Program, information processing system, and information processing device

Abstract

To provide a program, etc., which enable analysis of an impact of a given information page (e.g., advertisement) on sales without having to individuality access whether or not users are using an application via the given information page.SOLUTION: An estimation model is used to estimate a consumption compensation per unit of an application in contact with an information page during a given period, and a consumption compensation per unit of the application not in contact with the information page during the given period.SELECTED DRAWING: Figure 6

Description

Technical Field

[0001] The present invention relates to a program, an information processing system, an information processing apparatus, and an information processing method.

Background Art

[0002] Analyzing the impact of advertisements on sales based on data obtained from users is very useful for formulating more effective advertising strategies.

[0003] As a conventional technique, in an application such as a game, a mechanism for analyzing how much the advertisement has contributed to the sales of the application is known (for example, Patent Document 1).

[0004] For example, the conventional technique shown in Patent Document 1 discloses analyzing, for each user, how much sales are obtained with respect to the cost required for the advertisement (Return On Advertising Spend (ROAS)) in an application such as a game. ROAS is calculated from the amount of charge (number of used coins) for each user and information on whether the user has installed via an advertisement.

[0005] In order to perform analysis such as the conventional technique, in addition to grasping the amount of charge for the application for each user, it is necessary to individually grasp whether the user is using the application via an advertisement.

[0006] However, in recent years, from the viewpoint of personal information protection and the like, it has become difficult to individually grasp whether a user is using an application via an advertisement, and it has become difficult to accurately analyze the impact of advertisements on sales.

Prior Art Documents

Patent Documents

[0007]

Patent Document 1

SUMMARY OF THE INVENTION

PROBLEMS TO BE SOLVED BY THE INVENTION

[0008] The present invention has been made in view of the above problems, and an object thereof is to provide a program or the like that can analyze the influence of a predetermined information page on sales without individually determining whether a user is using an application via the predetermined information page (for example, an advertisement).

MEANS FOR SOLVING THE PROBLEMS

[0009] (1) The present invention relates to the number of terminal devices that installed the application by contacting a predetermined information page at a reference time point, the number of terminal devices that installed the application without contacting the information page at the reference time point, the total value of the consideration consumed for the application within a predetermined period from the reference time point for the terminal device that installed the application, Based on the above, using an estimation model, the consumption price per unit of the application that has contacted the information page during the predetermined period and the consumption price per unit of the application that has not contacted the information page during the predetermined period are estimated respectively. An estimation unit, which is a program characterized by causing a computer to function.

[0010] The present invention also relates to an information processing apparatus (for example, a terminal device, a server device) including each of the above units. The present invention also relates to an information processing system including each of the above units. The present invention also relates to an information storage medium readable by a computer, which stores a program for causing a computer to function as each of the above units.

[0011] The present invention also relates to The number of terminal devices that installed an application by contacting a predetermined information page at a reference time point, The number of terminal devices that installed the application without contacting the information page at the reference time point, Regarding the terminal devices that installed the application, the total value of the consideration consumed for the application within a predetermined period from the reference time point, Based on the above, using an estimation model, the consumption cost per unit of the application that contacted the information page within the predetermined period and the consumption cost per unit of the application that did not contact the information page within the predetermined period are respectively estimated. The present invention relates to an information processing method characterized by this.

[0012] Note that the "number related to the terminal device" may be the "number of terminal devices" itself, or the "number of users" who operate the terminal device. Also, the number related to the terminal device may be the number of service IDs (identification information) assigned by a service company that provides a given service.

[0013] Also, the consideration consumed for the application is, for example, the charged amount, points, etc.

[0014] Also, the "total value of the consideration" is, for example, the cumulative sales.

[0015] Also, the "consumption cost per unit of the application" may be the consumption cost per one terminal device that installed the application, the consumption cost per one user who installed the application, or the consumption cost per one service ID that installed the application.

[0016] Further, the "predetermined information page" is, for example, a page (information) of an advertisement (promotion, PR). Also, the "predetermined information page" is a page that can measure the number of contacts on the online (for example, whether there is a user operation such as a click on the information page) in a predetermined unit (for example, daily), or a page related to the measurement, or information (assuming the URL itself). Specifically, the "predetermined information page" refers to an advertisement page, an advertisement banner, a company's HP (home page), a banner to the HP, a banner leading to an app, an HP, etc.

[0017] According to the present invention, it is possible to analyze the influence of the information page on sales without individually grasping whether or not the user is using the application via the information page (for example, an advertisement).

[0018] (2) Further, the present invention The number related to the terminal device that logged in to the application after a predetermined elapsed period from the previous login time by contacting a predetermined information page at the reference time point, The number related to the terminal device that logged in to the application after a predetermined elapsed period from the previous login time without contacting the information page at the reference time point, Regarding the terminal device that logged in to the application, within a predetermined period from the reference time point The total value of the consideration consumed for the application, Based on the above, a program is characterized in that a computer functions as an estimation unit that estimates, using an estimation model, the consumption consideration per unit of the application that contacted the information page and the consumption consideration per unit of the application that did not contact the information page during the predetermined period, respectively.

[0019] Furthermore, the present invention relates to an information processing apparatus (e.g., a terminal device, a server device) including the above-described respective parts. The present invention also relates to an information processing system including the above-described respective parts. The present invention further relates to a computer-readable information storage medium storing a program for causing a computer to function as the above-described respective parts.

[0020] Furthermore, the present invention relates to the number of terminal devices that, at a reference time point, have logged in to an application after a predetermined elapsed period from the previous login time point while contacting a predetermined information page. relates to the number of terminal devices that, at the reference time point, have logged in to the application after a predetermined elapsed period from the previous login time point without contacting the information page. relates to the total value of the consideration consumed for the application by a terminal device that has logged in to the application within a predetermined period from the reference time point. Based on these, using an estimation model, the consumption consideration per unit of the application that has contacted the information page and the consumption consideration per unit of the application that has not contacted the information page within the predetermined period are respectively estimated. The present invention relates to an information processing method characterized by this.

[0021] Note that the "number related to a terminal device" may be the "number of terminal devices" itself, or the "number of users" who operate the terminal device. Also, the number related to a terminal device may be the number of service IDs (identification information) assigned by a service company that provides a given service.

[0022] Also, the consideration consumed for an application is, for example, the charged amount or points.

[0023] Also, the "total value of the consideration" is, for example, the cumulative sales.

[0024] Further, the "consumption cost per application unit" may be the consumption cost per one terminal device that has logged in to the application after a predetermined period has elapsed since the previous login, or the consumption cost per one user who has logged in to the application after a predetermined period has elapsed since the previous login, or the consumption cost per one service ID that has logged in to the application after a predetermined period has elapsed since the previous login.

[0025] Further, the "predetermined information page" is, for example, a page (information) of an advertisement (promotion, PR). Also, the "predetermined information page" is a page that can measure the number of contacts on the online (for example, whether there is a user operation such as clicking on the information page) in a predetermined unit (for example, daily), or a page related to the measurement, or information (assuming the URL itself). Specifically, the "predetermined information page" is an advertisement page, an advertisement banner, a company's HP (home page), a banner to the HP, a banner or HP that leads to an app, etc.

[0026] According to the present invention, it is possible to analyze the influence of the information page on sales without individually grasping whether a user is using the application via the information page (for example, an advertisement).

[0027] (3) Further, in the program, information processing system, information processing device, information processing method, and information storage medium according to the present invention, the estimation unit the number related to the terminal device when a terminal device that has never installed the application in the past contacts the information page and installs the application, the number related to the terminal device when a terminal device that has never installed the application in the past installs the application without contacting the information page, Regarding a terminal device that newly installs the application without having installed the application in the past, the total value of the consideration consumed for the application within the predetermined period from the reference time point. Based on , the consumption cost per unit of the application that has contacted the information page and the consumption cost per unit of the application that has not contacted the information page may be respectively estimated using an estimation model.

[0028] According to the present invention, it is possible to analyze new users. That is, the present invention can analyze the influence of the information page on sales without individually grasping whether a new user is using the application via the information page (for example, an advertisement).

[0029] (4) Further, in the program, information processing system, information processing device, information processing method, and information storage medium according to the present invention, the estimation unit The number of terminal devices regarding the case where a terminal device that has installed the application in the past contacts the information page and reinstalls the application. The number of terminal devices regarding the case where a terminal device that has installed the application in the past reinstalls the application without contacting the information page. Regarding a terminal device that has installed the application in the past and reinstalled the application, the total value of the consideration consumed for the application within the predetermined period from the reference time point. Based on , the consumption cost per unit of the application that has contacted the information page and the consumption cost per unit of the application that has not contacted the information page may be respectively estimated using an estimation model.

[0030] According to the present invention, it is possible to analyze a returning user. That is, the present invention can analyze the influence of an information page on sales without individually determining whether a returning user is using an application via the information page (e.g., an advertisement).

[0031] (5) Further, in the program, information processing system, information processing apparatus, information processing method, and information storage medium according to the present invention, the estimation unit may estimate, for each unit of the application that has contacted the information page, the consumption cost per unit of the application that has contacted the information page and the consumption cost per unit of the application that has not contacted the information page, respectively, for a terminal device for which a specific period or more has elapsed since the uninstallation time or the previous login time.

[0032] The specific period is the disengagement period of the application.

[0033] According to the present invention, it is possible to analyze a returning user for which a specific period or more has elapsed. possible.

[0034] (6) Further, in the program, information processing system, information processing apparatus, information processing method, and information storage medium according to the present invention, the estimation unit assuming that the time change is t, the consumption cost per unit of the application that has contacted the information page is ltv i,t and the consumption cost per unit of the application that has not contacted the information page is ltv 0,t and the total value of the cost is R t and the number of terminal devices that have contacted the information page is x i,t and the number of terminal devices that have not contacted the information page is x 0,t and when the number of information media is K, R t=ltv 0,t ·x 0,t +Σ[i=1→K]ltv i,t ·x i,t It may be estimated by the formula of.

[0035] According to the present invention, it can be appropriately estimated.

[0036] (7) Further, in the program, information processing system, information processing apparatus, information processing method, and information storage medium according to the present invention, The computer may be further caused to function as an output unit that outputs a time-series change in the consumption cost per unit of the application that has contacted the information page during the predetermined period for a plurality of reference time points.

[0037] According to the present invention, since the time-series change in the consumption cost per unit of the application that has contacted the information page (for example, an advertisement) during the predetermined period is output for a plurality of reference time points, analysis of the effect (for example, an advertisement effect) in the application becomes easy.

[0038] (8) Further, in the program, information processing system, information processing apparatus, information processing method, and information storage medium according to the present invention, The computer may be further caused to function as an output unit that outputs a time-series change in the consumption cost per unit of the application that has not contacted the information page during the predetermined period for a plurality of reference time points.

[0039] According to the present invention, since the time-series change in the consumption cost per unit of the application that has not contacted the information page (for example, an advertisement) during the predetermined period is output for a plurality of reference time points, analysis of the effect (for example, an advertisement effect) in the application becomes easy.

[0040] (9) Further, in the program, information processing system, information processing apparatus, information processing method, and information storage medium according to the present invention, The estimation unit When there are a plurality of information media, for each information media, the consumption cost per unit of the application that has contacted the information page during the predetermined period and the consumption cost per unit of the application that has not contacted the information page during the predetermined period may be estimated respectively.

[0041] According to the present invention, for each information media (for example, for each advertising media), the consumption cost per unit of the application that has contacted the information page during the predetermined period and the consumption cost per unit of the application that has not contacted the information page during the predetermined period are estimated respectively Therefore, in each information media, the effect (for example, advertising effect) in the application can be analyzed.

[0042] (10) Further, in the program, information processing system, information processing apparatus, information processing method, and information storage medium according to the present invention, For a plurality of reference time points, the consumption cost per unit of the application that has contacted the information page during the predetermined period is estimated for each information media, and the computer may be further caused to function as an output unit that outputs the time-series change of the consumption cost for each information media.

[0043] According to the present invention, for each information media (for example, for each advertising media), the time-series change of the consumption cost per unit of the application that has contacted the information page during the predetermined period is output for a plurality of reference time points, so that the analysis of the effect (for example, advertising effect) in the application for each information media becomes easy.

Brief Description of Drawings

[0044]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Mode for Carrying Out the Invention

[0045] Hereinafter, this embodiment will be described. Note that the embodiment described below does not unduly limit the content of the present invention described in the claims. Also, not all of the configurations described in this embodiment are essential constituent elements of the present invention.

[0046] [1] Description of the information processing system FIG. 1 is a diagram showing an example of the system configuration of the information processing system of this embodiment. The information processing system of this embodiment includes an information processing apparatus 10 and a terminal apparatus 20.

[0047] The information processing system of this embodiment is configured such that the information processing apparatus 10 and the terminal apparatus 20 can be connected to the Internet (an example of a network). The information processing system provides information regarding an application from the information processing apparatus 10 to the user's terminal apparatus 20. This information includes a program for operating the terminal apparatus 20 and update information of various data. The terminal apparatus 20 receives information regarding the application from the information processing apparatus 10.

[0048] The terminal apparatus 20 can receive the service of the application by installing the application. Also, the application can be uninstalled.

[0049] The information processing apparatus 10 is an information processing apparatus capable of providing a given service using the terminal apparatus 20. The information processing apparatus 10 may be referred to as a server apparatus, an analysis apparatus, or an estimation apparatus.

[0050] The terminal device 20 is an information processing device such as a home video game console, a personal computer (also referred to as a PC), a smartphone, a mobile phone, a PHS, a computer, a PDA, a portable game console, etc., an image generation device, a business game device, an arcade game device, a housing, etc., and can be connected to the information processing device 10 via a network such as the Internet (WAN) or LAN. It is a device that can do so.

[0051] The app providing server 30 is a server (information processing device, computer) that manages applications (hereinafter, "application" is also referred to as "app") to be provided to users in this embodiment. For example, the application may be, for example, a game app, but is not limited to game apps and may be various apps (video apps, manga providing apps, etc.).

[0052] The advertising medium server 41 is a first advertising medium (for example, Google (registered trademark)) and is an advertising platform.

[0053] The advertising medium server 42 is a second advertising medium (for example, facebook (registered trademark)) and is an advertising platform.

[0054] For example, the app providing server 30 can post an advertisement on the first advertising medium and post an advertisement on the second advertising medium. Note that the number of advertising media may be one, two, or three or more.

[0055] In this embodiment, it is assumed that the app providing server 30 provides apps to users on a specific OS (operating system). For example, the OS is iOS (registered trademark), but is not limited to this, and may be various OSs such as Android (registered trademark), Windows (registered trademark), etc.

[0056] [2] Functional configuration Next, the information processing device 10 of this embodiment will be described with reference to FIG. 2. Note that in this embodiment, a configuration in which a part of the components (each part) in FIG. 2 is omitted may be used.

[0057] In the information processing apparatus 10 of this embodiment, an input unit 160 for use by an administrator or other inputs, an information storage medium 180 in which predetermined information is stored, a communication unit 196 that communicates with a terminal device 20 or others, a processing unit 100 that mainly executes estimation processing, and a storage unit 170 that stores various data mainly used for estimation processing are included.

[0058] The input unit 160 is used by a system administrator or the like for settings and other necessary settings and data input.

[0059] The information storage medium 180 (a computer-readable medium) stores programs, data, etc., and its function is constituted by an optical disk (CD, DVD), a magneto-optical disk (MO), a magnetic disk, a hard disk, a magnetic tape, or a memory (ROM), etc.

[0060] The communication unit 196 performs various controls for communicating with the outside (for example, the terminal device 20, the application providing server 30, the advertising media servers 41, 42, etc.), and its function is constituted by hardware such as various processors or communication ASICs, programs, etc.

[0061] The storage unit 170 serves as a work area for the processing unit 100, the communication unit 196, etc., and its function is constituted by a RAM (VRAM), etc. Note that the information stored in the storage unit 170 may be managed by a database.

[0062] In addition, the storage unit 170 of this embodiment has, in addition to the main storage unit 171, a first measurement data storage unit 172, a second measurement data storage unit 173, an advertising information storage unit 174, and an estimation model storage unit 17 5.

[0063] In the first measurement data storage unit 172, data obtained by the first measurement function (first measurement data) is stored. The first measurement function is a function of acquiring data that can be used for measuring advertising effects from a terminal device (which may be each ID). Also, in the present embodiment, it is dedicated to an iOS (registered trademark) application and acquires the data, but it is not limited to this. In the case of implementing it for applications of other operating systems, the data corresponding to the operating system may be acquired.

[0064] In the second measurement data storage unit 173, data obtained by the second measurement function (second measurement data) is stored. The second measurement function is a function of acquiring data related to an application. Specifically, for example, the second measurement function is a function of acquiring data related to applications of a plurality of operating systems including Android (registered trademark) as well as iOS (registered trademark) applications. The data to be acquired is information such as the charging amount (consumption amount of consideration) in the application for each user's terminal device and the access log of the terminal device to the application.

[0065] Advertising information is stored in the advertising information storage unit 174. The advertising information includes, for example, information of the advertiser (company name, brand, product information), advertising costs (total costs for an advertising campaign, cost per click (CPC is the abbreviation of Cost Per Click), etc.).

[0066] Also, the advertising information may include a target audience (a specific user group targeted by the advertisement, data based on age, gender, region, interests and concerns, etc.).

[0067] Also, the advertising information may include an advertising format and types of advertisements (banner advertisement, video advertisement, interstitial advertisement, etc.).

[0068] In addition, the advertisement information includes advertisement placement information, for example, the location and timing where the advertisement is displayed within the application. Further, the advertisement information may include a campaign period (the period during which the advertisement campaign is executed).

[0069] In addition, the advertisement information may include performance indicators (the number of impressions (the number of times the advertisement is displayed), the number of clicks, the click-through rate, the number of conversions (such as app installations and purchases)).

[0070] In addition, the advertisement information may include creative materials (contents such as images, videos, and texts used in the advertisement).

[0071] In addition, the advertisement information may include the budget and the bidding strategy (for example, daily, weekly, monthly) of the advertisement budget and the bidding strategy.

[0072] The estimated model storage unit 175 stores the formula (function) of the estimated model used in the present embodiment. Note that the estimated model may be referred to as the estimated ROAS.

[0073] The processing unit 100 performs various processes using the main storage unit 171 in the storage unit 170 as a work area. The functions of the processing unit 100 can be realized by hardware such as various processors (CPU, DSP, etc.), ASIC (gate array, etc.), or a program.

[0074] The processing unit 100 performs various processes of the present embodiment based on the program (data) stored in the information storage medium 180. That is, the information storage medium 180 stores a program for causing a computer to function as each part of the present embodiment (a program for causing a computer to execute the processes of each part). A program for causing a computer to function as each part of the present embodiment (a program for causing a computer to execute the processes of each part) is stored.

[0075] For example, the processing unit 100 (processor) controls the entire information processing apparatus 10 based on a program stored in the information storage medium 180, and performs various processes such as control of data transfer between each part. Further, it performs processes for providing various services in response to requests from the terminal device 20.

[0076] Note that in the present embodiment, the information processing apparatus 10 may perform part or all of the processes of the processing unit 100, or the terminal device 20 may perform part of the processes of the processing unit 100.

[0077] The processing unit 100 includes an acquisition unit 110, an estimation unit 111, an output unit 112, a communication control unit 120, and a management unit 122.

[0078] The acquisition unit 110 acquires information such as first measurement data, second measurement data, and advertisement information. Note that the first measurement data acquired by the acquisition unit 110 may be stored in the first measurement data storage unit 172. The second measurement data acquired by the acquisition unit 110 may be stored in the second measurement data storage unit 173. The advertisement information acquired by the acquisition unit 110 may be stored in the advertisement information storage unit 174.

[0079] The estimation unit 111 estimates (infers, predicts, calculates) the customer lifetime value (LTV). As shown in FIG. 3, the estimation unit 111 calculates the estimated ROAS based on the first measurement data, the second measurement data, and the advertisement data (advertisement information). For example, the estimation unit 111 estimates the customer lifetime value per unit of the application via the information page (for example, advertisement). Further, the estimation unit 111 may calculate the ROAS by dividing the sales of the application via the advertisement by the advertisement cost.

[0080] For example, the estimation unit 111 is based on the number of terminal devices (e.g., the number of terminals, the number of unique users, the number of Apple (registered trademark) IDs) that contacted an information page (e.g., an advertisement) and installed an application at a reference time point (e.g., day t), the number of terminal devices (e.g., the number of terminals, the number of unique users, the number of Apple (registered trademark) IDs) that installed the application without contacting the information page (e.g., an advertisement) at the reference time point (e.g., day t), and the total value of the consideration consumed for the application within a predetermined period (e.g., 30 days) from the reference time point (e.g., day t) for the terminal devices on which the application was installed. Using an estimation model, the consideration consumed per unit of the application that contacted the information page (e.g., an advertisement) within the predetermined period (e.g., 30 days) (LTV) and the consideration consumed per unit of the application that did not contact the information page (e.g., an advertisement) within the predetermined period (e.g., 30 days) (LTV) are estimated respectively. Note that LTV is an abbreviation of Lifetime Value and may also be expressed as ltv.

[0081] For example, the estimation unit 111 sets the time change as t, sets the consideration consumed per unit of the application that contacted the information page (e.g., an advertisement) as ltv i,t and sets the consideration consumed per unit of the application that did not contact the information page (e.g., an advertisement) as ltv 0,t and sets the total value of the consideration as R t and sets the number of terminal devices that contacted the information page (e.g., an advertisement) as x i,t and sets the number of terminal devices that did not contact the information page (e.g., an advertisement) as x 0,t and when the number of information media (e.g., advertising media) is K, R t = ltv 0,t · x 0,t + Σ[i = 1 → K] ltv i,t · x i,t It is estimated by the formula described below. Details will be described later.

[0082] The output unit 112 outputs information regarding the customer lifetime value (LTV) (for example, the time-series change of the customer lifetime value). Details will be described later.

[0083] The communication control unit 120 establishes a connection (session or connection) with the outside (for example, the terminal device 20, the app providing server 30, the advertising media servers 41 and 42), and performs a process of communicating (transmitting and receiving) data via the network.

[0084] The management unit 122 manages various information, stores the information in the storage unit 170, deletes, updates, etc. the information stored in the storage unit 170.

[0085] [3] Overview In recent years, with the version upgrade of the operating system (for example, iOS (registered trademark)) of the terminal device 20, the protection of personal information of apps has been strengthened, and tracking restrictions for each user have been imposed. As a result, the contact data of online advertisements for each user is not shared with the app side, and the advertising effect (also referred to as ROAS) in the app cannot be directly measured. To solve this problem, by developing an estimation model (also referred to as a formula or function) called estimated ROAS, it becomes possible to estimate the ROAS for each advertising medium even after the tracking restrictions of iOS (registered trademark).

[0086] [4] Measurement of Conventional Advertising Effects Currently, advertising formats are shifting from offline advertising to "online advertising" along with the progress of digital technology. Online advertising refers to advertisements that can be published on the online web, where data can be confirmed in real-time, and it is possible to control the display or non-display of advertisements based on user behavior history. In particular, for smartphones, unlike offline advertising such as on TV, one of the advantages of online advertising is that user behavior can be tracked. By associating the advertising contact history linked to identifiers of the terminal such as cookies and advertising IDs with the terminal data of the installed app, it becomes possible to measure by linking information on whether the user has come into contact with the advertisement before installing the smartphone app and the subsequent app behavior.

[0087] Generally, as an indicator for calculating how much sales are obtained for advertising costs, Return On Advertising Spend (ROAS), which divides the sales through advertising by the advertising costs, is used to grasp the effectiveness of advertising. There are various media for conducting online advertising, such as google (registered trademark), facebook (registered trademark), etc. By calculating this ROAS, there exists a marketing method for optimizing the costs paid to each advertising medium.

[0088] However, in recent years, the concept of personal information protection has spread worldwide, and regulations have been introduced to restrict linking without obtaining user consent. In particular, smartphones equipped with iOS (registered trademark) provided by apple (registered trademark) have introduced a mechanism since iOS (registered trademark) version 14.5 that prohibits sharing identifiers of the terminal unless the user permits it themselves. As a result, on the app side, the user's behavior before installing the app is not shared, so it has become impossible to measure the advertising effect by the conventional method. This is a problem that occurs not only in smartphone games but also in all apps operating on iOS (registered trademark), and its influence is enormous.

[0089] [5] Method for grasping sales through advertising Concerned about the business losses due to the inability to fully measure advertising effectiveness, Apple (registered trademark) is providing data of the measurement function for iOS (registered trademark) apps (hereinafter referred to as "the first measurement function") while strengthening personal information protection. With the first measurement function, Even for iOS (registered trademark) versions 14.5 and later, the number of app installations can be measured.

[0090] However, since it is certain that the advertising exposure history will not be passed to the app, it is impossible to associate with users in the app, and the sales through advertising, which is the basis for ROAS calculation, remains unknown. Since the cost optimization of online advertising is considered based on sales rather than the number of installations, it is impossible to conduct marketing as before iOS (registered trademark) version 14.5.

[0091] Therefore, in this embodiment, an estimation model is adopted that estimates the sales through advertising, which cannot be measured on a per-user basis, using daily sales data and daily installation numbers. Note that the estimation model may also be referred to as an estimation formula, an estimation function, or an estimated ROAS. By this method, it is possible to know the sales through advertising as before, and the cost optimization of online advertising can be performed.

[0092] [6] Data and its formula used for estimated ROAS Organize the data and its formula used for estimated ROAS.

[0093] Let the number of unique users who installed a certain app on day t be UU(t). Also, let the cumulative sales R d (t) be the amount that this user paid on the app from day t to day d. The relationship between the two can be expressed as follows using the amount of money used per person over d days (LTV d (t)). Note that LTV is an abbreviation for Life Time Value and means "consumption price". It is an indicator that shows how much profit a user has brought to a company during the period from the start to the end of the transaction with the company.

[0094]

Number

[0095]

Number

[0096] It should be noted that different from the installation data, the cumulative sales data R d (t) cannot be separated between organic and advertisement-media users due to tracking limitations, and only the aggregated value of all users can be obtained from the app measurement data. The purpose of the estimated ROAS is to solve each LTV d (t) using these observed data.

[0097] [7] Solution [7.1] Specific Example of the Solution When solving the above formula (2), one assumption is made. The assumption is that the main component causing the time change of LTV d (t) is the event within the app, and the time fluctuations of each LTV d (t) are correlated. In fact, in the game of a smartphone app, in-game sales largely depend on in-game events. Also, the LTV generated from the specific origin of organic or advertising media dThe wave of the time variation of (t) has no reproducibility and is the same as random noise, and is not important when estimating the systematic effect of the advertising effect.

[0098] Using this assumption, Equation (2) is solved in the following steps.

[0099] First, the first term on the right side of Equation (2) is calculated for its wave of time change using the time-varying coefficient dynamic linear model.

[0100] Next, the LTV of the advertising medium is expressed as the product of the coefficient in the above wave of time change.

[0101] Then, the sum of the two is solved by MCMC (Markov chain Monte Carlo method).

[0102] Before explaining each step, Equation (2) is expressed in a more generalized form. In Equation (3), with organic being i = 0, each variable is defined as follows. The symbol corresponding to the time change is attached with t at the lower right of each variable.

[0103]

Number

[0104]

Number

[0105]

Number

Number

[0106] Note that

Number

Number

Number

[0107]

Number

[0108]

Number

[0109] Here, αi , γ i Since γ is a constant that does not change over time, there is an assumption in Equation (11) that the advertising effect does not change over time with respect to the regression coefficient of the organic matter in the calculation interval of t. In order to accommodate advertising media that do not satisfy this assumption, Equation (11) is extended by multiplying it with a function that changes over time, resulting in the following Equation (12).

[0110]

Equation

[0111]

Equation

[0112] Finally, the equation to be obtained is the combination of Equation (7) and Equation (12), which is the following Equation (14).

[0113]

Equation

[0114]

Equation

Equation

Equation

[0115] [7.2] Summary By adopting the above estimation formula (estimation model), the information processing apparatus 10 of the present embodiment can estimate the consumption price.

[0116] That is, the estimation unit 111 is based on (1) the number of unique users who installed the application by contacting a predetermined information page (for example, an advertisement) at the reference time point (t day), (2) the number of unique users who installed the application without contacting the information page at the reference time point (t day), and (3) the total value (Rt) of the price consumed for the application within a predetermined period (d days) from the reference time point (t day) for the terminal device on which the application was installed. Using the above estimation model (estimation formula), the consumption price per unique user (ltv) of the application that contacted the information page during the predetermined period (d days) and the consumption price per unique user (ltv) of the application that did not contact the information page during the predetermined period are estimated respectively. When there are a plurality of information media (for example, advertising media), the consumption price (ltv) is estimated for each information medium (for example, for each advertising media).

[0117] In addition, in this embodiment, the consumption price may be estimated for each terminal device. That is, the estimation unit 111 estimates, based on (1) the number of terminal devices that have installed an application by contacting a predetermined information page (for example, an advertisement) at a reference time point (t day), (2) the number of terminal devices that have installed the application without contacting the information page at the reference time point (t day), and (3) the total value (Rt) of the price consumed for the application within a predetermined period (d days) from the reference time point (t day) for the terminal devices that have installed the application, the consumption price (ltv) per terminal device of the application that has contacted the information page and the consumption price (ltv) per terminal device of the application that has not contacted the information page within the predetermined period (d days) using the above estimation model (estimation formula). When there are a plurality of information media (for example, advertising media), the consumption price (ltv) is estimated for each information medium (for example, each advertising media).

[0118] Also, in this embodiment, the consumption price may be estimated for each service ID (identification information) allocated by a service company that provides a given service. The service ID is an ID provided by the OS (for example, the ID of Apple (registered trademark)).

[0119] That is, the estimation unit 111 uses the above estimation model (estimation formula) based on (1) the number of service IDs that installed the application by contacting a predetermined information page (e.g., an advertisement) at the reference time point (t day), (2) the number of service IDs that installed the application without contacting the information page at the reference time point (t day), and (3) the total value (Rt) of the consideration consumed for the application within a predetermined period (d days) from the reference time point (t day) with respect to the terminal device on which the application was installed, to estimate the consumption consideration per service ID (ltv) of the application that contacted the information page and the consumption consideration per service ID (ltv) of the application that did not contact the information page during the predetermined period (d days). When there are a plurality of information media (e.g., advertising media), the consumption consideration (ltv) is estimated for each information medium (e.g., for each advertising media).

[0120] Also, when there are a plurality of information media (e.g., advertising media), the estimation unit 111 estimates the consumption consideration per unit of the application that contacted the information page and the consumption consideration per unit of the application that did not contact the information page during the predetermined period (d days) for each information medium (e.g., for each advertising media).

[0121] [8] Estimation in the case where the application has already been installed When the terminal device 20 has already installed the application, the consumption consideration is estimated based on the login status.

[0122] That is, the estimation unit 111, based on (1) the number of unique users who logged in to the application after a predetermined elapsed period (the second predetermined period) from the previous login time while contacting the information page (e.g., advertisement) at the reference time point (t day), (2) the number of unique users who logged in to the application after a predetermined elapsed period (the second predetermined period) from the previous login time without contacting the information page at the reference time point (t day), and (3) the total value (Rt) of the consideration consumed for the application within a predetermined period (d days) from the reference time point (t day) with respect to the terminal device that logged in to the application, uses an estimation model (estimation formula) to estimate, for each unique user of the application that contacted the information page within the predetermined period (d days), the consumption consideration per user (ltv), and, for each unique user of the application that did not contact the information page within the predetermined period, the consumption consideration per user (ltv). When there are a plurality of information media (e.g., advertising media), the consumption consideration (ltv) is calculated for each information medium (e.g., for each advertising media).

[0123] In addition, in the present embodiment, the consumption price may be estimated for each terminal device. That is, the estimation unit 111 estimates (1) the number of terminal devices that logged in to the application after a predetermined elapsed period (second predetermined period) from the previous login time while contacting an information page (e.g., an advertisement) at the reference time point (t day), (2) the number of terminal devices that logged in to the application after a predetermined elapsed period (second predetermined period) from the previous login time without contacting the information page at the reference time point (t day), and (3) based on the total value (Rt) of the price consumed for the application within a predetermined period (d days) from the reference time point (t day) for the terminal devices that logged in to the application, using the above estimation model (estimation formula), the consumption price (ltv) per one terminal device of the application that contacted the information page within the predetermined period (d days), and the consumption price (ltv) per one terminal device of the application that did not contact the information page within the predetermined period are estimated respectively. In the case where there are a plurality of information media (e.g., advertising media), the consumption price (ltv) is estimated for each information medium (e.g., for each advertising medium).

[0124] Also, in the present embodiment, the consumption price may be estimated for each service ID (identification information) allocated by a service company that provides a given service.

[0125] That is, the estimation unit 111 estimates (1) the number of service IDs that logged in to the application after a predetermined elapsed period (second predetermined period) from the previous login time while contacting an information page (e.g., an advertisement) at the reference time point (t day), (2) the number of service IDs that logged in to the application after a predetermined elapsed period (second predetermined period) from the previous login time without contacting the information page at the reference time point (t day), and (3) based on the total value (Rt) of the price consumed for the application within a predetermined period (d days) from the reference time point (t day) for the terminal devices that logged in to the application, using the above estimation model (estimation formula), the consumption price (ltv) per one service ID of the application that contacted the information page within the predetermined period (d days), and the consumption price (ltv) per one service ID of the application that did not contact the information Estimate the consumption cost (LTV) per service ID of an application that does not contact the page, respectively. When there are multiple information media (for example, advertising media), estimate the consumption cost (LTV) for each information media (for example, for each advertising media).

[0126] [9]New estimated ROAS [9.1]Estimation of consumption cost for new users The estimation unit 111 is based on (1) the number related to the terminal device (for example, the number of unique users, the number of terminal devices, the number of service IDs) when a terminal device that has never installed the application in the past contacts an information page (for example, an advertisement) and installs the application, (2) the number related to the terminal device (for example, the number of unique users, the number of terminal devices, the number of service IDs) when a terminal device that has never installed the application in the past does not contact the information page and installs the application, and (3) the total value of the cost consumed for the application within a predetermined period (d days) from the reference time point (t day) for a terminal device that has never installed the application in the past and newly installs the application. Using an estimation model (estimation formula), estimate the consumption cost per unit of the application that has contacted the information page (for example, the consumption cost per unique user, the consumption cost per terminal device, the consumption cost per service ID) and the consumption cost per unit of the application that has not contacted the information page (for example, the consumption cost per unique user, the consumption cost per terminal device, the consumption cost per service ID), respectively.

[0127] [9.2]Specific solution examples Let Rd(t) be the cumulative sales from t day to d days for the number of new unique users (NU) installed at the reference time point (t day). Assuming the number of advertising media is K, the following holds when expressing this Rd(t) in terms of the number of each new unique user (NU) and the LTV within d days after installation.

[0128] [Equation] Regarding the above (18), R d (t) represents the cumulative sales over d days. Also, LTV organic (t) represents the LTV of new unique users who do not come into contact with the advertisement over d days. Also, NU organic (t) represents the number of new unique users on day t who do not come into contact with the advertisement. Also, LTV i (t) represents the LTV of new unique users who have come into contact with the advertisement over d days. Also, NU i (t) represents the number of new unique users on day t who have come into contact with the advertisement. K represents the number of advertising media.

[0129] The necessary R in the above mathematical formula (18) d (t) is the second measurement data, and the NU of the advertising media is obtained from the first measurement data. However, the number of new unique users NU who have not come into contact with the advertisement (the number of new unique users NU who have not come into contact with the advertisement is also referred to as the "organic new unique user number NU") cannot be directly tabulated.

[0130] Therefore, as shown in the following mathematical formula (19), the number obtained by subtracting the number of new unique users NU who have come into contact with the advertising media (tabulated from the first measurement data) from the total number of new unique users NU on day t tabulated with the second measurement data is taken as the organic new unique user number NU.

[0131]

Number

[0132] [9.3] Data Aggregable for Estimating Consumption Consideration for New Users The data aggregable for estimating consumption consideration for new users is summarized as follows.

[0133] (1) The daily sales data for d days can be aggregated from the second measurement data.

[0134] (2) The daily number of new unique users can be aggregated from the second measurement data.

[0135] (3) The daily number of unique users installed via an advertising medium can be aggregated from the first measurement data.

[0136]

[10] Return Estimated ROAS [10.1] Estimation of Consumption Consideration for Returning Users The estimation unit 111 is based on (1) the number (e.g., the number of unique users, the number of terminal devices, the number of service IDs) regarding the terminal device when a terminal device that has installed the application in the past contacts an information page (e.g., an advertisement) and reinstalls the application, (2) the number (e.g., the number of unique users, the number of terminal devices, the number of service IDs) regarding the terminal device when a terminal device that has installed the application in the past reinstalls the application without contacting the information page, and (3) the total value of the consideration consumed for the application within a predetermined period (d days) from the reference time point (t day) for the terminal device that has installed the application in the past and reinstalled the application. Using an estimation model (estimation formula), the consumption consideration per unit of the application that has contacted the information page (e.g., the consumption consideration per unique user, the consumption consideration per terminal device, the consumption consideration per service ID) and the consumption consideration per unit of the application that has not contacted the information page (e.g., the consumption consideration per unique user, the consumption consideration per terminal device, the consumption consideration per service ID) may be estimated respectively.

[0137] Note that the estimation unit 111 may estimate, for a terminal device in which a specific period or more (for example, 7 days or more) has elapsed since the time of uninstallation of the application or the time of the previous login, the consumption price per unit of the application that has contacted the information page and the consumption price per unit of the application that has not contacted the advertisement, respectively.

[0138] [9.2] Specific solution examples Let R d (t) be the cumulative sales from day t to day d for the number of reinstalled unique users CU installed at the reference time point (day t).

[0139] Assuming the number of advertising media is K, the following holds when expressing this Rd(t) in terms of the respective number of reinstalled unique users CU and the LTV for d days after installation. Note that CU7d means that the user has returned after being away for 7 days or more (7 days or more have elapsed since the uninstallation time point or the previous login). It means that the user has returned after being away for 7 days or more (7 days or more have elapsed since the uninstallation time point or the previous login).

[0140] [Equation] The mathematical formula (20) for estimating the consumption price for returning users can be solved in the same way as the mathematical formula (18) for estimating the consumption price for new users.

[0141] Regarding the above (20), R d (t) indicates the cumulative sales for d days. Also, LTV organic (t) indicates the LTV for d days of a returning unique user who has not contacted the advertisement. Also, CU organic7d (t) indicates the number of returning unique users on day t who have passed 7 days since the previous login and have not contacted the advertisement. Also, LTV i (t) indicates the LTV for d days of a returning unique user who has contacted the advertisement. Also, CU 7di (t) indicates the number of returning unique users on day t who have passed 7 days since the previous login and have contacted the advertisement. K indicates the number of advertising media.

[0142] [10.3] Data Aggregable for Estimating Consumption Price for Returning Users The data aggregable for estimating the consumption price for returning users is summarized as follows.

[0143] (1) The sales data for d days of reinstalled users who have passed 7 days or more since the previous login can be aggregated from the second measurement data.

[0144] (2) The number of reinstallations of the advertising medium can be aggregated from the first measurement data.

[0145] (3) The number of unique users of reinstalled users who have passed 7 days or more since the previous login can be aggregated from the second measurement data.

[0146] (4) The rate of users who have passed 7 days since the previous login among reinstalled users can be aggregated from the second measurement data.

[0147] [10.4] Method for Determining Whether Reinstallation from Second Measurement Data In this embodiment, the second measurement data can identify the elapsed period since the previous login, etc., but it is not directly clear whether it is a reinstallation (redownload).

[0148] The second measurement data has columns representing the number of app launches, session_count and lifetime_session_count, and when the second measurement function is reset, session_count is also reset. Therefore, the timing when session_count = 1 is regarded as the timing of reinstallation. As a result, the estimation of reinstallation becomes unnecessary.

[0149] Figure 4 shows an example of the second measurement data, indicating the history of the login time (create_date) of id=001 and the number of app launches (session_count). As shown in Figure 4, for the user with id=001, on October 20, 2023, the value of session_count was "1", and before October 20, 2023, there was a history of session_count values greater than or equal to 2 (for example, on October 11, 2023, the value of session_count was "101" and there was a history of session_count values greater than or equal to 2). Therefore, it means that the user with id=001 has reinstalled the application.

[0150] In addition, when the value of the user's session_count is "1" and there is no history of session_count values greater than or equal to 2 before the day with the value of session_count being "1", it means that the application has not been reinstalled, that is, it has been newly installed.

[0151] [10.5] Method for Discriminating the Breakdown of Reinstallations from the First Measurement Data In this embodiment, it can be recognized that it is a reinstallation (redownload) from the first measurement data. However, the breakdown cannot be directly discriminated. Therefore, the breakdown is grasped by calculating according to the following mathematical formulas (21) to (23).

[0152] In addition, in the following mathematical formulas (21) to (23), R d (t) is the cumulative sales from the reference time point (t days) of the returnees who have passed 7 days since the previous login for d days. Also, CU all7d (t) is the number of unique users of all returnees who have passed 7 days since the previous login (for example, the number of unique users). Also, CU all (t) is the number of unique users of all returnees (for example, the number of unique users). Also, CU i (t) is the number of unique users of the returnees who have contacted the advertisement (for example, the number of unique users).

[0153]

Number

Number

Number

[11] Output of the time - series change of the consumption price In this embodiment, (for example, the output unit 112) outputs the time - series change of the consumption price per unit of the application that has come into contact with the advertisement within a predetermined period (for example, 30 days) for a plurality of reference time points (for example, the t - th day, the (t + 1)-th day, etc.).

[0154] FIG. 5 is a diagram showing the time - series change of the consumption price per unit of the application that has come into contact with the advertisement within a predetermined period (for example, 30 days) for a plurality of reference time points (for example, each day from October 1, 2023 to October 30, 2023). Note that the horizontal axis of FIG. 5 indicates time, and the vertical axis indicates the LTV (the consumption price per unit of the application that has come into contact with the advertisement within d days).

[0155] FIG. 5 shows the time - series change of the consumption price per unit of the application for each of the first advertising medium and the second advertising medium. For example, graph G1 shows the time - series change of the consumption price per unit of the application of the first advertising medium. Graph G2 shows the time - series change of the consumption price per unit of the application of the second advertising medium.

[0156] Although not shown, the graph of the time - series change output in this embodiment may show the time - series change of the average value of the consumption price per unit of all applications of the first and second advertising media.

[0157] In this embodiment, for an advertiser (for example, the administrator of an app providing server), the time-series change in the consumption price per unit of the application that has come into contact with the advertisement is output (provided, presented, notified). Note that "output" means outputting (displaying, presenting, providing) the time-series change on a screen viewable by the administrator (for example, the advertiser), outputting (notifying, presenting, providing) the time-series change to the administrator (for example, the advertiser) by email, outputting the content of the time-series change to the administrator (for example, the advertiser) by voice, outputting the image of the time-series change by a printer, and the like.

[0158] For example, to explain the time point of day t (for example, October 1, 2023), it shows the consumption price per one terminal device (user, service ID) in a predetermined period (for example, 30 days) from day t (October 1, 2023) after installing the application by coming into contact with the advertisement. Also, to explain the time point of the next day (t + 1) (for example, October 2, 2023), it shows the consumption price per one terminal device (user, service ID) in 30 days from day (t + 1) (October 2, 2023) after installing the application by coming into contact with the advertisement.

[0159] In this way, since the time-series change in the consumption price per unit of the application that has come into contact with the advertisement in a predetermined period (for example, 30 days) is displayed at a plurality of reference time points (for example, each day from October 1, 2023 to October 30, 2023), the advertiser can easily analyze to what extent the advertisement has affected the sales of the application based on this information.

[0160] Further, in this embodiment, for a plurality of reference time points, the consumption price per unit of the application that has come into contact with the advertisement in a predetermined period may be estimated for each advertising medium, and the time-series change in the consumption price for each advertising medium may be output.

[0161] For example, as shown in FIG. 5, since it shows the consumption price per one terminal device (user, service ID) for each of the first advertising medium and the second advertising medium, it is possible to easily analyze to what extent each advertising medium has affected the sales of the application.

[0162] Although not shown, in this embodiment, (for example, the output unit 112) may output the time-series change of the consumption price per unit of the application that does not contact the advertisement in a predetermined period for a plurality of reference time points.

[0163]

[12] Flowchart An example of the processing flow of this embodiment will be described with reference to the flowchart of FIG. 6. First, with respect to the number of terminal devices that installed the application by contacting the advertisement at the reference time point, the number of terminal devices that installed the application without contacting the advertisement at the reference time point, and the total value of the price consumed for the application from the reference time point within a predetermined period for the terminal devices that installed the application, are referred to (step S1).

[0164] These data may refer to the data stored in the first measurement data storage unit 172, the second measurement data storage unit 173, and the advertisement information storage unit 174, or the first measurement data obtained by the first measurement function and the second measurement data obtained by the second measurement function may be acquired from a given device or system (external server, protocol, etc.).

[0165] Next, using the estimation model, the consumption price per unit of the application that contacted the advertisement in a predetermined period and the consumption price per unit of the application that did not contact the advertisement in the predetermined period are respectively estimated (step S2).

[0166] For a plurality of reference time points, the time-series change of the consumption price per unit of the application that contacted the advertisement in a predetermined period is output (step S3). The processing ends here.

[0167]

[13] Application Example When there are multiple types of applications adopted in this embodiment (for example, when there are multiple types such as a battle game application and a quiz game application), the above estimation is performed for each application.

[0168] Also, although the reference time point (t) has been described in terms of days, it may be in terms of minutes or seconds.

[0169] Also, although an example where the number of advertising media is mainly two has been described, the number of advertising media may be one or three or more.

[0170] In this embodiment, the terminal device 20 may perform all or part of the processing of the processing unit 100 of the information processing device 10. That is, the terminal device 20 may acquire (refer to, store) all or part of the data in the storage unit 170 and estimate the consumption cost using an estimation model like the information processing device 10.

[0171]

[14] Others The present invention is not limited to those described in the above embodiments, and various modifications can be made. For example, terms cited as broad or synonymous terms in the description in the specification or drawings can be replaced with broad or synonymous terms in other descriptions in the specification or drawings.

[0172] The present invention includes configurations that are substantially the same as the configurations described in the embodiments (for example, configurations having the same functions, methods, and results, or configurations having the same purposes and effects). Further, the present invention includes configurations in which non-essential parts of the configurations described in the embodiments are replaced. Also, the present invention includes configurations that exhibit the same operational effects as the configurations described in the embodiments or configurations that can achieve the same purposes. Also, the present invention includes configurations in which known technologies are added to the configurations described in the embodiments.

[0173] As described above, the embodiments of the present invention have been described in detail, but it will be easily understood by those skilled in the art that many modifications can be made without substantially departing from the novel matters and effects of the present invention. Therefore, all such modified examples are considered to be included in the scope of the present invention.

Description of Signs

[0174] 10 Information processing device, 20, 20A, 20B, 20C Terminal devices, 30 App providing server, 41, 42 Advertising media servers, 100 Processing unit, 110 Acquisition unit, 111 Estimation unit, 112 Output unit, 113 Communication control unit, 114 Management unit, 160 Input unit, 170 Storage unit, 171 Main storage unit, 172 First measurement data storage unit, 173 Second measurement data storage unit, 174 Advertising information storage unit, 175 Estimation model storage unit, 180 Information storage medium, 196 Communication unit,

Claims

1. The number of terminal devices that installed an application by contacting a predetermined information page at a reference time point, The number of terminal devices that installed the application without contacting the information page at the reference time point, Based on the total value of the consideration consumed for the application within a predetermined period from the reference time point for the terminal devices that installed the application, A program that causes a computer to function as an estimation unit that estimates, using an estimation model, the consumption cost per unit of the application that contacted the information page and the consumption cost per unit of the application that did not contact the information page within the predetermined period, respectively.

2. The number of terminal devices that logged in to an application after a predetermined elapsed period from the previous login time by contacting a predetermined information page at a reference time point, The number of terminal devices that logged in to the application after a predetermined elapsed period from the previous login time without contacting the information page at the reference time point, Based on the total value of the consideration consumed for the application within a predetermined period from the reference time point for the terminal devices that logged in to the application, A program that causes a computer to function as an estimation unit that estimates, using an estimation model, the consumption cost per unit of the application that contacted the information page and the consumption cost per unit of the application that did not contact the information page within the predetermined period, respectively.

3. In Claim 1, The estimation unit, The number of terminal devices when a terminal device that has never installed the application before contacts the information page and installs the application, The number of terminal devices when a terminal device that has never installed the application before installs the application without contacting the information page, Based on the total value of the consideration consumed for the application within the predetermined period from the reference time point for the terminal devices that have never installed the application before and newly install the application, Based on this, using an estimation model, the consumption cost per unit of the application that has contacted the information page and the consumption cost per unit of the application that has not contacted the information page are respectively estimated. A program characterized by this.

4. In Claim 1, The estimation unit The number regarding the terminal device when the terminal device that has installed the application in the past contacts the information page and reinstalls the application, The number regarding the terminal device when the terminal device that has installed the application in the past reinstalls the application without contacting the information page, Regarding the terminal device that has installed the application in the past and reinstalled the application, the total value of the cost consumed for the application during the predetermined period from the reference time point, Based on this, using an estimation model, the consumption cost per unit of the application that has contacted the information page and the consumption cost per unit of the application that has not contacted the information page are respectively estimated. A program characterized by this.

5. In Claim 4, The estimation unit Regarding a terminal device for which a specific period or more has elapsed since the uninstall time or the previous login time, the consumption cost per unit of the application that has contacted the information page and the consumption cost per unit of the application that has not contacted the information page are respectively estimated. A program characterized by this.

6. In Claim 1 or 2, The estimation unit When the time change is t and The consumption price per unit of the application that has contacted the information page is defined as ltv ｉ,ｔ and The consumption cost per unit of the application that does not contact the information page is defined as ltv ０,ｔ and Let the total value of the consideration be R ｔ and Let x be the number of terminal devices that have contacted the information page ｉ，ｔ and Let x be the number of terminal devices that do not contact the information page ０，ｔ and the number of information media is K, R ｔ = ltv ０,ｔ · x ０，ｔ + Σ[i = 1 → K] ltv ｉ,ｔ · x ｉ，ｔ Estimated by the formula of. A program characterized by this.

7. In Claim 1 or 2, A program characterized by further causing a computer to function as an output unit that outputs the time-series change of the consumption cost per unit of the application that has contacted the information page during the predetermined period for a plurality of reference time points.

8. In Claim 1 or 2, A program characterized by further causing a computer to function as an output unit that outputs the time-series change of the consumption cost per unit of the application that has not contacted the information page during the predetermined period for a plurality of reference time points.

9. In Claim 1 or 2, The estimation unit When there are a plurality of information media, for each information media, estimating the consumption cost per unit of the application that has contacted the information page during the predetermined period and the consumption cost per unit of the application that has not contacted the information page during the predetermined period, respectively. A program characterized by this.

10. In claim 9, For a plurality of reference time points, estimating the consumption cost per unit of the application that has contacted the information page during the predetermined period for each information media, and further causing a computer to function as an output unit that outputs the time-series change of the consumption cost for each information media. A program characterized by this.

11. The number of terminal devices that have installed an application by contacting a predetermined information page at a reference time point, The number of terminal devices that have installed the application without contacting the information page at the reference time point, Regarding the terminal device that has installed the application, a predetermined period from the reference time point The total value of the cost consumed for the application during the period, Based on this, an information processing system including an estimation unit that estimates, using an estimation model, the consumption cost per unit of the application that has contacted the information page during the predetermined period and the consumption cost per unit of the application that has not contacted the information page during the predetermined period, respectively. A system characterized by this.

12. The number of terminal devices that have logged in to an application after a predetermined elapsed period from the previous login time point by contacting a predetermined information page at a reference time point, The number of terminal devices that have logged in to the application after a predetermined elapsed period from the previous login time point without contacting the information page at the reference time point, Regarding the terminal device that has logged in to the application, the total value of the cost consumed for the application during a predetermined period from the reference time point, Based on this, an information processing system including an estimation unit that estimates, using an estimation model, the consumption cost per unit of the application that has contacted the information page during the predetermined period and the consumption cost per unit of the application that has not contacted the information page during the predetermined period, respectively. A system characterized by this.

13. The number of terminal devices that have installed an application by contacting a predetermined information page at a reference time point, The number of terminal devices that installed the application without contacting the information page at the reference time point, For the terminal devices that installed the application, the total value of the consideration consumed for the application within a predetermined period from the reference time point, Based on the above, an estimation unit that uses an estimation model to estimate, respectively, the consumption cost per unit of the application that contacted the information page and the consumption cost per unit of the application that did not contact the information page during the predetermined period. An information processing apparatus characterized by including.

14. The number of terminal devices that logged in to the application after a predetermined elapsed period from the previous login time by contacting a predetermined information page at the reference time point, The number of terminal devices that logged in to the application after a predetermined elapsed period from the previous login time without contacting the information page at the reference time point, For the terminal devices that logged in to the application, the total value of the consideration consumed for the application within a predetermined period from the reference time point, Based on the above, an estimation unit that uses an estimation model to estimate, respectively, the consumption cost per unit of the application that contacted the information page and the consumption cost per unit of the application that did not contact the information page during the predetermined period. An information processing apparatus characterized by including.

15. The number of terminal devices that installed the application by contacting a predetermined information page at the reference time point, The number of terminal devices that installed the application without contacting the information page at the reference time point, For the terminal devices that installed the application, the total value of the consideration consumed for the application within a predetermined period from the reference time point, Based on the above, an information processing method characterized by using an estimation model to estimate, respectively, the consumption cost per unit of the application that contacted the information page and the consumption cost per unit of the application that did not contact the information page during the predetermined period.

16. The number of terminal devices that logged in to the application after a predetermined elapsed period from the previous login time by contacting a predetermined information page at the reference time point, The number of terminal devices that logged in to the application after a predetermined period had elapsed since the previous login without accessing the information page at the reference time point, For the terminal devices that logged in to the application, the total value of the consideration consumed for the application within a predetermined period from the reference time point, Based on the above, using an estimation model, the consumption price per unit of the application that accessed the information page and the consumption price per unit of the application that did not access the information page within the predetermined period are respectively estimated. An information processing method characterized by this.

Images