15 specifies the purchase factors presumed to lead to the purchase behavior of the customers belonging to the group as described above for each of the divided periods set in this way. As an example of the method of specifying this purchase factor, the same AI technology as the AI technology used in the method of specifying the variation factors can be mentioned.

[0041] The specific unit 15 stores, in the database 50, the information on the identified variable factors and purchase factors as described above, in association with the information for identifying related groups and the information representing the analysis unit period or the classified period.

[0042] The output unit 17 outputs, as variable factor information, the information representing the variable factors identified by the specific unit 15. Further, the output unit 17 outputs, as purchase factor information, the information on the purchase factors identified by the specific unit 15. As an output destination, for example, it is the terminal device 20 connected to the purchase behavior analysis device 1, and the output timing is, for example, the timing triggered by receiving an information provision request from the terminal device 20. The information provision request is a command for requesting information provision, and the target person information, type information, and target period information are associated with the information provision request. The target person information is the information representing the group related to the information for which provision is requested. The type information is the information representing the type of the information for which provision is requested. Here, the types of information include types such as variable factor information (that is, the information representing the variable factors in the analysis unit period) and purchase factor information (that is, the information representing the purchase factors in a plurality of classified periods in the analysis unit period). The target period information is the information representing the analysis unit period of the information for which provision is requested.

[0043] For example, it is assumed that purchase factor information is requested for a group of customers whose rank has risen from D to B due to their purchase behavior in the analysis unit period from April to June in ○○○○. In this case, the target period information associated with the information provision request is the information representing the analysis unit period from April to June in ○○○○. Further, the target person information is the information representing the group of customers whose rank has risen from D to B due to their purchase behavior in that analysis unit period. Furthermore, the type information is the information for identifying the purchase factor information.

[0044] Upon receiving an information provision request from the terminal device 20, the output unit 17 searches the database 50 for the information of the information provision target using the target person information, type information, and target period information associated with the information provision request. Then, the output unit 17 reads out the retrieved information from the database 50 and returns the information to the terminal device 20, which is the source of the information provision request. In the terminal device 20, the variation factor information or purchase factor information received from the purchase behavior analysis device 1 is displayed on the display device 21. The display mode for displaying the variation factor information and purchase factor information is not limited here as long as it is a display mode that takes into account factors such as readability and understandability of the information. Specific examples are shown in FIGS. 9 to 14.

[0045] For example, FIG. 9 is a diagram showing an example of the display mode of the variation factor information. In the example of FIG. 9, the variation factor information is information regarding a group of customers whose rank has risen from D to B due to purchase behavior during the analysis unit period. The information on the variation factors included in the variation factor information is presented in a table format (or also referred to as a list). In the example of FIG. 9, the variation factor information is associated with an image representing the variation factor and detailed information on the target group. For example, when an information reading section where link information for calling such an image and detailed information is set is specified on the display screen, the detailed information on the group is displayed.

[0046] FIG. 10 is a diagram showing another example of the display mode of the variation factor information. FIG. 10 schematically represents the variation factor information. FIG. 10 is a diagram regarding the variation factor information of a group of customers whose rank has risen from D to B due to purchase behavior during the analysis unit period. Also in the example of FIG. 10, similar to the example of FIG. 9, it is configured to be able to display the detailed information on the group.

[0047] FIG. 11 is a diagram showing an example of a display mode for displaying purchase factor information. The groups shown in the example of FIG. 11 are a group whose rank has risen from D to B and a group ranked with a rank other than from D to B according to purchase behavior during the analysis unit period. For those groups, purchase factors have been identified for each of a plurality of divided periods set in the analysis unit period. In the example of FIG. 11, character marks representing the identified purchase factors are arranged in time series. That is, in the example of FIG. 11, the purchase factor information is displayed in the form of a purchase factor time series pattern in which the purchase factors are arranged in time series. Further, in the example of FIG. 11, the thickness of the line connecting between the marks of the purchase factors arranged in this way is the thickness according to the number of people. That is, the number of customers (number of subjects) having the same combination of purchase factors in adjacent divided periods is represented by the thickness of the line. Here, the thickness of the line becomes thicker as the number of customers increases. That is, in this example, the purchase factor information output by the output unit 17 includes information representing the number of subjects having the same combination of adjacent purchase factors.

[0048] Furthermore, in the example of FIG. 11, for customers whose rank based on their purchasing behavior in the previous analysis unit period was D, in the classification period A centered around the day 75 days back from the current time, the product classifications of daily necessities, noodles, and meat are identified as the main purchasing factors. Further, it is shown that for the customers for whom meat was a purchasing factor, the purchasing factors in the next classification period B (i.e., the classification period centered around the day 50 days back from the current time) are not identified. Further, in the example of FIG. 11, many of the customers for whom meat was a purchasing factor in the classification period A follow one of the following two shifts in purchasing factors. That is, one of the shifts in purchasing factors is a shift where the purchasing factor in the classification period A is meat, the purchasing factor in the classification period C (i.e., the classification period centered around the day 2 weeks back from the current time) is vegetables, and the rank at the current time has risen to B. The other shift in purchasing factors is a shift where the purchasing factor in the classification period A is meat, the purchasing factor in the classification period D (i.e., the classification period centered around the day 1 week back from the current time) is miscellaneous goods, and the rank at the current time is a rank other than B. Note that, as the information on the number of people related to the thickness of the line connecting the marks of the purchasing factors shown in FIG. 11 is included in the purchasing factor information, the purchasing factor information output from the purchasing behavior analysis device 1 to the terminal device 20 includes the information necessary for the display of the said information.

[0049] In FIG. 11, the information on the number of people represented by the thickness of the line connecting the marks of the purchasing factors may be displayed. Further, the purchasing factor information may be displayed in the form of a table or a list instead of a schematic diagram as shown in FIG. 11.

[0050] The output unit 17 may further output the following information based on the purchase status information generated by the generation unit 13 to, for example, the terminal device 20. The information based on the purchase status information is, for example, aggregated information about the ranks of customers at the store of interest. Such aggregated information is output to the terminal device 20, for example, according to a request, and is displayed on the display device 21 by the display control of the terminal device 20. FIGS. 12 and 13 each show a specific example of the aggregated information. In the example of FIG. 12, the aggregated information is information representing the aggregated count of customers by ranking in a certain analysis unit period at the store of interest, and this aggregated information is displayed on the screen in the form of a table. In the example of FIG. 13, the aggregated information includes, for a certain analysis unit period, the ratio (number ratio) of the number of customers in each rank to the total number of customers at the store of interest, and the ratio (amount ratio) of the total amount of purchases of customers in each rank to the total amount of purchases of all customers. The aggregated information shown in FIG. 13 is also presented in the form of a table. Note that the display mode of the aggregated information is not limited to the examples of FIGS. 12 and 13, and an appropriate display mode (display configuration) considering clarity and the like may be adopted.

[0051] The output unit 17 may further output information such as customer attribute information to the terminal device 20 according to a request. The customer attribute information output to the terminal device 20 is, for example, displayed on the display device 21 in the form of a table.

[0052] The purchase behavior analysis device 1 of the first embodiment has the configuration as described above. Next, an example of the operation related to the analysis process of the purchase behavior in this purchase behavior analysis device 1 will be described with reference to the flowchart of FIG. 14. Note that FIG. 14 is also a diagram showing an example of the purchase behavior analysis method in the purchase behavior analysis device 1.

[0053] For example, assume that the database 50 stores the purchase history information of customers at the store of interest. In such a situation, upon receiving a command to start analysis from the terminal device 20, the generation unit 13 generates purchase situation information (step 101). That is, the generation unit 13 calculates the total purchase amount for each customer for each analysis unit period using the purchase history information of the customers at the store of interest, and generates information including the history information of the total purchase amount as the purchase situation information. Also, here, the generation unit 13 calculates the number of purchases for each analysis unit period, and includes the information on the number of purchases in the purchase situation information. Further, the generation unit 13 ranks each customer for each analysis unit period, and includes the information on the rank determined thereby in the purchase situation information. Using such rank information, customer grouping is performed using the trend of rank fluctuations.

[0054] Subsequently, for example, for the analysis unit period specified using the terminal device 20 as the analysis target, the specifying unit 15 specifies the factors causing fluctuations in the purchase situation of the group in which the purchase situation has fluctuated, for example, by using a factor - specifying model, using the purchase history information and purchase situation information of the customers, thereby specifying the factors causing fluctuations in the analysis unit period of the customers in the group (step 102). Further, the specifying unit 15 may also specify the purchase factors in the classification period set in the analysis unit period. The specified information is stored in the database 50.

[0055] Thereafter, when receiving a request for information provision from the terminal device 20, the output unit 17 reads out the information including the factors causing fluctuations corresponding to the request from the database 50, and returns (outputs) the read - out information as the factors - causing - fluctuations information to the terminal device 20, which is the source of the information - provision request. Also, at this time, if the provision of information on the purchase factors is also requested, the output unit 17 reads out the information on the purchase factors corresponding to the request from the database 50, and outputs the read - out information as the purchase - factors information to the terminal device 20, which is the source of the information - provision request. The factors - causing - fluctuations information and purchase - factors information output to the terminal device 20 in this way are displayed on the display device 21 as a screen.

[0056] The purchase behavior analysis device 1 of the first embodiment is configured to identify, using purchase history information and purchase situation information, a product or product classification, which is a factor causing a change in the purchase situation, as a change factor, and output change factor information representing the identified change factor. It can also be said that the change factor information is information representing a product or product classification that mainly contributed to the change in the purchase behavior of customers (the entities that purchased the products) during the analysis unit period. Such information is useful when considering marketing strategies and is considered to contribute to the promotion of the development of marketing strategies.

[0057] In addition, the purchase behavior analysis device 1 of the first embodiment is configured to be able to identify purchase factor information. The purchase factor information can also be said to be information representing a purchase pattern that focuses on the change in the products purchased by customers during the analysis unit period (in other words, the time-series change of key driver products), and it is information that represents the purchase behavior of customers in more detail. Such information is also useful when considering the development of marketing strategies.

[0058] <Modification example of the output destination> In the example described above, the terminal device 20 is cited as the destination to which the output unit 17 outputs information. The output destination to which the output unit 17 outputs information is not limited to the terminal device 20, and the following output destinations may be used. For example, as shown in FIG. 15, the output unit 17 may output information to the customer management system 60, the POS (Point Of Sales) system 70, or the order system 80.

[0059] The customer management system 60 is, for example, a system that manages customers who are ordinary consumers, and it also has functions as a marketing system, for example. In the customer management system 60, for example, a marketing strategy (sales promotion measures) using the information output from the output unit 17 is executed. Examples of such a marketing strategy include generating (selecting) information for sales promotion suitable for customers using the information output from the output unit 17 and displaying the information on the customer's mobile terminal 90, or sending an e-mail for sales promotion to the customer.

[0060] The POS system 70 is a system that manages information at the time when products in a retail store are sold, and includes a POS terminal. In the POS system 70, for example, the information output from the output unit 17 of the purchase behavior analysis device 1 is used for selecting target products for coupon tickets issued from the POS terminal, determining the content of a promotion message to be conveyed to store staff using the POS terminal, and so on.

[0061] The ordering system 80 is a system that is connected to a retail store and executes processing related to ordering products. In the ordering system 80, for example, the information output from the output unit 17 of the purchase behavior analysis device 1 is used in the process of determining the order quantity of products.

[0062] As described above, the customer management system 60, the POS system 70, and the ordering system 80 are configured to execute processing related to promotion and ordering using the information output from the output unit 17 of the purchase behavior analysis device 1. In other words, it can be said that the promotion and ordering processing in the customer management system 60, the POS system 70, and the ordering system 80 is controlled using the information output from the output unit 17 of the purchase behavior analysis device 1.

[0063] <Modification Example of Specific Processing in a Specific Part> In the example described above, the specific part 15 identifies variation factors and sales factors using AI technology. Instead of this, the specific part 15 may identify variation factors and sales factors using statistical processing. For example, in this case, examples of statistical processing include cross-tabulation by customer attribute information such as age, and items such as sales floors and products.

[0064] ≪Second Embodiment≫ Hereinafter, a second embodiment according to the present disclosure will be described. In the description of the second embodiment, the same names as those with reference numerals used in the description of the first embodiment are given the same reference numerals, and duplicate descriptions thereof are omitted.

[0065] As shown in FIG. 16, the purchase behavior analysis system 100 in the second embodiment includes a plurality of computer devices 110. As shown in FIG. 17, the computer device 110 includes a processor 111 and a storage device 112. The processor 111 and the storage device 112 each have the same configuration as the arithmetic device (processor) 10 and the storage device 30 described in the first embodiment. That is, the processor 111 can have functions according to the program by executing the program stored in the storage device 112. In the example of FIG. 16, two computer devices are illustrated as the computer devices 110 that constitute the purchase behavior analysis system 100, but the number of computer devices 110 that constitute the purchase behavior analysis system 100 is not limited as long as it is plural.

[0066] The purchase behavior analysis system 100 in the second embodiment has the same functions as the generation unit 13, the identification unit 15, and the output unit 17 described in the first embodiment. In the purchase behavior analysis system 100 of the second embodiment, a plurality of computer devices 110 execute processing in a distributed manner to realize the functions of such a generation unit 13, identification unit 15, and output unit 17. Here, the allocation method and allocation content for determining which of the plurality of computer devices 110 to allocate each of the plurality of processes for realizing the functions of the generation unit 13, the identification unit 15, and the output unit 17 are not limited and the description thereof is omitted.

[0067] The purchase behavior analysis system 100 of the second embodiment has a configuration for realizing the functions of the generation unit 13, the identification unit 15, and the output unit 17 described in the first embodiment by using a plurality of computer devices 110. That is, since the purchase behavior analysis system 100 of the second embodiment has the same functions as the generation unit 13, the identification unit 15, and the output unit 17 as in the first embodiment, it can achieve the same effects as the first embodiment.

[0068] ≪Other Embodiments≫ The present disclosure is not limited to the first and second embodiments, and various embodiments can be adopted. For example, one or both of the analysis unit period and the division period may be set using AI technology. For example, from the purchase history information of all customers at the store of interest, the tendency of purchase behavior fluctuations is detected by AI technology, and the analysis unit period or the division period may be set by a computer from the detected fluctuation tendency.

[0069] In addition to the configurations of the first and second embodiments, the following configuration may be further provided. That is, an analysis unit 19 as shown in FIG. 18 may be provided. The analysis unit 19 uses the attribute information of customers (subjects) who purchased the product or product classification to be analyzed (for example, a promotional product or a product in a promotional product classification) to identify the characteristics of people (customers) who purchased the product or product classification to be analyzed. As a method for identifying the characteristics of these customers, a method using a model generated by AI technology (hereinafter also referred to as a feature identification model) can be cited as an example. The feature identification model is, for example, a model that has learned data in which the attribute information of the subject (customer) who purchased the product or product classification to be analyzed is associated with information indicating whether or not the subject (customer) has purchased the product or product classification to be analyzed. The input of this model is information representing the product or product classification to be analyzed, and the output of the model is information representing the characteristics of people (customers) who purchased the product or product classification to be analyzed.

[0070] When the analysis unit 19 is provided, the output unit 17 outputs information (purchaser characteristic information) representing the analysis result by the analysis unit 19 according to a request (information request). For example, assume that information on product classification, which is a purchase factor as shown in FIG. 19, is displayed on the display device 21 of the terminal device 20. In this case, for example, assume that "fresh fish" is specified by the user as the product classification to be analyzed using the screen display. In response to the specified "fresh fish" being output from the terminal device 20 as information representing the product classification to be analyzed, the output unit 17 returns to the terminal device 20 information representing the characteristics of customers who purchased the product classification "fresh fish" to be analyzed by the analysis unit 19. That is, the output unit 17 returns to the terminal device 20 purchaser characteristic information corresponding to the information request that requests information representing the characteristics of people who purchased the product classification "fresh fish". As a result, the characteristics of customers who purchased the product classification "fresh fish" to be analyzed are displayed on the display device 21 of the terminal device 20.

[0071] Also, as another embodiment, the purchase behavior analysis device in the present disclosure may adopt a configuration as shown in FIG. 20, for example. That is, the purchase behavior analysis device 200 of another embodiment includes a generation unit 201, a specification unit 202, and an output unit 203. The generation unit 201 generates purchase situation information including information on the total purchase amount, which is the total amount of the purchase amount of the subject for each analysis unit period, using purchase history information including the subject who purchased the product, the timing of purchasing the product, and the purchase amount of the product. The specification unit 202 specifies, as a variation factor, a product or product classification that is the factor for the variation in the purchase situation of the subject whose purchase situation has changed, using the purchase history information and the purchase situation information. The output unit 203 outputs variation factor information representing the specified variation factor.

[0072] Next, an example of the operation of the purchase behavior analysis process in the purchase behavior analysis device 200 will be described with reference to FIG. 21. FIG. 21 can also be said to be a diagram for explaining the purchase behavior analysis method in the purchase behavior analysis device 200.

[0073] For example, the generation unit 201 uses purchase history information including the entity that purchased the product, the timing of the purchase, and the purchase amount of the product to generate purchase situation information including information on the total purchase amount, which is the total amount of the purchase amount for each entity in each analysis unit period (step 201). After the purchase situation information is generated in this way, the specifying unit 202 uses the purchase history information and the generated purchase situation information to specify, as a change factor, the product or product classification that is the factor for the change in the purchase situation for the entity whose purchase situation has changed (step 202). Then, the output unit 203 outputs change factor information representing the specified change factor (step 203).

[0074] As described above, the purchase behavior analysis device 200 is configured to generate purchase situation information and specify, as a change factor, the product or product classification that is the factor for the change in the purchase situation for the entity whose purchase situation has changed. That is, the purchase behavior analysis device 200 can provide information focusing on the product or product classification that affects the purchase behavior, rather than simply information such as the purchase amount and the number of purchases of the entity (for example, a customer). In other words, the purchase behavior analysis device 200 can provide information from a new perspective that can contribute to the formulation of marketing strategies.

[0075] Some or all of the above embodiments can be described as follows in the following supplementary notes, but are not limited thereto. [Supplementary Note 1] A generation unit that generates purchase situation information including information on the total purchase amount, which is the total amount of the purchase amount for each entity in each analysis unit period, using purchase history information including the entity that purchased the product, the timing of the purchase, and the purchase amount of the product; A specifying unit that uses the purchase history information and the purchase situation information to specify, as a change factor, the product or product classification that is the factor for the change in the purchase situation for the entity whose purchase situation has changed; An output unit that outputs change factor information representing the specified change factor A purchase behavior analysis device comprising: [Supplementary Note 2] For each of a plurality of divided periods set in the analysis unit period, the specific unit further specifies, as a purchase factor, a product or product classification presumed to lead to the purchase behavior of the subject. The output unit further outputs, as purchase factor information, information representing the purchase factor for each divided period in the analysis unit period to be displayed on a display device, which is the purchase behavior analysis device described in Supplementary Note 1. [Supplementary Note 3] The specific unit specifies the variation factor using a variation factor specific model that takes at least the purchase history information and the purchase situation information as input and outputs information on the variation factor. The variation factor specific model is generated by learning purchase history information of a subject showing a variation in which the purchase situation is focused and purchase history information of a subject in which the variation in the purchase situation is otherwise, which is the purchase behavior analysis device described in Supplementary Note 1. [Supplementary Note 4] The device further includes an analysis unit that specifies characteristics of a person who has purchased a product or product classification to be analyzed, using attribute information of the subject who has purchased the product or product classification to be analyzed. The output unit outputs purchaser characteristic information of a purchaser who has purchased a product or product classification to be analyzed in response to an information request for requesting purchaser characteristic information representing characteristics of a person who has purchased a product, which is the purchase behavior analysis device described in Supplementary Note 1. [Supplementary Note 5] The purchase factor information output by the output unit includes information for causing a display device to display a purchase factor time series pattern in which purchase factors for each divided period in the analysis unit period are arranged in time series, which is the purchase behavior analysis device described in Supplementary Note 2. [Supplementary Note 6] The purchase factor information output by the output unit includes information representing the number of subjects for which combinations of adjacent purchase factors are the same in the purchase factor time series pattern, which is the purchase behavior analysis device described in Supplementary Note 5. [Supplementary Note 7] The device further includes an analysis unit that specifies characteristics of a person who has purchased a product or product classification to be analyzed, using attribute information of the subject who has purchased the product or product classification to be analyzed. The output unit outputs purchaser characteristic information in response to an information request for requesting information representing the characteristics of a person who has purchased a product, which is a purchase factor, displayed on a display device. The purchase behavior analysis apparatus according to Supplementary Note 2. [Supplementary Note 8] Using purchase history information including the entity that purchased the product, the timing of the purchase of the product, and the purchase amount of the product, generation means for generating purchase situation information including information on the total purchase amount, which is the total amount of the purchase amounts in the entity for each analysis unit period; Using the purchase history information and the purchase situation information, specifying means for specifying, as a variation factor, a product or product classification that is a factor in the variation of the purchase situation in an entity in which the purchase situation has changed; Output means for outputting variation factor information representing the specified variation factor A purchase behavior analysis system comprising: [Supplementary Note 9] By a computer, Using purchase history information including the entity that purchased the product, the timing of the purchase of the product, and the purchase amount of the product, generating purchase situation information including information on the total purchase amount, which is the total amount of the purchase amounts in the entity for each analysis unit period; Using the purchase history information and the purchase situation information, specifying, as a variation factor, a product or product classification that is a factor in the variation of the purchase situation in an entity in which the purchase situation has changed; A purchase behavior analysis method for outputting variation factor information representing the specified variation factor. [Supplementary Note 10] A process of generating purchase situation information including information on the total purchase amount, which is the total amount of the purchase amounts in the entity for each analysis unit period, using purchase history information including the entity that purchased the product, the timing of the purchase of the product, and the purchase amount of the product; A process of specifying, as a variation factor, a product or product classification that is a factor in the variation of the purchase situation in an entity in which the purchase situation has changed, using the purchase history information and the purchase situation information; A process of outputting variation factor information representing the specified variation factor A computer program for causing a computer to execute.

[0076] Note that some or all of the configurations described in Appendices 2 to 7, which are subordinate to Appendix 1 described above, may be subordinate to each of Appendices 8 to 10 in the same subordinate relationship as Appendices 2 to 7. Furthermore, not limited to Appendix 1 and Appendices 8 to 10, within the scope not departing from the above-described embodiments, similarly for various hardware, software, various recording means for recording software, or systems, some or all of the configurations described as appendices may be made subordinate.

[0077] As described above, the present disclosure has been described with reference to the embodiments. However, the present disclosure is not limited to the above-described embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. And each embodiment can be combined with other embodiments as appropriate.

Description of Reference Numerals

[0078] 1,200 Purchase Behavior Analysis Device 13,201 Generation Unit 15,202 Identification Unit 17,203 Output Unit

Claims

1. A generation unit that generates purchase situation information including information on the total purchase amount, which is the total amount of purchase amounts for the subject in each analysis unit period, using purchase history information including the subject who purchased the product, the timing of purchasing the product, and the purchase amount of the product; A specifying unit that specifies, as a variation factor, a product or product classification that is a factor in the variation of the purchase situation for a subject whose purchase situation has changed, using the purchase history information and the purchase situation information; An output unit that outputs variation factor information representing the specified variation factor A purchase behavior analysis device comprising:

2. The specifying unit further specifies, as a purchase factor, a product or product classification that is presumed to be related to the purchase behavior of the subject for each of a plurality of divided periods set in the analysis unit period, The output unit further outputs, as purchase factor information, information representing the purchase factor for each divided period in the analysis unit period for display on a display device. The purchase behavior analysis device according to claim 1.

3. The specifying unit specifies the variation factor using a variation factor specifying model that takes at least the purchase history information and the purchase situation information as input and outputs information on the variation factor, The variation factor specifying model is generated by learning purchase history information of a subject showing a variation in the purchase situation of interest and purchase history information of a subject whose purchase situation variation is otherwise. The purchase behavior analysis device according to claim 1.

4. Further comprising an analysis unit that specifies characteristics of a person who purchased a product or product classification to be analyzed, using attribute information of the subject who purchased the product or product classification to be analyzed, The output unit outputs purchaser characteristic information of a purchaser who purchased a product or product classification to be analyzed in response to an information request requesting purchaser characteristic information representing characteristics of a person who purchased the product. The purchase behavior analysis device according to claim 1.

5. The purchase factor information output by the output unit includes information for causing a display device to display a purchase factor time series pattern in which purchase factors for each divided period in the analysis unit period are arranged in time series. The purchase behavior analysis device according to claim 2.

6. The purchase factor information output by the output unit includes information representing the number of subjects for which combinations of adjacent purchase factors are the same in the purchase factor time series pattern. The purchase behavior analysis device according to claim 5.

7. Furthermore, an analysis unit is provided that identifies the characteristics of a person who has purchased the product or product category to be analyzed, using the attribute information of the entity that has purchased the product or product category to be analyzed. The purchasing behavior analysis apparatus according to claim 2, wherein the output unit outputs purchaser characteristic information in response to an information request that requests information representing the characteristics of a person who has purchased a product that is a purchasing factor displayed on a display device.

8. Generating means for generating purchase situation information including information on the total purchase amount, which is the total amount of purchase amounts for the entity for each analysis unit period, using purchase history information including the entity that has purchased a product, the timing of purchase of the product, and the purchase amount of the product; Specifying means for specifying, as a variation factor, a product or product category that is a factor in the variation of the purchase situation for an entity whose purchase situation has varied, using the purchase history information and the purchase situation information; Output means for outputting variation factor information representing the specified variation factor A purchasing behavior analysis system comprising:

9. By a computer, Generating purchase situation information including information on the total purchase amount, which is the total amount of purchase amounts for the entity for each analysis unit period, using purchase history information including the entity that has purchased a product, the timing of purchase of the product, and the purchase amount of the product; Specifying, as a variation factor, a product or product category that is a factor in the variation of the purchase situation for an entity whose purchase situation has varied, using the purchase history information and the purchase situation information; A purchasing behavior analysis method for outputting variation factor information representing the specified variation factor.

10. A process of generating purchase situation information including information on the total purchase amount, which is the total amount of purchase amounts for the entity for each analysis unit period, using purchase history information including the entity that has purchased a product, the timing of purchase of the product, and the purchase amount of the product; A process of specifying, as a variation factor, a product or product category that is a factor in the variation of the purchase situation for an entity whose purchase situation has varied, using the purchase history information and the purchase situation information; A process of outputting variation factor information representing the specified variation factor A computer program for causing a computer to execute.

Images