Information processing device, information processing method and information processing program

Abstract

To solve "the problems in a method of setting a non-touch group in advertisement touch group / non-touch group comparison and its accuracy" and the problem in "accuracy of targeting to be measured" that occur in analysis in which advertisement effect is measured.SOLUTION: A purchase analysis support system includes: an advertisement distribution target group extraction unit that extracts targets to be used for advertisement distribution and purchase analysis from member data and / or purchase fact data based on predetermined segment conditions such as member status and purchase loyalty by user's purchase starting points, which are in line with needs of an advertiser; a distribution group / control group extraction unit that generates an accurate and bias-free control group from an advertisement distribution target group; and an analysis unit that analyzes changes in purchase behaviors of members before and after provision of advertisements, based on the member data and / or the purchase fact data, for an advertisement touch group and an advertisement non-touch group identified from the advertisement distribution target group.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program for supporting purchasing analysis of products and services. [Background technology]

[0002] It is generally known that analysis is carried out on products and services. For example, sales promotions are carried out for a company's products and services, and which products sell well to which segments of users, or the strengths and areas for improvement of the company's products and services are analyzed for target users.

[0003] Online advertisements provided on the Internet and the like, as well as offline advertisements using television, newspapers, flyers, or advertising media attached to trains, buses, stations, etc., are used as a means of promoting the sale of products and services to users. In recent years, advertisements using video display devices such as digital signage have also become widespread. There has been a need to measure the effectiveness of advertisements in order to understand the extent to which such advertisements have attracted users' interest, or whether advertisements that match the users' preferences are being provided.

[0004] In ad distribution, targets that are likely to have a good effect on sales promotion are set in advance, and ads are distributed to users who meet those conditions. Therefore, when measuring the effectiveness of an ad, by subdividing the target groups into the same groups as those used at the time of ad distribution and analyzing them, it is possible to understand whether the targets set before distribution were satisfactory and which target group contributed the most to sales promotion. In particular, advertisers want to change the means and methods of communication depending on the user's purchasing status of their own products or their competitors' products, so there is a high demand for setting targets based on user purchasing behavior information.

[0005] On the other hand, when measuring the effectiveness of marketing using advertising, it is effective to use the purchasing behavior information of users who received the advertisement as an index (called purchasing analysis). A known method is to analyze how the purchasing behavior of users regarding the company's products and services changed before and after the advertisement of the company's products and services is presented. It is also effective to compare the advertising exposure group (hereinafter also referred to as the "exposure group") with the advertising non-exposure group (hereinafter also referred to as the "non-exposure group") and measure the change in purchasing behavior after the advertisement is delivered. In other words, if the purchasing growth rate of those exposed to the advertisement (change in purchasing behavior due to the effect of the advertisement) is higher than the purchasing growth rate of the non-advertisement group (change in purchasing behavior due to factors other than the advertisement, such as the market) before and after the advertisement is delivered, then the advertisement can be evaluated as having had an impact on purchasing.

[0006] Patent Literature 1 discloses a technique for enabling measurement of the effectiveness of an advertising campaign based on online advertising targeted at members of an online system. Patent Literature 1 describes a technique for measuring the effectiveness of an advertisement by presenting an advertisement to one set of members of the online system such that one set of members is permitted to view the advertisement and another set of members is not permitted to view the advertisement under the same conditions as the first set, and presenting questions to these members aimed at measuring the effectiveness of the advertising campaign.

[0007] Patent Document 2 describes a method of dividing a group of potential customers, which is a group of individuals who may visit a target store, into a group of customers who have contacted the store content of the target store and a group of customers who have not contacted the store content of the target store based on the probability of the group of potential customers viewing the advertisement flyer, and verifying the effectiveness of the advertisement flyer. Patent Document 2 also describes a method of classifying the group of potential customers according to an index of the ease of contact with the store content based on the number of times the group of potential customers has visited the target store in the past and the gender and age of the group of potential customers, and obtaining the difference between the group of customers and the group of customers who have not contacted the store content among the customers of the target store for each of the hierarchical levels of the ease of contact with the store content, that is, Light, Middle, and Heavy, as the effect of the advertisement flyer. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] Special Publication No. 2012-528394 [Patent Document 2] Patent No. 7124240 Summary of the Invention [Problem to be solved by the invention]

[0009] Issues with conventional technology in measuring advertising effectiveness include the "precision of targeting the items being measured" and "problems with the method of setting the non-exposed group when comparing exposed and non-exposed groups, and the accuracy of that method."

[0010] First, let us consider the "precision of targeting to be measured." In conventional technology, targets for ad delivery and effectiveness measurement are extracted from members registered in an online system or from a group of individuals who are likely to visit a target store. For example, a platform provider that provides a specific online system creates targets based on user registration information, web action data, or estimations made from such data. Or, as in Patent Document 2, targets may be created based on the number of past visits to a target store or the gender and age information of potential store visitors. Such methods in the prior art cannot be said to accurately capture the purchasing behavior of users that advertisers are looking for, and there were problems with the accuracy of the targeting created.

[0011] Next, we will discuss "problems with the method of setting the non-contact group in the comparison of the ad contact group and the non-contact group and its accuracy." The ad contact group, which is necessary to carry out a measurement method for comparing the ad contact group and the ad contact group, is almost never managed on the online system that distributes the ads. Therefore, it is necessary to set the ad contact group in some way, as in Patent Document 1. However, in the method of setting the ad contact group in the conventional technology, the member group is dependent on a specific online system or store, so there are cases where bias occurs in the ad evaluation, or cases where bias occurs due to the extraction method, making it difficult to make a proper comparison with the ad contact group. In addition, some platforms do not have the function to set the ad contact group in the first place. Another setting method is to conduct research or questionnaires on members, and set members who answer "I have not seen an advertisement (or I feel like I have not seen an advertisement)" as the ad contact group based on the results, but this method cannot be said to be reliable in terms of accuracy.

[0012] The present invention has been made in light of the above-mentioned circumstances, and provides an information processing device, an information processing method, and an information processing program that, in advertising delivery and effectiveness measurement, set advertising exposure / non-exposure groups that are not dependent on a specific environment and that have reduced bias, and further support highly accurate purchasing analysis in segments that truly capture user behavior by using purchasing fact data. [Means for solving the problem]

[0013] In order to solve the above problem, a first aspect of an information processing device supporting purchasing analysis according to the present disclosure is characterized in that it comprises an advertising delivery target group extraction unit that divides members who are candidates for advertising delivery (referred to as an advertising delivery candidate group) from a predetermined member group into predetermined segments and extracts an advertising delivery target group, and an analysis unit that analyzes changes in purchasing behavior before and after the provision of advertising for an advertising exposure group and an advertising non-exposure group identified from the advertising delivery target group based on member data and / or transaction data.

[0014] A second aspect of the information processing device of the present disclosure is characterized in that the advertising delivery target group extraction unit divides the member group subject to purchasing analysis into the specified segments based on the member data and / or transaction data.

[0015] A third aspect of the information processing device according to the present disclosure is characterized in that the analysis unit analyzes changes in purchasing behavior based on the advertisement exposure group, the advertisement non-exposure group, and the specified segment.

[0016] A fourth aspect of the information processing device according to the present disclosure is characterized in that the analysis unit displays a purchase lift before and after providing the advertisement.

[0017] A fifth aspect of the information processing device according to the present disclosure is characterized in that the predetermined segments include existing, dormant, and new categories.

[0018] A sixth aspect of the information processing device according to the present disclosure is characterized in that the specified segments include divisions based on purchasing loyalty (a measurement of the degree of loyalty based on user purchasing behavior such as purchasing frequency and purchasing period).

[0019] A seventh aspect of the information processing device of the present disclosure is characterized in that when a control group to which advertisements are not delivered is set to the advertising delivery target group, the control group is randomly extracted from the advertising delivery target group. Effect of the Invention

[0020] According to the first aspect of the present disclosure, targets to be set during advertisement delivery can be selected based on predetermined conditions using purchase fact data, so targets that meet the needs of advertisers can be generated with high accuracy compared to conventional techniques. Similarly, by analyzing using purchase fact data, it is possible to measure the effectiveness of advertisements based on the purchasing behavior of users, which is the final result of advertisements. Furthermore, since information on people who have purchased competing products can be obtained, it is possible to appeal to such purchasers about the company's own products.

[0021] According to the second aspect of the present disclosure, it is possible to set conditions for extracting targets for purchasing analysis from member data and / or purchasing fact data associated with a member ID, thereby making it possible to set a group of advertising delivery targets that match the analysis purpose.

[0022] According to a third aspect of the present disclosure, by dividing an ad delivery candidate group into predetermined segments and generating an ad delivery target group, the ad delivery target group can be subdivided into an ad exposure group and an ad non-exposure group, and during purchase analysis, it is possible to perform detailed and multifaceted analysis of the fluctuations in purchase rates of each ad delivery target group before and after ad delivery, and the ad exposure group / non-exposure group.

[0023] According to the fourth aspect of the present disclosure, by calculating the lift in purchase rate, it is possible to present in graphs and numerical values ​​which categories of members the advertising was effective for, making it easy to compare the advertising effectiveness for each advertising delivery target group.

[0024] According to a fifth aspect of the present disclosure, by dividing the group of candidates for ad delivery into existing, dormant, and new segments based on purchasing fact data, it becomes possible to grasp the chronological changes in user status (referred to as membership status) regarding purchases of the company's own or competitors' products, etc., and this can be utilized in a variety of ways for ad delivery and purchasing analysis.

[0025] According to the sixth aspect of the present disclosure, by dividing the group of candidates for ad delivery into segments based on purchase loyalty based on purchase fact data, it becomes possible to grasp the degree of loyalty, such as purchase frequency, of users who use the company's own or competitor's products, which can be utilized in a variety of ways for ad delivery and purchase analysis.

[0026] According to the seventh aspect of the present disclosure, by setting a control group for an ad delivery target group with the same segment conditions, the ad exposure group / non-exposure group can be measured fairly, and furthermore, since the control group can be randomly extracted based on the gender and age ratio of the ad delivery target group, data bias can be suppressed. In addition, since the ad non-exposure group can be set without depending on a specific environment and is clearly distinguished from the ad delivery group, it is guaranteed that the group is not exposed to the ad. This allows the ad non-exposure group to be set accurately, and the advertising effect can be measured without market influence. [Brief description of the drawings]

[0027] [Figure 1] FIG. 1 is a bird's-eye view showing an example of the overall configuration of a purchase analysis support system 1 to which an information processing device according to the first embodiment of the present disclosure is applied. [Diagram 2] FIG. 2 is a block diagram showing an example of a functional configuration of the purchase analysis device 2 according to the first embodiment of the present disclosure. [Diagram 3] FIG. 3 is a diagram illustrating an example of a member data table according to the first embodiment of the present disclosure. [Figure 4] FIG. 4 is a diagram illustrating an example of a transaction data table according to the first embodiment of the present disclosure. [Diagram 5] FIG. 5 is a diagram illustrating an example of an advertisement delivery result table according to the first embodiment of the present disclosure. [Figure 6] FIG. 6 is a diagram showing an example of a segment based on purchase royalty according to the first embodiment of the present disclosure. [Figure 7] FIG. 7 is a diagram for explaining definitions of member statuses according to the first embodiment of the present disclosure. [Figure 8] FIG. 8 is a diagram showing an example of segments based on chronological member status (existing, dormant, new) according to the first embodiment of the present disclosure. [Figure 9] FIG. 9 is a block diagram showing an example of a functional configuration of the transaction collection device 3 according to the first embodiment of the present disclosure. [Figure 10]FIG. 10 is a diagram illustrating an example of a functional configuration of the user-side advertisement display processing device 4 according to the first embodiment of the present disclosure. [Figure 11] FIG. 11 is a flowchart for explaining an example of the operation of the purchase analysis device 2 according to the first embodiment of the present disclosure. [Figure 12] FIG. 12 is a diagram illustrating an example of an analytical tally table according to the first embodiment of the present disclosure. [Figure 13] FIG. 13 is a diagram showing an example of a purchase analysis result according to the first embodiment of the present disclosure. [Figure 14] FIG. 14 is a diagram for explaining the reverse analysis according to the second embodiment of the present disclosure. [Figure 15] FIG. 15 is a flowchart for explaining an example of the operation of the purchase analysis device 2 using reverse lookup analysis according to the second embodiment of the present disclosure. [Figure 16] FIG. 16 is a diagram showing an example of an analysis result by the reverse analysis according to the second embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0028] Hereinafter, multiple embodiments of the present disclosure will be described in detail with reference to the drawings. In the following description, configurations and elements that are the same as or similar to configurations that have already been described will be given the same reference numerals and descriptions thereof will be omitted.

[0029] First Embodiment An example of a purchase analysis support system 1 according to the first embodiment will be described below with reference to the drawings.

[0030] (1) Overall configuration of purchasing analysis support system 1 Fig. 1 is a bird's-eye view showing an example of the overall configuration of a purchase analysis support system 1 to which an information processing device, an information processing method, and an information processing program according to a first embodiment of the present disclosure are applied. The purchase analysis support system 1 shown in Fig. 1 includes a purchase analysis device 2, transaction collection devices 3A and 3B associated with store A and store B, and a user-side advertisement display processing device 4 used by a user (also called a "member," "consumer," or "customer," depending on the situation), and these information processing devices are connected via a network 5 such as the Internet.

[0031] The purchase analysis device 2 is assumed to manage a point service, for example, a P point service, operated by a specific point issuing company. Store A and Store B also participate in the P point service, and when a user purchases a product or service at Store A or Store B, P points are given according to the purchase amount, and the user can use the P points for purchases. Store A and Store B may be stores of any type, including, for example, small stores run by individuals or private companies, large retail stores such as shopping centers, department stores, and the like. In the case of a franchise formed, such as a convenience store, the stores may be directly managed stores or affiliated stores managed by a specific franchiser. Furthermore, the stores are not limited to physical stores, and may be virtual stores on a network.

[0032] The purchase analysis device 2 communicates between the transaction collection devices 3A, 3B and the user-side advertisement display processing device 4, and is responsible for managing the entire purchase analysis support system 1. The purchase analysis device 2 acquires transaction data from the transaction collection devices 3A, 3B. The transaction data is data managed by the transaction collection device 3, and includes at least IDPOS data and point management data. The IDPOS data is POS data linked to customer information by an identifier that identifies the customer. Here, for example, an ID that identifies a member of the P point service can be used as an identifier that identifies the customer. The POS data includes purchase fact data such as the store used by the customer, date and time, quantity, and price. The point management data includes the customer's point usage history, such as points given to the customer at the time of purchase and points used by the customer.

[0033] By using IDPOS data, transaction data including POS data and point management data can be linked to membership data including attribute information such as the user's gender and age, and purchase history information, via the P Point Service member ID. This allows members with linked personal attribute information and purchase information to be the subject of analysis, allowing purchase analysis to be performed with a high degree of freedom from a variety of perspectives. Note that in cases where the member ID and the customer ID assigned by the store are associated with each other through a correspondence table or the like, the POS data may include the customer ID instead of the member ID. In other words, the IDPOS data may be data that allows the member data and transaction data to be associated directly or indirectly with each other through an identifier that identifies the user.

[0034] The purchasing analysis device 2 extracts a group of advertising delivery candidates, extracts a group of advertising delivery targets, extracts an advertising delivery group and a control group (also called a control group) from the group of advertising delivery targets, generates an advertising delivery list, delivers advertisements, stores advertising delivery results in an advertising delivery result table, collects purchasing fact data, and performs purchasing analysis.

[0035] The purchase analysis device 2 extracts a group of candidates for advertisement delivery from a predetermined group of members based on predetermined extraction conditions. Here, the predetermined group of members is a collection of members managed by IDs that identify the members, such as a P point service. The predetermined extraction conditions are conditions for extracting a group of members who are candidates for advertisement delivery from member data or transaction data, and the member data or transaction data is searched for based on the predetermined extraction conditions, and members who satisfy the extraction conditions are extracted as targets for purchase analysis.

[0036] The purchase analysis device 2 divides the ad delivery candidate group into predetermined segments, which are used as ad delivery target groups for ad delivery and purchase analysis. The predetermined segments are set so as to be able to divide the ad delivery target group according to the purpose of ad delivery and purchase analysis. Once the predetermined segments are defined, segment conditions are set to identify members who fall within the segment from the ad delivery candidate group. Then, the ad delivery candidate group is divided into predetermined segments by searching for members who satisfy the segment conditions from member data or transaction data from among the ad delivery candidate group.

[0037] The purchase analysis device 2 identifies a delivery group to which an advertisement is delivered from the advertisement delivery target group. At this time, a delivery group to which an advertisement is delivered and a control group to which an advertisement is not delivered may be extracted from the advertisement delivery target group, or a control group may not be set and the entire advertisement delivery target group may be set as the delivery group. When a control group is set for the advertisement delivery target group, the control group may be randomly extracted from the advertisement delivery target group.

[0038] The purchase analysis device 2 generates an advertisement distribution list and executes advertisement distribution to the advertisement distribution group. The advertisement distribution list is a list of advertisement distributors in each advertisement distribution target group. When requesting the platformer to distribute advertisements, the advertisement distribution list can be presented and the advertisements can be distributed to those matching the platformer's users. The advertisement distribution result table is a table for recording advertisement distribution information after advertisement distribution, and includes distribution information such as the advertisement distribution destination for each member and whether the member has come into contact with the advertisement. The advertisement distribution result table may be held by the platformer. Furthermore, although the control group is not included in the advertisement distribution list, it may be configured to be managed in association with the purchase analysis device 2 and extracted and used during analysis.

[0039] The predetermined segment may be a segment based on purchase loyalty. Purchase loyalty is an index of a user's purchasing behavior among customer loyalty, which indicates the degree of trust, attachment, satisfaction, and loyalty to a company's products or brand. For example, segment conditions based on the members' purchasing frequency can be set, and the members can be divided into heavy, middle, and light segments. The predetermined segment may also be a segment based on member status, and for example, the members can be divided into existing, dormant, and new segments. Customer loyalty and member status are not limited to the company's products, and a competitor's products or brand may be set as a segment condition to understand the situation of other companies. Furthermore, the segments are not limited to classification based on purchase loyalty and member status, and any segment can be defined according to the purpose of the purchase analysis.

[0040] After the advertisement delivery period ends, the purchase analysis device 2 collects purchase fact data including data before and after the advertisement delivery period. The purchase analysis device 2 acquires transaction data for a predetermined analysis period from the transaction collection device 3, and extracts purchase fact data of members in the advertisement delivery target group. Using the purchase fact data, the purchasing behavior of each member before and after advertisement delivery can be analyzed.

[0041] Furthermore, the purchasing analysis device 2 identifies an advertisement exposure group and an advertisement non-exposure group from the advertising delivery target group. When the advertising delivery target group is divided into a "delivery group" and a "control group", the purchasing analysis device 2 determines members from the "delivery group" who have been exposed to the advertisement as the advertising exposure group, and treats the "control group" as the advertising non-exposure group. On the other hand, when a control group is not set for the advertising delivery target group, the purchasing analysis device 2 determines members from the advertising delivery target group who have been exposed to the advertisement, and identifies the "exposure group". The fact that a member has been exposed to an advertisement is determined using the "advertisement exposure" flag recorded in the advertising delivery result table.

[0042] The purchase analysis device 2 analyzes changes in purchasing behavior before and after the provision of advertisements for an advertisement contact group and an advertisement non-contact group identified from an advertisement delivery target group, based on member data and / or transaction data. The purchase analysis device 2 collects purchasing behavior obtained from segment information, contact / non-contact information, and purchasing fact data set for each member of the advertisement delivery target group, and analyzes changes in purchasing behavior based on the advertisement contact group, advertisement non-contact group, and segments. For example, for each contact group and non-contact group, the purchase rate of members included in the segment is calculated before and after advertisement delivery, to calculate changes in purchasing behavior, and purchase lift before and after advertisement delivery is displayed. The purchase analysis device 2 may record the purchase analysis results in the form of a report.

[0043] The transaction collection devices 3A and 3B are installed in, for example, a store, and are information processing devices capable of collecting and managing transaction data such as IDPOS data and point management data. The transaction collection devices 3A and 3B include a management device 31 and a storage device 32, and are communicatively connected to the purchase analysis device 2 via a network 5. The transaction collection devices 3A and 3B do not have to be dedicated information processing devices, and may be incorporated into a POS device installed in a store, a management device (or a management server) that manages the POS device, or a product / inventory management device. The management device 31 and the storage device 32 may be communicatively connected, and for example, the storage device 32 may be installed in a store, or may be arranged on a server via a network.

[0044] The user-side advertisement display processing device 4 is an information processing device used by a user, and is communicably connected to the purchase analysis device 2 via a network 5. The user-side advertisement display processing device 4 acquires advertisements and landing pages (LPs) provided by the purchase analysis device 2, and performs display control or operation history management. A landing page is a page on which a user first lands via an advertisement or the like. The user-side advertisement display processing device 4 will be described as displaying advertisements and the like provided by the purchase analysis device 2 and recording a browsing history by an application installed in the user-side advertisement display processing device 4, but instead of such an application, it may be configured to access a Web server or an ad server including advertisements and the like provided by the purchase analysis device 2 by a web browser provided in the user-side advertisement display processing device 4. At this time, the Web server or the ad server may record the browsing history.

[0045] (2) Hardware Configuration (2-1)Purchase analysis device 2 The purchase analysis device 2 includes a processor 21, a memory 22, a storage 23, a communication interface (communication I / F) 24, an input / output interface (input / output I / F) 25, and a bus 26 connecting them. The processor 21 controls the operation of the entire purchase analysis device 2. The processor 21 may be a general-purpose processor such as a CPU, MPU, or GPU, but is not limited to a general-purpose processor and may be a dedicated processor such as an ASIC or FPGA. The memory 22 is a main storage device and includes a RAM, etc. The storage 23 is an auxiliary storage device and includes a non-volatile storage device such as a hard disk drive (HDD) or a solid state drive (SSD). The communication interface (communication I / F) 24 is a wired or wireless communication interface and is a module for communicating with the transaction collection device 3, the user side advertisement display processing device 4, etc. via the network 5. The network 5 is, for example, the Internet, and may include access networks such as a LAN, a WAN, a mobile communication network, a wired telephone network, a FTTH, and a CATV network. An input / output interface (input / output I / F) 25 receives input data from an external device and outputs output data to an external device.

[0046] (2-2) Transaction collection device 3 The transaction collection device 3 includes at least a processor, a memory, a storage, a communication interface (communication I / F), an input / output interface (input / output I / F), and a bus connecting these. The transaction collection device 3 corresponds to, for example, a POS device installed in a store, or a management server located in a headquarters that is an upper authority of the store. The transaction collection device 3 does not have to be a single information processing device, and may be, for example, distributed among multiple information processing devices, or a single information processing device may be used in common for other processes.

[0047] (2-3) User-side advertisement display processing device 4 The user-side advertisement display processing device 4 includes a processor 41, a memory 42, a storage 43, a camera 44, a communication interface (communication I / F) 45, an input / output interface (input / output I / F) 46, a touch screen 47, and a bus 40 connecting these. Here, the storage 43 stores an operating system (OS), a web browser, an application program, various data, and the like. The touch screen 47 includes an input unit 48 and a display unit 49. The user-side advertisement display processing device 4 is, for example, a mobile terminal such as a smartphone, a tablet terminal, or a notebook personal computer. The user-side advertisement display processing device 4 may be a stationary computer such as a desktop personal computer. The user-side advertisement display processing device 4 may also be an information processing device that collectively refers to a form in which a mobile terminal and a stationary computer are combined, such as displaying an advertisement using a stationary computer and responding to the advertisement using a mobile terminal.

[0048] (3) Functional configuration (3-1)Purchase analysis device 2 2 is a block diagram showing an example of a functional configuration of the purchase analysis device 2 according to the first embodiment. The purchase analysis device 2 according to the first embodiment includes, as functional components, a purchase analysis unit 200, a storage unit 210, and a communication control unit 220. The communication control unit 220 controls communication with the transaction collection device 3 and the user-side advertisement display processing device 4 via the network 5 in accordance with instructions from the purchase analysis unit 200.

[0049] (3-1-1) Storage section 210 The memory unit 210 stores information for supporting purchase analysis, and includes a member data table storage unit 211 , a transaction data storage unit 212 , an advertisement delivery result table storage unit 213 , an analysis data storage unit 214 , and an analysis result storage unit 215 .

[0050] A member data table that records member data of the P point service is stored in the member data table storage unit 211. The member data table includes personal attribute data such as a member ID, a member's name, sex, date of birth, address, family structure, hobbies, and preferences.

[0051] FIG. 3 is a diagram illustrating a main part of a member data table. As shown in FIG. 3, in the member data table, a record of member data is formed for each member ID, and the main contents of the record include a member ID for uniquely identifying a member, a key type and key for identifying the member's contact information, and the member's gender and age. The member ID is, for example, an identifier for identifying a member of a P point service. The key type is an item that explains the meaning of the following key, TEL means that the following number is a telephone number, the mobile advertising ID means that the following key is a mobile advertising identifier, and MAIL indicates that the following key is an email address. When delivering advertisements, the key type and key are used as the advertisement delivery destination. It goes without saying that the member data table in FIG. 3 shows only the main data, and may include other data related to the member.

[0052] The transaction data storage unit 212 stores a transaction data table including IDPOS data including member ID and POS data, and point management data. The POS data includes purchase fact data and / or point transactions when a user purchases a product or receives a service at a store, such as the store where the user purchased the product, date and time, quantity, and price. The purchase analysis device 2 acquires transaction data from the transaction collection device 3 periodically or at any time, and updates the transaction data table with the acquired transaction data.

[0053] FIG. 4 is a diagram illustrating the main part of the transaction data table. The transaction data table includes member ID, store ID, purchase date (YYYY MM / DD), time (HH:MM:SS), product name, JAN code, number of items, unit price, point type (granted, used, cancelled), and points. "Member ID" is the member ID of the P point service, and this member ID associates records in the transaction data table with records of the same member ID in the member data table. "Store ID" is an identifier for identifying the store that obtained the transaction data. "Purchase date", "Purchase time", "JAN code", "number of items", and "unit price" are examples of purchase fact data by members. Among these, "JAN code (Japanese Article Number)" is an international standard product identification code that indicates which business operator and which product. "Point type" and "point" are data related to point transactions, and "point type" indicates whether points were granted, used, or cancelled when purchasing a product, and "point" records the number of points granted, used, or cancelled depending on the point type.

[0054] The member data table in Figure 3 and the transaction data table in Figure 4 are linked by the P Point Service member ID to form a relational database. In this way, linking the two tables with a single source ID makes it possible to combine and extract the necessary data from these tables. Note that a relational database may also be formed from three or more tables. This makes it possible to use purchasing fact data in addition to member attributes, and makes it possible to measure effectiveness using indicators that are more in line with the user's actual situation than with existing technology.

[0055] The advertisement delivery result table storage unit 213 stores advertisement delivery result information. The advertisement delivery result table is used to record delivery information such as delivery destination information when the purchase analysis device 2 delivers an advertisement to an advertisement delivery target group, and advertisement contact information by members. The delivery information recorded in the advertisement delivery result table is used in purchase analysis.

[0056] The analytical data storage unit 214 stores data required for purchase analysis by the analysis unit 207 as an analytical aggregation table. The analytical aggregation table stores, for example, member information collected from the member data table, delivery information collected from the ad delivery result table, segment information set by the ad delivery target group extraction unit 202, purchase fact data aggregated by the purchase fact data collection unit 206, and the like.

[0057] The analysis result storage unit 215 stores a report analyzed by the analysis unit 207. The report includes data and graphs related to purchase lift.

[0058] (3-1-2) Purchasing Analysis Department 200 The purchasing analysis unit 200 functions as an advertising delivery candidate group extraction unit 201, an advertising delivery target group extraction unit 202, a delivery group / control group extraction unit 203, an advertising delivery list generation unit 204, an advertising delivery unit 205, a purchasing fact data collection unit 206, and an analysis unit 207 by the processor 21 executing the program stored in the memory 22.

[0059] The ad delivery candidate group extraction unit 201 searches the member data table and the transaction data table according to predetermined extraction conditions, and extracts members who satisfy the predetermined extraction conditions as an ad delivery candidate group. The ad delivery candidate group is a member group that is a candidate for the ad delivery target group, and the ad delivery target group to which the advertisement is actually delivered is determined from this member group. The extraction conditions are set for extracting a group that is significant for purchase analysis, and for example, members who have not made a purchase for more than two years or members whose correct contact information is not registered are not included in the ad delivery candidate group.

[0060] The ad delivery target group extraction unit 202 divides the ad delivery candidate group into predetermined segments. The ad delivery target group extraction unit 202 searches for members who satisfy the segment conditions using member data or transaction data for members included in the ad delivery candidate group. The segment conditions are conditions for extracting members that correspond to a segment using member data or transaction data. Then, for members who satisfy the segment conditions, the corresponding segment is assigned. For example, as a segment condition for purchase loyalty, a purchase frequency of a specific product of 10 or more per month is set as a heavy layer, 9 to 3 per month as a middle layer, and 2 or less per month as a light layer. The purchase analysis device 2 uses purchase fact data of each member included in the transaction data to determine whether each member matches the segment condition of the heavy layer, middle layer, or light layer, and assigns the corresponding segment to each member.

[0061] Figure 6 is a diagram showing the distribution status of the ad delivery target group when purchasing loyalty is divided into segments of heavy layer (H), middle layer (M), and light layer (L). As shown in Figure 6, the ad delivery target group is divided into an ad exposure group that was exposed to advertisements and an ad non-exposure group that was not exposed to advertisements. Then, each of the ad exposure group and the ad non-exposure group can be divided into the heavy layer (H), middle layer (M), and light layer (L) segments.

[0062] Segments based on purchase loyalty can be set not only based on purchase frequency, but also on point usage, the number of points awarded, and the like. Furthermore, purchase loyalty segments can also be calculated for specific products or brands. In this way, segments can be defined from multiple perspectives regarding purchase loyalty. Note that segments may be defined in advance, or may be defined by an operator by selecting from multiple segments at his / her discretion. Furthermore, new purchase loyalty can also be defined by combining multiple elements such as purchase frequency and the number of points awarded.

[0063] FIG. 7 is a diagram for explaining "existing," "dormant," and "new" as examples of segments based on member status. These segments can be set by observing the purchasing status of members in a chronological order. When viewing the purchasing status of members going back two years from the advertisement delivery, for example, a member who made a purchase one to six months before the advertisement delivery is classified as "existing," a member who did not make a purchase one to six months before the advertisement delivery but made a purchase six months to two years before the advertisement delivery is classified as "dormant," and a member who did not make a purchase one to two years before the advertisement delivery is classified as "new." The relationship between these purchasing statuses and segments can be set as segment conditions. The purchasing analysis device 2 uses the purchasing fact data of each member included in the transaction data to determine whether each member matches the segment condition of existing, dormant, or new, and assigns the corresponding segment to each member.

[0064] Fig. 8 is a diagram showing the distribution status of the ad delivery target group when the membership status is divided into segments of "existing", "dormant", and "new". As shown in Fig. 8, the ad delivery target group is divided into an ad-exposed group and an ad-unexposed group, and the ad-exposed and ad-unexposed groups can be further divided into "existing", "dormant", and "new" segments. The advertising effect on the ad delivery target group is analyzed according to the six categories shown in Fig. 8.

[0065] As for segments based on membership status, in addition to "existing," "dormant," and "new," segments can be defined from multiple perspectives, such as membership period, premium membership, etc. In this case, a configuration may be made in which it is possible to select which of these segments to use as the membership status, or a new membership status segment can be defined by combining multiple elements such as existing, dormant, new, and membership period.

[0066] The segments based on membership status and purchase loyalty may be defined as separate segments, but they can also be defined as a combination of the two. For example, for members classified as "existing" in membership status, segmentation by purchase loyalty can be applied based on each member's purchase frequency and purchase amount, and they can be classified as "existing (high loyalty)" and "existing (low loyalty)".

[0067] The delivery group / control group extraction unit 203 selects one of two extraction methods depending on whether or not the operator has specified to set a control group for the ad delivery target group. The first extraction method is a method of extracting a delivery group and a control group to which no advertisement is delivered from the ad delivery target group. The second extraction method is a method of extracting the ad delivery target group as a delivery group as it is.

[0068] When the control group setting is specified, the delivery group / control group extraction unit 203 randomly extracts about 10 to 20 percent of the ad delivery target group and sets them as a control group to which the ad is not delivered. Then, the group of members remaining after removing the control group from the ad delivery target group is set as the delivery group. By setting the control group, the group of members who are not exposed to the ad can be clearly set in the purchase analysis process. As a method for extracting the control group, in addition to random extraction from the entire ad delivery target group, it may be randomly extracted according to the gender and age ratio of the ad delivery target group. Alternatively, the control group may be randomly extracted from members of the ad delivery target group to whom delivery cannot actually be made, for example, members who do not hold information agreed upon by the platformer when delivery is made by the platformer.

[0069] On the other hand, when the control group setting is not specified, the delivery group / control group extraction unit 203 sets the advertisement delivery target group as the delivery group as it is, and does not set a control group.

[0070] The advertisement distribution list generating unit 204 generates an advertisement distribution list in which distribution information of members to whom advertisements are to be distributed is recorded for each advertisement distribution target group.

[0071] When the advertisement distribution unit 205 receives a distribution instruction command including the distribution start date and time, distribution period, content of the advertisement to be distributed, distribution medium, etc. from the operator, it distributes the advertisement instructed to be distributed to the members of the distribution group at the specified distribution start date and time. The advertisement distribution unit 205 acquires the corresponding "key type" and "key" based on the member ID included in the distribution group of the advertisement distribution list generated by the advertisement distribution list generation unit 204. The advertisement distribution unit 205 generates a distribution destination address of the advertisement using the "key type" and "key", and distributes the advertisement to the user-side advertisement display processing device 4. Note that advertisement distribution may be requested to a third party having an advertisement distribution surface such as a platformer. Furthermore, it is not limited to electronic distribution, and may be distribution by paper medium.

[0072] The advertisement distribution unit 205 can determine that an impression has occurred when an advertisement is displayed on an application or browser being viewed by a user. Alternatively, it may detect that a member has performed some operation on the distributed advertisement information. For example, it can determine that a member has been in contact with an advertisement when it is detected that a user has opened a distributed advertisement page or when a user operation on a distributed advertisement is detected. Also, a landing page (LP) for directing a user to an advertisement-related site may be embedded in the distributed advertisement, and an operation on the advertisement (Click) may be determined when a member clicks on a link to the LP displayed on the user-side advertisement display processing device 4. When the advertisement distribution unit 205 determines that a member has been in contact with an advertisement, it sets the "advertisement exposure" flag in the advertisement distribution result table to 1.

[0073] The advertisement distribution unit 205 uses the "advertisement contact" flag to determine whether or not a member has actually been contacted with an advertisement, and a member with an "advertisement contact" flag set to 1 can be considered to have been contacted with the advertisement.

[0074] FIG. 5 is a diagram showing an example of an advertisement delivery result table. The member ID is a member ID of the P point service. The destination information of the advertisement delivery destination is a transcription of the key type and key contents recorded in the member data table, and types such as telephone (Tel), mobile advertising ID (Mobile Advertising ID: MAID), and email (Mail) are set as key types, and telephone numbers, mobile advertising IDs, and email addresses are recorded as keys. The advertisement contact flag is set to 1 if the member has contacted the delivered advertisement, and 0 if the member has not contacted the advertisement. Note that the table configuration of the advertisement delivery result table in FIG. 5 is merely an example, and any other table configuration may be adopted as long as the necessary delivery information is included.

[0075] The purchasing fact data collection unit 206 collects transaction data collected during an analysis period required for purchasing analysis from the transaction collection device 3. The analysis period is set as a period including before and after advertisement delivery. The transaction data collected from the transaction collection device 3 is recorded in a transaction data table, but is also transcribed into an analysis aggregation table and used as purchasing fact data in the analysis unit 207.

[0076] The analysis unit 207 matches the purchase fact data recorded in the transaction data table with the contact / non-contact group of each segment of the ad delivery target group for each member ID, and records the member IDs of the members included in each ad delivery target group and the purchase amounts before and after the advertisement for each member ID in the analysis aggregation table. In addition, the analysis unit 207 records the segments assigned to each member by the ad delivery target group extraction unit 202 and the "advertisement contact" flag recorded in the ad delivery result table in the analysis aggregation table.

[0077] The analysis unit 207 uses the analysis tally table to analyze changes in purchasing behavior before and after advertisement delivery. The changes in purchasing behavior can be measured, for example, as changes in purchase rate, and can be indexed as purchase lift.

[0078] The purchase rate is an index that indicates the percentage of users who actually purchased a product among users who visited a store. For example, the purchase rate of a certain target product can be calculated as the ratio of the number of users who purchased the target product to the total number of target users, as shown in the following formula. Purchase rate = Number of users who purchased the target product / Total number of target users Purchase lift is an index showing the degree to which a delivered advertisement has influenced a user's purchasing behavior, and is calculated as the percentage change in purchase rate before and after the advertisement is delivered. For example, the purchase lift of a target product can be calculated as the increase (or decrease) in purchase rate before and after the advertisement is delivered, as shown in the following formula. Purchase lift = Purchase rate after ad delivery / Purchase rate before ad delivery By measuring the purchase lift by segment for each contacted and non-contacted person, it is possible to analyze which segment of the advertising distribution target group the advertisement was effective for. The analysis unit 207 records the results of the analysis as a report in the analysis result storage unit 215.

[0079] (3-2) Transaction collection device 3 FIG. 9 is a block diagram showing an example of a functional configuration of the transaction collection device 3. As shown in FIG. The transaction collection device 3 includes a transaction processing unit 300, a transaction data storage unit 310, and a communication control unit 320. The transaction data storage unit 310 records transaction data collected by the transaction collection device 3, and in response to a request to acquire transaction data from the purchase analysis device 2, reads out the requested transaction data. FIG. 9 illustrates a configuration in which the transaction collection device 3 includes the transaction data storage unit 310, but the configuration is not limited to this. For example, if a store forms a franchise, the transaction data storage unit 310 may be installed in a server at a headquarters that manages the franchise store. That is, the transaction data storage unit 310 can be installed in any server that is communicably connected via the network 5.

[0080] The transaction processing unit 300 functions as a transaction manager 301 and a transaction provider 302 by the processor executing a program stored in the memory. The transaction management unit 301 acquires transaction data generated by a user's purchase of a product or service or the use of points from a POS device installed in the store, and stores the data in the transaction data storage unit 310. At this time, if it can be determined that the user is a member of the P point service by the user presenting a point card, etc., the transaction management unit 301 adds the member ID of the P point service to the transaction data and stores the transaction data in the transaction data storage unit 310.

[0081] When the transaction providing unit 302 receives a request to acquire transaction data generated during the purchase analysis period from the purchase analysis device 2, it authenticates the legitimacy of the purchase analysis device 2. If the transaction providing unit 302 authenticates the legitimacy of the purchase analysis device 2, in response to the request to acquire transaction data, it reads the requested transaction data from the transaction data storage unit 310 and transmits it to the purchase analysis device 2 via the communication control unit 320.

[0082] (3-3) User-side advertisement display processing device 4 FIG. 10 is a block diagram showing an example of a functional configuration of the user-side advertisement display processing device 4. As shown in FIG. The user-side advertisement display processing device 4 includes an advertisement display processing unit 400, a browsing history storage unit 410, and a communication control unit 420. The browsing history storage unit 410 stores information indicating that a user has viewed an advertisement displayed on the display unit 49 of the user-side advertisement display processing device 4 as a browsing history.

[0083] The advertisement display processing unit 400 functions as an advertisement receiving unit 401, an advertisement display control unit 402, and a browsing history management unit 403 by the processor 41 executing an application program stored in the memory 42. The advertisement receiving unit 401 receives an advertisement delivered from the purchase analysis device 2 to a telephone number, a mobile advertisement ID, or an email address. At this time, the advertisement itself may be received, or a link destination to the advertisement may be received. When the user performs an operation to view the advertisement delivered to the user-side advertisement display processing device 4, the advertisement display control unit 402 displays the received advertisement on the display unit 49. The browsing history management unit 403 records the user's operation on the advertisement, and transmits the browsing history to the purchase analysis device 2 in response to a request from the purchase analysis device 2. The user's operation on the advertisement may be an operation indicating that the user has come into contact with the advertisement, and includes, for example, an operation of clicking on the advertisement itself or a link to the advertisement, or an operation of clicking, scrolling, swiping, flicking, dragging, or character input or selection input on the displayed advertisement.

[0084] (4) Operation of Purchase Analysis Support System 1 (4-1) Operation of Purchase Analysis Support System 1 11 is a diagram showing an example of a schematic flowchart of a purchase analysis program implemented in the purchase analysis device 2. The purchase analysis program will be described as a computer program installed in the purchase analysis device 2, but it may be executed by one computer or multiple computers in the purchase analysis device 2. Also, instead of implementing all the computer programs in the purchase analysis device 2, part of the processing by the computer program may be distributed to other information processing devices.

[0085] The purchasing analysis program causes one or more computers to execute each of the steps of: extracting an advertising delivery candidate group (S1); extracting an advertising delivery target group (S2); extracting an advertising delivery group / control group (S3, S4); generating an advertising delivery list (S5, S6); implementing advertising delivery (S7); collecting purchasing data (S8); and analyzing purchasing lift (S9, S10).

[0086] In the ad delivery candidate group extraction step (S1), based on predetermined extraction conditions, members who satisfy the extraction conditions are searched for from the member data table stored in member data table storage unit 211 or the transaction data table stored in transaction data storage unit 212, and set as an ad delivery candidate group that is a candidate for ad delivery destinations. The predetermined extraction conditions are set for extracting a group that is significant for purchase analysis, and for example, members who have not made a purchase for more than two years or members whose correct contact information is not registered are not included in the ad delivery candidate group.

[0087] In the step (S2) of extracting an ad delivery target group, the members included in the ad delivery candidate group are divided into segments based on the member data or transaction data, and an ad delivery target group is extracted.

[0088] Segments based on purchase loyalty are defined as heavy (H), middle (M), and light (L) users based on purchasing conditions such as purchase frequency in a unit period, purchase amount per visit, point usage, etc. When an operator specifies segment conditions that correspond to this definition, the transaction data table is searched based on the segment conditions, members who satisfy the segment conditions are searched for, and the relevant segment is assigned to the member.

[0089] For example, if the segments heavy (H), middle (M), and light (L) are defined based on the purchase frequency of a specific product, segment conditions can be set such that if the purchase frequency is x times / month or more, the member is in the heavy tier, if it is between y times and less than x times, the member is in the middle tier, and if it is less than y times, the member is in the light tier. The monthly purchase frequency of a specific product can be calculated for each member ID from the "purchase date" and "product name" in the transaction data table shown in Figure 4. Then, by comparing the monthly purchase frequency with the segment conditions, each member can be assigned to one of the heavy (H), middle (M), or light (L) segments.

[0090] Segments based on membership status are defined as "existing," "dormant," or "new" based on when the member's purchase date occurred. In accordance with this definition, for example, the following segment conditions are set: "existing" if the purchase date recorded in the transaction data table is between 1 month and 6 months prior to ad delivery, "dormant" if it is between 6 months and 2 years prior, and "new" if there is no purchase in either period. By searching the transaction data table based on these segment conditions, each member can be assigned to one of the "existing," "dormant," or "new" segments. The membership status calculated for each member is recorded in the member data table, etc.

[0091] Instead of segmentation such as purchase loyalty or membership status, the operator may be allowed to specify segment conditions for arbitrary segmentation. When segment conditions are specified by the operator, the purchase analysis device 2 can classify the group of delivery candidates into arbitrary segments by searching for members who meet the specified segment conditions from the membership data or transaction data.

[0092] In the steps of setting the ad delivery group / control group (S3, S4), first, it is determined whether or not an instruction to set a control group has been given (S3). If the setting of the control group has been given, the process proceeds to step S4, and if not, step S6 is executed. In step S4, for example, 10 to 20% of the ad delivery target group is randomly extracted as the control group. The ratio of the control group to the ad delivery target group, the method of extracting the control group, such as whether or not to randomly extract, may be arbitrarily specified by the operator, for example.

[0093] In the ad delivery list generation steps (S5, S6), different ad delivery lists are generated depending on whether or not a control group setting is specified in step S3. If a control group setting is specified in step S3, the ad delivery list is generated by setting the member group obtained by excluding the control group from the ad delivery target group as the delivery group (S5). On the other hand, if a control group setting is not specified in step S3, the ad delivery list is generated by setting all members included in the ad delivery target group as the delivery group (S6).

[0094] In the advertisement distribution step (S7), advertisements are distributed based on the advertisement distribution list. Advertisements may be distributed online using telephone numbers, mobile advertisement IDs, email addresses, etc., or may be distributed offline, such as by flyers or direct mail. Advertisements are distributed during a preset advertising period or advertising time.

[0095] In the advertisement delivery step (S7), it is also possible to determine whether a member has come into contact with a delivered advertisement. For example, it is determined whether a member included in the delivery group viewed the delivered online advertisement on an application or browser, or clicked on a displayed advertisement as another operation, and a flag is set in the "advertisement contact" field in the advertisement delivery result table.

[0096] In the purchase fact data collection step (S8), the purchase analysis device 2 requests each transaction collection device 3 to submit transaction data for the target product during the analysis period. The transaction data includes purchase fact data indicating, for example, when, who (member ID), what, how much, how many items were purchased, whether points were used, etc. When the purchase analysis device 2 receives the transaction data from the transaction collection device 3, it records it in a transaction data table or the like.

[0097] In the matching step (S9), the contact group and non-contact group based on the advertisement contact information acquired in the advertisement delivery step (S7) are matched with the transaction data by member ID, the member to be analyzed is identified, and registered in the member ID field of the analysis tally table. In addition, for the analysis tally table in which the member IDs are set in the matching step (S9), the advertisement contact information acquired in the advertisement delivery step (S7), the segment information set in the step of extracting advertisement delivery target groups (S2), and the purchase data before and after advertisement delivery collected in the purchase data collection step (S8) are recorded in the analysis tally table for each member ID, and stored in the analysis data storage unit 214.

[0098] FIG. 12 is a diagram showing an example of the main part of the analysis aggregation table. The analysis aggregation table is a table that aggregates data used for purchase analysis, and includes member ID, purchase loyalty (segment), member status (segment), presence or absence of exposure to advertisement, purchase amount before advertisement, and purchase amount after advertisement. In "member ID", the member ID included in the ad delivery target group is recorded using the ad delivery list and the control group. In "purchase loyalty" and "member status", the result of segmentation performed in step S7 is recorded. In "advertisement exposure", the judgment result of the ad exposure recorded in the ad delivery result table (FIG. 5) is transcribed. In "purchase amount before advertisement" and "purchase amount after advertisement" of the analysis aggregation table, the purchase amount of the survey target product, service, etc. in the transaction table (FIG. 4) during a specified period before and after the ad delivery is aggregated and recorded. Note that FIG. 12 illustrates an analysis aggregation table including segments of purchase loyalty and member status, but it goes without saying that any other segment may be set and analysis may be performed.

[0099] In the analysis aggregation table of FIG. 12, it is recorded that a member with a member ID of 001 is an existing member in the heavy demographic who was exposed to the advertisement, and before the advertisement, he purchased the target product for 2,000 yen and a competing product for 600 yen, and after the advertisement was distributed, he purchased the target product for 3,000 yen and a competing product for 200 yen. Also, a member with a member ID of 002 is a dormant member in the light demographic who was exposed to the advertisement but did not purchase the target product. Also, a member with a member ID of 010 is an existing member in the heavy demographic who was not exposed to the advertisement, and before the advertisement, he purchased the target product for 1,000 yen, but after the advertisement, he purchased 0 yen.

[0100] In the purchase lift analysis step (S10), purchase lift is tabulated by contact / non-contact group and by advertising delivery target, and compared and analyzed. The purchase analysis program analyzes the purchasing behavior of members belonging to a specific advertising delivery target using an analysis tabulation table (FIG. 12) in which, for example, purchasing loyalty (heavy layer (H), middle layer (M), light layer (L)), membership status (existing, dormant, new), and purchasing fact data before and after advertising delivery are recorded for each member ID. Based on the "purchase amount before advertising" and "purchase amount after advertising" in the analysis tabulation table, the ratio of the number of users who purchased the target product among the specific advertising delivery target, that is, the purchase rate, can be calculated. In addition, the extent to which the purchase rate changed before and after advertising can be calculated as purchase lift.

[0101] Figure 13 shows an example in which the advertising delivery target group is divided by membership status into existing, dormant, and new, and the existing membership status is then further divided into high loyalty and low loyalty, with the purchase rate and purchase lift calculated for each classification.

[0102] In Figure 13, comparing the purchase rate of users classified as "existing (high loyalty)" who have been exposed to advertisements and those who have not been exposed to advertisements, the purchase rate of those who have been exposed to advertisements is 85.6%, while that of those who have not been exposed to advertisements is 84.6%, meaning that the purchase rate of those who have been exposed to advertisements is 1.01 times that of those who have not been exposed to advertisements (85.6 / 84.6). Similarly, the purchase rate of "existing (low loyalty)" is 1.15 times (56.8 / 49.4), that of "dormant" is 1.31 times (35.7 / 27.2), and that of "new" is 3.49 times (10.0 / 2.9). Focusing on "dormant" users, the purchase rate of those who have been exposed to advertisements is 35.7%, while that of those who have not been exposed to advertisements is 27.2%, meaning that the purchase rate of those who have been exposed to advertisements is 1.31 times that of those who have not been exposed to advertisements. Focusing on "new" users, 10.0% were exposed to advertisements and 2.9% were not, with the purchase rate of those exposed to advertisements being 3.49 times (10.0 / 2.9) higher than those not exposed to advertisements.

[0103] In the analysis results shown in Figure 13, the purchase rate of those exposed to advertisements was 85.6% for users classified as "existing (high loyalty)," which is the highest in absolute terms of purchase rate. However, looking at the purchase lift for those exposed to advertisements and those not exposed to advertisements, for users classified as "new," the purchase lift was 10.0% for those with ad delivery and 2.9% for those without ad delivery, meaning that the purchase lift for those exposed to advertisements was 3.49 times that of those without ad delivery. In this way, it is estimated that the effect of advertising on "new" users is the highest compared to other categories.

[0104] In FIG. 13, the purchasing lift of the target product was calculated by distributing the advertisement, but the purchasing lift that the distribution of the company's advertisement has given to the competing product may be evaluated by comparing the change in the purchasing rate of the competing product of others between those exposed to the advertisement and those not exposed to the advertisement. In addition to analyzing how the purchasing rate of the target product has changed before and after the advertisement distribution, it is also possible to analyze how the purchase amount or number of items purchased by members for the target product or competing products has changed. In this case, the purchasing lift for the purchase amount or number of items purchased can be calculated.

[0105] <Second embodiment> A purchase analysis support system in the second embodiment will be described. In the second embodiment, a configuration for performing reverse analysis is further adopted in addition to the purchase analysis device 2 in the first embodiment. Here, reverse analysis is a method for searching for a segment with better results by adopting a segment other than the initial segment in the subsequent purchase analysis while looking at the purchase analysis results by the segment that was the advertising delivery target group in the initial advertising delivery.

[0106] In the second embodiment, the system includes an advertising delivery target group extraction unit that extracts an advertising delivery target group from a specified advertising delivery candidate group based on specified segment conditions, and an analysis unit that analyzes changes in purchasing behavior before and after the provision of advertisements for an advertising exposure group and an advertising non-exposure group identified from the advertising delivery target group based on member data and / or transaction data, and is characterized in that the analysis unit is configured to set conditions different from the segment conditions set at the time of advertising delivery and to perform purchasing analysis again with new segment conditions.

[0107] FIG. 14 is a diagram for explaining the reverse analysis in the second embodiment, and is a diagram comparing the schematic flow of the purchase analysis method in the first embodiment with that in the second embodiment. FIG. 14(a) shows a schematic flow of the purchase analysis method in the first embodiment. In FIG. 14(a), an ad delivery candidate group is divided into segments, and an ad delivery target group that satisfies conditions A / B / C is extracted, and a purchase analysis is performed after ad delivery. For example, when an extraction condition is set such that the member status is existing (condition A), dormant (condition B), and new (condition C), an ad delivery target group (A / B / C) that satisfies this extraction condition is set. Then, after ad delivery to each of the ad delivery target groups (A / B / C), a purchase analysis is performed for the contact group and non-contact group of each of the ad delivery target groups (A / B / C). In this way, in FIG. 14(a), an ad delivery target group is identified according to the segment conditions set at the time of ad delivery, and then ad delivery and purchase analysis of the advertising effect are performed.

[0108] FIG. 14(b) shows a schematic flow of the purchase analysis support system in the second embodiment, and is configured to perform reverse analysis when a new segment is to be grasped according to the result of purchase analysis of the ad delivery target group extracted at the time of ad delivery, or when a superior segment other than the ad delivery target group is to be searched for. In the second embodiment, after performing analysis of the ad delivery target group set at the time of ad delivery in the analysis unit, conditions (segment conditions) for dividing the ad delivery target group into new segments for reverse analysis can be set. Therefore, as shown in FIG. 14(b), after analyzing the ad delivery target group extracted by the segment conditions A / B / C, purchase analysis can be performed again with new segment conditions X / Y. Note that, although it has been described that reverse analysis is performed after analyzing the ad delivery target group set at the time of ad delivery, it is not necessarily necessary to analyze the ad delivery target group set at the time of ad delivery first, and if a superior segment can be predicted in advance, reverse analysis may be performed first.

[0109] As shown in Figure 14(b), if the reverse analysis produces effective results for new segment conditions X / Y, it is possible to generate an ad delivery target group using these as segment conditions and deliver ads separately. In addition, as another reverse analysis, it is also possible to set new segment conditions X / Y and perform a purchasing analysis using the new segment conditions X / Y while reusing the ad delivery target group (A / B / C) extracted in the previous analysis. Note that the next purchasing analysis may be performed using conditions in which new conditions are added to the previous extraction conditions or segment conditions.

[0110] FIG. 15 is a flowchart for explaining an example of the operation of the purchase analysis device using reverse analysis according to the second embodiment. The flowchart in FIG. 15 is obtained by adding reverse analysis steps S11 and S12 to the flowchart (FIG. 11) of the purchase analysis device 2 according to the first embodiment. The purchase analysis program according to the second embodiment executes each of the steps of the ad delivery candidate group extraction step (S1), the ad delivery target group extraction step (S2), the ad delivery group / control group extraction step (S3, S4), the ad delivery list generation step (S5, S6), the ad delivery step (S7), the purchase data collection step (S8), and the purchase lift analysis step (S9, S10) in the first embodiment, and then executes steps (S11, S12) of setting a new segment by using reverse analysis. When performing reverse analysis, new segment conditions are obtained from the operator, and the process branches to the purchase data collection step S8 for reverse analysis as shown in branch 2. Since the segment conditions that have produced good results can be used as targets for the next ad delivery, the process branches to the step of extracting candidates for ad delivery (S1) via branch 1 (S12).

[0111] FIG. 16 is a diagram showing the purchase analysis results when reverse analysis is performed. Figure 16(a) shows the results of the first ad delivery and purchase analysis (let's call it Measure A), in which the ad delivery target group was extracted using the segment condition of members who use restaurant chains as the industry, and purchase analysis was performed on the exposed and non-exposed groups to the ads of Measure A. After that, as a reverse analysis, the ad delivery target group was set with new segment conditions of "purchases by day of the week and time of day at the supermarket" and "gender", and the purchase analysis was performed again. As a result of the reverse analysis, the purchase rate of "women who use supermarkets during the day on weekdays" was 0.36% for those who were not exposed to the ad, and 0.88% for those who were exposed to the ad, resulting in a purchase lift of 2.4 times. Meanwhile, the purchase rate of other segments other than "women who use supermarkets during the day on weekdays" was 0.30% for those who were not exposed to the ad, and 0.61% for those who were exposed to the ad, resulting in a purchase lift of 2.0 times. Therefore, it was found that the advertising effect was higher for "women who use supermarkets during the day on weekdays" among those who were exposed to Measure A for restaurant chains than for other segments.

[0112] Figure 16(b) shows the results of the first ad delivery and purchase analysis (let's call this Measure B), in which the ad delivery target group was extracted using the segment condition of members who use sports shops as the industry, and purchase analysis was performed on the exposed and non-exposed groups to the ads of Measure B. After that, as a reverse analysis, the ad delivery target group was set with the new segment conditions of "use of gas stations by day of the week" and "amount of fuel filled up", and the purchase analysis was performed again. As a result of the reverse analysis, the purchase rate of the segment "uses gas stations on holidays and fills up a lot of gas" was 0.47% for those not exposed to the ad and 0.57% for those exposed to the ad, resulting in a purchase lift of 1.2 times. On the other hand, the purchase rate of other segments other than "uses gas stations on holidays and fills up a lot of gas" was 0.32% for those not exposed to the ad and 0.31% for those exposed to the ad, resulting in a purchase lift of 0.97 times. Therefore, it was found that the advertising effect was higher for "uses gas stations on holidays and fills up a lot of gas" among those exposed to Measure B for members who use sports shops than for other segments.

[0113] Reverse analysis allows exploratory analysis of the segment with the highest purchase rate lift while looking at the results of multiple analyses, not limited to the segment conditions set before ad delivery. This is because a unified and continuous member ID is held, and member data and / or transaction data that serve as segment conditions are managed in association with it.

[0114] In the first and second embodiments, an advertisement management system including one purchase analysis device 2, two transaction collection devices 3, and one user-side advertisement display processing device 4 has been exemplified, but the number of these devices may be arbitrary. Also, while a configuration has been exemplified in which each of the purchase analysis device 2 and the transaction collection device 3 includes a storage unit and specific data is stored in the storage unit, the present invention is not limited to such a configuration and may be configured so long as required data can be accessed over a network.

[0115] In the first and second embodiments, an ad delivery target group is extracted from a member group managed by member data, and ad delivery and purchase analysis are performed, but this is not limited to member data. The ad delivery target group is a user group managed by user ID, and as long as the user ID is associated with purchase data, it may be any group of users, not limited to members.

[0116] The present disclosure is not limited to the above-described embodiments as they are, and in the implementation stage, the components can be modified and embodied without departing from the gist of the present disclosure. In addition, various inventions can be formed by appropriately combining multiple components disclosed in the above-described embodiments. For example, some components may be deleted from all the components shown in the embodiments. Furthermore, components across different embodiments may be appropriately combined. [Explanation of symbols]

[0117] 1...Purchase analysis support system, 2...Purchase analysis device, 21...Processor, 22...Memory, 23...Storage, 24...Communication I / F, 25...Input / output I / F, 3...Transaction collection device, 31...Management device, 32...Storage device, 4...User-side advertisement display processing device, 40...Bus, 41...Processor, 42...Memory, 43...Storage, 44...Camera, 45...Communication I / F, 46...Input / output I / F, 47...Touch screen, 48...Input unit, 49...Display unit, 5...Network, 200...Purchase analysis unit, 201...Advertisement delivery candidate group extraction unit, 202...Advertisement delivery target group extraction unit, 203...Delivery group / control group extraction unit, 204...Advertisement Distribution list generation unit, 205... advertisement distribution unit, 206... purchasing fact data collection unit, 207... analysis unit, 210... storage unit, 211... member data table storage unit, 212... transaction data storage unit, 213... advertisement distribution result table storage unit, 214... analysis data storage unit, 215... analysis result storage unit, 220... communication control unit, 300... transaction processing unit, 301... transaction management unit, 302... transaction providing unit, 310... transaction data storage unit, 320... communication control unit, 400... advertisement display processing unit, 401... advertisement receiving unit, 402... advertisement display control unit, 403... browsing history management unit, 420... communication control unit

Claims

1. an advertisement delivery target group extraction unit that divides members who are candidates for advertisement delivery from among a predetermined member group into predetermined segments and extracts an advertisement delivery target group; An information processing device comprising: an analysis unit that analyzes changes in purchasing behavior before and after the provision of advertisements for an advertisement exposure group and an advertisement non-exposure group identified from the advertisement delivery target group based on member data and / or transaction data.

2. The information processing device according to claim 1 , wherein the advertisement delivery target group extraction unit classifies a member group that is the subject of purchase analysis into the predetermined segments based on the member data and / or transaction data.

3. The information processing device according to claim 1 , wherein the analysis unit analyzes the change in purchasing behavior based on the advertisement exposure group, the advertisement non-exposure group, and the predetermined segment.

4. The information processing device according to claim 1 , wherein the analysis unit displays a purchase lift before and after the advertisement is provided.

5. The information processing apparatus according to claim 3 , wherein the predetermined segments include existing, dormant, and new categories.

6. The information processing device according to claim 3 , wherein the predetermined segments include a division based on purchase royalties.

7. The information processing device according to claim 1 , wherein, in a case where a control group to which no advertisement is delivered is set to the advertisement delivery target group, the control group is randomly extracted from the advertisement delivery target group.

8. Dividing a predetermined group of members into predetermined segments and extracting an advertising delivery target group; Analyzing changes in purchasing behavior before and after the provision of advertisements based on member data and / or transaction data for an advertisement exposure group and an advertisement non-exposure group identified from the advertisement delivery target group; An information processing method comprising:

9. A computer program causing a computer to execute each of the parts of claims 1 to 7.

Images