JP7203788B2 - Information processing program, information processing apparatus, and information processing method - Google Patents

Description

The present disclosure relates to an information processing program, an information processing apparatus, and an information processing method.

Techniques for generating a cost-conversion graph and displaying the generated graph on display means are known in order to make people recognize appropriate costs according to the results of advertising.

JP 2016-045542 A

However, as described above, in the prior art, the relationship between the result of the advertisement and the cost required for the advertisement is determined based on the advertisement as a whole, so it is difficult to realize efficient advertisement.

Accordingly, in one aspect, an object of the present invention is to appropriately process various information related to advertisements so as to realize efficient advertisements.

In one aspect, an information processing program that causes a computer to execute a predetermined process,
The predetermined process is
Get information about your advertising-attributable performance;
Acquiring information about the cost required for the advertisement;
An information processing program is provided that includes a process of associating a first index value representing the relationship between the result and the cost with each piece of user information.

In one aspect, according to the present invention, it is possible to appropriately process various information related to advertisements so as to implement efficient advertisements.

1 is a block diagram of a system according to this embodiment; FIG. It is a functional block diagram about an advertisement-related function. FIG. 4 is an explanatory diagram of basic data in user information data in a user information storage unit; FIG. 4 is an explanatory diagram of data of information related to an application among user information data in a user information storage unit; FIG. 4 is an explanatory diagram of data in an advertisement-related data storage unit; FIG. 10 is an explanatory diagram of an example of installation number data; FIG. 10 is an explanatory diagram showing an example of a method of calculating an advertisement cost for each advertisement type; FIG. 4 is an explanatory diagram of a method of calculating an index value IDX1; FIG. 4 is an explanatory diagram of a method of calculating an index value IDX2; FIG. 4 is an explanatory diagram of data in an index value data storage unit; FIG. 11 is an explanatory diagram of another example of a method of calculating an index value IDX1; FIG. 10 is a functional block diagram relating to advertisement-related functions among various functions realized by the server device according to the second embodiment; FIG. 11 is an explanatory diagram showing an example of a method of calculating an index value IDX1 by an index value calculation unit according to the second embodiment; FIG. 11 is an explanatory diagram showing another example of a method of calculating the index value IDX1 by the index value calculator according to the second embodiment; 13 is a schematic flow chart showing another example of a method of calculating a real ad placement cost and a method of calculating an index value IDX1 according to the second embodiment; FIG. 4 is an explanatory diagram of an index value IDX1 calculated for each user and for each attribute of primary results; FIG. 11 is a functional block diagram relating to advertisement-related functions among various functions realized by the server device according to Embodiment 3; It is a schematic flowchart which shows the process of the advertisement target user extraction part by a modification. It is a schematic flowchart which shows the process of the advertisement target user extraction part by another modification.

Hereinafter, each embodiment will be described in detail with reference to the accompanying drawings.

(system overview)
An overview of a system 1 according to the present embodiment will be described with reference to FIG. FIG. 1 is a block diagram of a system 1 according to this embodiment. The system 1 includes a server device 10 and one or more terminal devices 20 . Although three terminal devices 20 are illustrated in FIG. 1 for the sake of simplicity, the number of terminal devices 20 may be two or more.

The server device 10 is an information processing device such as a server managed by a service operator, for example. In this embodiment, as an example, the server device 10 implements a so-called service platform and provides various services via the service platform. Specifically, as an example, the server device 10 serves as a service that provides digital content such as an application portal site (hereinafter also simply referred to as “site”) and application software (hereinafter also simply referred to as “application”). , and social network functions (hereinafter also simply referred to as “SNS functions”) to registered users (and guest users as necessary).

The number of sites provided by the server device 10 is arbitrary, but in this embodiment, as an example, it is assumed that there is one site (hereinafter also referred to as "site A"). Note that in a variant, site A may include multiple sites, including related sites.

The type and number of digital contents provided by the server device 10 are arbitrary, but in this embodiment, as an example, the digital content provided by the server device 10 is one or more game applications (hereinafter simply " (also referred to as "game application"), an application for providing information, an electronic book, and the like. A game application or the like may be made executable via site A, or may be made executable without going through a predetermined site such as site A (for example, may be made executable offline). ). That is, various applications as digital content provided by the server device 10 may be installed in the terminal device 20 and used as so-called native applications. Hereinafter, a game refers to a game related to a game application provided by the server device 10 . Applications for providing information are related to information providing services of specific genres (information providing services related to travel, housing, food, fashion, health, beauty, etc.), broadcasting services by specific users (for example, YouTube (registered trademark)), and the like. can be anything. Also, the specific application provided by the server device 10 may be a combination of a plurality of applications (for example, a hierarchical application).

The type and number of SNS functions provided by the server device 10 are arbitrary, but here, diary, chat, message board/bulletin board (community), comment, messenger, friend request, object transmission (gift transmission), etc. It may be a function such as SNS functionality may be enabled via Site A.

The terminal device 20 is an information processing device used by a user, such as a mobile phone, a smart phone, a tablet terminal, a PC (Personal Computer), or a game device. The terminal device 20 can execute various applications according to this embodiment. Various applications may be received by the terminal device 20 from the server device 10 or another predetermined application distribution server via the network 3, or stored in a storage device provided in the terminal device 20 or a memory readable by the terminal device 20. It may be stored in advance in a storage medium such as a card. The server device 10 and the terminal device 20 are communicably connected via the network 3 . For example, the server device 10 and the terminal device 20 cooperate to execute various processes related to services.

Note that the network 3 may include a wireless communication network, the Internet, a VPN (Virtual Private Network), a WAN (Wide Area Network), a wired network, or any combination thereof.

(Configuration of server device)
The configuration of the server device 10 will be specifically described. The server device 10 is configured by a server computer. The server device 10 may be implemented in cooperation with a plurality of server computers.

The server device 10 includes a server communication section 11 , a server storage section 12 and a server control section 13 .

The server communication unit 11 includes an interface that performs wireless or wired communication with an external device to transmit and receive information. The server communication unit 11 may include, for example, a wireless LAN (Local Area Network) communication module or a wired LAN communication module. The server communication unit 11 can transmit and receive information to and from the terminal device 20 via the network 3 .

The server storage unit 12 includes, for example, a primary storage device and a secondary storage device. For example, the server storage unit 12 may include semiconductor memory, magnetic memory, optical memory, or the like. The server storage unit 12 stores various information and programs used for service-related processing. At least part of the information and programs stored in the server storage unit 12 may be shared and synchronized with the terminal device 20 .

Server control unit 13 includes one or more processors. The processor may include a general-purpose processor that implements a specific function by reading a specific program, and a dedicated processor that specializes in specific processing. The server control unit 13 controls the operation of the server device 10 as a whole. Details of the server control unit 13 will be described later.

(Configuration of terminal device)
The configuration of the terminal device 20 will be specifically described. As shown in FIG. 1 , the terminal device 20 includes a terminal communication section 21 , a terminal storage section 22 , a display section 23 , an input section 24 and a terminal control section 25 .

The terminal communication unit 21 includes an interface that communicates with an external device wirelessly or by wire to transmit and receive information. The terminal communication unit 21 may include a wireless communication module, a wireless LAN communication module, a wired LAN communication module, or the like compatible with mobile communication standards such as LTE (Long Term Evolution) (registered trademark) and 5G (5th Generation). good. The terminal communication unit 21 can transmit and receive information to and from the server device 10 via the network 3 .

The terminal storage unit 22 includes, for example, a primary storage device and a secondary storage device. For example, the terminal storage unit 22 may include semiconductor memory, magnetic memory, optical memory, or the like. The terminal storage unit 22 stores various information and programs received from the server device 10 and used for service-related processing. Information and programs used for service-related processing may be acquired from an external device via the terminal communication unit 21 . For example, a game application may be acquired from a predetermined application distribution server.

The display unit 23 includes a display device such as a liquid crystal display or an organic EL (Electro-Luminescence) display. The display unit 23 can display various screens.

The input unit 24 includes an input interface including a touch panel provided integrally with the display unit 23, for example. The input unit 24 can accept user input to the terminal device 20 . Also, the input unit 24 may include physical keys, and may further include any input interface including a pointing device such as a mouse.

Terminal control unit 25 includes one or more processors. The terminal control unit 25 controls the operation of the terminal device 20 as a whole.

The terminal control section 25 transmits and receives information via the terminal communication section 21 . For example, the terminal control unit 25 receives various information and programs used for service-related processing from at least one of the server device 10 and other external servers. The terminal control section 25 stores the received information and program in the terminal storage section 22 .

The terminal control unit 25 activates the site A and various applications (for example, a native application on the terminal device 20, a game application in the site A, etc.) according to the user's operation. The terminal control unit 25 cooperates with the server device 10 to execute processing related to services. For example, the terminal control unit 25 causes the display unit 23 to display various screens related to service-related processing. A GUI (Graphic User Interface) that detects user operations, for example, may be displayed on the screen. The terminal control unit 25 can detect user operations on the screen via the input unit 24 .

Also, the terminal control unit 25 causes the display unit 23 to display an advertisement based on the advertisement information provided via the advertisement media server 16 or the like. Advertisements may be displayed in any manner, but in the present embodiment, clickable advertisements are displayed, as described below. For example, the user can change the screen on the display unit 23 to a new site related to the advertisement, an advertisement screen, or the like by clicking (touching, etc.) the advertisement.

(advertisement-related functions)
The server control unit 13 of the server device 10 further has a function of processing various information related to advertisements (hereinafter also referred to as "advertisement-related functions") so as to realize efficient advertisements. The advertisement is arbitrary, but is, for example, an advertisement for the user to use the site A or various services available at the site A. In the following, for convenience of understanding, the advertisement relates to one application (hereinafter referred to as "target application 7") within the site A unless otherwise specified.

The advertisement may be any advertisement for promoting usage of the target application 7 . In this embodiment, unless otherwise specified, the advertisement is, for example, intended to get the user who sees the advertisement to install the target application 7. shall be an advertisement for inducing (promoting) Note that other advertisements can be advertisements or the like that target other uses (for example, secondary results) of the target application 7 described later.

The target application 7 is optional, but in this embodiment is of the type with intra-application billing. Note that the target application 7 may be installed for a fee or may be installed free of charge.

The output form of the advertisement is arbitrary. For example, the advertisement may be output at site A, may be output at another site, etc., may be output as a paper medium by mail, etc., or may be output as an outdoor advertisement (OHH). Alternatively, it may be output using radio waves such as television or radio. In this embodiment, an online type advertisement will be mainly described as an example. The online type advertisement is output (displayed) on the terminal device 20 via the advertisement media server 16 (see FIG. 1). In this case, the advertisement may be output in a clickable manner, and when clicked, a screen transition to a screen related to the target application 7 (for example, an installation screen of the target application 7, etc.) may be realized. Note that the screen transition may involve movement between different sites. In addition, in the case of outdoor advertisements, advertisements may be output via paper media, digital signage, or the like, or may be output via goods or the like. In the case of advertisements using radio waves such as television and radio, the advertisements may be output in the form of television commercials.

When the advertisement placement request is received from the server device 10, the advertisement media server 16 creates an advertisement and outputs the created advertisement to the target user via the advertisement medium. Note that the advertising medium server 16 may be provided for each advertising medium. The advertising media can be various sites on the Internet, such as sites provided by Google (registered trademark), Facebook (registered trademark), Twitter (registered trademark), and the like. The target user may be instructed by the server device 10 together with the advertisement placement request. The target user may be identified by, for example, an advertisement ID associated with a user ID described later. In this case, the server device 10 extracts the target users and notifies the advertisement media server 16 of the advertisement ID for each target user.

Next, advertisement-related functions of the server device 10 will be further described by dividing into several embodiments.

[Embodiment 1]
FIG. 2 is a functional block diagram relating to advertisement-related functions among various functions realized by the server device 10 according to the first embodiment. FIG. 3A is an explanatory diagram of basic data in the user information data in the user information storage unit 30. FIG. FIG. 3B is an explanatory diagram of data of information related to applications among the user information data in the user information storage unit 30. As shown in FIG. FIG. 4A is an explanatory diagram of data in the advertisement-related data storage unit 32. FIG. FIG. 4B is an explanatory diagram of an example of installation number data. FIG. 5 is an explanatory diagram of a method of calculating the index value IDX1. FIG. 6 is an explanatory diagram of a method of calculating the index value IDX2. FIG. 7 is an explanatory diagram of data in the index value data storage unit 40. As shown in FIG. In FIGS. 3A, 4A, etc., "***" means that some information is stored, and "..." means repetition of similar information.

As shown in FIG. 2, the server device 10 includes a user information storage unit 30, an advertisement placement request generation unit 31, an advertisement-related data storage unit 32, a primary result information acquisition unit 33, a cost information acquisition unit 34, It includes a first index value calculation unit 36 , a secondary result information acquisition unit 38 , an index value data storage unit 40 , a second index value calculation unit 42 , and an advertisement target user extraction unit 44 . In this case, storage units such as the user information storage unit 30 and the advertisement-related data storage unit 32 can be realized by the server storage unit 12 shown in FIG. The processing unit can be implemented by the server control unit 13 and can communicate with the outside via the server communication unit 11 .

The user information storage unit 30 stores user information related to users who use the site A. FIG. User information includes, as basic information, a user ID (an example of user identification information) and an advertisement ID, as shown in FIG. Includes coins, information on SNS, preference information, and the like. User information may also include gender, age, address, IP address, and the like. User information may include, for example, initially registered information such as a user ID and user name, as well as post-accumulated information such as game-related information. User information is typically generated for each user.

A user ID is information that can uniquely identify a user. A user ID may be in the form of a user account. Hereinafter, the user ID will be simply referred to as a user.

The advertisement ID corresponds to the user ID on a one-to-one basis and is set for the advertisement medium. In addition, when there are a plurality of advertisement media servers 16, a plurality of advertisement IDs may be set for each user.

A user name is information indicating a user's name. A user name, unlike a user ID, may not uniquely identify a user. The user name may be arbitrarily determined and changed according to user operations on the terminal device 20 .

The password is a password that enables login to site A using the user name. A plurality of types of passwords may be set according to usage.

The phone number is, for example, a phone number for short message service (SMS) authentication.

Possessed coins represent the number of coins owned by the user (the number of coins currently available). Further, in the present embodiment, as an example, it is assumed that the coin is a coin that can be used at site A (hereinafter referred to as "A coin"). You can receive various compensation within the application. For example, in the case of a game application, the user can obtain desired game media, obtain desired items, and the like by consuming the A coins possessed by the user. In addition, by consuming the A coin possessed, for example, a gacha can be played. Note that, in a modified example, the A coin may be usable only within the target application 7 .

The information about SNS may include various types of information related to SNS functions. For example, the SNS-related information may include avatar-related information, friend information, and the like for each user ID. The information about the avatar may include clothing information for changing clothes, and the like. The friend information may represent user IDs and group IDs connected via the SNS function. The friend information may or may not be the same as the friend information within a particular application.

The preference information includes information representing user's preference or information from which user's preference can be calculated. The information representing the user's preferences includes information such as what services and privileges the user prefers, what services and privileges the user does not like, what services and privileges the user is interested in, and so on. It may be information representing what kind of services or benefits the user is not interested in, and the like. Preference information may be generated, for example, based on a simple questionnaire survey at the time of initial registration. Also, the information from which user preferences can be calculated may be user log information or the like. In this case, the log information to be accumulated may be all log information that can be related to the user's preferences, or may be specific log information (for example, log information related to operation logs). The log information represents various activities (login, logout, use of digital content, access/use of SNS functions, billing, etc.) at site A based on each user ID. Also, log information to be accumulated may include log information related to game play.

The user information may further include, as information on applications, information on various applications related to site A for each user ID, as shown in FIG. 3B. Specifically, the information about the application includes information about the installed application, the usage status, the number of A coins used, the timing of installation, other detailed information, and the like.

The information about the installed application is information specifying the application installed for each user ID. For example, in the example shown in FIG. 3B, a user whose user ID is "U01" has installed applications A0, A1, A2, and the like. Information about the target application 7 is associated with the user who has installed the target application 7 . For example, application A0 may correspond to target application 7 .

The usage state is the usage state of each application. For example, the usage state represents usage frequency, and in FIG. 3B, it is classified into dormant state, high frequency, normal frequency, and the like. Note that the usage status may be classified, for example, based on the log information within the most recent fixed period. The usage state may be used when determining the user to whom the advertisement for each application is to be output (see the advertisement target user extraction unit 44 described later).

The number of A coins used represents the number (number of sheets) of A coins used in each application. The number of A coins used may be used to calculate an index value IDX2, which will be described later.

The installation trigger may include "advertisement-induced" indicating that the installation was caused by an advertisement, and "natural inflow" indicating that the user installed without being caused by an advertisement (natural inflow user).

Detailed information is optional, but may include, for example, various application-specific information. For example, if the application A0 is a game app, the detailed information may include information about the game. The information about the game may include rank, information about owned game media, and the like.

The advertisement placement request generating unit 31 generates an advertisement placement request based on an input from, for example, an administrator (for example, an advertiser) of the server device 10 and transmits the advertisement placement request to the advertising media server 16 . The advertisement placement request generated and transmitted by the advertisement placement request generation unit 31 includes the advertisement placement request related to the advertisement of the target application 7 described above. The advertisement placement request includes an advertisement ID for which an advertisement is to be output. The advertisement placement request generation unit 31 generates an advertisement placement request including the advertisement ID based on the advertisement ID specified by the advertisement target user extraction unit 44, which will be described later. For other advertisements, the advertisement placement request generation unit 31 may generate an advertisement placement request so that the advertisement is output for the advertisement IDs of all users, if necessary. In addition, when the advertising medium server 16 (see FIG. 1) receives such an advertisement placement request, based on a plurality of advertisement IDs included in the advertisement placement request, the advertisement is linked to each advertisement ID, and the corresponding advertisement medium Output the advertisement above. The advertisement output timing, period, number of times of output, etc. are appropriately determined by the advertisement media server 16 and are optional. Also, the output timing, period, number of times of output, etc. of the advertisement may be specified in the advertisement placement request.

The advertisement-related data storage unit 32 stores advertisement-related data related to various advertisements output in response to the above-described advertisement placement request. In the advertisement-related data, as shown in FIG. 4A, for each advertisement placement ID assigned to each advertisement placement request generated by the advertisement placement request generation unit 31, the placement cost, the advertisement ID included in the advertisement placement request, the number of clicks or imp (impression), the number of installs, etc. are associated with each other. The install number data in the advertisement-related data storage unit 32 is based on the primary result information acquired by the primary result information acquisition unit 33 .

The number of installs data in the advertisement-related data storage unit 32 is associated with the number of installs for each advertisement placement ID. In this case, the number of installs to be associated is preferably segmented by period after the advertisement was posted (or output). Alternatively, the number of associated installs may be segmented by time period since the advertisement was clicked for each install. In the example shown in FIG. 4B, as the period after clicking, "1st day (first day)", "2nd day", "3rd day", etc. are set. is associated with the number of installs obtained within the corresponding period. In addition, instead of the period after clicking, installs obtained within the corresponding period for each of multiple periods including the advertisement output period and the period after the advertisement output period (multiple periods) A number may be associated. Hereinafter, the install count data in which the install count is associated with each of a plurality of periods in this manner is also referred to as "period management type install count data". Note that such period management type installation number data may be associated with each corresponding advertisement placement ID as shown in FIG. 4B.

The primary result information acquisition unit 33 acquires information (hereinafter also referred to as "primary result information") related to results (hereinafter also referred to as "primary results" for distinction) (an example of the first result) resulting from advertisements. get. In this embodiment, the advertisement is associated with the installation of the target application 7 as described above, and thus the primary outcome attributed to the advertisement is associated with the installation of the target application 7 . Primary performance depends on the purpose of the advertisement and may have multiple attributes.

In this embodiment, as an example, there are two types of attributes of the primary result, the first attribute being the installation of the target application 7, and the second attribute being the This is resumption of use of the target application 7 (hereinafter also simply referred to as “return”).

The predetermined period ΔT1 is arbitrary as long as it is a relatively long period representing a so-called dormant state, and may be on the order of months, for example. Note that the return may be realized while the target application 7 is still installed in the terminal device 20 . In this case, returning often involves downloading (updating) an updated version of the target application 7 . Alternatively, returning may involve reinstalling the target application 7 on the terminal device 20 . Note that some users may return multiple times. Further, the use related to the return may be the use accompanied by login, or the play of the game when the target application 7 is a game application. Alternatively, the use related to the return may be a use that generates a secondary result, which will be described later.

In this case, the primary result information indicates users who have contributed to the primary result for each attribute. Specifically, the primary result information may include a user ID that contributed to the primary result of the first attribute and a user ID that contributed to the primary result of the second attribute. In addition, the primary result information may include information on the time contributed to the primary result of the first attribute and/or the second attribute (for example, the date of installation in the case of the first attribute) for each user ID.

The primary result information acquisition unit 33 preferably acquires the primary result information in a manner that allows the primary result information to be associated with each advertisement. In this case, the unit of advertisement may be, for example, each advertisement placement request. Thereby, it is possible to grasp which primary result was obtained due to the advertisement related to which advertisement placement request. In other words, it is possible to reduce the possibility that the information related to the spontaneous user (that is, the user who installs the target application 7 without being attributed to the advertisement) is included in the primary result information.

The cost information acquisition unit 34 acquires information about the cost required for advertising (hereinafter also referred to as "cost information"). The cost information is preferably obtained in a manner associated with each advertisement. Note that the unit of advertisement may be the same as in the case of the primary result information described above, and may be for each advertisement placement request, for example. This will give you an idea of how much each ad costs. The cost required for the advertisement may be the cost paid by the advertiser to the outside (expense based on the contract), and furthermore, the personnel cost etc. required by the advertiser to place the order for the advertisement may be added. Alternatively, advertising costs for forms that do not use the advertising media server 16, such as television commercials and outdoor advertisements, may also be added. In this embodiment, as an example, it is assumed that the cost required for one advertisement is the advertising expense related to the one advertisement.

The placement cost for one advertisement may be set appropriately as described above, but may be set in the manner shown in FIG. 4C, for example. In the example shown in FIG. 4C, the advertisement placement ID is also assigned separately depending on the type of advertisement, and the placement cost is calculated in a different manner according to the type of advertisement. Specifically, if the type of advertisement is online ("YES" in step S1402), the advertisement cost is calculated based on the cost to be paid to the advertising media server 16 (step S1404). On the other hand, if the type of advertisement is an offline type ("NO" in step S1402) and is an outdoor advertisement ("YES" in step S1406), the advertising cost is calculated for each region (for example, by prefecture or municipality). Divide and allocate (step S1408). If the type of advertisement is an offline type ("NO" in step S1402) and is a television commercial ("YES" in step S1410), the advertising cost is divided and allocated for each broadcasting station (step S1412). . In addition, in the case of a nationally broadcast program, the advertisement cost may be divided equally among the regions. If it is another type of advertisement ("NO" in step S1410), an appropriate advertising fee is allocated according to the type. Although three representative types are described in FIG. 4C, there may be two types, or four or more types. In the following, unless otherwise specified, the advertising expenses may be the advertising expenses related to online advertisements or the advertising expenses related to offline advertisements. However, there may be a case where an advertisement placement ID to which an advertisement placement cost is allocated for each region is treated differently from an advertisement placement ID that is not otherwise treated, and such a case will be explained as appropriate.

The first index value calculation unit 36 expresses the relationship between the primary result and the cost based on the primary result information acquired by the primary result information acquisition unit 33 and the cost information acquired by the cost information acquisition unit 34. An index value IDX1 (an example of a first index value) is calculated for each user ID and associated with each user ID. Specifically, the first index value calculator 36 stores the calculated index value IDX1 in the index value data storage unit 40 in association with each user ID (see FIG. 7). The index value IDX1 is an index value widely known in the field of advertising (for example, CPI: Cost Per Install, CPA: Cost Per Acquisition, CTR: Click Through Rate) or the like (for example, CPA: Cost Per Action, CPC: Cost Per Click, CPM: Cost Per Mille).

For example, when the index value IDX1 is CPI, the primary result may be the installation of the target application 7, and the cost may be the advertising cost required to obtain the primary result. In this case, the user ID associated with the index value IDX1 includes the user ID of the user who installed the target application 7 . For example, when a user having a certain user ID installs the target application 7, the index value IDX1 is associated with the one user ID. At this time, the index value IDX1 associated with the one user ID is based on the cost (posting cost) required for the advertisement posted in association with a plurality of advertisement IDs including the advertisement ID associated with the one user ID. , is calculated. For example, if the cost is C1 and the number of installed user IDs is M, the index value IDX1 associated with the one user ID may be calculated as IDX1=C1/M.

In addition, in the case of an advertisement ID for which an advertising fee is allocated for each region, if a user with a certain user ID installs the target application 7, the advertising fee related to the region associated with the one user may be used as the cost C1. Note that the area associated with one user may be determined based on user information. For example, an area (for example, prefecture) associated with one user may be identified based on an address or IP address.

Further, for example, when the index value IDX1 is CPI, the primary result is a click on a specific advertisement (access to a site or landing page where the target application 7 can be installed), and the cost is It may be the cost associated with a particular advertisement spent on In this case, the user ID associated with the index value IDX1 includes the user ID of the user who clicked on the specific advertisement. For example, when a user with a certain user ID clicks a specific advertisement, the index value IDX1 is associated with the user ID. At this time, the index value IDX1 associated with the one user ID is calculated based on the cost required for a specific advertisement posted in association with a plurality of advertisement IDs including the advertisement ID associated with the one user ID. be. For example, if the cost is C2 and the number of user IDs that have clicked on a specific advertisement is M2, the index value IDX1 associated with the one user ID may be calculated as IDX1=C2/M2. .

In this way, when the primary results can be expressed numerically, the number of primary results is Mi, and the advertising cost required to obtain the primary results is Ci. Each contributing user ID may be associated with an index value IDX1=Ci/Mi. Another calculation method for the index value IDX1 when the primary result can be represented by a number will be described later.

Also, the index value IDX1 can be similarly calculated when the primary result cannot be expressed numerically, but can be expressed quantitatively. For example, when Li is the amount of the primary result and Ci is the advertising cost required to obtain the primary result, each user ID that contributed to the primary result has an index value IDX1=Ci/Li may be associated.

Here, a more specific example of the method for calculating the index value IDX1 will be described with reference to FIG. FIG. 5 exemplifies various numerical values for two representative user IDs. Here, a method of calculating the index value IDX1 associated with each user ID will be described with reference to FIG. 5 with respect to two user IDs as representatives.

In the example shown in FIG. 5, the user with the user ID "U01" has installed the target application 7 through the advertisement, and the cost required for the advertisement is 1000 yen. Note that no cost is incurred after the target application 7 is installed. In this case, the index value IDX1 associated with the user ID "U01" may be 1000 yen.

Further, in the example shown in FIG. 5, for the user with the user ID "U02", costs are incurred for each attribute of the primary result. Specifically, first, the user with the user ID "U02" has installed the target application 7 through an advertisement, and the cost of the advertisement is 1000 yen. The user with user ID "U02" has left after installing the target application 7, and has returned twice through advertisements. In this case, the advertising cost for the first return is 200 yen, and the advertising cost for the second return is 150 yen.

Note that, in the example shown in FIG. 5, the costs of 1000 yen and 200 yen are based on the cost information acquired by the cost information acquisition unit 34 . Information such as that the user with the user ID “U01” or the user ID “U02” has installed the target application 7 through an advertisement is based on the primary result information acquired by the primary result information acquisition unit 33 .

In such a case, the first index value calculator 36 calculates the index value IDX1 for the user with the user ID "U01" based on the cost of 1000 yen. For example, the first index value calculator 36 may set the index value IDX1=1000. On the other hand, for the user with the user ID "U02", the first index value calculation unit 36 calculates the index value IDX1 based on the installation cost of 1000 yen and the recovery cost of 200 yen and 150 yen. calculate. For example, the first index value calculator 36 may set the index value IDX1=1350 (=1000+200+150). That is, in the present embodiment, the first index value calculation unit 36 adds up all the costs for each attribute of these primary results to calculate the index value IDX1. However, in the modified example, for the user with the user ID “U02”, the first index value calculation unit 36 calculates a combination of two or more of these costs (for example, a combination of installation and most recent return). Based on this, the index value IDX1 may be calculated.

In this way, the method of calculating the index value IDX1 as shown in FIG. Including calculating. The calculation method includes calculating the index value IDX1 for each user ID based on the total value for each user ID (total cost value for each attribute). In this way, according to the present embodiment, the index value IDX1 is obtained for each user ID, so not only the advertising effectiveness of the advertisement as a whole but also the advertising effectiveness of each user can be evaluated.

The secondary result information acquisition unit 38 obtains information (hereinafter referred to as " (also referred to as "secondary result information"). In this embodiment, the primary result is related to the installation of the target application 7 as described above, so the secondary result is the result after the installation of the target application 7 . A secondary result may be any result that occurs in connection with the use of the target application 7 . This type of secondary result information is information obtained for each user.

In this embodiment, as an example, the secondary result is the number of A coins used in relation to the target application 7 (ie, the number of A coins used). The number of A coins used may be the number of coins accumulated after installation, or may be the number of coins used within a predetermined period (for example, the most recent predetermined period). Also, the number of A coins to be used may include the number of coins to be used that are purchased for the target application 7 . In the modified example, equivalently, instead of the number of used A coins, the billing amount converted into the number of used A coins may be used. The secondary results are the advertising income that can be generated by each user's use of the target application 7, the degree of use of the target application 7 (continuation rate, consent to the terms of use of the target application 7, and the tutorial related to the target application 7). completed or not).

The second index value calculator 42 calculates an index value IDX2 representing the relationship between the index value IDX1 and the secondary result for each user ID, and associates it with each user ID. Specifically, the second index value calculator 42 stores the calculated index value IDX2 in the index value data storage unit 40 in association with each user ID (see FIG. 7). The index value IDX2 may be an index value widely known in the field of advertising (eg ROAS: Return On Advertising Spend, ROI: Return On Investment) or the like.

Here, an example of a method of calculating the index value IDX2 will be described with reference to FIG. FIG. 6 exemplifies various numerical values for two representative user IDs. Here, a method of calculating the index value IDX2 associated with each user ID will be described with reference to FIG. 6 with respect to two user IDs as representatives.

The example shown in FIG. 6 is a continuation of the example shown in the previous FIG. is 1350 as described above. Further, the user with the user ID "U01" uses 10000 A coins in the target application 7 after installing the target application 7, and the user with the user ID "U02" uses the target application 7 after installing the target application 7. Assume that application 7 similarly uses 10,000 A coins.

In the example shown in FIG. 6, the number of A coins used, such as 10000, is based on the secondary result information acquired by the secondary result information acquisition unit 38 .

In such a case, the second index value calculation unit 42 calculates the index value IDX2 for the user with the user ID “U01” by dividing the number of A coins used, 10000, by the index value IDX1 “1000”. In this case, the index value IDX2=10. Similarly, the second index value calculator 42 calculates the index value IDX2 for the user with the user ID “U02” by dividing the number of A coins used, 10000, by the index value IDX1 “1350”. In this case, the index value IDX2=7.41.

As described above, the method of calculating the index value IDX2 as shown in FIG. 6 calculates the index value IDX2 for each user ID based on the relationship between the index value IDX1 associated with each user ID and the number of A coins used. Including calculating. Therefore, since the index value IDX2 is obtained for each user ID, it is possible to evaluate not only the advertising effectiveness of the advertisement as a whole but also the advertising effectiveness of each user.

The index value data storage unit 40 stores the above-described various index values IDX1 and IDX2 in association with user IDs. FIG. 7 shows the index value data related to the target application 7, but when advertising for another application, the same index value IDX1 and index value IDX2 are calculated for the other application as well. Value data may be stored in the index value data storage unit 40 .

The index value data in the index value data storage unit 40 may be updated regularly and/or irregularly. For example, when there is a user whose primary result information has changed (for example, when there is a user who has returned or a new user who has installed the target application 7), the index value IDX1 related to the user and An index value IDX2 may be generated/updated. Also, when there is a user whose secondary result information described above has changed (for example, when there is a user whose usage number of A coins has increased), the index value IDX2 related to the user may be updated. Alternatively, the index value data in the index value data storage unit 40 may be periodically updated at the beginning of the month, or when determining the advertisement ID for outputting the advertisement of another application different from the target application 7. may be updated to

The advertisement target user extraction unit 44 extracts one or more advertisement IDs to be included in the advertisement placement request generated by the advertisement placement request generation unit 31 from among the advertisement IDs related to all users associated with the index value IDX2. do. A method for extracting one or more advertisement IDs may be defined based on the index value IDX2. For example, when the number of advertisement IDs to be included in the advertisement placement request generated by the advertisement placement request generation unit 31 is a fixed number N, the advertisement target user extraction unit 44 performs From among the advertisement IDs, N advertisement IDs are extracted (determined) in descending order of index value IDX2.

Note that the advertisement target user extraction unit 44 may narrow down the advertisement IDs to be extracted for each advertisement according to the purpose of the advertisement. For example, if the purpose of the advertisement is to promote a return to the target application 7, the advertisement target user extraction unit 44 first selects, from among the advertisement IDs related to all the users associated with the index value IDX2, Extract advertisement IDs related to users in the state (users who have left for a predetermined period of time ΔT1 or more), and then extract N advertisement IDs from the extracted advertisement IDs in descending order of index value IDX2. good too. Further, when the purpose of the advertisement is to increase the secondary results related to the target application 7, the advertisement target user extraction unit 44 selects, from among the advertisement IDs related to all the users associated with the index value IDX2, Advertisement IDs associated with users logged in within a predetermined period ΔT1 may be extracted, and then N advertisement IDs may be extracted from the extracted advertisement IDs in descending order of the index value IDX2.

In this manner, according to the present embodiment, when an advertisement is output in association with an advertisement ID, which advertisement ID (that is, the user targeted by the advertisement) to be associated with is determined based on the index value IDX2. can. Specifically, by linking the user IDs (advertisement IDs) associated with the high index value IDX2 to the advertisements to be posted in order, it is possible to output advertisements limited to users who are highly likely to have high advertising effectiveness. . As a result, efficient advertisement can be realized. In this manner, according to the present embodiment, it is possible to appropriately process various types of information related to advertisements so as to implement efficient advertisements.

By the way, since the index value IDX2 is used to extract the target user of the next advertisement as described above, it is desirable that the level of the advertising effect is linked with the level of the value with high accuracy. That is, the index value IDX2 is more useful as the correlation (interlocking) with the degree of advertising effectiveness is higher. For example, if there is a relationship that the higher the index value IDX2, the higher the advertising effectiveness, it is desirable that the relationship holds for each index value IDX2 associated with as many users as possible.

In this respect, according to the present embodiment, the index value IDX2 for each user ID is calculated using the index value IDX1 for each user ID calculated as described above. Therefore, by appropriately calculating the index value IDX1 for each user ID, the index value IDX2 for each user ID can provide an effective index that is highly correlated (linked) with the level of advertising effectiveness.

Here, with reference to FIG. 5, the effect of the method of calculating the index value IDX1 described above will be described in comparison with another example of the method of calculating the index value IDX1 shown in FIG. 8 (hereinafter also referred to as "comparative example"). do. FIG. 8 is an explanatory diagram of a method of calculating the index value IDX1 according to a comparative example.

In the comparative example, the index value IDX1 is calculated by considering only the cost of the most recent primary result. Therefore, in the example shown in FIG. 5, for the user with the user ID "U01", IDX1=1000 is calculated based on the cost of 1000 yen for the most recent installation, and for the user with the user ID "U02" Based on the cost of 150 yen for the most recent return, IDX1=150 is calculated. In such a case, the user with the user ID “U02” is more likely to be evaluated as having a higher advertising effect than the user with the user ID “U01”.

Specifically, as shown in FIG. 8, for the user with the user ID "U01", the index value IDX2 is obtained by dividing the number of A coins used, 10000, by the index value IDX1 "1000". Similarly, for the user with the user ID "U02", the index value IDX2 is calculated as 66.67. As a result, the index value IDX2 for the user ID "U02" becomes significantly larger than the index value IDX2 for the user ID "U01".

However, unlike the user with the user ID of "U02", the user with the user ID of "U01" does not incur the cost of returning, and the advertising effect is higher than that of the user with the user ID of "U02". Probability is high.

On the other hand, in the method of calculating the index value IDX1 as shown in FIG. Then, the index value IDX1 is calculated. Therefore, in the method of calculating the index value IDX1 as shown in FIG. It is possible to obtain an index value IDX1 with high compatibility (interlocking). In other words, it is possible to obtain the highly useful index value IDX1 by taking into consideration the costs related to the past primary results.

Specifically, in the method of calculating the index value IDX1 as shown in FIG. = higher than 1000. As a result, contrary to the case of the comparative example, the index value IDX2 for the user ID "U01" becomes significantly larger than the index value IDX1 for the user ID "U02". Specifically, as shown in FIG. 6, for the user with the user ID “U01”, the index value IDX2=10.00 is similarly calculated, but for the user with the user ID “U02”, the index value The value IDX2=7.41 is calculated.

In this way, in the method of calculating the index value IDX1 as shown in FIG. An index value IDX1 with minutes added can be obtained. As a result, based on the index value IDX2 calculated based on the index value IDX1, it is possible to provide an effective index (index for each user) highly correlated (interlocking) with the level of advertising effectiveness.

In addition, in the method of calculating the index value IDX1 as shown in FIG. 5, the index value IDX1 is calculated by totaling the costs for each attribute of the primary result, but the method is not limited to this. For example, equivalently, the index value IDX1 may be calculated by summing the costs for each attribute of the advertisement. The attribute of the advertisement is, for example, an attribute related to the purpose, and may be, for example, an advertisement for newly installing the target application 7 or an advertisement for encouraging a dormant user to return.

[Embodiment 2]
FIG. 9 is a functional block diagram relating to advertisement-related functions among various functions realized by the server device 10A according to the second embodiment. Below, the same reference numerals may be attached to components that may be the same as those in the first embodiment, and the description thereof may be omitted.

As shown in FIG. 9, in the server device 10A according to the second embodiment, the first index value calculation unit 36 is replaced with a first index value calculation unit 36A in the server device 10 according to the first embodiment described above, and the measurement period The difference is that a setting unit 46, a proportional division ratio calculation unit 47, and a cost calculation unit 48 are added.

The measurement period setting unit 46 sets a primary result measurement period (an example of a predetermined measurement period) used for calculating the index value IDX1 by the first index value calculation unit 36A. The measurement period setting unit 46 may set the measurement period for each advertisement placement request (advertisement placement ID). The measurement period for one advertisement (advertisement placement ID) is set to include a period during which the primary results resulting from the one advertisement are appropriately (for example, completely) measured. Therefore, the measurement period for one advertisement (advertisement placement ID) preferably includes the output period of the one advertisement (the period during which it can be output to each user). Also, the measurement period for one advertisement (advertisement placement ID) may further include a period following the output period of the one advertisement (the period during which the advertisement can be output to each user). This is because, as will be described later, there are many users who install the software immediately after seeing the advertisement, but there are also users who install the software after a relatively long period of time. In this case, one measurement period may be set for one advertisement placement ID, or a plurality of measurement periods may be set so as not to overlap each other. Note that when a plurality of measurement periods are set, the length of each measurement period may be the same or different.

Note that the measurement period setting unit 46 may set the measurement period in a different manner according to the attribute of the advertisement for each advertisement placement ID. For example, the measurement period setting unit 46 sets a single measurement period for one advertisement (advertisement placement ID), whereas for another advertisement (advertisement placement ID) , a plurality of measurement periods may be set. Further, the measurement period setting unit 46 sets a single relatively long measurement period for one advertisement (advertisement placement ID), whereas for another advertisement (advertisement placement ID) Alternatively, a single relatively short measurement period may be set.

Moreover, the measurement period setting unit 46 may set the measurement period in a manner dependent on a predetermined parameter. For example, a measurement period may end when the value of a given parameter reaches a given value. In this case, the next one measurement period is started, and the next one measurement period may likewise end when the value of the predetermined parameter reaches the predetermined value. The predetermined parameter may be the number of advertisement clicks, the number of imps, the number of installations, or the like.

When the measurement period setting unit 46 sets a plurality of measurement periods for one advertisement (advertisement placement ID), the proportional division ratio calculation unit 47 calculates and associates a proportional division ratio with each measurement period. The proportional division ratio calculation unit 47 calculates each proportional division ratio so that the sum of the proportional division ratios associated with each measurement period is 100%. Note that when a single measurement period is set for one advertisement (advertisement placement ID) by the measurement period setting unit 46, a proportional division rate of 100% is assigned to the measurement period.

The method of calculating the proportional division ratio associated with each measurement period is arbitrary, but performance data of primary results in past measurement periods with the same attribute may be used. A specific example of the method of calculating the proportional division ratio will be described later with reference to FIG.

When a plurality of measurement periods are set for one advertisement (advertisement placement ID) by the measurement period setting section 46, the cost calculation section 48 proportionally divides the cost required for the one advertisement for each measurement period. associated with the costs incurred. In the present embodiment, as an example, the cost required for the one advertisement is the advertising cost, and the advertising cost is based on the cost information acquired by the cost information acquisition unit 34 described above. Note that, in a modified example, the cost required for one advertisement may include costs other than the advertising cost, as described above.

The cost calculation unit 48 calculates the cost associated with each measurement period based on the proportional division ratio of each measurement period associated by the proportional distribution ratio calculation unit 47 . Specifically, the cost associated with one measurement period is calculated based on the proportional division ratio associated with the one measurement period. For example, if the cost required for one advertisement is C0 and the apportionment ratio associated with one measurement period is α%, the cost associated with the one measurement period is calculated as C0×α/100. be.

In the present embodiment, the first index value calculation unit 36A, for each measurement period set by the measurement period setting unit 46, based on the relationship between the cost for each measurement period and the primary result for each measurement period, An index value IDX1 for each user ID is calculated and associated with each user ID. A specific example of a method of calculating the index value IDX1 by the first index value calculator 36A will be described with reference to FIGS. 10 and 11. FIG.

FIG. 10 is an explanatory diagram showing an example of a method of calculating the index value IDX1 by the first index value calculator 36A. FIG. 10 shows an example of a method of calculating the index value IDX1 when the measurement period set by the measurement period setting unit 46 is set as one period after placing an advertisement. It should be noted that the “(one) period after placing the ad” does not have to start immediately after the ad is placed. good too.

In FIG. 10, it is assumed that an advertisement placement request is generated for the advertisement media server 16 with an advertisement cost of 100,000 yen. Also, in this case, the output period is one day of 1/1 day of a certain year.

In the example shown in FIG. 10, the advertisement was clicked 2000 times and the target application 7 was installed 100 times on January 1st. Since the output period is 1/1 day, the number of clicks is 0 after 1/2 day. However, users who clicked on the advertisement on January 1st but did not install the target application 7 on January 1st and installed the target application 7 after January 2nd (hereafter referred to as "installation outside the output period (also referred to as “users who have For example, in the example shown in FIG. 10, 20 installations occur on 1/2 day, 10 installations occur on 1/3 day, and so on. It should be noted that such an installation number is based on the primary result information acquired by the primary result information acquisition unit 33 .

In this case, in the method of calculating the index value IDX1 (denoted as “single-day CPI” in FIG. 10) considering only 1/1 day, the index value IDX1 is calculated by dividing the advertising cost of 100,000 yen by 1/1. By dividing by the number of installs per day (=100), the index value IDX1 is calculated as 1000 yen (=100,000 yen/100). In this case, the primary result (the number of installations) used to calculate the index value IDX1 is the number of installations during one day (one day including the current day, that is, one day on 1/1) after placing the advertisement. Corresponds to the number of users (an example of the number of results).

In this case, the calculated index value IDX1 may be associated with each of the users (that is, 100 users) who installed the target application 7 on 1/1. Also, in this case, the calculated index value IDX1 is for each of the users who installed the target application 7 outside the output period, that is, the users (45 users) who installed the target application 7 during eight days from 1/2 to 1/8. may also be associated with Alternatively, in this case, an index value IDX1 different from the calculated index value IDX1 may be associated with each user who installed the software outside the output period.

Alternatively, a method of calculating the index value IDX1 (denoted as “real CPI” in FIG. 10) considering a longer period from 1/1 day may be used. For example, in the case of 8 days from 1/1 (8 days including 1/1, that is, 8 days from 1/1 to 1/8), the index value IDX1 is calculated as follows: By dividing by the number of installs (=145) for eight days from 1st to 1/8th, the index value IDX1 is calculated as 689 yen (=100,000 yen/145). In this case also, the calculated index value IDX1 may be associated with each of the users who installed it outside the output period, and the index value IDX1 different from the calculated index value IDX1 may be installed outside the output period. It may be associated with each of the users.

In this way, according to the example shown in FIG. 10, the primary result used for calculating the index value IDX1 is that each primary result occurs within one measurement period after placing the advertisement. Significantly different timings are treated in the same way. Therefore, the index value IDX1 obtained by relatively simple processing can be associated with each user.

FIG. 11 is an explanatory diagram showing another example of the method of calculating the index value IDX1 by the first index value calculator 36A. FIG. 11 assumes the same case as FIG. FIG. 11 shows an example of a method of calculating the index value IDX1 when a plurality of measurement periods are set as periods that do not overlap each other after an advertisement is placed.

In the example shown in FIG. 11, the measurement period is each day of eight days from 1/1 to 1/8. In this example, the plurality of measurement periods each have the same length (24 hours). good too.

Further, in the example shown in FIG. 11, the proportional division ratio calculated by the proportional division ratio calculation unit 47 is associated with each measurement period. Specifically, in the example shown in FIG. 11, a 55% apportionment rate is associated with 1/1 day, a 14% apportionment rate is associated with 1/2 day, and a 10% apportionment rate is associated with 1/3 day. A proportional division ratio is associated, and so on. In this way, the example shown in FIG. 11 includes processing for associating (assigning) a proportional division rate to each of a plurality of measurement periods.

The method of calculating the proportional division ratio (method of allocating the proportional division ratio) by the proportional division ratio calculation unit 47 is arbitrary, but may be equal division, for example. Alternatively, the apportionment rate associated with each of the plurality of measurement periods may be derived based on past performance data of the number of installations. For example, period management type installation number data (performance data) as shown in FIG. 4B may be preferably used. In this case, the number of installs occurring in each of the multiple measurement periods is predicted (derived) based on period management type install number data, and based on their ratio, the apportionment ratio associated with each of the multiple measurement periods may be determined. For example, let N1 be the number of installations on the first day (the first day) as the period after clicking, and similarly, the number of installs in the kth day (k is a natural number of 2 or more) as the period after clicking. When the number of installations is Nk, the proportional division ratios α1 to αk associated with each of the k measurement periods may be calculated as N1:Nk=α1:αk. In this case, the proportional division ratio α1 is associated with the most recent period after the advertisement was placed (for example, in the example shown in FIG. 11, 1/1 day), and the proportional division ratio α2 (k=2) is , is associated with the next most recent period after placing the advertisement (for example, in the example shown in FIG. 11, 1/2 day), and so on.

Also, the proportional division ratio associated with each of the plurality of measurement periods may be derived using a machine learning model learned based on past performance data of the number of installations. In this case, the machine learning model may be learned based on period management type installation number data (performance data) as shown in FIG. 4B.

Also, in the example shown in FIG. 11, the costs obtained by proportionally dividing the advertising cost of 100,000 yen by each proportional division ratio (denoted as “actual advertising costs” in FIG. 11) are associated with each measurement period. Specifically, in the example shown in FIG. 11, 55,000 yen, which is 55% of 100,000 yen, is associated with 1/1 day, and 14% of 100,000 yen is associated with 1/2 day. A certain $14,000 is associated, and so on.

In the example shown in FIG. 11, the index value IDX1 is calculated for each measurement period. Specifically, for a day on 1/1, by dividing the actual advertising cost of 55,000 yen by the number of installs on 1/1 (=100), the index value IDX1 = 550 yen. For the 1/2 day, the index value IDX1 is calculated as 700 yen by dividing the real ad placement cost of 14,000 yen by the number of installs on the 1/2 day (=20). , and so on.

In this case, the index value IDX1 calculated for one day on January 1st may be associated with each of the users (that is, 100 users) who installed the target application 7 on January 1st. In addition, the index value IDX1 calculated for the 1/2 day may be associated with each of the users who installed the target application 7 on the 1/2 day (that is, 20 users). In this manner, in the example shown in FIG. 11, for each measurement period, the index value IDX1 calculated in association with the period is associated with the user who installed the target application 7 during the period.

As described above, according to the method of calculating the index value IDX1 shown in FIG. 11, the cost used to calculate the index value IDX1 is based on the cost portion obtained by proportionally dividing the advertising cost (cost) by each proportional division ratio for each measurement period. It is determined. Accordingly, different index values IDX1 can be associated with users who installed the software outside the output period and users other than that (that is, users who installed the software on 1/1). This makes it possible to increase the reliability of the index value IDX2 calculated based on the index value IDX1, compared to a configuration in which the same index value IDX1 is associated with these users (see the example shown in FIG. 10). . That is, it is possible to increase the possibility of deriving the index value IDX2 having a higher correlation (interlocking) with the degree of advertising effectiveness.

FIG. 11A is a schematic flow chart showing another example of the method of calculating the actual advertising expenses (cost portion) and the method of calculating the index value IDX1 shown in FIG.

In step S1100, the cost calculation unit 48 selects one advertisement placement ID to be processed.

In step S1102, the cost calculation unit 48 determines whether or not the advertisement placement ID selected in step S1100 is an advertisement placement ID to which the placement cost is allocated for each region. The advertisement placement ID to which the placement cost is allocated for each region can be, for example, an advertisement ID related to outdoor advertisements, as described above. If the determination result is "YES", the process proceeds to step S1104; otherwise, the process proceeds to step S1120.

In step S1104, the cost calculation unit 48 selects one region. Although the order of selection is arbitrary, for example, in the case of prefectures in Japan, selection may be made in order from the north (in order from Hokkaido).

In step S<b>1106 , the cost calculation unit 48 selects one measurement period from the measurement periods set by the measurement period setting unit 46 . Note that the order of selection is arbitrary, but may be, for example, an order close to the output period of the advertisement associated with the selected advertisement placement ID. For example, in the example shown in FIG. 11, the cost calculation unit 48 selects one day of 1/1.

In step S<b>1108 , the cost calculation unit 48 calculates the actual advertising expenses for the selected one measurement period. For example, in the case of outdoor advertising, the real advertising expenses are calculated by proportionally dividing the advertising expenses related to the selected area by the proportional division ratio according to the selected one measurement period. In this case, the proportional division rate corresponding to the one selected measurement period may be daily (equally divided by number of days). It should be noted that the value calculated by the method described above with reference to FIG. 4C may be used as the advertisement cost for the selected region.

In step S1110, the first index value calculation unit 36A extracts all users who have installed the target application 7 during the selected measurement period and by accessing from the selected region. It should be noted that the area from which the access is made may be determined based on the IP address or the like, as described above.

In step S1111, the first index value calculation unit 36A equally divides the actual advertising expenses calculated in step S1108 by the number of users extracted in step S1110.

In step S1112, the first index value calculation unit 36A associates the values equally divided in step S1111 as index values IDX1 with the users extracted in step S1110.

In step S1113, the cost calculation unit 48 determines whether or not all measurement periods have been selected. If the determination result is "YES", the process proceeds to step S1116; otherwise, the process proceeds to step S1114.

In step S1114, the cost calculation unit 48 selects a new measurement period from among the measurement periods set by the measurement period setting unit 46, and returns to step S1108.

In step S1116, the cost calculation unit 48 determines whether or not all regions have been selected. If the determination result is "YES", the process ends; otherwise, the process advances to step S1118.

In step S1118, the cost calculation unit 48 selects a new area and returns to step S1106.

In step S<b>1120 , similarly to step S<b>1104 , the cost calculation unit 48 selects one measurement period from among the measurement periods set by the measurement period setting unit 46 .

In step S<b>1122 , the cost calculation unit 48 calculates the actual advertising expenses for the selected one measurement period. This calculation method may be, for example, the method described above with reference to FIG. For example, 55,000 Yen, which is 55% of 100,000 Yen, is calculated for one day of 1/1.

In step S1124, the first index value calculation unit 36A extracts all users who installed the target application 7 during the selected one measurement period.

In step S1125, the first index value calculation unit 36A equally divides the real ad placement expenses calculated in step S1122 by the number of users extracted in step S1124.

In step S1126, the first index value calculation unit 36A associates the values equally divided in step S1125 as index values IDX1 with each user extracted in step S1124.

In step S1127, the cost calculation unit 48 determines whether or not all measurement periods have been selected. If the determination result is "YES", the process ends; otherwise, the process proceeds to step S1128.

In step S1128, the cost calculation unit 48 selects one new measurement period from among the measurement periods set by the measurement period setting unit 46, and returns to step S1122.

Such processing as shown in FIG. 11A is executed for each advertisement placement ID, so that the index value IDX1 can be associated with each user for each advertisement placement ID. The process shown in FIG. 11A is suitable when the output periods of advertisements associated with a plurality of advertisement placement IDs do not overlap. Alternatively, the processing as shown in FIG. 11A can be performed even if the output periods of advertisements associated with a plurality of advertisement IDs overlap, and the user installs the target application 7 due to the advertisement associated with which advertisement ID. It is suitable when it is possible to identify whether In this case, the user extracted in steps S1110 and S1124 may be the user who installed the target application 7 due to the advertisement associated with the advertisement placement ID selected in step S1100. Note that what kind of advertisement the user saw and installed may be determined based on the results of a questionnaire. In a modified example, the users extracted in step S1110 may include natural users. This is because a natural influx of users may have been interested in the target application 7 after seeing an offline advertisement such as an outdoor advertisement. This is because there are many cases where it is unknown whether the software has been installed.

It should be noted that the index value IDX1 may be associated with the same user by processing related to a plurality of advertisement placement IDs. In this case, if the advertisement output periods for the plurality of advertisement IDs do not overlap, the user may be associated with the total value of the plurality of index values IDX1 (see FIG. 12 described later). ).

Note that the processing shown in FIG. 11A is based on the premise that there is an advertisement placement ID to which the placement cost is allocated for each region, but the present invention is not limited to this. For example, as described above, the advertisement costs for offline advertisements such as outdoor advertisements may not be considered, or may be considered in addition to the advertisement costs for online advertisements. good too. In this case, for example, in the example shown in FIG. 11, when the advertising cost “100,000 yen” is a value that only considers the advertising cost related to the online type advertisement, the actual advertising cost associated with each measurement period The cost may be added by the amount of the advertisement cost related to the offline advertisement associated with the corresponding measurement period. In this case, the advertising expenses for offline type advertisements may be equally divided by the number of measurement periods as described above, or if the measurement period lengths are different, a proportional division rate may be used according to the length. may be divided (calculated).

Note that the method of calculating the index value IDX1 according to the second embodiment described above may be applied to each attribute of the primary result. FIG. 12 is an explanatory diagram of the index value IDX1 calculated for each user and for each attribute of the primary result.

In the example shown in FIG. 12, the user with the user ID "U010" has installed the target application 7 through an advertisement, and the index value IDX1 associated with the advertisement at the time of installation is 1000 yen. It should be noted that no cost is incurred after the target application 7 is installed. In this case, the index value IDX1 associated with the user ID "U010" may be 1000 yen.

Further, in the example shown in FIG. 12, for the user with the user ID "U020", costs are generated for each attribute of the primary result. Specifically, first, the user with the user ID “U020” has installed the target application 7 through an advertisement, and the index value IDX1 associated with the advertisement at the time of installation is 1000 yen. The user with user ID "U020" has left after installing the target application 7, and has returned twice via advertisements. In this case, the index value IDX1 associated with the advertisement for the first return is 300 yen, and the index value IDX1 associated with the advertisement for the second return is 250 yen.

In such a case, the first index value calculation unit 36A may associate the index value IDX1=1000 yen for the user with the user ID “U010”. On the other hand, the first index value calculation unit 36A may associate the total value of the index values IDX1 (1550 yen in this case) for each attribute of the primary results for the user with the user ID “U020”.

[Embodiment 3]
FIG. 13 is a functional block diagram relating to advertisement-related functions among various functions realized by the server device 10B according to the third embodiment. Below, the same reference numerals may be attached to components that may be the same as those in the first embodiment, and the description thereof may be omitted.

The server device 10B according to the third embodiment provides an index value for not only the user who installed the target application 7 due to the advertisement, but also the user who installed the target application 7 not due to the advertisement (that is, the natural flow user). IDX1 and the corresponding index value IDX2 are associated. Hereinafter, the user who installed the target application 7 due to the advertisement is also referred to as the first user, and the user who installed the target application 7 not due to the advertisement is also referred to as the second user.

In the server device 10B according to the third embodiment, the first index value calculation unit 36 is replaced with the first index value calculation unit 36B, and the similarity determination unit 49 is added to the server device 10 according to the first embodiment. The difference is that

The similarity determination unit 49 extracts, from among the plurality of first users, first users having similarity to one second user. Whether the user who installed the target application 7 is the first user or the second user can be determined based on the user information in the user information storage unit 30 (for example, "installation opportunity" in FIG. 3B). .

The similarity determination unit 49 may determine similarity between users by any method. For example, the similarity determination unit 49 may determine similarity between users based on demographic data. In this case, the user information data in the user information storage unit 30 may be used as the demographic data. Also, a Jaccard coefficient or the like may be used to determine similarity between users.

Note that the similarity determination unit 49 may extract a plurality of first users having similarity to one second user. For example, the similarity determination unit 49 may extract a predetermined number of first users in descending order of similarity with respect to one second user, or extract all first users having similarity higher than a predetermined threshold. may be extracted.

The first index value calculator 36B includes an index value calculator 361 for the first user and an index value calculator 362 for the second user.

The index value calculator 361 for the first user calculates the index value IDX1 for each first user in the same manner as the first index value calculators 36 and 36A described in the first and second embodiments. , the index value IDX1 may be associated with each of the first users.

The index value calculation unit 362 for the second user calculates or determines the index value IDX1 associated with one second user based on the index value IDX1 associated with the first user extracted by the similarity determination unit 49. do. For example, the index value calculation unit 362 for the second user may directly associate the index value IDX1 associated with the first user extracted by the similarity determination unit 49 with one second user. Alternatively, the index value calculation unit 362 for the second user gives one second user a new index value IDX1 calculated based on the index value IDX1 associated with the first user extracted by the similarity determination unit 49. may be associated. For example, when a plurality of first users are extracted by the similarity determination unit 49 for one second user, the index value calculation unit 362 for the second user calculates the An average value of the index values IDX1 may be calculated, and the average value may be associated with the one second user.

As described above, according to the present embodiment, it is possible to associate the index value IDX1 with a user who installed the target application 7 not due to an advertisement, that is, a user who was a natural inflow. As a result, the index value IDX2 can be calculated based on the index value IDX1 associated with the user, even for such a natural inflow user. The method itself for calculating the index value IDX2 based on the index value IDX1 may be as described above. Therefore, according to the present embodiment, the advertisement target user extraction unit 44 selects from among a larger number of users (users associated with the index value IDX2), including natural inflow users, by the advertisement placement request generation unit 31 An advertisement ID to be included in the generated advertisement placement request can be extracted.

Further, according to the present embodiment, as described above, the index value IDX1 associated with the natural inflow user (second user) is the index value associated with the first user having similarity based on demographic data or the like. It is determined or calculated based on the value IDX1. As a result, the reliability of the index value IDX1 associated with the natural inflow user (second user) can be enhanced. As a result, the reliability of the index value IDX2 calculated based on the index value IDX1 can be improved, and as a result, efficient advertising can be realized based on the index value IDX2.

Although each embodiment has been described in detail above, it is not limited to a specific embodiment, and various modifications and changes are possible within the scope described in the claims. It is also possible to combine all or more of the constituent elements of the above-described embodiments.

For example, in the above-described embodiment, the index value IDX1 and the index value IDX2 related to the advertisement for one target application 7 are calculated for each user so as to realize efficient advertisement for one target application 7. However, it is not limited to this. The index value IDX1 and the index value IDX2 related to the advertisement related to one target application 7 may be calculated for each user so that efficient advertisement can be realized for applications other than the target application 7 . That is, the index value IDX1 and/or the index value IDX2 associated with each user in the manner described above with respect to the target application 7 may be used when placing advertisements regarding other applications. For example, when an application having similarity to the target application 7 is newly released, the advertisement related to the application is displayed based on the index value IDX2 associated with each user in the manner described above with respect to the target application 7. may be preferentially output to a plurality of users (targeting) extracted in descending order of .

Such processing may be realized by the processing shown in FIG. 14, for example. In FIG. 14, first, the advertised user extraction unit 44 extracts other applications with the highest similarity based on the similarities between the advertised application and other applications (step S1500). Next, the advertisement target user extracting unit 44 extracts a predetermined number of users in descending order of the index value IDX2 based on the index value IDX2 related to other applications with the highest similarity (step S1502). The predetermined number is arbitrary and may be given in the input. Then, the advertisement target user extraction unit 44 determines the advertisement IDs of the extracted predetermined number of users as one or more advertisement IDs to be included in the advertisement placement request generated by the advertisement placement request generation unit 31 (step S1504). .

In this case, in step S1500, similarities between applications may be determined artificially or a priori. Further, after the release, the index values IDX1 and IDX2 for each user may be corrected/corrected by using information related to two or more similar applications. In this case, for example, by re-analyzing the advertisement effect for a predetermined period after the release and the information of the users who naturally flowed in, it is possible to correct the similarity determination logic and update the targeting as needed.

Further, in the above-described embodiment, the index value IDX1 is not associated with each of the users associated with the advertisement IDs that did not install the target application 7 among the advertisement IDs included in the advertisement placement request. is not limited to For example, such information (information indicating that no results were obtained from the advertisement) itself can be effective, so each user associated with the advertisement ID has a specific flag (for example, a flag indicating that the advertisement was not effective). may be associated. The presence or absence of such a specific flag is used for appropriately selecting (extracting) target users for which advertisements related to the application are to be output when a new application having similarity to the target application 7 is released. good too. For example, with respect to an application having similarity to the target application 7, target users for whom advertisements related to the application are output may be extracted from a group of users not associated with a specific flag.

Such processing may be implemented by, for example, the processing shown in FIG. In FIG. 15, first, the advertised user extraction unit 44 extracts other applications with the highest similarity based on the similarities between the advertised application and other applications (step S1600). Next, the advertisement target user extracting unit 44 extracts a group of users for which a specific flag is not associated with other applications having the highest similarity (step S1602). Then, the advertisement target user extraction unit 44 selects the advertisement IDs of a predetermined number of users from the extracted user group as one or more advertisement IDs to be included in the advertisement placement request generated by the advertisement placement request generation unit 31. Determine (step S1604). In step S1604, the advertisement target user extracting unit 44 uses the advertisement IDs of all users in the extracted user group as one or more advertisement IDs to be included in the advertisement placement request generated by the advertisement placement request generation unit 31. may decide.

Also, in the above-described embodiment, the index value IDX2 is calculated based on the index value IDX1, but the present invention is not limited to this. For example, the index value IDX1 may be calculated using the secondary results described above instead of the primary results described above. In this case, the calculated index value IDX1 is substantially the same as the index value IDX2, so the calculation of the index value IDX2 may be omitted. For example, in the case of the index value IDX1 using the secondary result, the cost used to calculate the index value IDX1 may be the cost related to the specific advertisement required to obtain the secondary result, or It may be the cost of a particular advertisement required to obtain the primary result of the previous stage, or it may be the cost of a combination thereof. In this case, the user IDs associated with the index value IDX1 may include user IDs that have contributed to secondary results through clicking on a specific advertisement. For example, when a user with a certain user ID contributes to a secondary result by clicking on a specific advertisement, the user ID is associated with an index value IDX1. In this case, for example, if the above-mentioned cost is C3 and the amount of secondary results (for example, the number of A coins used) associated with the one user ID is M3, the index value associated with the one user ID IDX1 may be calculated as IDX1=C3/M3. The index value IDX1 calculated in this manner may be used by the advertisement target user extraction unit 44 or the like instead of the index value IDX2 described above.

In addition, the following additional remarks are disclosed regarding the above embodiment.

[Appendix 1]
An information processing program that causes a computer to execute a predetermined process,
The predetermined process is
Get information about your advertising-attributable performance;
Acquiring information about the cost required for the advertisement;
An information processing program including a process of associating a first index value representing the relationship between the result and the cost with each piece of user information.

[Appendix 2]
The predetermined process is
Each user information is associated with a cost for each attribute of the result,
Calculate the total cost for each attribute for each user information,
The information processing program according to appendix 1, further comprising a process of calculating the first index value for each piece of user information based on the total value for each piece of user information.

[Appendix 3]
the results include a first result for an application;
Attributes of the first outcome are:
installing the application;
The information processing program according to appendix 2, further comprising resuming use of the application after withdrawal for a predetermined period or longer.

[Appendix 4]
3. The information processing program according to appendix 3, wherein the cost required for the advertisement includes a first cost related to the advertisement related to the application posted in association with a plurality of pieces of first user information.

[Appendix 5]
The first index value for each user information is calculated based on the number of results,
5. The information processing program according to appendix 4, wherein the number of results corresponds to the number of pieces of user information that contributed to the results within a predetermined measurement period after placing the advertisement.

[Appendix 6]
The predetermined measurement period is set in a plurality of periods that do not overlap each other after the advertisement is placed,
The predetermined process is
Associate a cost portion obtained by proportionally dividing the first cost to each of the plurality of periods;
Further comprising a process of calculating the first index value for each user information based on the number of results in each of the plurality of periods and the cost associated with each of the plurality of periods. 6. The information processing program according to 5.

[Appendix 7]
The predetermined process is
further comprising a process of associating the first cost apportionment rate with each of the plurality of periods;
7. The information processing program according to appendix 6, wherein a cost portion obtained by proportionally dividing the first cost is calculated based on the proportional division rate associated with each of the plurality of periods.

[Appendix 8]
8. The information processing program according to appendix 7, wherein the predetermined process further includes a process of determining the proportional division rate based on past performance data of the number of results.

[Appendix 9]
9, wherein the user information associated with the first index value includes a plurality of the first user information and a plurality of second user information different from the first user information, out of appendices 4 to 8 The information processing program according to any one of .

[Appendix 10]
The predetermined process is
extracting one or more pieces of first user information having similarity to one piece of second user information from among the plurality of pieces of first user information;
calculating or determining the first index value associated with the one piece of second user information based on the first index value associated with the extracted one or more pieces of first user information; The information processing program according to appendix 8, further comprising:

[Appendix 11]
11. The information processing program according to appendix 10, wherein the similarity is determined based on demographic data.

[Appendix 12]
The predetermined process is
Obtaining information about a second outcome that may be obtained subsequent to the outcome;
12. The information processing according to any one of appendices 1 to 11, further comprising a process of associating a second index value representing the relationship between the first index value and the second outcome with each piece of user information. program.

[Appendix 13]
The predetermined process is
13. The information processing program according to supplementary note 12, further including a process of extracting user information linked to a new advertisement different from the advertisement based on the second index value.

[Appendix 14]
An information processing device that executes a predetermined process,
The predetermined process is
Get information about your advertising-attributable performance;
Acquiring information about the cost required for the advertisement;
An information processing apparatus, comprising a process of associating a first index value representing the relationship between the result and the cost with each piece of user information.

[Appendix 15]
Get information about your advertising-attributable performance;
Acquiring information about the cost required for the advertisement;
An information processing method executed by a computer, including associating a first index value representing the relationship between the result and the cost with each piece of user information.

1 system 3 network 7 target application 10, 10A server device 11 server communication unit 12 server storage unit 13 server control unit 16 advertising media server 20 terminal device 21 terminal communication unit 22 terminal storage unit 23 display unit 24 input unit 25 terminal control unit 30 User information storage unit 31 Ad placement request generation unit 32 Advertisement related data storage unit 33 Primary result information acquisition unit 34 Cost information acquisition units 36, 36A, 36B First index value calculation unit 361 Index value calculation unit 362 Index value calculation unit 38 Next result information acquisition unit 40 Index value data storage unit 42 Second index value calculation unit 44 Advertising target user extraction unit 46 Measurement period setting unit 47 Proportion ratio calculation unit 48 Cost calculation unit 49 Similarity determination unit

Claims (15)

An information processing program that causes a computer to execute a predetermined process,
The predetermined process includes, as an advertisement-related function,
Obtaining information about a first outcome resulting from an advertisement to facilitate utilization of a processed subject ;
Acquiring information about the cost required for the advertisement;
A first index value representing the relationship between the first result and the cost is associated with each user information related to a user who uses the site provided with the advertisement-related function ;
Obtaining information about a second outcome that can be obtained subsequent to the first outcome;
a process of associating a second index value representing the relationship between the first index value and the second result with each of the user information ;
Corresponding a cost for each attribute of the first result to each of the user information,
Calculate the total cost for each attribute for each user information,
An information processing program comprising a process of calculating the first index value for each piece of user information based on the total value for each piece of user information . An information processing program that causes a computer to execute a predetermined process,
The predetermined process includes, as an advertisement-related function,
Obtaining information about a first outcome resulting from an advertisement to facilitate utilization of a processed subject ;
Acquiring information about the cost required for the advertisement;
A first index value representing the relationship between the first result and the cost is associated with each user information related to a user who uses the site provided with the advertisement-related function ;
Obtaining information about a second outcome that can be obtained subsequent to the first outcome;
a process of associating a second index value representing the relationship between the first index value and the second result with each of the user information ;
An information processing program including a process of extracting user information linked to a new advertisement different from the advertisement based on the second index value . the processing target is an application ;
Attributes of the first outcome are:
installing the application;
3. The information processing program according to claim 1 , further comprising: resuming use of said application after leaving said application for a predetermined period of time or more. 4. The information processing program according to claim 3, wherein the cost required for the advertisement includes a first cost related to the advertisement related to the application posted in association with a plurality of pieces of first user information. The first index value for each user information is calculated based on the number of results related to the first result,
5. The information processing program according to claim 4, wherein said number of results corresponds to the number of pieces of user information that contributed to said first result within a predetermined measurement period after placing said advertisement. The predetermined measurement period is set in a plurality of periods that do not overlap each other after the advertisement is placed,
The predetermined process is
Associate a cost portion obtained by proportionally dividing the first cost to each of the plurality of periods;
further comprising calculating the first index value for each user information based on the number of results in each of the plurality of periods and the cost associated with each of the plurality of periods; Item 6. The information processing program according to item 5. The predetermined process is
further comprising a process of associating the first cost apportionment rate with each of the plurality of periods;
7. The information processing program according to claim 6, wherein a cost portion obtained by apportioning said first cost is calculated based on said apportionment rate associated with each of said plurality of periods. 8. The information processing program according to claim 7, wherein said predetermined process further includes a process of determining said proportional division rate based on performance data of said number of results in the past. 9. The user information of claims 4 to 8, wherein the user information associated with the first index value includes a plurality of the first user information and a plurality of second user information different from the first user information. The information processing program according to any one of the items. The predetermined process is
extracting one or more pieces of first user information having similarity to one piece of second user information from among the plurality of pieces of first user information;
calculating or determining the first index value associated with the one piece of second user information based on the first index value associated with the extracted one or more pieces of first user information; 9. The information processing program according to claim 8, further comprising: 11. The information processing program according to claim 10, wherein said similarity is determined based on demographic data. An information processing device that executes a predetermined process,
The predetermined process includes, as an advertisement-related function,
Obtaining information about a first outcome resulting from an advertisement to facilitate utilization of a processed subject ;
Acquiring information about the cost required for the advertisement;
A first index value representing the relationship between the first result and the cost is associated with each user information related to a user who uses the site provided with the advertisement-related function ;
Obtaining information about a second outcome that can be obtained subsequent to the first outcome;
a process of associating a second index value representing the relationship between the first index value and the second result with each of the user information ;
Corresponding a cost for each attribute of the first result to each of the user information,
Calculate the total cost for each attribute for each user information,
An information processing apparatus, comprising a process of calculating the first index value for each piece of user information based on the total value for each piece of user information . An information processing method in which a predetermined process is executed by a computer,
The predetermined process includes, as an advertisement-related function,
Obtaining information about a first outcome resulting from an advertisement to facilitate utilization of a processed subject ;
Acquiring information about the cost required for the advertisement;
A first index value representing the relationship between the first result and the cost is associated with each user information related to a user who uses the site provided with the advertisement-related function ;
Obtaining information about a second outcome that can be obtained subsequent to the first outcome;
a process of associating a second index value representing the relationship between the first index value and the second result with each of the user information ;
Corresponding a cost for each attribute of the first result to each of the user information,
Calculate the total cost for each attribute for each user information,
A process of calculating the first index value for each user information based on the total value for each user information
A computer-implemented information processing method comprising: An information processing device that executes a predetermined process,
The predetermined process includes, as an advertisement-related function,
Obtaining information about a first outcome resulting from an advertisement to facilitate utilization of a processed subject ;
Acquiring information about the cost required for the advertisement;
A first index value representing the relationship between the first result and the cost is associated with each user information related to a user who uses the site provided with the advertisement-related function ;
Obtaining information about a second outcome that can be obtained subsequent to the first outcome;
a process of associating a second index value representing the relationship between the first index value and the second result with each of the user information ;
An information processing apparatus including a process of extracting user information linked to a new advertisement different from the advertisement based on the second index value . An information processing method in which a predetermined process is executed by a computer,
The predetermined process includes, as an advertisement-related function,
Obtaining information about a first outcome resulting from an advertisement to facilitate utilization of a processed subject ;
Acquiring information about the cost required for the advertisement;
A first index value representing the relationship between the first result and the cost is associated with each user information related to a user who uses the site provided with the advertisement-related function ;
Obtaining information about a second outcome that can be obtained subsequent to the first outcome;
a process of associating a second index value representing the relationship between the first index value and the second result with each of the user information ;
An information processing method executed by a computer, including a process of extracting user information linked to a new advertisement different from the advertisement based on the second index value .

Images