WO2017141837A1 - Program, information processing device, and information processing method - Google Patents

Program, information processing device, and information processing method Download PDF

Info

Publication number
WO2017141837A1
WO2017141837A1 PCT/JP2017/005001 JP2017005001W WO2017141837A1 WO 2017141837 A1 WO2017141837 A1 WO 2017141837A1 JP 2017005001 W JP2017005001 W JP 2017005001W WO 2017141837 A1 WO2017141837 A1 WO 2017141837A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
advertisement
result
control unit
correlation
Prior art date
Application number
PCT/JP2017/005001
Other languages
French (fr)
Japanese (ja)
Inventor
喜昭 平尾
利貢 岩澤
Original Assignee
株式会社サイカ
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from JP2016111015A external-priority patent/JP6364581B2/en
Application filed by 株式会社サイカ filed Critical 株式会社サイカ
Publication of WO2017141837A1 publication Critical patent/WO2017141837A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising

Definitions

  • the present invention relates to a program, an information processing apparatus, and an information processing method.
  • Patent Document 1 the relationship from a promotion element to a sales share is determined in a multi-stage manner regarding a plurality of promotion elements related to advertising activities, evaluations collected through questionnaires, purchase motivation degrees, and sales shares of products or services.
  • a promotion evaluation device that calculates by linear regression is disclosed.
  • Patent Document 1 the consumer's evaluation degree, purchase motivation degree, and the like are collected by a questionnaire, and it is difficult to say that the request for quickness is satisfied.
  • the present invention has been made under such circumstances, and an object thereof is to provide a program or the like that supports efficient advertising activities.
  • the program according to the present invention acquires, in a computer, advertisement information indicating an advertisement viewing status of a product or service and result information indicating a result of an advertising activity, and based on the advertisement information and the result information, A process of generating correlation information indicating a correlation with the result is executed.
  • the program according to the present invention is characterized in that the correlation information is generated by executing a covariance structure analysis including a multistage regression analysis based on the advertisement information and the result information.
  • the advertisement information is passive viewing information when a consumer's advertisement viewing mode is passive, indirect viewing information when active and indirect, and active and direct.
  • Direct viewing information and confirmation information indicating whether or not the consumer confirms the product or service
  • the result information is purchase information indicating a purchase status of the product or service
  • the passive viewing information, indirect Regression analysis is performed on each element included in viewing information, direct viewing information, confirmation information, and purchase information.
  • the program according to the present invention is characterized by outputting a sequence screen indicating a correlation between elements included in the advertisement information and the result information based on the correlation information.
  • the program according to the present invention is characterized by acquiring cost information related to the advertising activity and generating the correlation information based on the advertising information, result information, and cost information.
  • the program according to the present invention receives a condition input for the result information or cost information, calculates a composition ratio satisfying the received condition for each advertising measure related to the advertising activity based on the correlation information, It is characterized by outputting a budget plan indicating the composition ratio of the advertising measure.
  • the program according to the present invention is characterized by outputting a progress screen indicating a budget consumption status and the results for the advertising activity based on the budget plan.
  • the information processing apparatus includes an acquisition unit that acquires advertisement information indicating an advertisement viewing status of a product or service and result information indicating a result of an advertising activity, and the advertisement viewing based on the advertisement information and the result information.
  • a generating unit that generates correlation information indicating a correlation between the situation and the result is provided.
  • the information processing method acquires, in a computer, advertisement information indicating an advertisement viewing status of a product or service and result information indicating a result of an advertisement activity, and based on the advertisement information and the result information, the advertisement viewing A process for generating correlation information indicating a correlation between a situation and the result is executed.
  • FIG. 10 is an explanatory diagram for explaining statistical analysis according to an ad stock period according to Embodiment 2.
  • FIG. 10 is a flowchart illustrating an example of a processing procedure of a correlation information generation subroutine according to the second embodiment.
  • 10 is an explanatory diagram illustrating an example of a planning screen according to Embodiment 3.
  • FIG. 12 is an explanatory diagram illustrating an example of a progress screen according to Embodiment 3.
  • FIG. 12 is a flowchart illustrating an example of a processing procedure executed by a server according to the third embodiment.
  • FIG. 1 is a schematic diagram illustrating a configuration example of an information processing system.
  • an information processing system for realizing a service for performing a statistical evaluation related to the advertising effect of an advertising activity will be described as an example for an advertising activity carried out by a customer company that provides products or services.
  • the information processing system includes an information processing apparatus 1, viewing terminals 2, 2, 2,..., A customer server 3, a service center 4, a customer terminal 5, and the like.
  • Each device transmits and receives information to and from each other via a network N such as the Internet.
  • the information processing apparatus 1 is an information processing apparatus that stores various types of information and transmits and receives information via the network N.
  • the information processing apparatus 1 is, for example, a server computer or a personal computer.
  • the information processing apparatus 1 is assumed to be a server computer, and will be read as the server 1 below.
  • the server 1 acquires various types of information via the network N and performs statistical analysis related to the advertising effect of the advertising activity. This process will be described in detail later.
  • the viewing terminal 2 is a terminal device that can view advertisements, and is, for example, a personal computer, a smartphone, a television receiver, or the like.
  • the customer server 3 is a server device of a customer company that provides goods or services to consumers, and stores various information such as purchase status of goods or services, expenses spent for advertising activities, and the like.
  • the customer server 3 may be a personal computer, for example.
  • the service center 4 is a server device, for example, and aggregates the audience rating related to the television receiver.
  • the customer terminal 5 is an information processing terminal of a customer company, for example, a personal computer. The customer terminal 5 receives and displays various information related to the analysis result of the statistical analysis from the server 1.
  • FIG. 2 is a block diagram illustrating a configuration example of the server 1.
  • the server 1 includes a control unit 11, a storage unit 12, a communication unit 13, and a mass storage device 14.
  • the control unit 11 includes an arithmetic processing device such as a CPU (Central Processing Unit) or an MPU (Micro-Processing Unit).
  • the control unit 11 reads out and executes the program P stored in the storage unit 12 to perform various information processing or control processing related to the server 1.
  • the storage unit 12 includes memory elements such as RAM (Random Access Memory) and ROM (Read Only Memory).
  • the storage unit 12 stores a program P or data necessary for the control unit 11 to execute the processing according to the present embodiment.
  • the storage unit 12 temporarily stores data and the like necessary for the control unit 11 to execute arithmetic processing.
  • the communication unit 13 includes a processing circuit for performing processing related to communication, and transmits / receives information to / from the viewing terminal 2, the customer server 3, and the like via the network N.
  • the large-capacity storage device 14 is a large-capacity storage device including, for example, a hard disk.
  • the mass storage device 14 stores an advertisement DB 141, a result DB 142, a cost DB 143, and the like.
  • the advertisement DB 141 stores advertisement information indicating the viewing status of advertisements.
  • the result DB 142 stores result information indicating the result of the advertising activity.
  • the result DB 142 stores purchase information indicating the purchase status of the product or service related to the advertising activity.
  • the expense DB 143 stores expense information related to advertising activities.
  • the storage unit 12 and the mass storage device 14 may be configured as an integrated storage device. Further, the mass storage device 14 may be constituted by a plurality of storage devices.
  • the mass storage device 14 may be an external storage device connected to the server 1.
  • the server 1 is not limited to the above configuration, and may include, for example, an input unit that receives an operation input, a display unit that displays information related to the server 1, and the like.
  • FIG. 3 is an explanatory diagram illustrating an example of a record layout of the advertisement DB 141.
  • the advertisement DB 141 includes a customer column, a date column, a passive viewing column, an indirect viewing column, a direct viewing column, and a confirmation column.
  • the customer column stores an ID for identifying a customer company.
  • the date string stores the date during the advertising activity period in association with the ID.
  • the passive viewing column stores passive viewing information described later in association with the ID and date.
  • the passive viewing column includes the number of imps (impressions) of online advertising displays, the number of imps of video advertisements, the number of households watching television commercials, and general keywords (for example, “ The number of imps of listing advertisements displayed by Internet search based on “automobile” and the like is stored.
  • each passive viewing information is stored after being grouped according to the advertisement mode (for example, classified into “imp” group when displaying advertisement on the Internet).
  • the passive viewing information is not limited to the above, and includes, for example, the number of sessions to owned media (in-house website, blog, etc.). Note that the number of visits may be stored instead of the number of sessions.
  • information is stored for each advertisement medium (for example, whether the viewing terminal 2 is a personal computer or a smartphone) in the actual passive viewing sequence. The same applies to the following indirect viewing column, direct viewing column, and confirmation column.
  • the indirect viewing column stores indirect viewing information described later in association with the ID and date. Specifically, the indirect viewing column is searched and displayed based on the number of clicks on display ads, the number of clicks on video ads, the number of clicks on listing ads searched based on general keywords, and the designated keywords (for example, specific product names). The number of imps of listing advertisements, the number of imps of retargeting (registered trademark) advertisements, and the like are stored.
  • the indirect viewing information is not limited to the above, and includes, for example, the number of inflows from the external site to the owned media.
  • the direct viewing column stores direct viewing information described later in association with the ID and date.
  • the direct viewing column is the number of clicks on listing ads searched based on designated keywords, the number of clicks on retargeting ads, the number of inflows to owned media based on organic search, and the inflow to owned media due to direct traffic. I remember numbers.
  • the confirmation column stores confirmation information described later in association with the ID and date. Specifically, the confirmation column stores the number of sessions to a website such as online shopping that sells products or services. Note that the confirmation information is not limited to the above, and includes, for example, the number of sessions to individual landing pages on the Web site.
  • the advertisement information stored in the advertisement DB 141 may include not only information related to the company advertisement but also information related to advertising activities of competitors, for example. Further, the advertisement information may indicate not only the viewing status of the advertisement via the viewing terminal 2, but also the advertisement viewing status not via the viewing terminal 2, such as the number of visitors to the sales store, traffic advertisements, and the like. Good.
  • FIG. 4 is an explanatory diagram showing an example of the record layout of the result DB 142.
  • the result DB 142 includes a customer column, a date column, and a purchase column.
  • the customer column stores the ID of the customer company.
  • the date string stores the date during the advertising activity period in association with the ID.
  • the purchase column stores purchase information indicating the purchase status of the product or service of the customer company in association with the ID and date.
  • the purchase column stores, for example, the number of purchases of goods or services online, the number of purchases of goods or services offline (for example, the number of sales at stores), and the like. Note that the purchase information may be sales amount instead of the number of purchases.
  • FIG. 5 is an explanatory diagram showing an example of a record layout of the expense DB 143.
  • the cost DB 143 includes a customer column, a date column, and a cost column.
  • the customer column stores the ID of the customer company.
  • the date string stores the date during the advertising activity period in association with the ID.
  • the cost column stores cost information related to the advertising activity in association with the ID and date.
  • the expense column stores expenses for each advertising measure, such as display advertisements, video advertisements, television commercials, listing advertisements, retargeting advertisements, and advertisements on websites.
  • the cost information may be stored as a quantity (for example, the number of imps and the number of clicks) instead of the amount of money, and the server 1 may calculate the cost based on the quantity and the unit price of each advertising measure.
  • FIG. 6 is an explanatory diagram for explaining an outline of advertisement evaluation. Below, the outline
  • the processing entity of the server 1 is the control unit 11.
  • the control part 11 performs the process which evaluates effectiveness about an advertising activity of a customer company. That is, the control unit 11 evaluates how effective the advertisement has led to the consumer's purchasing activity based on the advertisement information indicating the viewing status of the advertisement and the result information indicating the result of the advertising activity.
  • the control unit 11 acquires advertisement information via the network N and stores the advertisement information in the advertisement DB 141.
  • the control unit 11 may accept manual input of advertisement information via the customer terminal 5 or the like.
  • the advertisement information is a group of information indicating how much the consumer has viewed the product or service of the customer company. Specifically, the advertisement information is the so-called online advertisements such as display advertisements, video advertisements, listing advertisements, retargeting advertisements, the number of imps and clicks, the number of sessions to the owned media or a predetermined website, so-called offline advertisements. Time-series data including the number of households viewing television commercials.
  • the control unit 11 classifies the advertisement information into four stages and stores the advertisement information in the advertisement DB 141 according to the consumer's advertisement viewing mode. Specifically, the control unit 11 classifies and stores advertisement information into passive viewing information, indirect viewing information, direct viewing information, and confirmation information.
  • Passive viewing information is advertising information when a consumer views an advertisement in a passive manner without the purpose of viewing the advertisement of a specific product or service. For example, when a display advertisement is considered, when a consumer browses a predetermined website on the viewing terminal 2, the display advertisement is displayed regardless of the intention of the consumer, and the consumer passively views the display advertisement. Become.
  • the control part 11 which concerns on this Embodiment judges the advertisement viewing condition in a passive aspect as mentioned above as passive viewing information.
  • the indirect viewing information is advertising information when a consumer views an advertisement in an active and indirect manner, although the consumer is not intended to view the advertisement of a specific product or service. For example, if the product of a customer company is a car, and a consumer searches for a general keyword “car” in a search engine, a listing advertisement is displayed in conjunction with the search result, and the consumer Will be watched. In the case where the consumer 11 does not intend to view the advertisement of the specific product or service as described above, but the consumer indirectly views the advertisement as a result of the information search according to his / her interest. Is determined to be indirect viewing information.
  • Direct viewing information is advertising information when a consumer actively and directly views an advertisement for the purpose of viewing an advertisement for a specific product or service. For example, when a consumer searches on a search engine based on a nominated keyword associated with a product name of a customer company, a listing advertisement related to the product is displayed, and the consumer views the listing advertisement. As described above, the control unit 11 determines that the advertisement information when directly viewing the advertisement is the direct viewing information as a result of the information search intended for the consumer to view the advertisement of the specific product or service as described above.
  • Confirmation information is advertisement information indicating whether or not the product or service related to the advertisement is confirmed by the consumer.
  • the terminal 2 accesses a link destination embedded in the listing advertisement (for example, a website that sells products or services) and displays information on the products or services.
  • the consumer views information on the product or service and confirms the content of the product or service.
  • the control unit 11 of the server 1 determines the advertisement information as confirmation information when the consumer confirms the product or service triggered by the advertisement as described above.
  • the number of sessions to a specific Web page is taken as an example.
  • the confirmation information may include the number of visitors to an actual store.
  • the control unit 11 classifies advertisement information into passive viewing information, indirect viewing information, direct viewing information, and confirmation information, so that the consumer is interested in the product or service related to the advertisement at the stage when the advertisement is recognized.
  • the advertisement viewing status of the consumer is classified at the stage of checking the goods or services related to the advertisement and the stage of checking the goods or services related to the advertisement.
  • the control unit 11 receives result information indicating the result of the advertising activity from the customer server 3.
  • the result information is time series data representing the advertising effect of the advertisement of the client company, that is, the result.
  • the result information is purchase information related to a product or service.
  • the purchase information is time-series data indicating sales etc. related to a product or service of a customer company.
  • each information is further subdivided according to a specific viewing mode of the advertisement.
  • the illustrated content of FIG. 6 is an exemplification, and each element group related to the advertisement information is not limited to this.
  • the control unit 11 performs a statistical analysis to be described later, and how the consumer is guided from “recognition” to “interest”, “search, etc.” and “confirmation” by the advertisement to reach “result (purchasing)”. To evaluate.
  • control unit 11 receives cost information related to the advertising activity from the customer server 3.
  • the cost information is information indicating, for each measure, the cost spent on the advertising activity by the client company.
  • the control unit 11 stores cost information in the cost DB 143.
  • FIG. 7 is an explanatory diagram for explaining an outline of processing executed by the information processing system.
  • the control unit 11 acquires advertisement information, result information, cost information, and the like via the network N.
  • the control unit 11 automatically acquires the imp number from an ASP (Affiliate Service Provider) server that publishes the display advertisement on the Internet.
  • ASP Application Service Provider
  • the control unit 11 accepts manual input via the customer terminal 5, for example.
  • the controller 11 performs integrated management and output processing of each information by accumulating each acquired information in a database.
  • the control unit 11 outputs each information to the customer terminal 5.
  • the control unit 11 performs a statistical analysis based on the acquired information, performs a statistical analysis on each correlation such as an advertising activity, a purchasing activity, and a necessary cost, and performs a process of outputting an analysis result.
  • FIG. 8 is an explanatory diagram illustrating an example of a dashboard screen.
  • the control unit 11 In response to an output request from the customer terminal 5, the control unit 11 generates a dashboard screen illustrated in FIG. 8 based on various information stored in the advertisement DB 141, the result DB 142, and the expense DB 143, and outputs the dashboard screen to the terminal. To do.
  • the dashboard screen is a screen for browsing each information accumulated in the advertisement DB 141, the result DB 142, and the expense DB 143.
  • the control unit 11 outputs each information in a table format and a graph format according to the advertisement medium and the advertisement mode. For example, the control part 11 displays the combination graph 81 which shows the time series data which concern on advertisement information, result information, etc. with a line or a bar graph.
  • the left vertical axis represents the number of advertisements such as the number of imps and clicks
  • the right vertical axis represents the number of purchases related to “result”
  • the horizontal axis represents the time axis.
  • the control unit 11 outputs the combination graph 81 by changing the parameter related to the right vertical axis in accordance with the selection input to the change column 81a shown in the upper left of FIG. Further, for example, the control unit 11 displays a bar graph 82 indicating the composition ratio for each advertising measure with respect to the information (cost information in FIG. 8) selected via the information selection column 82a.
  • control unit 11 displays a tabulation table 83 that displays a tabulation result relating to the advertisement viewing status, result (purchase status), cost, and the like for the advertisement measure selected via the type selection column 83a and the advertisement selection column 83b.
  • the control unit 11 refers to the advertisement DB 141 and the like for listing advertisements that are a type of online advertisement. The number of imps, the number of clicks, the cost, and other information are displayed in the tabulation table 83.
  • the control unit 11 may further display the total result of the entire measure related to the online advertisement or the offline advertisement. As described above, the control unit 11 can integrate and visualize the acquired various types of information, and the customer company can quickly confirm the activity status of the advertising activity.
  • FIG. 9 is an explanatory diagram for explaining the covariance structure analysis.
  • the control unit 11 performs statistical analysis of various pieces of information that are integrated and managed. Specifically, the control unit 11 analyzes the correlation between elements for each element included in passive viewing information, indirect viewing information, direct viewing information, confirmation information, purchase information, and the like. For example, the control unit 11 performs analysis by covariance structure analysis. Covariance structure analysis is a statistical tool for examining the relationship between multiple elements. In the present embodiment, the control unit 11 repeatedly performs multi-stage multiple regression analysis between a plurality of elements, thereby generating correlation information indicating a correlation between the elements.
  • control unit 11 selects an objective variable related to the regression analysis from the information related to one stage, and selects an explanatory variable related to the regression analysis from the information related to the previous stage. For example, in FIG. 9, regression analysis is performed using indirect viewing information related to “listing general word click” as an objective variable and each passive viewing information related to “cognition” as an explanatory variable. Details of the regression analysis will be described later.
  • the control unit 11 repeats regression analysis between each element to generate correlation information. Note that the control unit 11 not only has a relationship between successive elements such as a correlation between passive viewing information and indirect viewing information, but also a correlation between passive viewing information and direct viewing information, for example. The correlation between each element may be analyzed regardless of the continuity of the stages.
  • FIG. 10 is an explanatory diagram for explaining the regression analysis.
  • the x-axis represents the number of clicks on the listing advertisement based on the designated word (hereinafter referred to as the number of clicks on the designated listing), and the y-axis represents the number of online product purchases.
  • the control unit 11 calculates an approximate straight line shown in FIG. 10 by regression analysis.
  • the slope a of the slope line indicates the influence of the explanatory variable x on the objective variable y
  • the control unit 11 replaces the explanatory variable and the objective variable according to various information, repeats the regression analysis, and generates correlation information indicating a statistical correlation.
  • FIG. 10 illustrates the case of two variables, that is, the case of simple regression, but in practice, statistical processing is executed with multiple variables of third order or higher. That is, the control unit 11 defines a plurality of explanatory variables x1, x2,... With respect to one objective variable y, and performs a third or higher order multiple regression analysis. As a result, for example, as illustrated in FIG. 9, the control unit 11 performs one objective variable y (for example, “listing general word click” related to “interest”) and a plurality of explanatory variables x1, x2, x3 (for example, “recognition”).
  • one objective variable y for example, “listing general word click” related to “interest”
  • explanatory variables x1, x2, x3 for example, “recognition”.
  • control unit 11 also analyzes the correlation between the cost information and other information. That is, the control unit 11 also analyzes the correlation with respect to the cost effectiveness of the advertising activity.
  • the control unit 11 stores the correlation information that is the analysis result in the mass storage device 14.
  • FIG. 11 is an explanatory diagram for explaining an explanatory variable selection process. Although the outline of the covariance structure analysis has been described above, the control unit 11 performs the analysis process by appropriately selecting information to be used as the explanatory variable when executing the analysis process.
  • the control unit 11 selects one objective variable and specifies a candidate for an explanatory variable for the objective variable (hereinafter referred to as “candidate variable” for convenience of explanation).
  • the candidate variable is advertisement information for each advertisement mode (TV, Web, video), and the objective variable is purchase information (sales).
  • the storage unit 12 stores a file that defines an analysis policy related to statistical analysis, and the control unit 11 refers to the file to identify candidate variables.
  • the control unit 11 selects one or a plurality of explanatory variables optimal for explaining the objective variable from the candidate variables.
  • control unit 11 selects one variable from the candidate variables as an explanatory variable, and executes regression analysis.
  • the control unit 11 first executes regression analysis with “TV” as an explanatory variable.
  • the control unit 11 calculates an index for determining the validity of the explanatory variable based on the analysis result related to the regression analysis. Specifically, the control unit 11 calculates an index value CF (Candidate Factor) by the following equation (1).
  • R2adj is a coefficient of freedom adjusted for regression analysis
  • p is a P value
  • se is a standard error.
  • the control unit 11 calculates the index value CF with the explanatory variable “TV”. Further, the control unit 11 temporarily sets the candidate variable “TV” used in the regression analysis as an adoption candidate to be finally adopted as an explanatory variable.
  • the set explanatory variable is referred to as an “adoption candidate variable”.
  • control unit 11 adds another candidate variable to the explanatory variable in addition to the adoption candidate variable. Then, the control unit 11 executes the regression analysis again using the adoption candidate variable and the newly added candidate variable as explanatory variables.
  • the control unit 11 performs a regression analysis by adding another candidate variable “Web” to the adoption candidate variable “TV”. Based on the analysis result, the control unit 11 calculates the index value CF in the same manner as described above.
  • the control unit 11 compares the previously calculated index value CF (explanation variable is “TV”) with the current index value CF (explanation variable is “TV” “Web”) calculated by newly adding a candidate variable. Then, it is determined whether or not the current index value CF is larger than the previous index value CF. When it is determined that the current index value CF is larger, the control unit 11 sets an explanatory variable related to the current regression analysis as an adoption candidate variable. On the other hand, when it is determined that the current index value CF is smaller, the control unit 11 continues to set the explanatory variable related to the previous regression analysis as the adoption candidate variable. In the example of FIG. 11, the control unit 11 determines that the index value CF (the explanatory variable is “TV” “Web”) calculated this time is smaller, and continues to set the variable “TV” as an adoption candidate variable.
  • the control unit 11 repeats the same processing. For example, as illustrated in FIG. 11, the control unit 11 performs regression analysis on “TV”, which is the first candidate for adoption, by sequentially replacing other candidate variables to be combined, and compares the index values CF in magnitude. When all combinations of “TV” as the initial employment candidate are considered, as illustrated in FIG. 11, the control unit 11 once clears the explanatory variable and selects another candidate variable “Web” as a new employment candidate. The same processing is performed. Then, the control unit 11 selects the combination of explanatory variables having the maximum index value CF, that is, the optimal analysis result (“TV” and “moving image” in FIG. 11). Through the above selection process, the control unit 11 excludes “Web” defined as the explanatory variable candidate as having low correlation, and adopts “TV” and “moving image” as the explanatory variable.
  • FIG. 12 is an explanatory diagram showing an example of a navigation screen.
  • the control unit 11 outputs the correlation information generated by the series of processes to the navigation screen shown in FIG.
  • the navigation screen includes passive viewing information related to “recognition”, indirect viewing information related to “interest”, direct viewing information related to “Search & Retage”, confirmation information related to “confirmation”, and result information related to “result” ( It is a screen which shows a correlation with purchase information) with a sequence diagram.
  • the box 121 is a display field that shows an outline of the elements included in the advertisement information or the result information
  • the detail box 122 is a display field that shows elements obtained by further subdividing the elements shown in the box 121.
  • Each element related to the box 121 includes, for example, an advertisement mode (a mode of an advertisement viewed on various types of viewing terminals 2) by an icon such as “PC” or “SP” for the advertising medium (type of the viewing terminal 2) related to each element. ) Is indicated by the text in the box.
  • the control unit 11 accepts a selection input to one box 121 via the customer terminal 5.
  • the control unit 11 displays the correlation between the selected box 121 and the other box 121 by a solid line or a numerical value.
  • the control unit 11 indicates the magnitude of the influence by the thickness of the solid line connecting the boxes 121, and shows the breakdown of the influence of the explanatory variable on the objective variable by the numerical value attached to the solid line.
  • the control unit 11 shows the correlation of not only the outline of each element, that is, the detailed box 122 more detailed than the box 121 as well as the box 121.
  • the customer terminal 5 accepts a designation input for each box 121 generally showing various information, and notifies the server 1 of the designated content.
  • the control unit 11 of the server 1 outputs a more detailed detail box 122 for the specified box 121 to the customer terminal 5.
  • the control unit 11 displays a detailed box 122 in a pull-down format, and each advertisement related to “interest” and advertisements in other stages Display the correlation.
  • the client company can grasp the outline of the correlation with respect to its advertising activity and can easily grasp the detailed relation.
  • the control unit 11 displays the contribution value of the other box 121 or the detail box 122 with respect to the selected box 121 in a graph in the contribution value graph 124 shown in the lower side of FIG.
  • the vertical axis represents the contribution value (percentage).
  • the control part 11 shows the contribution value of the other box 121 or the detail box 122 with a bar graph, respectively.
  • the control unit 11 displays the element name of the box 121 or the detailed box 122 related to the bar graph (“Y company listing general word” in FIG. 12). To do.
  • the control unit 11 also displays a combination graph 81 shown at the lower right of FIG.
  • the combination graph 81 is a graph showing time series data related to the advertisement information and the result information, and is the same graph as the combination graph 81 shown in FIG. Thereby, the customer company can confirm the actual totaling result while browsing the analysis result of the advertising activity on the navigation screen.
  • FIG. 13 is a flowchart illustrating an example of a processing procedure executed by the server 1. Based on FIG. 13, the processing content which the server 1 performs is demonstrated.
  • the control unit 11 of the server 1 acquires advertisement information indicating the viewing status of the advertisement, result information indicating the result of the advertisement activity, and cost information related to the advertisement activity (step S11).
  • the advertisement information includes the above-described passive viewing information, indirect viewing information, direct viewing information, and confirmation information.
  • the result information is purchase information of goods or services.
  • the control unit 11 acquires each information via the network N. Note that the control unit 11 may acquire each information by manual input.
  • the control unit 11 stores the acquired advertisement information in the advertisement DB 141, the result information in the result DB 142, and the cost information in the cost DB 143 in association with the date.
  • the control unit 11 determines whether or not to output the dashboard screen to the customer terminal 5 (step S12). For example, the control unit 11 determines whether an output request has been received from the customer terminal 5. When it is determined that the dashboard screen is output (S12: YES), the control unit 11 refers to the advertisement DB 141, the result DB 142, and the cost DB 143 to obtain the advertisement information, cost information, and result information according to the advertisement medium and the advertisement mode. A dashboard screen is generated and output to the customer terminal 5 (step S13). Specifically, the control unit 11 includes a combination graph 81 indicating time series data such as advertisement information, a bar graph 82 indicating a composition ratio of each advertisement measure with respect to cost information, etc., advertisement viewing status, purchase status, cost, etc. for each advertisement measure. A dashboard screen including a summary table 83 and the like for displaying detailed summary results is generated and output.
  • the control unit 11 determines whether a predetermined analysis period has elapsed (step S14).
  • the analysis period is a period during which, for example, an advertising activity is performed. Note that the control unit 11 may accept a setting change for the analysis period regardless of the advertisement activity period.
  • the control part 11 returns a process to step S11.
  • the control part 11 shows the correlation information which shows the correlation of advertisement viewing condition, a result, and advertising expense based on advertisement information, result information, and expense information.
  • the process to generate is executed (step S15).
  • the control unit 11 generates a correlation information by executing a covariance structure analysis including a multistage regression analysis based on each information.
  • the control part 11 outputs the navigation screen which shows the correlation between the elements contained in advertisement information and result information based on the produced
  • the control unit 11 receives an output request from the customer terminal 5
  • the correlation of each element related to the stages of passive viewing information, indirect viewing information, direct viewing information, confirmation information, and purchase information is shown by a sequence diagram.
  • a navigation screen is generated and output to the customer terminal 5.
  • the control unit 11 shows an outline of elements included in each piece of information on the navigation screen by a box 121, shows the magnitude of influence between elements by a solid line connecting the boxes 121, and A numerical display of the breakdown of the influence of other elements on the element of.
  • the control unit 11 determines whether or not a designation input has been received for each element indicated on the navigation screen via the customer terminal 5 (step S17). For example, the control unit 11 determines whether or not an operation input to the box 121 is received at the customer terminal 5. When it determines with not receiving the designation
  • the control unit 11 determines whether or not to end the output of the navigation screen (step S19). For example, the control unit 11 determines whether or not the customer terminal 5 has received an input related to a predetermined end instruction. When it determines with not complete
  • FIG. 14 is a flowchart illustrating an example of a processing procedure of a correlation information generation subroutine. Based on FIG. 14, the correlation information generation subroutine of step S15 will be described.
  • the control unit 11 of the server 1 refers to the advertisement DB 141 or the result DB 142 and selects an objective variable from elements included in the advertisement information or the result information (Step S31). For example, the control unit 11 selects one objective variable from elements included in passive viewing information, indirect viewing information, direct viewing information, confirmation information, or purchase information.
  • the control unit 11 refers to the advertisement DB 141 or the cost DB 143, and specifies candidate explanatory variables for the objective variable selected in Step S31 from elements included in the advertisement information or the cost information (Step S32).
  • the control unit 11 refers to a file that defines an analysis policy related to statistical analysis, and identifies candidate variables that are candidates for explanatory variables in accordance with the analysis policy.
  • the control unit 11 selects one explanatory variable from the candidate variables identified in step S32 (step S33).
  • the control unit 11 performs regression analysis on the objective variable selected in step S31 and the explanatory variable selected in step S33 (step S34).
  • the control unit 11 calculates an index for determining the validity of the explanatory variable related to the regression analysis performed in step S34 (step S35). Specifically, the control unit 11 calculates the index value CF from Equation (1) based on the degree-of-freedom-adjusted determination coefficient R2adj, P value, standard error se, and the like calculated by regression analysis.
  • the control unit 11 determines whether or not the regression analysis executed in step S34 is the first analysis (step S36). When it determines with it being the first analysis (S36: YES), the control part 11 transfers a process to step S38.
  • the control unit 11 uses the index calculated this time in step S35 for the regression analysis performed on the explanatory variable set as the adoption candidate in step S38 described later. It is determined whether it is larger (step S37). That is, the control unit 11 determines whether or not the newly calculated current index is larger than the index related to the explanatory variable held as the adoption candidate. When it is determined that the current index is larger (S37: YES), the control unit 11 sets the explanatory variable related to the current index as an adoption candidate (step S38). Specifically, the control unit 11 temporarily stores one or a plurality of explanatory variables used in the regression analysis executed in step S34 as final adoption candidates. As a result, the control unit 11 holds a combination of more appropriate elements as explanatory variables based on the index value CF as an adoption candidate.
  • the control unit 11 sets the one explanatory variable selected in step S33 as another candidate variable. It is determined whether all the combinations have been examined (step S39). Specifically, the control unit 11 determines whether or not the index value CF is calculated and the determination process of step S37 is performed for all combinations of one explanatory variable selected in step S33 and other candidate variables. If it is determined that all are not considered (S39: NO), the control unit 11 adds one variable from the candidate variables identified in step S32 to the explanatory variable (step S40), and returns the process to step S34. As a result, the control unit 11 performs regression analysis on the explanatory variable obtained by adding another candidate variable to the adoption candidate variable, and calculates an index for the combination.
  • step S41 determines whether or not all other candidate variables have been selected in the explanatory variable selection process in step S33. That is, the control unit 11 selects each candidate variable as an initial explanatory variable, and determines whether or not the processing of steps S34 to S39 has been executed based on the selected candidate variable. If it is determined that there is a candidate variable that has not been selected (S41: NO), the control unit 11 returns the process to step S33. As a result, the control unit 11 selects another explanatory variable from the candidate variables, executes the processes of steps S34 to S39 again, and examines the validity of all combinations of elements related to the advertisement information.
  • step S42 the control unit 11 determines whether or not the processes of steps S32 to S41 have been executed for all elements that can be objective variables such as passive viewing information, indirect viewing information, direct viewing information, confirmation information, and purchase information.
  • the control unit 11 returns the process to step S31. Thereby, the control part 11 selects another objective variable, performs a regression analysis again, and analyzes a covariance structure.
  • the control part 11 returns a process.
  • the server 1 classifies the advertisement information into four stages as described above, the present embodiment is not limited to this.
  • the server 1 may regard the direct viewing information for searching for a product or service and the confirmation information for confirming the product or service as one information group, and classify the advertisement information in three stages.
  • the server 1 may subdivide passive viewing information, indirect viewing information, and the like.
  • the server 1 may classify the advertisement information into three levels or less, or may classify the advertisement information into five levels or more. That is, the server 1 only needs to be able to generate correlation information by multistage regression analysis.
  • the server 1 regards the result information indicating the result of the advertising activity as the purchase information of the product or service, and performs a series of processes.
  • the server 1 may perform a series of processes by regarding the confirmation information (for example, the number of sessions to an online store that sells the product or service) indicating whether the consumer confirms the product or service as the result information.
  • the server 1 defines passive viewing information, indirect viewing information, and direct viewing information as advertisement information, and defines confirmation information as result information.
  • the server 1 analyzes the correlation between the advertisement information and the result information, that is, the confirmation information.
  • the result information can be changed according to the analysis policy and may be information that can be defined as the result of the advertising activity.
  • the server 1 may determine the correlation regarding the viewing state of the advertisement in consideration of other external factors.
  • External factors are, for example, weather, product or service campaigns, and the like.
  • the server 1 may correct the advertisement information, the result information, etc. according to the external factor and perform the covariance structure analysis.
  • the server 1 automatically acquires information related to the advertising activity from each device and generates an analysis result related to the statistical analysis, thereby supporting efficient advertising activity. be able to.
  • the correlation between the advertising activity and the result of the advertising activity can be appropriately grasped by the covariance structure analysis including the multi-step regression analysis.
  • advertisement information is classified into four stages of confirmation such as consumer recognition, interest, search, etc., and the result information is statistically analyzed as sales of goods or services, whereby advertisement The process from recognition to outcome can be analyzed appropriately.
  • the first embodiment it is possible to easily grasp the correlation of each element related to the advertising activity on the navigation screen (sequence screen).
  • the cost-effectiveness of the advertising activity can be grasped.
  • Embodiment 2 In the present embodiment, a mode that considers the ad stock effect will be described. In addition, about the content which overlaps with Embodiment 1, the same code
  • FIG. 15 is an explanatory diagram for explaining the statistical analysis according to the ad stock period according to the second embodiment.
  • the server 1 performs a process according to an ad stock period (hereinafter referred to as an ad stock period).
  • the server 1 performs an analysis process in consideration of ad stock for statistical analysis related to offline advertising. Note that the server 1 may perform the following processing in consideration of ad stock not only for offline advertisements but also for online advertisements.
  • the passive viewing information related to the number of TV commercial viewing households is used as an explanatory variable, and the direct viewing information related to the number of imps of a listing advertisement based on a designated word is used as a target variable, and the TV commercial is displayed on a predetermined date.
  • the control unit 11 of the server 1 sets candidates for a plurality of periods (for example, 1 to 13 weeks) as the ad stock period. Then, as shown in FIG. 15, the control unit 11 assumes that the passive viewing information (in the case of FIG. 15 the number of viewing households) continues to attenuate according to the plurality of periods, and responds to each ad stock period. Time series model data is generated.
  • the control unit 11 when the ad stock period is one week, the control unit 11 generates model data in which the number of viewing households is attenuated to 10% at the end of the broadcast one week after the end of the broadcast. Similarly, the controller 11 generates model data of passive viewing information according to the ad stock period.
  • the control unit 11 performs model analysis using regression analysis based on each passive viewing information generated for each period and direct viewing information.
  • the control unit 11 compares the model accuracy for the results of each model analysis for each period. For example, the control unit 11 compares the model accuracy according to the magnitude of the degree-of-freedom adjusted determination coefficient R2adj related to the model analysis.
  • the control unit 11 selects the period related to the most accurate model data as the ad stock period. When there are a plurality of model data having the same accuracy, the control unit 11 selects the longest period as the ad stock period.
  • the control unit 11 employs the model data related to the selected ad stock period as an explanatory variable, and performs statistical analysis.
  • FIG. 16 is a flowchart illustrating an example of a processing procedure of a correlation information generation subroutine according to the second embodiment. Based on FIG. 16, the correlation information generation subroutine of step S15 according to the present embodiment will be described.
  • the control unit 11 of the server 1 After identifying the explanatory variable candidates for the objective variable (step S32), the control unit 11 of the server 1 executes the following processing.
  • the control unit 11 determines whether or not there is advertisement information related to the offline advertisement among the candidates for the explanatory variable specified in Step S32 (Step S201).
  • the element related to the offline advertisement is, for example, advertisement information related to a television commercial, a traffic advertisement, a magazine advertisement, or the like.
  • the control unit 11 If it is determined that there is advertisement information related to the offline advertisement (S201: YES), the control unit 11 generates model data corresponding to the ad stock period for the advertisement information related to the offline advertisement (step S202). For example, the control unit 11 generates model data in which the numerical value of the time series data related to the offline advertisement is attenuated during the ad stock period from the end of the offline advertisement. Specifically, the control unit 11 assumes that the ad stock period is 1 to 13 weeks, and generates 13 model data according to each ad stock period. For example, the control unit 11 generates model data on the assumption that the data is attenuated to 10% after the ad stock period has elapsed since the end of the offline advertisement. When it determines with there being no advertisement information which concerns on an offline advertisement (S201: NO), or after performing the process of step S202, the control part 11 transfers a process to step S33.
  • control unit 11 After selecting one explanatory variable from candidate variables (step S33) or adding another candidate variable to the explanatory variable (step S40), the control unit 11 executes the following processing.
  • the control unit 11 determines whether or not the advertisement information related to the offline advertisement is included in the explanatory variable selected in Step S33 or added in Step S40 (Step S203). When it determines with not being contained (S203: NO), the control part 11 transfers a process to step S34. When it determines with being contained (S203: YES), the control part 11 performs a regression analysis about each model data produced
  • the control unit 11 selects model data related to the ad stock period with the highest model accuracy as an explanatory variable (step S205). Specifically, the control unit 11 compares, for example, the degree-of-freedom-adjusted determination coefficient R2adj related to each analysis result, and determines model data having the largest coefficient R2adj. The control unit 11 determines that the ad stock period related to the model data is most appropriate, and selects model data generated according to the ad stock period as an explanatory variable. The control part 11 transfers a process to step S35.
  • Embodiment 3 a mode for planning a new advertising activity based on an analysis result related to statistical analysis and presenting a progress status of the advertising activity performed based on the planning will be described.
  • symbol is attached
  • FIG. 17 is an explanatory diagram illustrating an example of a planning screen according to the third embodiment.
  • the control unit 11 of the server 1 according to the present embodiment outputs a planning screen shown in FIG. 17 in response to an output request from the customer terminal 5.
  • the planning screen is a screen for creating a budget or the like related to a newly executed advertising activity based on the analysis result related to the advertising activity executed in the past.
  • the planning screen includes a correlation graph 171, a result comparison graph 172, a budget comparison graph 173, a budget table 174, and the like.
  • the correlation graph 171 is a graph showing the correlation between the result information and the cost information with respect to the correlation information that has already been generated with respect to the advertisement activity that has been performed in the past.
  • the correlation graph 171 includes a switching button 1711 and a setting object 1712 in addition to a graph having the number of purchases (results) and cost (budget) as axes.
  • the switch button 1711 is a button for switching the display of the correlation graph 171, and each axis of the correlation graph 171 is reversed depending on which of the “result” or “budget” switch buttons 1711 is designated. For example, in FIG. 17, since the “result” switching button 1711 is currently designated, the horizontal axis represents the number of purchases and the vertical axis represents the cost.
  • the control unit 11 accepts condition input related to result information (purchase information) or expense information in the correlation graph 171. Specifically, the control unit 11 receives an input regarding a result target (target number of purchases) or an upper limit cost, that is, a target value of sales or an upper limit value of a budget, regarding an advertising activity newly performed by a customer company.
  • the customer terminal 5 that is outputting the planning screen receives an operation input to the setting object 1712. Specifically, for the setting object 1712 shown in FIG. 17, the customer terminal 5 accepts an operation for shifting the position of the setting object horizontally along the horizontal axis “result”.
  • the control unit 11 acquires the operation content from the customer terminal 5.
  • the control unit 11 accepts the number of purchases indicated by the position as a condition value related to the result target according to the setting position of the setting object 1712 in the correlation graph 171, that is, the operation position of the setting object 1712 with respect to the horizontal axis “result”. As shown in FIG. 17, the control unit 11 displays the analysis accuracy in a label along with the horizontal axis of the correlation graph 171.
  • the label display indicates, for example, a numerical range of the number of purchases or expenses that have been analyzed in the past.
  • the client company can set the condition value within a range having high statistical reliability with reference to the label display.
  • the control unit 11 calculates an optimal budget plan that satisfies the accepted conditions and outputs it to the planning screen. Specifically, the control unit 11 calculates an optimal composition ratio that satisfies the received condition for each measure related to the advertising activity based on the generated correlation information. For example, the control unit 11 reads the coefficient a included in the generated correlation information. Further, the control unit 11 refers to the cost DB 143 and calculates the advertising unit price of each measure implemented in the past. Then, the control unit 11 calculates the composition ratio of each measure that satisfies the accepted condition and maximizes the cost effectiveness. For example, the control unit 11 performs calculation using a Lagrange undetermined multiplier method.
  • the control unit 11 when a condition input is received for the result target, the control unit 11 refers to the correlation information, and calculates the composition ratio of each measure that satisfies the result target and minimizes the advertising cost. For example, as shown in FIG. 17, when each measure is “G company listing”, “Y company listing”, etc., the control unit 11 fixes the result (the number of purchases) to the result target (the target number of purchases), The composition ratio of each measure is calculated based on the analysis result of the advertising activity, that is, the correlation information. In this case, the control unit 11 refers to the advertising unit price of each measure, and calculates the composition ratio so that the necessary cost is minimized. That is, the control unit 11 calculates a configuration ratio that maximizes cost effectiveness. When the condition input is received for the upper limit cost, the control unit 11 fixes the total cost for the advertising activity to the specified upper limit cost, and calculates the composition ratio that maximizes the result.
  • the control unit 11 outputs a budget plan indicating the calculated composition ratio. Specifically, the control unit 11 outputs the composition ratio as a budget distribution for each measure in the budget table 174.
  • the budget table 174 is a table showing details of the budget plan for each measure related to the advertising activity. As shown in FIG. 17, in the budget table 174, the control unit 11 outputs the name of each measure, the past (previous month) advertising cost, and the measure-specific budget.
  • the control unit 11 also outputs a result comparison graph 172 and a budget comparison graph 173.
  • the result comparison graph 172 and the budget comparison graph 173 are graphs showing comparison data between the results and costs related to past advertising activities and the expected results (expected number of purchases) and expected costs related to the budget plan.
  • the budget comparison graph 173 further shows the budget allocation for each measure in the expected cost.
  • control unit 11 receives a condition input regarding the composition ratio of each measure in the budget table 174.
  • the control unit 11 receives a budget allocation change based on an operation input to the simulation field 1741 of the budget table 174.
  • the control unit 11 calculates the budget plan again after fixing the cost related to the measure. That is, the control unit 11 recalculates the budget allocation for other measures other than the designated measure, and outputs it to the simulation column 1714.
  • the control unit 11 automatically links the result comparison graph 172 and the budget comparison graph 173 and outputs the calculation result again.
  • the budget plan can be automatically re-output according to the request of the client company.
  • the control unit 11 receives an operation input to the confirmation button 175 on the planning screen, the control unit 11 confirms the budget plan. Thereby, the customer company newly starts an advertising activity based on the generated budget plan.
  • the control unit 11 acquires advertisement information and the like as in the first embodiment, and outputs the following progress screen to the customer terminal 5.
  • FIG. 18 is an explanatory diagram showing an example of the progress screen according to the third embodiment.
  • the progress screen is a screen showing the status of the budget consumption and the result of the advertising activity, that is, the purchase status of the product or service, for the advertising activity based on the budget plan.
  • the progress screen includes approximate data 181, activity status data 182, and a detailed table 183.
  • the control unit 11 refers to the acquired cost information, and calculates the cost (budget) for each measure that has been spent since the start of the advertising activity. In addition, the control unit 11 refers to the result information (purchase information), and performs aggregation related to the results (the number of purchases) from the start of the advertising activity.
  • the control unit 11 outputs a comparison between the cost and result and the expected cost and the expected result related to the budget plan as the approximate data 181.
  • the approximate data 181 is a bar graph showing the budget consumption status indicating the current cost amount with respect to the expected cost amount, and the achievement target achievement level indicating the current result with respect to the result target.
  • control unit 11 outputs an overall rough estimate of the budget consumption status and the advertising activity result (purchase status of goods or services) in the approximate data 181. Moreover, the control part 11 shows the time-sequential budget consumption situation and result regarding an advertising activity in the activity status data 182.
  • the activity status data 182 is a graph showing the cost amount for each measure and the number of purchases on each date.
  • control unit 11 shows the detailed budget consumption status on each date in the detailed table 183.
  • the detailed table 183 is a table showing the target cost to be consumed in one day, the cost currently being consumed, and the target cost to be consumed at a later date for each measure.
  • the customer company can confirm the digestion status of the budget in more detail than the approximate data 181 and the activity status data 182 with reference to the detailed table 183.
  • the control unit 11 simply displays a navigation screen (sequence diagram) indicating the correlation between the measures. Note that the control unit 11 may output not only the budget consumption status but also the results of advertising activities (the number of purchases) in the detailed table 183.
  • FIG. 19 is a flowchart illustrating an example of a processing procedure executed by the server 1 according to the third embodiment. Based on FIG. 19, the processing content which the server 1 which concerns on this Embodiment performs is demonstrated. It is assumed that the server 1 has already generated correlation information by the processing described in the first embodiment.
  • the control unit 11 of the server 1 refers to the correlation information generated with respect to the past advertising activity, generates a planning screen for determining a budget plan related to the advertising activity, and outputs the planning screen to the customer terminal 5 (step S301).
  • the control unit 11 refers to past advertisement information, result information, and cost information, and generates the planning screen illustrated in FIG.
  • the planning screen shows, for example, a correlation graph 171 indicating the correlation between the result information related to past advertising activities and the cost information, and the comparison between the results related to past advertising activities (number of purchases) and the expected results (number of expected purchases).
  • Result comparison graph 172 budget comparison graph 173 showing a comparison between costs (budgets) related to past advertising activities and expected costs
  • a budget table 174 for showing details of a budget plan generated by the processing in step S304 described later, etc. including.
  • the control unit 11 outputs the generated planning screen to the customer terminal 5 via the network N.
  • the control part 11 receives the condition input regarding a budget plan about result information or expense information (step S302). Specifically, the control unit 11 accepts a setting input of a target purchase quantity that is a result target, for example, a sales target, in the correlation graph 171 on the planning screen. Or the control part 11 receives the setting input of the upper limit expense spent on advertising activity in the correlation graph 171. That is, the control unit 11 receives a condition input for either the achievement target or the upper limit cost.
  • the control unit 11 calculates the composition ratio of each measure that satisfies the condition received in step S302 based on the correlation information generated for the past advertising activity (step S303). Specifically, the control unit 11 refers to the cost DB 143 and calculates the cost unit price of each measure. In addition, the control unit 11 reads the correlation information that has been generated for the past advertising activity from the mass storage device 14. Specifically, the control unit 11 reads the coefficient a included in the correlation information. And the control part 11 calculates the composition ratio (budget allocation) of each measure which satisfy
  • the control unit 11 calculates the composition ratio of each measure that satisfies the result target and minimizes the advertising cost.
  • the control unit 11 calculates the composition ratio of each measure that satisfies the upper limit cost and maximizes the result. That is, the control unit 11 calculates the composition ratio of each measure that satisfies the condition of the result information (result target) or the cost information (budget) and maximizes the cost effectiveness.
  • the control unit 11 performs the calculation process using, for example, a Lagrange undetermined multiplier method.
  • the control unit 11 outputs a budget plan indicating the calculated composition ratio of each measure (step S304). For example, the control unit 11 outputs a budget plan indicating budget allocation for each measure in the budget table 174 on the planning screen. In addition, the control unit 11 outputs comparison data between the results and costs related to past advertising activities and the predicted results and costs related to the budget plan to the result comparison data 172 and the budget comparison data 173.
  • the control unit 11 receives a condition input for the composition ratio of each measure (step S305). Specifically, the control unit 11 receives an input of a change regarding the expected cost of each measure in the simulation column 1741 of the budget table 174. The control unit 11 determines whether or not to finalize the budget plan (step S306). For example, the control unit 11 determines whether or not an operation input to the confirmation button 175 displayed on the planning screen at the customer terminal 5 has been received. If it is determined not to finalize the budget plan (S306: NO), the control unit 11 returns the process to step S303. When receiving the composition ratio condition input in step S305, the control unit 11 calculates the composition ratio of each measure again including the condition and outputs a budget plan.
  • the control unit 11 acquires each piece of information such as advertisement information, result information, and cost information from each device (step S307). Thereby, the control part 11 collects information regarding the advertising activity of the client company based on the budget plan.
  • the control unit 11 refers to the acquired cost information, and calculates a digestion budget and the like related to the budget digest status (step S308). For example, the control unit 11 calculates the expenses spent until the present after the start of the advertising activity based on the budget plan for each measure. Further, the control unit 11 compares the expected cost related to the budget plan with the cost after the start of the advertising activity, and calculates the budget digestion rate for each measure.
  • the control unit 11 calculates the result of the advertising activity (step S309). For example, the control unit 11 counts the results (the number of purchases) from the start of the advertising activity to the present based on the proposal. Further, the control unit 11 compares the result target related to the budget plan with the result after the start of the advertising activity, and calculates the target achievement rate for the result of the advertising activity.
  • the control unit 11 outputs a progress screen showing the budget consumption status and the result of the advertising activity for the advertising activity based on the budget plan (step S310). Specifically, the control unit 11 displays information such as the cost and digestion rate calculated in step S308 and information such as the result and target achievement rate calculated in step S309 on the progress screen. More specifically, the control unit 11 displays, on the progress screen, rough data 181 indicating the budget consumption status and target achievement level, activity status data indicating the results of advertising activities (the number of products or services purchased) and costs for each measure in time series. 182 and a detailed table 183 showing details of the state of budget consumption are output.
  • the control unit 11 determines whether or not the scheduled period of the advertising activity has passed (step S311).
  • the scheduled period is an activity period of advertising activity set in advance, and is, for example, one month. If it is determined that the scheduled period has not elapsed (S311: NO), the control unit 11 returns the process to step S307. When it determines with the scheduled period having passed (S311: YES), the control part 11 complete
  • the server 1 executes various processes related to the planning screen and the progress screen as described above, the present embodiment is not limited to this.
  • the server 1 transmits generated correlation information or the like to the customer terminal 5, and the customer terminal 5 generates each screen, accepts a condition related to a budget plan, calculates a budget plan, calculates a digest budget related to progress, etc. May be performed. That is, the processing according to the present embodiment may be executed by a terminal device such as the customer terminal 5.

Abstract

Provided is a program and the like for supporting efficient advertisement activity. This program (P) is characterized by executing, in a computer, processing for acquiring advertisement information that indicates a product or service advertisement viewing status and results information that indicates advertisement activity results, and on the basis of said advertisement information and said results information, generating correlation information that indicates the correlation between said advertisement viewing status and said results. It is thus possible to perform a suitable analysis regarding advertisement activity and support efficient advertisement activity.

Description

プログラム、情報処理装置及び情報処理方法Program, information processing apparatus and information processing method
　本発明は、プログラム、情報処理装置及び情報処理方法に関する。 The present invention relates to a program, an information processing apparatus, and an information processing method.
　従来、商品又はサービスの広告活動について様々な評価手法が提案されている。例えば特許文献１では、広告活動に係る複数のプロモーション要素、アンケートにより収集した評価度、購入意欲度、及び商品又はサービスの販売シェアについて、プロモーション要素から販売シェアに至るまでの関係性を多段階の線形回帰により算出するプロモーション評価装置が開示されている。 Conventionally, various evaluation methods for advertising activities of products or services have been proposed. For example, in Patent Document 1, the relationship from a promotion element to a sales share is determined in a multi-stage manner regarding a plurality of promotion elements related to advertising activities, evaluations collected through questionnaires, purchase motivation degrees, and sales shares of products or services. A promotion evaluation device that calculates by linear regression is disclosed.
特開２００６－７２６４９号公報JP 2006-72649 A
　一方で、広告活動の評価分析には迅速性も要求される。特許文献１では消費者の評価度、購入意欲度等をアンケートにより収集しており、迅速性の要求を満たしているとは言い難い。 On the other hand, rapid evaluation is also required for evaluation and analysis of advertising activities. In Patent Document 1, the consumer's evaluation degree, purchase motivation degree, and the like are collected by a questionnaire, and it is difficult to say that the request for quickness is satisfied.
　本発明は斯かる事情によりなされたものであって、その目的とするところは、効率的な広告活動を支援するプログラム等を提供することにある。 The present invention has been made under such circumstances, and an object thereof is to provide a program or the like that supports efficient advertising activities.
　本発明に係るプログラムは、コンピュータに、商品又はサービスの広告視聴状況を示す広告情報と、広告活動の成果を示す成果情報とを取得し、前記広告情報及び成果情報に基づき、前記広告視聴状況と前記成果との相関関係を示す相関情報を生成する処理を実行させることを特徴とする。 The program according to the present invention acquires, in a computer, advertisement information indicating an advertisement viewing status of a product or service and result information indicating a result of an advertising activity, and based on the advertisement information and the result information, A process of generating correlation information indicating a correlation with the result is executed.
　本発明に係るプログラムは、前記広告情報及び成果情報に基づき、多段階の回帰分析を含む共分散構造分析を実行して前記相関情報を生成することを特徴とする。 The program according to the present invention is characterized in that the correlation information is generated by executing a covariance structure analysis including a multistage regression analysis based on the advertisement information and the result information.
　本発明に係るプログラムは、前記広告情報は、消費者の広告視聴態様が受動的である場合の受動視聴情報、能動的かつ間接的である場合の間接視聴情報、能動的かつ直接的である場合の直接視聴情報、及び前記消費者による前記商品又はサービスの確認の有無を示す確認情報を含み、前記成果情報は、前記商品又はサービスの購買状況を示す購買情報であり、前記受動視聴情報、間接視聴情報、直接視聴情報、確認情報、及び購買情報に含まれる各要素について回帰分析を行うことを特徴とする。 In the program according to the present invention, the advertisement information is passive viewing information when a consumer's advertisement viewing mode is passive, indirect viewing information when active and indirect, and active and direct. Direct viewing information and confirmation information indicating whether or not the consumer confirms the product or service, and the result information is purchase information indicating a purchase status of the product or service, and the passive viewing information, indirect Regression analysis is performed on each element included in viewing information, direct viewing information, confirmation information, and purchase information.
　本発明に係るプログラムは、前記相関情報に基づき、前記広告情報及び成果情報に含まれる要素間の相関関係を示すシーケンス画面を出力することを特徴とする。 The program according to the present invention is characterized by outputting a sequence screen indicating a correlation between elements included in the advertisement information and the result information based on the correlation information.
　本発明に係るプログラムは、前記広告活動に関する費用情報を取得し、前記広告情報、成果情報及び費用情報に基づき、前記相関情報を生成することを特徴とする。 The program according to the present invention is characterized by acquiring cost information related to the advertising activity and generating the correlation information based on the advertising information, result information, and cost information.
　本発明に係るプログラムは、前記成果情報又は費用情報について条件入力を受け付け、前記相関情報に基づき、前記広告活動に係る各広告施策について、受け付けた条件を満たす構成比を算出し、算出した前記各広告施策の構成比を示す予算案を出力することを特徴とする。 The program according to the present invention receives a condition input for the result information or cost information, calculates a composition ratio satisfying the received condition for each advertising measure related to the advertising activity based on the correlation information, It is characterized by outputting a budget plan indicating the composition ratio of the advertising measure.
　本発明に係るプログラムは、前記予算案に基づく前記広告活動について、予算の消化状況と、前記成果とを示す進捗画面を出力することを特徴とする。 The program according to the present invention is characterized by outputting a progress screen indicating a budget consumption status and the results for the advertising activity based on the budget plan.
　本発明に係る情報処理装置は、商品又はサービスの広告視聴状況を示す広告情報と、広告活動の成果を示す成果情報とを取得する取得部と、前記広告情報及び成果情報に基づき、前記広告視聴状況と前記成果との相関関係を示す相関情報を生成する生成部とを備えることを特徴とする。 The information processing apparatus according to the present invention includes an acquisition unit that acquires advertisement information indicating an advertisement viewing status of a product or service and result information indicating a result of an advertising activity, and the advertisement viewing based on the advertisement information and the result information. A generating unit that generates correlation information indicating a correlation between the situation and the result is provided.
　本発明に係る情報処理方法は、コンピュータに、商品又はサービスの広告視聴状況を示す広告情報と、広告活動の成果を示す成果情報とを取得し、前記広告情報及び成果情報に基づき、前記広告視聴状況と前記成果との相関関係を示す相関情報を生成する処理を実行させることを特徴とする。 The information processing method according to the present invention acquires, in a computer, advertisement information indicating an advertisement viewing status of a product or service and result information indicating a result of an advertisement activity, and based on the advertisement information and the result information, the advertisement viewing A process for generating correlation information indicating a correlation between a situation and the result is executed.
　本発明によれば、効率的な広告活動を支援することができる。 According to the present invention, efficient advertising activities can be supported.
情報処理システムの構成例を示す模式図である。It is a schematic diagram which shows the structural example of an information processing system. サーバの構成例を示すブロック図である。It is a block diagram which shows the structural example of a server. 広告ＤＢのレコードレイアウトの一例を示す説明図である。It is explanatory drawing which shows an example of the record layout of advertisement DB. 成果ＤＢのレコードレイアウトの一例を示す説明図である。It is explanatory drawing which shows an example of the record layout of result DB. 費用ＤＢのレコードレイアウトの一例を示す説明図である。It is explanatory drawing which shows an example of the record layout of expense DB. 広告評価の概要を説明するための説明図である。It is explanatory drawing for demonstrating the outline | summary of advertisement evaluation. 情報処理システムが実行する処理の概要を説明するための説明図である。It is explanatory drawing for demonstrating the outline | summary of the process which an information processing system performs. ダッシュボード画面の一例を示す説明図である。It is explanatory drawing which shows an example of a dashboard screen. 共分散構造分析を説明するための説明図である。It is explanatory drawing for demonstrating a covariance structure analysis. 回帰分析を説明するための説明図である。It is explanatory drawing for demonstrating regression analysis. 説明変数の選択処理について説明するための説明図である。It is explanatory drawing for demonstrating the selection process of an explanatory variable. ナビゲーション画面の一例を示す説明図である。It is explanatory drawing which shows an example of a navigation screen. サーバが実行する処理手順の一例を示すフローチャートである。It is a flowchart which shows an example of the process sequence which a server performs. 相関情報生成のサブルーチンの処理手順の一例を示すフローチャートである。It is a flowchart which shows an example of the process sequence of the subroutine of correlation information generation. 実施の形態２に係るアドストック期間に応じた統計分析について説明するための説明図である。10 is an explanatory diagram for explaining statistical analysis according to an ad stock period according to Embodiment 2. FIG. 実施の形態２に係る相関情報生成のサブルーチンの処理手順の一例を示すフローチャートである。10 is a flowchart illustrating an example of a processing procedure of a correlation information generation subroutine according to the second embodiment. 実施の形態３に係るプランニング画面の一例を示す説明図である。10 is an explanatory diagram illustrating an example of a planning screen according to Embodiment 3. FIG. 実施の形態３に係る進捗画面の一例を示す説明図である。12 is an explanatory diagram illustrating an example of a progress screen according to Embodiment 3. FIG. 実施の形態３に係るサーバが実行する処理手順の一例を示すフローチャートである。12 is a flowchart illustrating an example of a processing procedure executed by a server according to the third embodiment.
　以下、本発明をその実施の形態を示す図面に基づいて詳述する。
（実施の形態１）
　図１は、情報処理システムの構成例を示す模式図である。本実施の形態では、商品又はサービスを提供する顧客企業が実施する広告活動について、広告活動の宣伝効果に係る統計的な評価を行うサービスを実現するための情報処理システムを一例に説明を行う。情報処理システムは、情報処理装置１、視聴端末２、２、２…、顧客サーバ３、サービスセンタ４、顧客端末５等を含む。各装置は、インターネット等のネットワークＮを介して相互に情報の送受信を行う。 Hereinafter, the present invention will be described in detail with reference to the drawings illustrating embodiments thereof.
(Embodiment 1)
FIG. 1 is a schematic diagram illustrating a configuration example of an information processing system. In the present embodiment, an information processing system for realizing a service for performing a statistical evaluation related to the advertising effect of an advertising activity will be described as an example for an advertising activity carried out by a customer company that provides products or services. The information processing system includes an information processing apparatus 1, viewing terminals 2, 2, 2,..., A customer server 3, a service center 4, a customer terminal 5, and the like. Each device transmits and receives information to and from each other via a network N such as the Internet.
　情報処理装置１は、種々の情報を記憶すると共に、ネットワークＮを介して情報の送受信を行う情報処理装置である。情報処理装置１は、例えばサーバコンピュータ、パーソナルコンピュータ等である。本実施の形態において情報処理装置１はサーバコンピュータであるものとし、以下ではサーバ１と読み替える。サーバ１は、ネットワークＮを介して各種情報を取得し、広告活動の宣伝効果に係る統計分析を行う。当該処理について詳しくは後述する。 The information processing apparatus 1 is an information processing apparatus that stores various types of information and transmits and receives information via the network N. The information processing apparatus 1 is, for example, a server computer or a personal computer. In this embodiment, the information processing apparatus 1 is assumed to be a server computer, and will be read as the server 1 below. The server 1 acquires various types of information via the network N and performs statistical analysis related to the advertising effect of the advertising activity. This process will be described in detail later.
　視聴端末２は、広告を視聴等することが可能な端末装置であり、例えばパーソナルコンピュータ、スマートフォン、テレビジョン受信機等である。顧客サーバ３は、商品又はサービスを消費者に提供する顧客企業のサーバ装置であり、商品又はサービスの購買状況、広告活動に費やす費用等の各種情報を記憶している。なお、顧客サーバ３は例えばパーソナルコンピュータ等であってもよい。サービスセンタ４は例えばサーバ装置であり、テレビジョン受信機に係る視聴率の集計を行う。顧客端末５は顧客企業の情報処理端末であり、例えばパーソナルコンピュータである。顧客端末５は、サーバ１から統計分析の分析結果等に係る種々の情報を受信し、表示する。 The viewing terminal 2 is a terminal device that can view advertisements, and is, for example, a personal computer, a smartphone, a television receiver, or the like. The customer server 3 is a server device of a customer company that provides goods or services to consumers, and stores various information such as purchase status of goods or services, expenses spent for advertising activities, and the like. The customer server 3 may be a personal computer, for example. The service center 4 is a server device, for example, and aggregates the audience rating related to the television receiver. The customer terminal 5 is an information processing terminal of a customer company, for example, a personal computer. The customer terminal 5 receives and displays various information related to the analysis result of the statistical analysis from the server 1.
　図２は、サーバ１の構成例を示すブロック図である。サーバ１は、制御部１１、記憶部１２、通信部１３、大容量記憶装置１４を含む。
　制御部１１は、ＣＰＵ（Central Processing Unit）又はＭＰＵ（Micro-Processing 　Unit）等の演算処理装置を含む。制御部１１は、記憶部１２に記憶されたプログラムＰを読み出して実行することにより、サーバ１に係る種々の情報処理又は制御処理等を行う。 FIG. 2 is a block diagram illustrating a configuration example of the server 1. The server 1 includes a control unit 11, a storage unit 12, a communication unit 13, and a mass storage device 14.
The control unit 11 includes an arithmetic processing device such as a CPU (Central Processing Unit) or an MPU (Micro-Processing Unit). The control unit 11 reads out and executes the program P stored in the storage unit 12 to perform various information processing or control processing related to the server 1.
　記憶部１２は、ＲＡＭ（Random Access Memory）、ＲＯＭ（Read Only Memory）等のメモリ素子を含む。記憶部１２は、制御部１１が本実施の形態に係る処理を実行するために必要なプログラムＰ又はデータ等を記憶している。また、記憶部１２は、制御部１１が演算処理を実行するために必要なデータ等を一時的に記憶する。 The storage unit 12 includes memory elements such as RAM (Random Access Memory) and ROM (Read Only Memory). The storage unit 12 stores a program P or data necessary for the control unit 11 to execute the processing according to the present embodiment. The storage unit 12 temporarily stores data and the like necessary for the control unit 11 to execute arithmetic processing.
　通信部１３は通信に関する処理を行うための処理回路等を含み、ネットワークＮを介して視聴端末２、顧客サーバ３等と情報の送受信を行う。 The communication unit 13 includes a processing circuit for performing processing related to communication, and transmits / receives information to / from the viewing terminal 2, the customer server 3, and the like via the network N.
　大容量記憶装置１４は、例えばハードディスク等を含む大容量の記憶装置である。大容量記憶装置１４は、広告ＤＢ１４１、成果ＤＢ１４２、費用ＤＢ１４３等を記憶している。広告ＤＢ１４１は、広告の視聴状況を示す広告情報を記憶している。成果ＤＢ１４２は、広告活動の成果を示す成果情報を記憶しており、本実施の形態においては広告活動に係る商品又はサービスの購買状況を示す購買情報を記憶している。費用ＤＢ１４３は、広告活動に関する費用情報を記憶している。
　なお、本実施の形態において記憶部１２及び大容量記憶装置１４は一体の記憶装置として構成されていてもよい。また、大容量記憶装置１４は複数の記憶装置により構成されていてもよい。また、大容量記憶装置１４はサーバ１に接続された外部記憶装置であってもよい。 The large-capacity storage device 14 is a large-capacity storage device including, for example, a hard disk. The mass storage device 14 stores an advertisement DB 141, a result DB 142, a cost DB 143, and the like. The advertisement DB 141 stores advertisement information indicating the viewing status of advertisements. The result DB 142 stores result information indicating the result of the advertising activity. In the present embodiment, the result DB 142 stores purchase information indicating the purchase status of the product or service related to the advertising activity. The expense DB 143 stores expense information related to advertising activities.
In the present embodiment, the storage unit 12 and the mass storage device 14 may be configured as an integrated storage device. Further, the mass storage device 14 may be constituted by a plurality of storage devices. The mass storage device 14 may be an external storage device connected to the server 1.
　なお、本実施の形態においてサーバ１は上記の構成に限られず、例えば操作入力を受け付ける入力部、サーバ１に係る情報を表示する表示部等を含んでもよい。 In the present embodiment, the server 1 is not limited to the above configuration, and may include, for example, an input unit that receives an operation input, a display unit that displays information related to the server 1, and the like.
　図３は、広告ＤＢ１４１のレコードレイアウトの一例を示す説明図である。広告ＤＢ１４１は、顧客列、日付列、受動視聴列、間接視聴列、直接視聴列、確認列を含む。顧客列は、顧客企業を識別するためのＩＤを記憶している。日付列は、ＩＤと対応付けて、広告活動期間中の日付を記憶している。受動視聴列は、ＩＤ及び日付と対応付けて、後述する受動視聴情報を記憶している。具体的に受動視聴列は、オンライン上での広告ディスプレイのｉｍｐ数（impression、表示回数）、動画広告のｉｍｐ数、テレビジョンコマーシャルの視聴世帯数、一般的キーワード（例えば商品が自動車であれば「自動車」等）に基づきインターネット検索で表示されたリスティング広告のｉｍｐ数等を記憶している。なお、例えば図３に示すように、各受動視聴情報は広告態様に応じてグループ化（例えばインターネット上で広告表示する場合は「ｉｍｐ」のグループに分類）して記憶されている。以下の間接視聴列、直接視聴列、確認列も同様である。
　なお、受動視聴情報は上記に限定されるものではなく、例えばオウンドメディア（自社ホームページ、ブログ等）へのセッション数等も含まれる。なお、セッション数ではなく訪問数を記憶してもよい。また、図３では簡潔のため図示していないが、実際の受動視聴列には広告媒体別（例えば視聴端末２がパーソナルコンピュータであるか、スマートフォンであるか）に情報が記憶される。以下の間接視聴列、直接視聴列、確認列も同様である。 FIG. 3 is an explanatory diagram illustrating an example of a record layout of the advertisement DB 141. The advertisement DB 141 includes a customer column, a date column, a passive viewing column, an indirect viewing column, a direct viewing column, and a confirmation column. The customer column stores an ID for identifying a customer company. The date string stores the date during the advertising activity period in association with the ID. The passive viewing column stores passive viewing information described later in association with the ID and date. Specifically, the passive viewing column includes the number of imps (impressions) of online advertising displays, the number of imps of video advertisements, the number of households watching television commercials, and general keywords (for example, “ The number of imps of listing advertisements displayed by Internet search based on “automobile” and the like is stored. For example, as shown in FIG. 3, each passive viewing information is stored after being grouped according to the advertisement mode (for example, classified into “imp” group when displaying advertisement on the Internet). The same applies to the following indirect viewing column, direct viewing column, and confirmation column.
The passive viewing information is not limited to the above, and includes, for example, the number of sessions to owned media (in-house website, blog, etc.). Note that the number of visits may be stored instead of the number of sessions. Although not shown in FIG. 3 for the sake of brevity, information is stored for each advertisement medium (for example, whether the viewing terminal 2 is a personal computer or a smartphone) in the actual passive viewing sequence. The same applies to the following indirect viewing column, direct viewing column, and confirmation column.
　間接視聴列は、ＩＤ及び日付と対応付けて、後述する間接視聴情報を記憶している。具体的に間接視聴列は、ディスプレイ広告のクリック数、動画広告のクリック数、一般的キーワードに基づき検索表示されたリスティング広告のクリック数、指名キーワード（例えば具体的な商品名）に基づき検索表示されたリスティング広告のｉｍｐ数、リターゲティング（登録商標）広告のｉｍｐ数等を記憶している。なお、間接視聴情報は上記に限定されるものではなく、例えば外部サイトからオウンドメディアへの流入数等も含まれる。
　直接視聴列は、ＩＤ及び日付と対応付けて、後述する直接視聴情報を記憶している。具体的に直接視聴列は、指名キーワードに基づき検索表示されたリスティング広告のクリック数、リターゲティング広告のクリック数、オーガニック検索に基づくオウンドメディアへの流入数、ダイレクトトラフィックによるオウンドメディアへの流入数等を記憶している。
　確認列は、ＩＤ及び日付と対応付けて、後述する確認情報を記憶している。具体的に確認列は、商品又はサービスを販売するオンラインショッピング等のＷｅｂサイトへのセッション数等を記憶している。なお、確認情報は上記に限定されるものではなく、例えば上記Ｗｅｂサイトにおける個別のランディングページへのセッション数等も含まれる。 The indirect viewing column stores indirect viewing information described later in association with the ID and date. Specifically, the indirect viewing column is searched and displayed based on the number of clicks on display ads, the number of clicks on video ads, the number of clicks on listing ads searched based on general keywords, and the designated keywords (for example, specific product names). The number of imps of listing advertisements, the number of imps of retargeting (registered trademark) advertisements, and the like are stored. The indirect viewing information is not limited to the above, and includes, for example, the number of inflows from the external site to the owned media.
The direct viewing column stores direct viewing information described later in association with the ID and date. Specifically, the direct viewing column is the number of clicks on listing ads searched based on designated keywords, the number of clicks on retargeting ads, the number of inflows to owned media based on organic search, and the inflow to owned media due to direct traffic. I remember numbers.
The confirmation column stores confirmation information described later in association with the ID and date. Specifically, the confirmation column stores the number of sessions to a website such as online shopping that sells products or services. Note that the confirmation information is not limited to the above, and includes, for example, the number of sessions to individual landing pages on the Web site.
　なお、図３では簡潔のため図示していないが、広告ＤＢ１４１に記憶される広告情報は自社広告に関する情報だけでなく、例えば競合企業の広告活動に係る情報を含めてもよい。また、広告情報は視聴端末２を介した広告の視聴状況だけでなく、例えば販売店舗への来客数、交通広告等のように、視聴端末２を介しない広告視聴状況を示すものであってもよい。 Although not shown in FIG. 3 for the sake of brevity, the advertisement information stored in the advertisement DB 141 may include not only information related to the company advertisement but also information related to advertising activities of competitors, for example. Further, the advertisement information may indicate not only the viewing status of the advertisement via the viewing terminal 2, but also the advertisement viewing status not via the viewing terminal 2, such as the number of visitors to the sales store, traffic advertisements, and the like. Good.
　図４は、成果ＤＢ１４２のレコードレイアウトの一例を示す説明図である。成果ＤＢ１４２は、顧客列、日付列、購買列を含む。顧客列は、顧客企業のＩＤを記憶している。日付列は、ＩＤと対応付けて、広告活動期間中の日付を記憶している。購買列は、ＩＤ及び日付と対応付けて、顧客企業の商品又はサービスの購買状況を示す購買情報を記憶している。具体的に購買列は、例えばオンライン上での商品又はサービスの購買数、オフラインでの商品又はサービスの購買数（例えば店頭での売上数）等を記憶している。なお、購買情報は購買数ではなく売上金額等であってもよい。 FIG. 4 is an explanatory diagram showing an example of the record layout of the result DB 142. The result DB 142 includes a customer column, a date column, and a purchase column. The customer column stores the ID of the customer company. The date string stores the date during the advertising activity period in association with the ID. The purchase column stores purchase information indicating the purchase status of the product or service of the customer company in association with the ID and date. Specifically, the purchase column stores, for example, the number of purchases of goods or services online, the number of purchases of goods or services offline (for example, the number of sales at stores), and the like. Note that the purchase information may be sales amount instead of the number of purchases.
　図５は、費用ＤＢ１４３のレコードレイアウトの一例を示す説明図である。費用ＤＢ１４３は、顧客列、日付列、費用列を含む。顧客列は、顧客企業のＩＤを記憶している。日付列は、ＩＤと対応付けて、広告活動期間中の日付を記憶している。費用列は、ＩＤ及び日付と対応付けて、広告活動に関する費用情報を記憶している。具体的に費用列は、ディスプレイ広告、動画広告、テレビジョンコマーシャル、リスティング広告、リターゲティング広告、Ｗｅｂサイトでの広告等の、広告施策別の費用をそれぞれ記憶している。なお、費用情報は金額ではなく数量（例えばｉｍｐ数、クリック数等）で記憶しておき、サーバ１は当該数量と、各広告施策の単価とにより費用を算出するようにしてもよい。 FIG. 5 is an explanatory diagram showing an example of a record layout of the expense DB 143. The cost DB 143 includes a customer column, a date column, and a cost column. The customer column stores the ID of the customer company. The date string stores the date during the advertising activity period in association with the ID. The cost column stores cost information related to the advertising activity in association with the ID and date. Specifically, the expense column stores expenses for each advertising measure, such as display advertisements, video advertisements, television commercials, listing advertisements, retargeting advertisements, and advertisements on websites. Note that the cost information may be stored as a quantity (for example, the number of imps and the number of clicks) instead of the amount of money, and the server 1 may calculate the cost based on the quantity and the unit price of each advertising measure.
　図６は、広告評価の概要を説明するための説明図である。以下では情報処理システムが実行する処理の概要について説明する。なお、以下では説明の便宜のため、サーバ１の処理主体は制御部１１であるものとする。
　制御部１１は、顧客企業の広告活動について有効性を評価する処理を実行する。すなわち制御部１１は、広告の視聴状況を示す広告情報、及び広告活動の成果を示す成果情報に基づき、広告がどの程度の有効性をもって消費者の購買活動に繋がったかを評価する。 FIG. 6 is an explanatory diagram for explaining an outline of advertisement evaluation. Below, the outline | summary of the process which an information processing system performs is demonstrated. In the following, for convenience of explanation, it is assumed that the processing entity of the server 1 is the control unit 11.
The control part 11 performs the process which evaluates effectiveness about an advertising activity of a customer company. That is, the control unit 11 evaluates how effective the advertisement has led to the consumer's purchasing activity based on the advertisement information indicating the viewing status of the advertisement and the result information indicating the result of the advertising activity.
　制御部１１は、ネットワークＮを介して広告情報を取得し、広告ＤＢ１４１に記憶する。なお、制御部１１は顧客端末５等を介して広告情報の手動入力を受け付けてもよい。
　広告情報は、消費者が顧客企業の商品又はサービスをどの程度視聴したかを示す情報群である。具体的に広告情報は、いわゆるオンライン広告であるディスプレイ広告、動画広告、リスティング広告、リターゲティング広告等のｉｍｐ数及びクリック数、オウンドメディア又は所定のＷｅｂサイトへのセッション数、いわゆるオフライン広告であるテレビジョンコマーシャルの視聴世帯数等を含む時系列データである。制御部１１は、消費者の広告視聴の態様に応じて、広告情報を四段階に分類して広告ＤＢ１４１に記憶する。具体的に制御部１１は広告情報を、受動視聴情報、間接視聴情報、直接視聴情報、及び確認情報に分類して記憶する。 The control unit 11 acquires advertisement information via the network N and stores the advertisement information in the advertisement DB 141. The control unit 11 may accept manual input of advertisement information via the customer terminal 5 or the like.
The advertisement information is a group of information indicating how much the consumer has viewed the product or service of the customer company. Specifically, the advertisement information is the so-called online advertisements such as display advertisements, video advertisements, listing advertisements, retargeting advertisements, the number of imps and clicks, the number of sessions to the owned media or a predetermined website, so-called offline advertisements. Time-series data including the number of households viewing television commercials. The control unit 11 classifies the advertisement information into four stages and stores the advertisement information in the advertisement DB 141 according to the consumer's advertisement viewing mode. Specifically, the control unit 11 classifies and stores advertisement information into passive viewing information, indirect viewing information, direct viewing information, and confirmation information.
　受動視聴情報は、消費者が特定の商品又はサービスの広告視聴を目的とせずに、受動的な態様で広告を視聴する場合の広告情報である。例えばディスプレイ広告を考えた場合、消費者は視聴端末２において所定のＷｅｂサイトを閲覧した場合、消費者の意図に関わらずディスプレイ広告が表示され、消費者は受動的にディスプレイ広告を視聴することになる。本実施の形態に係る制御部１１は、上記のように受動的態様での広告視聴状況を受動視聴情報として判断する。 Passive viewing information is advertising information when a consumer views an advertisement in a passive manner without the purpose of viewing the advertisement of a specific product or service. For example, when a display advertisement is considered, when a consumer browses a predetermined website on the viewing terminal 2, the display advertisement is displayed regardless of the intention of the consumer, and the consumer passively views the display advertisement. Become. The control part 11 which concerns on this Embodiment judges the advertisement viewing condition in a passive aspect as mentioned above as passive viewing information.
　間接視聴情報は、消費者が特定の商品又はサービスの広告視聴を目的とはしていないが、能動的かつ間接的な態様で広告を視聴する場合の広告情報である。例えば顧客企業の商品が自動車である場合であって、消費者が検索エンジンにおいて「自動車」という一般的キーワードを検索した場合、検索結果に連動してリスティング広告が表示され、消費者は当該リスティング広告を視聴することになる。制御部１１は、上記のように消費者が特定の商品又はサービスの広告視聴を意図していないが、消費者が自らの興味に応じて情報検索をした結果、間接的に広告を視聴する場合の広告情報を間接視聴情報と判断する。 The indirect viewing information is advertising information when a consumer views an advertisement in an active and indirect manner, although the consumer is not intended to view the advertisement of a specific product or service. For example, if the product of a customer company is a car, and a consumer searches for a general keyword “car” in a search engine, a listing advertisement is displayed in conjunction with the search result, and the consumer Will be watched. In the case where the consumer 11 does not intend to view the advertisement of the specific product or service as described above, but the consumer indirectly views the advertisement as a result of the information search according to his / her interest. Is determined to be indirect viewing information.
　直接視聴情報は、消費者が特定の商品又はサービスの広告視聴を目的として能動的かつ直接的に広告を視聴する場合の広告情報である。例えば消費者が検索エンジンにおいて顧客企業の商品名に係る指名キーワードに基づき検索をした場合、当該商品に係るリスティング広告が表示され、消費者は当該リスティング広告を視聴することになる。制御部１１は、上記のように消費者が特定の商品又はサービスの広告視聴を意図して情報検索をした結果、直接的に広告を視聴する場合の広告情報を直接視聴情報と判断する。 Direct viewing information is advertising information when a consumer actively and directly views an advertisement for the purpose of viewing an advertisement for a specific product or service. For example, when a consumer searches on a search engine based on a nominated keyword associated with a product name of a customer company, a listing advertisement related to the product is displayed, and the consumer views the listing advertisement. As described above, the control unit 11 determines that the advertisement information when directly viewing the advertisement is the direct viewing information as a result of the information search intended for the consumer to view the advertisement of the specific product or service as described above.
　確認情報は、広告に係る商品又はサービスについて消費者による確認の有無を示す広告情報である。例えば端末２はリスティング広告への操作入力を受け付けた場合、当該リスティング広告に埋め込まれたリンク先（例えば商品又はサービスを販売するＷｅｂサイト）にアクセスし、商品又はサービスに関する情報を表示する。この結果、消費者は商品又はサービスに関する情報を視聴し、当該商品又はサービスの内容を確認することになる。サーバ１の制御部１１は、上記のように消費者が広告をきっかけに商品又はサービスを確認する場合の広告情報を確認情報と判断する。なお、上記は特定のＷｅｂページへのセッション数を一例として挙げたが、例えば確認情報は、実店舗への来客数などを含めてもよい。 Confirmation information is advertisement information indicating whether or not the product or service related to the advertisement is confirmed by the consumer. For example, when the terminal 2 receives an operation input to the listing advertisement, the terminal 2 accesses a link destination embedded in the listing advertisement (for example, a website that sells products or services) and displays information on the products or services. As a result, the consumer views information on the product or service and confirms the content of the product or service. The control unit 11 of the server 1 determines the advertisement information as confirmation information when the consumer confirms the product or service triggered by the advertisement as described above. In the above, the number of sessions to a specific Web page is taken as an example. However, for example, the confirmation information may include the number of visitors to an actual store.
　上記のように制御部１１は広告情報を受動視聴情報、間接視聴情報、直接視聴情報、確認情報に分類することで、消費者が広告を認知した段階、広告に係る商品又はサービスに興味を持っている段階、広告に係る商品又はサービスを調べる段階、広告に係る商品又はサービスを確認する段階で、消費者の広告視聴状況を区分する。さらに制御部１１は、顧客サーバ３から広告活動の成果を示す成果情報を受信する。成果情報は、顧客企業の広告による宣伝効果、すなわち成果を表す時系列データである。本実施の形態において成果情報は、商品又はサービスに関する購買情報であるものとする。購買情報は、顧客企業の商品又はサービスに係る売上等を示す時系列データである。図６において「認知」は受動視聴情報を示し、「興味」は間接視聴情報を示し、「検索等」は直接視聴情報を示し、「確認」は確認情報を示し、「成果」は購買情報を示す。図６に示すように、各情報は広告の具体的な視聴態様によってさらに細分化される。なお、図６の図示内容は例示であって、広告情報に係る各要素群はこれに限るものではない。制御部１１は後述する統計分析を行い、消費者が広告によってどのように「認知」から「興味」、「検索等」、「確認」へと誘導され、「成果（購買）」へと至ったかを評価する。 As described above, the control unit 11 classifies advertisement information into passive viewing information, indirect viewing information, direct viewing information, and confirmation information, so that the consumer is interested in the product or service related to the advertisement at the stage when the advertisement is recognized. The advertisement viewing status of the consumer is classified at the stage of checking the goods or services related to the advertisement and the stage of checking the goods or services related to the advertisement. Further, the control unit 11 receives result information indicating the result of the advertising activity from the customer server 3. The result information is time series data representing the advertising effect of the advertisement of the client company, that is, the result. In the present embodiment, the result information is purchase information related to a product or service. The purchase information is time-series data indicating sales etc. related to a product or service of a customer company. In FIG. 6, “recognition” indicates passive viewing information, “interest” indicates indirect viewing information, “search etc.” indicates direct viewing information, “confirmation” indicates confirmation information, and “result” indicates purchase information. Show. As shown in FIG. 6, each information is further subdivided according to a specific viewing mode of the advertisement. The illustrated content of FIG. 6 is an exemplification, and each element group related to the advertisement information is not limited to this. The control unit 11 performs a statistical analysis to be described later, and how the consumer is guided from “recognition” to “interest”, “search, etc.” and “confirmation” by the advertisement to reach “result (purchasing)”. To evaluate.
　また、制御部１１は顧客サーバ３から広告活動に関する費用情報を受信する。費用情報は、顧客企業が広告活動に費やした費用等を施策別に示す情報である。制御部１１は費用情報を費用ＤＢ１４３に記憶する。 Further, the control unit 11 receives cost information related to the advertising activity from the customer server 3. The cost information is information indicating, for each measure, the cost spent on the advertising activity by the client company. The control unit 11 stores cost information in the cost DB 143.
　図７は、情報処理システムが実行する処理の概要を説明するための説明図である。上記のように制御部１１は、広告情報、成果情報、費用情報等を、ネットワークＮを介して取得する。例えばディスプレイ広告のｉｍｐ数について、制御部１１は当該ディスプレイ広告をインターネット上で掲載するＡＳＰ（Affiliate Service Provider）サーバから当該ｉｍｐ数を自動的に取得する。また、広告情報に店舗来客数、交通広告、雑誌広告等の視聴端末２を介しない視聴状況を含める場合、例えば制御部１１は顧客端末５を介して手動入力を受け付ける。制御部１１は、取得した各情報をデータベースに蓄積することで、各情報の統合管理及び出力処理を行う。例えば制御部１１は、各情報を顧客端末５に出力する。さらに制御部１１は、取得した情報に基づいて統計分析を行い、広告活動、購買活動、必要費用等の各相関関係について統計分析を行い、分析結果を出力する処理を行う。 FIG. 7 is an explanatory diagram for explaining an outline of processing executed by the information processing system. As described above, the control unit 11 acquires advertisement information, result information, cost information, and the like via the network N. For example, for the display advertisement imp number, the control unit 11 automatically acquires the imp number from an ASP (Affiliate Service Provider) server that publishes the display advertisement on the Internet. Moreover, when including viewing conditions not via the viewing terminal 2 such as the number of store visitors, traffic advertisements, and magazine advertisements in the advertisement information, the control unit 11 accepts manual input via the customer terminal 5, for example. The controller 11 performs integrated management and output processing of each information by accumulating each acquired information in a database. For example, the control unit 11 outputs each information to the customer terminal 5. Further, the control unit 11 performs a statistical analysis based on the acquired information, performs a statistical analysis on each correlation such as an advertising activity, a purchasing activity, and a necessary cost, and performs a process of outputting an analysis result.
　図８は、ダッシュボード画面の一例を示す説明図である。制御部１１は、顧客端末５からの出力要求に応じて、広告ＤＢ１４１、成果ＤＢ１４２、及び費用ＤＢ１４３に記憶している各種情報に基づき、図８に示すダッシュボード画面を生成して当該端末に出力する。ダッシュボード画面は、広告ＤＢ１４１、成果ＤＢ１４２、及び費用ＤＢ１４３に蓄積された各情報を閲覧するための画面である。制御部１１は、広告媒体及び広告態様に応じて、各情報を表形式及びグラフ形式で出力する。例えば制御部１１は、広告情報、成果情報等に係る時系列データを折れ線又は棒グラフで示す組み合わせグラフ８１を表示する。なお、図８の組み合わせグラフ８１において左縦軸は広告のｉｍｐ数、クリック数等の数値を、右縦軸は「成果」に係る購買数を、横軸は時間軸を示す。なお、制御部１１は図８左上に示す変更欄８１ａへの選択入力により、右縦軸に係るパラメータを変更して組み合わせグラフ８１を出力する。
　また、例えば制御部１１は、情報選択欄８２ａを介して選択された情報（図８では費用情報）について広告施策毎の構成比を示す棒グラフ８２を表示する。
　さらに制御部１１は、種別選択欄８３ａ及び広告選択欄８３ｂを介して選択された広告施策について広告視聴状況、成果（購買状況）、費用等に係る集計結果を表示する集計表８３を表示する。例えば種別選択欄８３ａで「オンライン施策」が選択され、広告選択欄８３ｂで「リスティング広告」が選択された場合、制御部１１は広告ＤＢ１４１等を参照し、オンライン広告の一種であるリスティング広告について、ｉｍｐ数、クリック数、費用その他の情報を集計表８３に表示する。なお、図８中段の表に示すように、制御部１１はさらにオンライン広告又はオフライン広告に係る施策全体での集計結果を表示してもよい。以上より、制御部１１は取得した各種情報を統合、可視化することができ、顧客企業は広告活動の活動状況を迅速に確認することができる。 FIG. 8 is an explanatory diagram illustrating an example of a dashboard screen. In response to an output request from the customer terminal 5, the control unit 11 generates a dashboard screen illustrated in FIG. 8 based on various information stored in the advertisement DB 141, the result DB 142, and the expense DB 143, and outputs the dashboard screen to the terminal. To do. The dashboard screen is a screen for browsing each information accumulated in the advertisement DB 141, the result DB 142, and the expense DB 143. The control unit 11 outputs each information in a table format and a graph format according to the advertisement medium and the advertisement mode. For example, the control part 11 displays the combination graph 81 which shows the time series data which concern on advertisement information, result information, etc. with a line or a bar graph. In the combination graph 81 in FIG. 8, the left vertical axis represents the number of advertisements such as the number of imps and clicks, the right vertical axis represents the number of purchases related to “result”, and the horizontal axis represents the time axis. The control unit 11 outputs the combination graph 81 by changing the parameter related to the right vertical axis in accordance with the selection input to the change column 81a shown in the upper left of FIG.
Further, for example, the control unit 11 displays a bar graph 82 indicating the composition ratio for each advertising measure with respect to the information (cost information in FIG. 8) selected via the information selection column 82a.
Further, the control unit 11 displays a tabulation table 83 that displays a tabulation result relating to the advertisement viewing status, result (purchase status), cost, and the like for the advertisement measure selected via the type selection column 83a and the advertisement selection column 83b. For example, when “online measure” is selected in the type selection field 83a and “listing advertisement” is selected in the advertisement selection field 83b, the control unit 11 refers to the advertisement DB 141 and the like for listing advertisements that are a type of online advertisement. The number of imps, the number of clicks, the cost, and other information are displayed in the tabulation table 83. In addition, as shown in the table in the middle of FIG. 8, the control unit 11 may further display the total result of the entire measure related to the online advertisement or the offline advertisement. As described above, the control unit 11 can integrate and visualize the acquired various types of information, and the customer company can quickly confirm the activity status of the advertising activity.
　図９は、共分散構造分析を説明するための説明図である。上述のごとく、制御部１１は統合管理する各種情報の統計分析を行う。具体的に制御部１１は、受動視聴情報、間接視聴情報、直接視聴情報、確認情報、及び購買情報等に含まれる各要素について、要素間の相関関係を分析する。例えば制御部１１は、共分散構造分析による分析を行う。共分散構造分析は、複数要素間の関係を検討するための統計手段である。本実施の形態では、制御部１１は複数要素間について多段階の重回帰分析を繰り返し実行することで、各要素間の相関関係を示す相関情報を生成する。 FIG. 9 is an explanatory diagram for explaining the covariance structure analysis. As described above, the control unit 11 performs statistical analysis of various pieces of information that are integrated and managed. Specifically, the control unit 11 analyzes the correlation between elements for each element included in passive viewing information, indirect viewing information, direct viewing information, confirmation information, purchase information, and the like. For example, the control unit 11 performs analysis by covariance structure analysis. Covariance structure analysis is a statistical tool for examining the relationship between multiple elements. In the present embodiment, the control unit 11 repeatedly performs multi-stage multiple regression analysis between a plurality of elements, thereby generating correlation information indicating a correlation between the elements.
　具体的に制御部１１は、一の段階に係る情報から回帰分析に係る目的変数を選択すると共に、当該段階の前の段階に係る情報から回帰分析に係る説明変数を選択する。例えば図９において、「リスティング一般ワードｃｌｉｃｋ」に係る間接視聴情報を目的変数とし、「認知」に係る各受動視聴情報を説明変数として回帰分析を行う。回帰分析の内容については後述する。制御部１１は各要素間について回帰分析を繰り返し、相関情報を生成していく。なお、制御部１１は、例えば受動視聴情報と間接視聴情報との相関関係というように段階が連続する要素間の関係だけでなく、例えば受動視聴情報と直接視聴情報との相関関係というように、段階の連続性に関わらず各要素間の相関関係を分析してよい。 Specifically, the control unit 11 selects an objective variable related to the regression analysis from the information related to one stage, and selects an explanatory variable related to the regression analysis from the information related to the previous stage. For example, in FIG. 9, regression analysis is performed using indirect viewing information related to “listing general word click” as an objective variable and each passive viewing information related to “cognition” as an explanatory variable. Details of the regression analysis will be described later. The control unit 11 repeats regression analysis between each element to generate correlation information. Note that the control unit 11 not only has a relationship between successive elements such as a correlation between passive viewing information and indirect viewing information, but also a correlation between passive viewing information and direct viewing information, for example. The correlation between each element may be analyzed regardless of the continuity of the stages.
　図１０は、回帰分析を説明するための説明図である。図１０のグラフにおいて、ｘ軸は指名ワードに基づくリスティング広告のクリック数（以下、リスティング指名クリック数という）を示し、ｙ軸がオンラインでの商品の購買数を示している。図１０のプロットはリスティング指名クリック数及び商品の購買数に係る実際の取得データの分布を、傾斜線ｙ＝ａｘ＋ｂはリスティング指名クリック数と購買数との相関関係を示す近似直線を、それぞれ概念的に示している。リスティング指名クリック数（直接視聴情報）を説明変数ｘとし、購買数（購買情報）を目的変数ｙとした場合、制御部１１は回帰分析により図１０に示す近似直線を算出する。 FIG. 10 is an explanatory diagram for explaining the regression analysis. In the graph of FIG. 10, the x-axis represents the number of clicks on the listing advertisement based on the designated word (hereinafter referred to as the number of clicks on the designated listing), and the y-axis represents the number of online product purchases. The plot of FIG. 10 shows the distribution of actual acquired data related to the number of clicks for listing designation and the number of purchases of products, and the inclined line y = ax + b is a conceptual straight line representing the correlation between the number of clicks for listing designation and the number of purchases. It shows. When the number of listing designation clicks (direct viewing information) is an explanatory variable x and the number of purchases (purchase information) is an objective variable y, the control unit 11 calculates an approximate straight line shown in FIG. 10 by regression analysis.
　具体的に制御部１１は、選択した説明変数ｘ及び目的変数ｙについて最小二乗法を適用し、説明変数ｘと目的変数ｙとの相関関係を表す近似直線ｙ＝ａｘ＋ｂの傾き（係数）ａ及び切片（定数）ｂを算出する。図１０において傾斜線の傾きａが目的変数ｙに対する説明変数ｘの影響力を示し、傾斜線ｙ＝ａｘ＋ｂとプロットとの誤差が相関の強さを示している。制御部１１は、誤差の二乗和が最小となる傾きａ及び切片ｂを算出し、相関関係に係る近似直線ｙ＝ａｘ＋ｂを求める。制御部１１は各種情報に応じて説明変数及び目的変数を入れ替えて回帰分析を繰り返し、統計上の相関関係を示す相関情報を生成していく。 Specifically, the control unit 11 applies the least square method to the selected explanatory variable x and the objective variable y, and the slope (coefficient) a of the approximate line y = ax + b representing the correlation between the explanatory variable x and the objective variable y and The intercept (constant) b is calculated. In FIG. 10, the slope a of the slope line indicates the influence of the explanatory variable x on the objective variable y, and the error between the slope line y = ax + b and the plot indicates the strength of the correlation. The control unit 11 calculates a slope a and an intercept b that minimize the sum of squared errors, and obtains an approximate straight line y = ax + b related to the correlation. The control unit 11 replaces the explanatory variable and the objective variable according to various information, repeats the regression analysis, and generates correlation information indicating a statistical correlation.
　なお、図１０では説明の便宜のため、２変数の場合、すなわち単回帰の場合を説明したが、実際には３次以上の多変数により統計処理を実行する。すなわち制御部１１は、一の目的変数ｙに対して複数の説明変数ｘ１、ｘ２、…を規定し、３次以上の重回帰分析を行う。これにより、例えば図９で示したように、制御部１１は一の目的変数ｙ（例えば「興味」に係る「リスティング一般ワードｃｌｉｃｋ」）と複数の説明変数ｘ１、ｘ２、ｘ３（例えば「認知」に係る「ディスプレイ広告ｉｍｐ」「動画広告ｉｍｐ」「リスティング一般ワードｉｍｐ」）との相関関係、詳しくは近似式ｙ＝ａ１ｘ１＋ａ２ｘ２＋ａ３ｘ３＋ｂを求めることができる。なお、例えば間接視聴情報を介して受動視聴情報が直接視聴情報に及ぼす間接効果のように、三段階以上の要素に係る相関関係を考える場合、制御部１１は説明変数ｘ１を目的変数に置き換え、近似式ｘ１＝ａ′１ｘ′１＋ａ′２ｘ′２＋…＋ｂ′を算出することで間接効果を分析することができる。 For convenience of explanation, FIG. 10 illustrates the case of two variables, that is, the case of simple regression, but in practice, statistical processing is executed with multiple variables of third order or higher. That is, the control unit 11 defines a plurality of explanatory variables x1, x2,... With respect to one objective variable y, and performs a third or higher order multiple regression analysis. As a result, for example, as illustrated in FIG. 9, the control unit 11 performs one objective variable y (for example, “listing general word click” related to “interest”) and a plurality of explanatory variables x1, x2, x3 (for example, “recognition”). (Display advertisement imp "," moving picture advertisement imp "," listing general word imp "), and in detail, an approximate expression y = a1x1 + a2x2 + a3x3 + b can be obtained. Note that, for example, when considering the correlation related to three or more elements, such as the indirect effect of passive viewing information directly on viewing information via indirect viewing information, the control unit 11 replaces the explanatory variable x1 with an objective variable, The indirect effect can be analyzed by calculating the approximate expression x1 = a′1x′1 + a′2x′2 +... + B ′.
　また、上記では広告情報及び成果情報の相関関係を分析する処理について述べたが、制御部１１は併せて、費用情報とその他の情報との相関関係も分析する。すなわち制御部１１は、広告活動の費用対効果等についても相関関係を分析する。制御部１１は、分析結果である相関情報を大容量記憶装置１４に記憶しておく。 In the above description, the processing for analyzing the correlation between the advertisement information and the result information has been described. However, the control unit 11 also analyzes the correlation between the cost information and other information. That is, the control unit 11 also analyzes the correlation with respect to the cost effectiveness of the advertising activity. The control unit 11 stores the correlation information that is the analysis result in the mass storage device 14.
　図１１は、説明変数の選択処理について説明するための説明図である。上記では共分散構造分析の概要について述べたが、制御部１１は当該分析処理を実行する場合、説明変数に採用する情報を適宜に選択して分析処理を行う。 FIG. 11 is an explanatory diagram for explaining an explanatory variable selection process. Although the outline of the covariance structure analysis has been described above, the control unit 11 performs the analysis process by appropriately selecting information to be used as the explanatory variable when executing the analysis process.
　具体的に制御部１１は、一の目的変数を選択すると共に、該目的変数に対する説明変数の候補（以下では説明の便宜のため「候補変数」という）を特定する。なお、図１１では候補変数を広告態様別（ＴＶ、Ｗｅｂ、動画）の広告情報とし、目的変数を購買情報（売上）としてある。なお、記憶部１２には統計分析に係る分析方針を規定するファイルが記憶されており、制御部１１は当該ファイルを参照して候補変数を特定するものとする。制御部１１は、当該候補変数から、目的変数を説明するために最適な一又は複数の説明変数を選択する。 Specifically, the control unit 11 selects one objective variable and specifies a candidate for an explanatory variable for the objective variable (hereinafter referred to as “candidate variable” for convenience of explanation). In FIG. 11, the candidate variable is advertisement information for each advertisement mode (TV, Web, video), and the objective variable is purchase information (sales). The storage unit 12 stores a file that defines an analysis policy related to statistical analysis, and the control unit 11 refers to the file to identify candidate variables. The control unit 11 selects one or a plurality of explanatory variables optimal for explaining the objective variable from the candidate variables.
　まず制御部１１は、当該候補変数から一の変数を説明変数として選択し、回帰分析を実行する。図１１の例では、まず制御部１１は「ＴＶ」を説明変数として回帰分析を実行する。制御部１１は、当該回帰分析に係る分析結果に基づき、当該説明変数の妥当性を判断するための指標を算出する。具体的に制御部１１は、以下の式（１）により指標値ＣＦ（Candidate Factor）を算出する。 First, the control unit 11 selects one variable from the candidate variables as an explanatory variable, and executes regression analysis. In the example of FIG. 11, the control unit 11 first executes regression analysis with “TV” as an explanatory variable. The control unit 11 calculates an index for determining the validity of the explanatory variable based on the analysis result related to the regression analysis. Specifically, the control unit 11 calculates an index value CF (Candidate Factor) by the following equation (1).
　ＣＦ＝１－（１－Ｒ２ａｄｊ）×ｐ×ｓｅ…（１） CF = 1- (1-R2adj) × p × se ... (1)
　Ｒ２ａｄｊは回帰分析に係る自由度調整済み決定係数、ｐはＰ値、ｓｅは標準誤差である。制御部１１はまず、説明変数を「ＴＶ」として指標値ＣＦを算出する。また、制御部１１は上記の回帰分析に用いた候補変数「ＴＶ」を、最終的に説明変数として採用する採用候補として一時的に設定しておく。なお、以下では説明の便宜のため、設定した当該説明変数を「採用候補変数」という。 R2adj is a coefficient of freedom adjusted for regression analysis, p is a P value, and se is a standard error. First, the control unit 11 calculates the index value CF with the explanatory variable “TV”. Further, the control unit 11 temporarily sets the candidate variable “TV” used in the regression analysis as an adoption candidate to be finally adopted as an explanatory variable. Hereinafter, for convenience of explanation, the set explanatory variable is referred to as an “adoption candidate variable”.
　次に制御部１１は、採用候補変数以外に、別の候補変数を新たに説明変数に追加する。そして制御部１１は、採用候補変数と、新たに追加した候補変数とを説明変数として、再度回帰分析を実行する。図１１の例では、制御部１１は採用候補変数「ＴＶ」に別の候補変数「Ｗｅｂ」を加えて回帰分析を実行する。制御部１１は当該分析結果に基づき、上記と同様に指標値ＣＦを算出する。 Next, the control unit 11 adds another candidate variable to the explanatory variable in addition to the adoption candidate variable. Then, the control unit 11 executes the regression analysis again using the adoption candidate variable and the newly added candidate variable as explanatory variables. In the example of FIG. 11, the control unit 11 performs a regression analysis by adding another candidate variable “Web” to the adoption candidate variable “TV”. Based on the analysis result, the control unit 11 calculates the index value CF in the same manner as described above.
　制御部１１は、前回算出した指標値ＣＦ（説明変数が「ＴＶ」）と、新たに候補変数を追加して算出した今回の指標値ＣＦ（説明変数が「ＴＶ」「Ｗｅｂ」）とを比較し、今回の指標値ＣＦが前回の指標値ＣＦよりも大きいか否かを判定する。今回の指標値ＣＦの方が大きいと判定した場合、制御部１１は、今回の回帰分析に係る説明変数を採用候補変数に設定する。一方で、今回の指標値ＣＦの方が小さいと判定した場合、制御部１１は前回の回帰分析に係る説明変数を引き続き採用候補変数に設定しておく。図１１の例では、制御部１１は今回算出した指標値ＣＦ（説明変数が「ＴＶ」「Ｗｅｂ」）の方が小さいと判定し、引き続き変数「ＴＶ」を採用候補変数に設定しておく。 The control unit 11 compares the previously calculated index value CF (explanation variable is “TV”) with the current index value CF (explanation variable is “TV” “Web”) calculated by newly adding a candidate variable. Then, it is determined whether or not the current index value CF is larger than the previous index value CF. When it is determined that the current index value CF is larger, the control unit 11 sets an explanatory variable related to the current regression analysis as an adoption candidate variable. On the other hand, when it is determined that the current index value CF is smaller, the control unit 11 continues to set the explanatory variable related to the previous regression analysis as the adoption candidate variable. In the example of FIG. 11, the control unit 11 determines that the index value CF (the explanatory variable is “TV” “Web”) calculated this time is smaller, and continues to set the variable “TV” as an adoption candidate variable.
　制御部１１は、同様の処理を繰り返す。例えば図１１に示すように、制御部１１は最初に採用候補とした「ＴＶ」について、組み合わせる他の候補変数を順々に入れ替えて回帰分析を行い、指標値ＣＦの大小を比較する。当初採用候補とした「ＴＶ」について全ての組み合わせを検討した場合、図１１に示すように、制御部１１は一度説明変数をクリアし、別の候補変数「Ｗｅｂ」を新たな採用候補に選択して同様の処理を行う。そして制御部１１は、指標値ＣＦが最大、すなわち分析結果として最適な説明変数の組み合わせ（図１１では「ＴＶ」「動画」）を選択する。以上の選択処理により、制御部１１は説明変数の候補に規定されている「Ｗｅｂ」は相関性が低いものとして除外し、「ＴＶ」「動画」を説明変数に採用する。 The control unit 11 repeats the same processing. For example, as illustrated in FIG. 11, the control unit 11 performs regression analysis on “TV”, which is the first candidate for adoption, by sequentially replacing other candidate variables to be combined, and compares the index values CF in magnitude. When all combinations of “TV” as the initial employment candidate are considered, as illustrated in FIG. 11, the control unit 11 once clears the explanatory variable and selects another candidate variable “Web” as a new employment candidate. The same processing is performed. Then, the control unit 11 selects the combination of explanatory variables having the maximum index value CF, that is, the optimal analysis result (“TV” and “moving image” in FIG. 11). Through the above selection process, the control unit 11 excludes “Web” defined as the explanatory variable candidate as having low correlation, and adopts “TV” and “moving image” as the explanatory variable.
　図１２は、ナビゲーション画面の一例を示す説明図である。制御部１１は、顧客端末５からの出力要求に応じて、上記一連の処理によって生成した相関情報を図１２に示すナビゲーション画面に出力する。ナビゲーション画面は、「認知」に係る受動視聴情報、「興味」に係る間接視聴情報、「検索＆リタゲ」に係る直接視聴情報、「確認」に係る確認情報、及び「成果」に係る成果情報（購買情報）との相関関係を、シーケンス図により示す画面である。 FIG. 12 is an explanatory diagram showing an example of a navigation screen. In response to the output request from the customer terminal 5, the control unit 11 outputs the correlation information generated by the series of processes to the navigation screen shown in FIG. The navigation screen includes passive viewing information related to “recognition”, indirect viewing information related to “interest”, direct viewing information related to “Search & Retage”, confirmation information related to “confirmation”, and result information related to “result” ( It is a screen which shows a correlation with purchase information) with a sequence diagram.
　図１２に示すように、上記の各種情報はボックス１２１又は詳細ボックス１２２で示され、各情報の相関関係が実線により示される。ボックス１２１は広告情報又は成果情報に含まれる要素の概要を示す表示欄であり、詳細ボックス１２２はボックス１２１で示す要素をより細分化した要素を示す表示欄である。ボックス１２１に係る各要素は、例えば各要素に係る広告媒体（視聴端末２の種別）が「ＰＣ」「ＳＰ」等のアイコンにより、広告態様（各種別の視聴端末２で視聴される広告の態様）がボックス内のテキストにより示される。例えば制御部１１は、顧客端末５を介して一のボックス１２１への選択入力を受け付ける。選択入力を受け付けた場合、制御部１１は、選択されたボックス１２１と他のボックス１２１との相関関係を実線又は数値により表示する。例えば制御部１１は、各ボックス１２１を繋ぐ実線の太さで影響力の大きさを、実線に付した数値で目的変数に対する説明変数の影響力の内訳を示す。 As shown in FIG. 12, the above-described various information is indicated by a box 121 or a detail box 122, and the correlation between the information is indicated by a solid line. The box 121 is a display field that shows an outline of the elements included in the advertisement information or the result information, and the detail box 122 is a display field that shows elements obtained by further subdividing the elements shown in the box 121. Each element related to the box 121 includes, for example, an advertisement mode (a mode of an advertisement viewed on various types of viewing terminals 2) by an icon such as “PC” or “SP” for the advertising medium (type of the viewing terminal 2) related to each element. ) Is indicated by the text in the box. For example, the control unit 11 accepts a selection input to one box 121 via the customer terminal 5. When the selection input is received, the control unit 11 displays the correlation between the selected box 121 and the other box 121 by a solid line or a numerical value. For example, the control unit 11 indicates the magnitude of the influence by the thickness of the solid line connecting the boxes 121, and shows the breakdown of the influence of the explanatory variable on the objective variable by the numerical value attached to the solid line.
　例えば制御部１１はナビゲーション画面において、各要素の概要、すなわちボックス１２１だけでなく、ボックス１２１よりも詳細な詳細ボックス１２２についても相関関係を示す。顧客端末５は、各種情報を概括的に示す各ボックス１２１に対する指定入力を受け付け、指定内容をサーバ１に通知する。当該指定内容を受信した場合、サーバ１の制御部１１は、指定されたボックス１２１について、より詳細な詳細ボックス１２２を顧客端末５に出力する。例えば「興味」における「Ｌｉｓｔ指名ｃｌｉｃｋ」のボックス１２１に対して操作入力を受け付けた場合、制御部１１は詳細ボックス１２２をプルダウン形式で表示し、「興味」に係る各広告と、他段階の広告との相関関係を表示する。これにより、顧客企業は自社の広告活動について相関関係の概要を把握できると共に、詳細な関係についても簡単に把握することができる。 For example, on the navigation screen, the control unit 11 shows the correlation of not only the outline of each element, that is, the detailed box 122 more detailed than the box 121 as well as the box 121. The customer terminal 5 accepts a designation input for each box 121 generally showing various information, and notifies the server 1 of the designated content. When the specified content is received, the control unit 11 of the server 1 outputs a more detailed detail box 122 for the specified box 121 to the customer terminal 5. For example, when an operation input is received for the box 121 of “List nomination click” in “interest”, the control unit 11 displays a detailed box 122 in a pull-down format, and each advertisement related to “interest” and advertisements in other stages Display the correlation. As a result, the client company can grasp the outline of the correlation with respect to its advertising activity and can easily grasp the detailed relation.
　さらに制御部１１は、図１２下側に示す貢献値グラフ１２４において、選択されたボックス１２１に対する他のボックス１２１又は詳細ボックス１２２の貢献値をグラフ表示する。なお、貢献値グラフ１２４において縦軸が貢献値（パーセント）を示す。ボックス１２１で示す各要素の貢献値は、回帰分析により求めた近似式ｙ＝ａ１ｘ１＋ａ２ｘ２＋ａ３ｘ３…＋ｂの各項ａ１ｘ１、ａ２ｘ２、ａ３ｘ３…に所定の係数を乗算することで求められる。例えば制御部１１は、他のボックス１２１又は詳細ボックス１２２の貢献値をそれぞれ棒グラフにより示す。一の棒グラフへのマウス入力を受け付けた場合、図１２に示すように、制御部１１は当該棒グラフに係るボックス１２１又は詳細ボックス１２２の要素名（図１２では「Ｙ社リスティング一般ワード」）を表示する。 Further, the control unit 11 displays the contribution value of the other box 121 or the detail box 122 with respect to the selected box 121 in a graph in the contribution value graph 124 shown in the lower side of FIG. In the contribution value graph 124, the vertical axis represents the contribution value (percentage). The contribution value of each element shown in the box 121 is obtained by multiplying each term a1x1, a2x2, a3x3... Of the approximate expression y = a1x1 + a2x2 + a3x3. For example, the control part 11 shows the contribution value of the other box 121 or the detail box 122 with a bar graph, respectively. When the mouse input to one bar graph is received, as shown in FIG. 12, the control unit 11 displays the element name of the box 121 or the detailed box 122 related to the bar graph (“Y company listing general word” in FIG. 12). To do.
　また、制御部１１は併せて図１２右下に示す組み合わせグラフ８１を表示する。組み合わせグラフ８１は、広告情報及び成果情報に係る時系列データを示すグラフであり、図８で示した組み合わせグラフ８１と同様のグラフである。これにより、顧客企業はナビゲーション画面において広告活動の分析結果を閲覧しながら、併せて実際の集計結果を確認することができる。 The control unit 11 also displays a combination graph 81 shown at the lower right of FIG. The combination graph 81 is a graph showing time series data related to the advertisement information and the result information, and is the same graph as the combination graph 81 shown in FIG. Thereby, the customer company can confirm the actual totaling result while browsing the analysis result of the advertising activity on the navigation screen.
　図１３は、サーバ１が実行する処理手順の一例を示すフローチャートである。図１３に基づいて、サーバ１が実行する処理内容について説明する。
　サーバ１の制御部１１は、広告の視聴状況を示す広告情報、広告活動の成果を示す成果情報、及び広告活動に関する費用情報を取得する（ステップＳ１１）。例えば広告情報は、上述の受動視聴情報、間接視聴情報、直接視聴情報、及び確認情報を含む。また、例えば成果情報は、商品又はサービスの購買情報である。制御部１１は各情報を、ネットワークＮを介して取得する。なお、制御部１１は手動入力により各情報を取得するようにしてもよい。制御部１１は、取得した広告情報を広告ＤＢ１４１に、成果情報を成果ＤＢ１４２に、費用情報を費用ＤＢ１４３に、それぞれ日付と対応付けて記憶する。 FIG. 13 is a flowchart illustrating an example of a processing procedure executed by the server 1. Based on FIG. 13, the processing content which the server 1 performs is demonstrated.
The control unit 11 of the server 1 acquires advertisement information indicating the viewing status of the advertisement, result information indicating the result of the advertisement activity, and cost information related to the advertisement activity (step S11). For example, the advertisement information includes the above-described passive viewing information, indirect viewing information, direct viewing information, and confirmation information. Further, for example, the result information is purchase information of goods or services. The control unit 11 acquires each information via the network N. Note that the control unit 11 may acquire each information by manual input. The control unit 11 stores the acquired advertisement information in the advertisement DB 141, the result information in the result DB 142, and the cost information in the cost DB 143 in association with the date.
　制御部１１は、ダッシュボード画面を顧客端末５に出力するか否かを判定する（ステップＳ１２）。例えば制御部１１は、顧客端末５から出力要求を受け付けたか否かを判定する。ダッシュボード画面を出力すると判定した場合（Ｓ１２：ＹＥＳ）、制御部１１は広告ＤＢ１４１、成果ＤＢ１４２、及び費用ＤＢ１４３を参照し、広告媒体及び広告態様に応じた広告情報、費用情報、及び成果情報を示すダッシュボード画面を生成して顧客端末５に出力する（ステップＳ１３）。具体的に制御部１１は、広告情報等の時系列データを示す組み合わせグラフ８１、費用情報等について広告施策毎の構成比を示す棒グラフ８２、各広告施策について広告視聴状況、購買状況、費用等の詳細な集計結果を表示する集計表８３等を含むダッシュボード画面を生成し、出力する。 The control unit 11 determines whether or not to output the dashboard screen to the customer terminal 5 (step S12). For example, the control unit 11 determines whether an output request has been received from the customer terminal 5. When it is determined that the dashboard screen is output (S12: YES), the control unit 11 refers to the advertisement DB 141, the result DB 142, and the cost DB 143 to obtain the advertisement information, cost information, and result information according to the advertisement medium and the advertisement mode. A dashboard screen is generated and output to the customer terminal 5 (step S13). Specifically, the control unit 11 includes a combination graph 81 indicating time series data such as advertisement information, a bar graph 82 indicating a composition ratio of each advertisement measure with respect to cost information, etc., advertisement viewing status, purchase status, cost, etc. for each advertisement measure. A dashboard screen including a summary table 83 and the like for displaying detailed summary results is generated and output.
　ダッシュボード画面を出力しないと判定した場合（Ｓ１２：ＮＯ）、又はステップＳ１３の処理を行った後で、制御部１１は、所定の分析期間が経過したか否かを判定する（ステップＳ１４）。分析期間は、例えば広告活動を実施する期間である。なお、制御部１１は分析期間について、広告活動期間に関わらず設定変更を受け付けるようにしてもよい。分析期間が経過していないと判定した場合（Ｓ１４：ＮＯ）、制御部１１は処理をステップＳ１１に戻す。分析期間が経過したと判定した場合（Ｓ１４：ＹＥＳ）、制御部１１は、広告情報、成果情報、及び費用情報に基づき、広告視聴状況、成果、及び広告宣伝費用の相関関係を示す相関情報を生成する処理を実行する（ステップＳ１５）。具体的に制御部１１は、各情報に基づき、多段階の回帰分析を含む共分散構造分析を実行して相関情報を生成する。制御部１１は、生成した相関情報に基づき、広告情報及び成果情報に含まれる要素間の相関関係を示すナビゲーション画面を顧客端末５に出力する（ステップＳ１６）。具体的に制御部１１は、顧客端末５から出力要求を受け付けた場合、受動視聴情報、間接視聴情報、直接視聴情報、確認情報及び購買情報の段階に係る各要素の相関関係をシーケンス図により示すナビゲーション画面を生成し、顧客端末５に出力する。例えば図１２で示したように、制御部１１はナビゲーション画面において各情報に含まれる要素の概要をボックス１２１で示し、各ボックス１２１を繋ぐ実線により要素間の影響力の大きさを示すと共に、一の要素について他の複数要素による影響力の内訳を数値表示する。 When it is determined that the dashboard screen is not output (S12: NO), or after performing the process of step S13, the control unit 11 determines whether a predetermined analysis period has elapsed (step S14). The analysis period is a period during which, for example, an advertising activity is performed. Note that the control unit 11 may accept a setting change for the analysis period regardless of the advertisement activity period. When it determines with the analysis period not having passed (S14: NO), the control part 11 returns a process to step S11. When it determines with the analysis period having passed (S14: YES), the control part 11 shows the correlation information which shows the correlation of advertisement viewing condition, a result, and advertising expense based on advertisement information, result information, and expense information. The process to generate is executed (step S15). Specifically, the control unit 11 generates a correlation information by executing a covariance structure analysis including a multistage regression analysis based on each information. The control part 11 outputs the navigation screen which shows the correlation between the elements contained in advertisement information and result information based on the produced | generated correlation information to the customer terminal 5 (step S16). Specifically, when the control unit 11 receives an output request from the customer terminal 5, the correlation of each element related to the stages of passive viewing information, indirect viewing information, direct viewing information, confirmation information, and purchase information is shown by a sequence diagram. A navigation screen is generated and output to the customer terminal 5. For example, as shown in FIG. 12, the control unit 11 shows an outline of elements included in each piece of information on the navigation screen by a box 121, shows the magnitude of influence between elements by a solid line connecting the boxes 121, and A numerical display of the breakdown of the influence of other elements on the element of.
　制御部１１は顧客端末５を介して、ナビゲーション画面で示す各要素について指定入力を受け付けたか否かを判定する（ステップＳ１７）。例えば制御部１１は、顧客端末５においてボックス１２１への操作入力を受け付けたか否かを判定する。指定入力を受け付けていないと判定した場合（Ｓ１７：ＮＯ）、制御部１１は処理をステップＳ１９に移行する。指定入力を受け付けたと判定した場合（Ｓ１７：ＹＥＳ）、制御部１１は、指定された要素に関してより詳細な要素を出力すると共に、当該詳細な要素と他の要素との相関関係をナビゲーション画面上に出力する（ステップＳ１８）。例えば制御部１１は、指定されたボックス１２１について詳細ボックス１２２を出力し、併せて他の要素との相関関係を実線等により出力する。なお、上述のごとく、制御部１１は指定された要素について、他の要素の貢献値を示す貢献値グラフ１２４を併せて表示する。 The control unit 11 determines whether or not a designation input has been received for each element indicated on the navigation screen via the customer terminal 5 (step S17). For example, the control unit 11 determines whether or not an operation input to the box 121 is received at the customer terminal 5. When it determines with not receiving the designation | designated input (S17: NO), the control part 11 transfers a process to step S19. When it is determined that the designated input has been received (S17: YES), the control unit 11 outputs a more detailed element with respect to the designated element and displays the correlation between the detailed element and other elements on the navigation screen. Output (step S18). For example, the control unit 11 outputs a detailed box 122 for the designated box 121 and outputs a correlation with other elements by a solid line or the like. As described above, the control unit 11 also displays the contribution value graph 124 indicating the contribution values of other elements for the designated element.
　制御部１１は、ナビゲーション画面の出力を終了するか否かを判定する（ステップＳ１９）。例えば制御部１１は、顧客端末５において所定の終了指示に係る入力を受け付けたか否かを判定する。終了しないと判定した場合（Ｓ１９：ＮＯ）、制御部１１は処理をステップＳ１７に戻す。終了すると判定した場合（Ｓ１９：ＹＥＳ）、制御部１１は一連の処理を終了する。 The control unit 11 determines whether or not to end the output of the navigation screen (step S19). For example, the control unit 11 determines whether or not the customer terminal 5 has received an input related to a predetermined end instruction. When it determines with not complete | finishing (S19: NO), the control part 11 returns a process to step S17. When it determines with complete | finishing (S19: YES), the control part 11 complete | finishes a series of processes.
　図１４は、相関情報生成のサブルーチンの処理手順の一例を示すフローチャートである。図１４に基づき、ステップＳ１５の相関情報生成のサブルーチンについて説明する。
　サーバ１の制御部１１は、広告ＤＢ１４１又は成果ＤＢ１４２を参照し、広告情報又は成果情報に含まれる要素から目的変数を選択する（ステップＳ３１）。例えば制御部１１は、受動視聴情報、間接視聴情報、直接視聴情報、確認情報又は購買情報に含まれる要素から一の目的変数を選択する。制御部１１は、広告ＤＢ１４１又は費用ＤＢ１４３を参照し、ステップＳ３１で選択した目的変数に対する説明変数の候補を、広告情報又は費用情報に含まれる要素から特定する（ステップＳ３２）。例えば制御部１１は、統計分析に係る分析方針を規定するファイルを参照して、当該分析方針に従い説明変数の候補である候補変数を特定する。 FIG. 14 is a flowchart illustrating an example of a processing procedure of a correlation information generation subroutine. Based on FIG. 14, the correlation information generation subroutine of step S15 will be described.
The control unit 11 of the server 1 refers to the advertisement DB 141 or the result DB 142 and selects an objective variable from elements included in the advertisement information or the result information (Step S31). For example, the control unit 11 selects one objective variable from elements included in passive viewing information, indirect viewing information, direct viewing information, confirmation information, or purchase information. The control unit 11 refers to the advertisement DB 141 or the cost DB 143, and specifies candidate explanatory variables for the objective variable selected in Step S31 from elements included in the advertisement information or the cost information (Step S32). For example, the control unit 11 refers to a file that defines an analysis policy related to statistical analysis, and identifies candidate variables that are candidates for explanatory variables in accordance with the analysis policy.
　制御部１１は、ステップＳ３２で特定した候補変数から一の説明変数を選択する（ステップＳ３３）。制御部１１は、ステップＳ３１で選択した目的変数、及びステップＳ３３で選択した説明変数について回帰分析を行う（ステップＳ３４）。制御部１１は、ステップＳ３４で実行した回帰分析に係る説明変数の妥当性を判断するための指標を算出する（ステップＳ３５）。具体的に制御部１１は、回帰分析により算出された自由度調整済み決定係数Ｒ２ａｄｊ、Ｐ値、標準誤差ｓｅ等に基づき、式（１）より指標値ＣＦを算出する。制御部１１は、ステップＳ３４で実行した回帰分析が一回目の分析であったか否かを判定する（ステップＳ３６）。一回目の分析であったと判定した場合（Ｓ３６：ＹＥＳ）、制御部１１は処理をステップＳ３８に処理を移行する。 The control unit 11 selects one explanatory variable from the candidate variables identified in step S32 (step S33). The control unit 11 performs regression analysis on the objective variable selected in step S31 and the explanatory variable selected in step S33 (step S34). The control unit 11 calculates an index for determining the validity of the explanatory variable related to the regression analysis performed in step S34 (step S35). Specifically, the control unit 11 calculates the index value CF from Equation (1) based on the degree-of-freedom-adjusted determination coefficient R2adj, P value, standard error se, and the like calculated by regression analysis. The control unit 11 determines whether or not the regression analysis executed in step S34 is the first analysis (step S36). When it determines with it being the first analysis (S36: YES), the control part 11 transfers a process to step S38.
　一回目の分析でなかったと判定した場合（Ｓ３６：ＮＯ）、制御部１１は、ステップＳ３５で今回算出した指標が、後述するステップＳ３８で採用候補に設定した説明変数について実行した回帰分析に係る指標より大きいか否かを判定する（ステップＳ３７）。すなわち制御部１１は、採用候補として保持してある説明変数に係る指標よりも、新たに算出した今回の指標の方が大きいか否かを判定する。今回の指標の方が大きいと判定した場合（Ｓ３７：ＹＥＳ）、制御部１１は、今回の指標に係る説明変数を採用候補に設定する（ステップＳ３８）。具体的に制御部１１は、今回ステップＳ３４で実行した回帰分析において使用した一又は複数の説明変数を、最終的な採用候補として一時的に記憶しておく。これにより制御部１１は、指標値ＣＦにより説明変数としてより適切な要素の組み合わせを採用候補として保持しておく。 When it is determined that the analysis is not the first time (S36: NO), the control unit 11 uses the index calculated this time in step S35 for the regression analysis performed on the explanatory variable set as the adoption candidate in step S38 described later. It is determined whether it is larger (step S37). That is, the control unit 11 determines whether or not the newly calculated current index is larger than the index related to the explanatory variable held as the adoption candidate. When it is determined that the current index is larger (S37: YES), the control unit 11 sets the explanatory variable related to the current index as an adoption candidate (step S38). Specifically, the control unit 11 temporarily stores one or a plurality of explanatory variables used in the regression analysis executed in step S34 as final adoption candidates. As a result, the control unit 11 holds a combination of more appropriate elements as explanatory variables based on the index value CF as an adoption candidate.
　今回の指標の方が小さいと判定した場合（Ｓ３７：ＮＯ）、又はステップＳ３８の処理を実行した後で、制御部１１は、ステップＳ３３で選択した一の説明変数について、他の候補変数との組み合わせを全て検討したか否かを判定する（ステップＳ３９）。具体的に制御部１１は、ステップＳ３３で選択した一の説明変数と他の候補変数との組み合わせ全てについて、指標値ＣＦを算出してステップＳ３７の判定処理を行ったか否かを判定する。全てを検討していないと判定した場合（Ｓ３９：ＮＯ）、制御部１１は、ステップＳ３２で特定した候補変数から一の変数を説明変数に追加し（ステップＳ４０）、処理をステップＳ３４に戻す。これにより制御部１１は、採用候補変数に別の候補変数を加えた説明変数について回帰分析を実行し、当該組み合わせについて指標を算出することになる。 When it is determined that the current index is smaller (S37: NO), or after executing the process of step S38, the control unit 11 sets the one explanatory variable selected in step S33 as another candidate variable. It is determined whether all the combinations have been examined (step S39). Specifically, the control unit 11 determines whether or not the index value CF is calculated and the determination process of step S37 is performed for all combinations of one explanatory variable selected in step S33 and other candidate variables. If it is determined that all are not considered (S39: NO), the control unit 11 adds one variable from the candidate variables identified in step S32 to the explanatory variable (step S40), and returns the process to step S34. As a result, the control unit 11 performs regression analysis on the explanatory variable obtained by adding another candidate variable to the adoption candidate variable, and calculates an index for the combination.
　全て検討したと判定した場合（Ｓ３９：ＹＥＳ）、制御部１１は、ステップＳ３３の説明変数の選択処理において、他の候補変数も全て選択したことがあるか否かを判定する（ステップＳ４１）。すなわち制御部１１は、候補変数それぞれを当初の説明変数として選択し、選択された候補変数を基準としてステップＳ３４～Ｓ３９の処理を実行したか否かを判定する。選択したことがない候補変数があると判定した場合（Ｓ４１：ＮＯ）、制御部１１は処理をステップＳ３３に戻す。これにより、制御部１１は候補変数から別の説明変数を選択し、再度ステップＳ３４～Ｓ３９の処理を実行して、広告情報に係る要素の全ての組み合わせについて妥当性を検討することになる。 If it is determined that all have been examined (S39: YES), the control unit 11 determines whether or not all other candidate variables have been selected in the explanatory variable selection process in step S33 (step S41). That is, the control unit 11 selects each candidate variable as an initial explanatory variable, and determines whether or not the processing of steps S34 to S39 has been executed based on the selected candidate variable. If it is determined that there is a candidate variable that has not been selected (S41: NO), the control unit 11 returns the process to step S33. As a result, the control unit 11 selects another explanatory variable from the candidate variables, executes the processes of steps S34 to S39 again, and examines the validity of all combinations of elements related to the advertisement information.
　全て選択したことがあると判定した場合（Ｓ４１：ＹＥＳ）、制御部１１は、ステップＳ３１の目的変数の選択処理について、目的変数となり得る広告情報及び成果情報の要素全てについて選択したことがあるか否かを判定する（ステップＳ４２）。すなわち制御部１１は、受動視聴情報、間接視聴情報、直接視聴情報、確認情報、購買情報等の目的変数となり得る要素全てについて、ステップＳ３２～Ｓ４１の処理を実行したか否かを判定する。目的変数に選択したことがない要素があると判定した場合（Ｓ４２：ＮＯ）、制御部１１は処理をステップＳ３１に戻す。これにより、制御部１１は別の目的変数を選択して再度回帰分析を行い、共分散構造を解析していく。全ての要素について選択したと判定した場合（Ｓ４２：ＹＥＳ）、制御部１１は処理をリターンする。 If it is determined that all have been selected (S41: YES), the control unit 11 has selected all the elements of the advertisement information and the result information that can be the objective variable for the objective variable selection process in step S31. It is determined whether or not (step S42). That is, the control unit 11 determines whether or not the processes of steps S32 to S41 have been executed for all elements that can be objective variables such as passive viewing information, indirect viewing information, direct viewing information, confirmation information, and purchase information. When it is determined that there is an element that has not been selected as the objective variable (S42: NO), the control unit 11 returns the process to step S31. Thereby, the control part 11 selects another objective variable, performs a regression analysis again, and analyzes a covariance structure. When it determines with having selected about all the elements (S42: YES), the control part 11 returns a process.
　なお、上記でサーバ１は広告情報を四段階に分類したが、本実施の形態はこれに限るものではない。例えばサーバ１は、商品又はサービスを検索等する直接視聴情報と、商品又はサービスを確認する確認情報とを一つの情報群として捉え、広告情報を三段階で分類してもよい。逆にサーバ１は、受動視聴情報、間接視聴情報等をより細分化してもよい。このように、サーバ１は広告情報を三段階以下に分類してもよいし、五段階以上に分類してもよい。すなわちサーバ１は、多段階の回帰分析により相関情報を生成可能であればよい。 Although the server 1 classifies the advertisement information into four stages as described above, the present embodiment is not limited to this. For example, the server 1 may regard the direct viewing information for searching for a product or service and the confirmation information for confirming the product or service as one information group, and classify the advertisement information in three stages. Conversely, the server 1 may subdivide passive viewing information, indirect viewing information, and the like. As described above, the server 1 may classify the advertisement information into three levels or less, or may classify the advertisement information into five levels or more. That is, the server 1 only needs to be able to generate correlation information by multistage regression analysis.
　また、上記でサーバ１は広告活動の成果を示す成果情報を商品又はサービスの購買情報と捉えて一連の処理を行ったが、本実施の形態はこれに限るものではない。例えばサーバ１は、消費者による商品又はサービスの確認の有無を示す確認情報（例えば商品又はサービスを販売するオンラインストアへのセッション数）を成果情報と捉えて一連の処理を行ってもよい。この場合、サーバ１は受動視聴情報、間接視聴情報、直接視聴情報を広告情報に規定し、確認情報を成果情報に規定する。そしてサーバ１は、広告情報と、成果情報、すなわち確認情報との相関関係を分析する。上述の如く、成果情報は分析方針によって変更可能であり、広告活動の成果として規定可能な情報であればよい。 In the above description, the server 1 regards the result information indicating the result of the advertising activity as the purchase information of the product or service, and performs a series of processes. However, the present embodiment is not limited to this. For example, the server 1 may perform a series of processes by regarding the confirmation information (for example, the number of sessions to an online store that sells the product or service) indicating whether the consumer confirms the product or service as the result information. In this case, the server 1 defines passive viewing information, indirect viewing information, and direct viewing information as advertisement information, and defines confirmation information as result information. Then, the server 1 analyzes the correlation between the advertisement information and the result information, that is, the confirmation information. As described above, the result information can be changed according to the analysis policy and may be information that can be defined as the result of the advertising activity.
　また、本実施の形態においてサーバ１は、広告の視聴状況について、その他の外部要因を考慮して相関関係を判断するようにしてもよい。外部要因は、例えば天候、商品又はサービスのキャンペーン等である。この場合、サーバ１は外部要因に応じて広告情報、成果情報等を補正し、共分散構造分析を行えばよい。 Further, in the present embodiment, the server 1 may determine the correlation regarding the viewing state of the advertisement in consideration of other external factors. External factors are, for example, weather, product or service campaigns, and the like. In this case, the server 1 may correct the advertisement information, the result information, etc. according to the external factor and perform the covariance structure analysis.
　以上より、本実施の形態１によれば、サーバ１は各装置から広告活動に係る情報を自動的に取得し、統計分析に係る分析結果を生成することで、効率的な広告活動を支援することができる。 As described above, according to the first embodiment, the server 1 automatically acquires information related to the advertising activity from each device and generates an analysis result related to the statistical analysis, thereby supporting efficient advertising activity. be able to.
　また、本実施の形態１によれば、多段階の回帰分析を含む共分散構造分析により、広告活動及び該広告活動の成果の相関関係を適切に把握することができる。 Further, according to the first embodiment, the correlation between the advertising activity and the result of the advertising activity can be appropriately grasped by the covariance structure analysis including the multi-step regression analysis.
　また、本実施の形態１によれば、広告情報を消費者の認知、興味、検索等、確認の四段階に分類すると共に、成果情報を商品又はサービスの売上として統計分析を行うことで、広告の認知から成果へと至る過程を適切に分析することができる。 Further, according to the first embodiment, advertisement information is classified into four stages of confirmation such as consumer recognition, interest, search, etc., and the result information is statistically analyzed as sales of goods or services, whereby advertisement The process from recognition to outcome can be analyzed appropriately.
　また、本実施の形態１によれば、ナビゲーション画面（シーケンス画面）により、広告活動に係る各要素の相関関係を容易に把握することができる。 Further, according to the first embodiment, it is possible to easily grasp the correlation of each element related to the advertising activity on the navigation screen (sequence screen).
　また、本実施の形態１によれば、統計分析に係る分析要素に費用情報を含めることにより、例えば広告活動の費用対効果等も把握することができる。 Further, according to the first embodiment, by including cost information in the analysis element related to the statistical analysis, for example, the cost-effectiveness of the advertising activity can be grasped.
　また、本実施の形態１によれば、説明変数を適宜に取捨選択することで統計分析に係る精度を上げることができる。 Further, according to the first embodiment, it is possible to increase the accuracy of statistical analysis by appropriately selecting explanatory variables.
（実施の形態２）
　本実施の形態では、アドストック効果を考慮した形態について述べる。なお、実施の形態１と重複する内容については同一の符号を付して説明を省略する。 (Embodiment 2)
In the present embodiment, a mode that considers the ad stock effect will be described. In addition, about the content which overlaps with Embodiment 1, the same code | symbol is attached | subjected and description is abbreviate | omitted.
　図１５は、実施の形態２に係るアドストック期間に応じた統計分析について説明するための説明図である。一般的に消費者が広告を視聴した場合に、消費者は視聴内容をすぐに忘れるわけではなく、時間の経過と共に徐々に忘れていく傾向がある。この概念を一般にアドストックという。サーバ１は上記の統計分析を行うに際して、アドストックの期間（以下ではアドストック期間という）に応じた処理を行う。本実施の形態でサーバ１は、オフライン広告に関する統計分析について、アドストックを考慮した分析処理を行う。なお、サーバ１はオフライン広告だけでなくオンライン広告についてもアドストックを考慮して以下の処理を行ってもよい。 FIG. 15 is an explanatory diagram for explaining the statistical analysis according to the ad stock period according to the second embodiment. In general, when a consumer views an advertisement, the consumer does not immediately forget the content to be viewed but tends to gradually forget with the passage of time. This concept is generally called ad stock. When performing the above statistical analysis, the server 1 performs a process according to an ad stock period (hereinafter referred to as an ad stock period). In the present embodiment, the server 1 performs an analysis process in consideration of ad stock for statistical analysis related to offline advertising. Note that the server 1 may perform the following processing in consideration of ad stock not only for offline advertisements but also for online advertisements.
　例えば図１５に示すように、テレビコマーシャルの視聴世帯数に係る受動視聴情報を説明変数とし、指名ワードに基づくリスティング広告のｉｍｐ数に係る直接視聴情報を目的変数として、所定の日付にテレビコマーシャルの放映が終了した場合を考える。この場合、サーバ１の制御部１１はアドストック期間として複数の期間（例えば１～１３週間）の候補を設定する。そして制御部１１は、図１５に示すように、当該複数の期間に応じて受動視聴情報（図１５の場合は視聴世帯数）が減衰しながら継続するものと仮定し、各アドストック期間に応じた時系列のモデルデータを生成する。例えばアドストック期間が１週間である場合、制御部１１は放映終了時から１週間後に、視聴世帯数が放映終了時の１０パーセントまで減衰するモデルデータを生成する。以下同様に、制御部１１はアドストック期間に応じて受動視聴情報のモデルデータを生成する。 For example, as shown in FIG. 15, the passive viewing information related to the number of TV commercial viewing households is used as an explanatory variable, and the direct viewing information related to the number of imps of a listing advertisement based on a designated word is used as a target variable, and the TV commercial is displayed on a predetermined date. Consider the case where the airing has ended. In this case, the control unit 11 of the server 1 sets candidates for a plurality of periods (for example, 1 to 13 weeks) as the ad stock period. Then, as shown in FIG. 15, the control unit 11 assumes that the passive viewing information (in the case of FIG. 15 the number of viewing households) continues to attenuate according to the plurality of periods, and responds to each ad stock period. Time series model data is generated. For example, when the ad stock period is one week, the control unit 11 generates model data in which the number of viewing households is attenuated to 10% at the end of the broadcast one week after the end of the broadcast. Similarly, the controller 11 generates model data of passive viewing information according to the ad stock period.
　制御部１１は、期間別に生成した各受動視聴情報と、直接視聴情報とに基づき、回帰分析を用いたモデル分析を行う。制御部１１は、期間別の各モデル分析の結果についてモデル精度を比較する。例えば制御部１１は、モデル分析に係る自由度調整済み決定係数Ｒ２ａｄｊの大小に応じてモデル精度を比較する。制御部１１は、最も精度の高いモデルデータに係る期間を、アドストック期間として選択する。なお、精度が同じモデルデータが複数ある場合、制御部１１は、最長の期間をアドストック期間として選択する。制御部１１は、選択したアドストック期間に係るモデルデータを説明変数に採用し、統計分析を実行する。 The control unit 11 performs model analysis using regression analysis based on each passive viewing information generated for each period and direct viewing information. The control unit 11 compares the model accuracy for the results of each model analysis for each period. For example, the control unit 11 compares the model accuracy according to the magnitude of the degree-of-freedom adjusted determination coefficient R2adj related to the model analysis. The control unit 11 selects the period related to the most accurate model data as the ad stock period. When there are a plurality of model data having the same accuracy, the control unit 11 selects the longest period as the ad stock period. The control unit 11 employs the model data related to the selected ad stock period as an explanatory variable, and performs statistical analysis.
　図１６は、実施の形態２に係る相関情報生成のサブルーチンの処理手順の一例を示すフローチャートである。図１６に基づき、本実施の形態に係るステップＳ１５の相関情報生成のサブルーチンについて説明する。
　目的変数に対する説明変数の候補を特定した後（ステップＳ３２）、サーバ１の制御部１１は以下の処理を実行する。制御部１１は、ステップＳ３２で特定した説明変数の候補に、オフライン広告に係る広告情報があるか否かを判定する（ステップＳ２０１）。オフライン広告に係る要素は、例えばテレビジョンコマーシャル、交通広告、雑誌広告等に係る広告情報である。オフライン広告に係る広告情報があると判定した場合（Ｓ２０１：ＹＥＳ）、制御部１１は、当該オフライン広告に係る広告情報について、アドストック期間に応じたモデルデータを生成する（ステップＳ２０２）。例えば制御部１１は、当該オフライン広告に係る時系列データについて、オフライン広告の終了時からアドストック期間だけデータ数値が減衰していくモデルデータを生成する。具体的に制御部１１は、アドストック期間を１～１３週間であると仮定し、各アドストック期間に応じて１３個のモデルデータを生成する。例えば制御部１１は、オフライン広告終了時からアドストック期間経過後にデータが１０パーセントまで減衰すると仮定し、モデルデータを生成する。オフライン広告に係る広告情報がないと判定した場合（Ｓ２０１：ＮＯ）、又はステップＳ２０２の処理を実行した後で、制御部１１は処理をステップＳ３３に移行する。 FIG. 16 is a flowchart illustrating an example of a processing procedure of a correlation information generation subroutine according to the second embodiment. Based on FIG. 16, the correlation information generation subroutine of step S15 according to the present embodiment will be described.
After identifying the explanatory variable candidates for the objective variable (step S32), the control unit 11 of the server 1 executes the following processing. The control unit 11 determines whether or not there is advertisement information related to the offline advertisement among the candidates for the explanatory variable specified in Step S32 (Step S201). The element related to the offline advertisement is, for example, advertisement information related to a television commercial, a traffic advertisement, a magazine advertisement, or the like. If it is determined that there is advertisement information related to the offline advertisement (S201: YES), the control unit 11 generates model data corresponding to the ad stock period for the advertisement information related to the offline advertisement (step S202). For example, the control unit 11 generates model data in which the numerical value of the time series data related to the offline advertisement is attenuated during the ad stock period from the end of the offline advertisement. Specifically, the control unit 11 assumes that the ad stock period is 1 to 13 weeks, and generates 13 model data according to each ad stock period. For example, the control unit 11 generates model data on the assumption that the data is attenuated to 10% after the ad stock period has elapsed since the end of the offline advertisement. When it determines with there being no advertisement information which concerns on an offline advertisement (S201: NO), or after performing the process of step S202, the control part 11 transfers a process to step S33.
　候補変数から一の説明変数を選択した後（ステップＳ３３）、又は説明変数に別の候補変数を追加した後（ステップＳ４０）、制御部１１は以下の処理を実行する。制御部１１は、ステップＳ３３で選択又はステップＳ４０で追加された説明変数に、オフライン広告に係る広告情報が含まれているか否かを判定する（ステップＳ２０３）。含まれていないと判定した場合（Ｓ２０３：ＮＯ）、制御部１１は処理をステップＳ３４に移行する。含まれていると判定した場合（Ｓ２０３：ＹＥＳ）、制御部１１は、ステップＳ２０２で生成した各モデルデータについて、回帰分析を行う（ステップＳ２０４）。すなわち制御部１１は、１～１３週間のアドストック期間に応じて生成した説明変数の各モデルデータについて、目的変数との相関関係を分析する。制御部１１は、ステップＳ２０４に係る各分析結果に基づき、最もモデル精度が高いアドストック期間に係るモデルデータを説明変数に選択する（ステップＳ２０５）。具体的に制御部１１は、例えば各分析結果に係る自由度調整済み決定係数Ｒ２ａｄｊを比較し、係数Ｒ２ａｄｊが最も大きいモデルデータを判別する。制御部１１は、当該モデルデータに係るアドストック期間が最も適切であると判断し、当該アドストック期間に応じて生成されたモデルデータを説明変数に選択する。制御部１１は、処理をステップＳ３５に移行する。 After selecting one explanatory variable from candidate variables (step S33) or adding another candidate variable to the explanatory variable (step S40), the control unit 11 executes the following processing. The control unit 11 determines whether or not the advertisement information related to the offline advertisement is included in the explanatory variable selected in Step S33 or added in Step S40 (Step S203). When it determines with not being contained (S203: NO), the control part 11 transfers a process to step S34. When it determines with being contained (S203: YES), the control part 11 performs a regression analysis about each model data produced | generated by step S202 (step S204). That is, the control unit 11 analyzes the correlation between the model data of the explanatory variables generated according to the ad stock period of 1 to 13 weeks and the objective variable. Based on each analysis result in step S204, the control unit 11 selects model data related to the ad stock period with the highest model accuracy as an explanatory variable (step S205). Specifically, the control unit 11 compares, for example, the degree-of-freedom-adjusted determination coefficient R2adj related to each analysis result, and determines model data having the largest coefficient R2adj. The control unit 11 determines that the ad stock period related to the model data is most appropriate, and selects model data generated according to the ad stock period as an explanatory variable. The control part 11 transfers a process to step S35.
　以上より、本実施の形態２によれば、広告のアドストック効果を考慮して適切な統計分析を行うことができる。 As described above, according to the second embodiment, appropriate statistical analysis can be performed in consideration of the ad stock effect of the advertisement.
（実施の形態３）
　本実施の形態では、統計分析に係る分析結果に基づき新たな広告活動のプランニングを行うと共に、当該プランニングに基づき実施する広告活動の進捗状況を提示する形態について述べる。なお、実施の形態１と重複する内容については同一の符号を付して説明を省略する。 (Embodiment 3)
In the present embodiment, a mode for planning a new advertising activity based on an analysis result related to statistical analysis and presenting a progress status of the advertising activity performed based on the planning will be described. In addition, about the content which overlaps with Embodiment 1, the same code | symbol is attached | subjected and description is abbreviate | omitted.
　図１７は、実施の形態３に係るプランニング画面の一例を示す説明図である。本実施の形態に係るサーバ１の制御部１１は、顧客端末５からの出力要求に応じて図１７に示すプランニング画面を出力する。プランニング画面は、過去に実施された広告活動に係る分析結果に基づき、新たに実施する広告活動に係る予算等を作成するための画面である。例えばプランニング画面は、相関グラフ１７１、成果比較グラフ１７２、予算比較グラフ１７３、予算表１７４等を含む。相関グラフ１７１は、過去に実施された広告活動に関し生成済みの相関情報について、成果情報と費用情報との相関関係を示すグラフである。なお、本実施の形態において成果情報は商品又はサービスの購買情報であるものとして説明を行うが、実施の形態１と同じく、成果情報は他の情報であってもよい。相関グラフ１７１は、購買数（成果）及び費用（予算）を各軸とするグラフの他に、切替ボタン１７１１、設定オブジェクト１７１２を含む。切替ボタン１７１１は、相関グラフ１７１の表示を切り替えるためのボタンであり、「成果」又は「予算」のいずれの切替ボタン１７１１が指定されるかに応じて相関グラフ１７１の各軸が逆になる。例えば図１７において、現在は「成果」の切替ボタン１７１１が指定されているため、横軸が購買数、縦軸が費用となっている。 FIG. 17 is an explanatory diagram illustrating an example of a planning screen according to the third embodiment. The control unit 11 of the server 1 according to the present embodiment outputs a planning screen shown in FIG. 17 in response to an output request from the customer terminal 5. The planning screen is a screen for creating a budget or the like related to a newly executed advertising activity based on the analysis result related to the advertising activity executed in the past. For example, the planning screen includes a correlation graph 171, a result comparison graph 172, a budget comparison graph 173, a budget table 174, and the like. The correlation graph 171 is a graph showing the correlation between the result information and the cost information with respect to the correlation information that has already been generated with respect to the advertisement activity that has been performed in the past. In the present embodiment, the description will be made assuming that the result information is purchase information of goods or services. However, as in the first embodiment, the result information may be other information. The correlation graph 171 includes a switching button 1711 and a setting object 1712 in addition to a graph having the number of purchases (results) and cost (budget) as axes. The switch button 1711 is a button for switching the display of the correlation graph 171, and each axis of the correlation graph 171 is reversed depending on which of the “result” or “budget” switch buttons 1711 is designated. For example, in FIG. 17, since the “result” switching button 1711 is currently designated, the horizontal axis represents the number of purchases and the vertical axis represents the cost.
　制御部１１は、相関グラフ１７１において、成果情報（購買情報）又は費用情報に関する条件入力を受け付ける。具体的に制御部１１は、顧客企業が新たに実施する広告活動に関し、成果目標（目標購買数）又は上限費用、すなわち売上の目標値又は予算の上限値について入力を受け付ける。例えばプランニング画面を出力中の顧客端末５は、設定オブジェクト１７１２への操作入力を受け付ける。具体的に顧客端末５は、図１７で示す設定オブジェクト１７１２について、横軸「成果」に沿って設定オブジェクトの位置を横にずらす操作を受け付ける。制御部１１は当該操作内容を顧客端末５から取得する。制御部１１は、相関グラフ１７１における設定オブジェクト１７１２の設定位置、すなわち横軸「成果」に対する設定オブジェクト１７１２の操作位置に応じて、当該位置が示す購買数を成果目標に係る条件値として受け付ける。なお、図１７に示すように、制御部１１は相関グラフ１７１の横軸に併せて分析精度をラベル表示する。当該ラベル表示は、例えば過去に分析実績がある購買数又は費用の数値範囲を示す。顧客企業は当該ラベル表示を参考に、統計的な信頼性が高い範囲で条件値を設定することができる。 The control unit 11 accepts condition input related to result information (purchase information) or expense information in the correlation graph 171. Specifically, the control unit 11 receives an input regarding a result target (target number of purchases) or an upper limit cost, that is, a target value of sales or an upper limit value of a budget, regarding an advertising activity newly performed by a customer company. For example, the customer terminal 5 that is outputting the planning screen receives an operation input to the setting object 1712. Specifically, for the setting object 1712 shown in FIG. 17, the customer terminal 5 accepts an operation for shifting the position of the setting object horizontally along the horizontal axis “result”. The control unit 11 acquires the operation content from the customer terminal 5. The control unit 11 accepts the number of purchases indicated by the position as a condition value related to the result target according to the setting position of the setting object 1712 in the correlation graph 171, that is, the operation position of the setting object 1712 with respect to the horizontal axis “result”. As shown in FIG. 17, the control unit 11 displays the analysis accuracy in a label along with the horizontal axis of the correlation graph 171. The label display indicates, for example, a numerical range of the number of purchases or expenses that have been analyzed in the past. The client company can set the condition value within a range having high statistical reliability with reference to the label display.
　制御部１１は、受け付けた条件を満たす最適な予算案を算出し、プランニング画面に出力する。具体的に制御部１１は、生成済みの相関情報に基づき、広告活動に係る各施策について、受け付けた条件を満たす最適な構成比を算出する。例えば制御部１１は、生成済みの相関情報に含まれる係数ａを読み出す。また、制御部１１は費用ＤＢ１４３を参照し、過去に実施された各施策の広告単価を算出する。そして制御部１１は、受け付けた条件を満たし、かつ、費用対効果が最大となる各施策の構成比を算出する。例えば制御部１１は、ラグランジュ未定乗数法を用いて算出を行う。 The control unit 11 calculates an optimal budget plan that satisfies the accepted conditions and outputs it to the planning screen. Specifically, the control unit 11 calculates an optimal composition ratio that satisfies the received condition for each measure related to the advertising activity based on the generated correlation information. For example, the control unit 11 reads the coefficient a included in the generated correlation information. Further, the control unit 11 refers to the cost DB 143 and calculates the advertising unit price of each measure implemented in the past. Then, the control unit 11 calculates the composition ratio of each measure that satisfies the accepted condition and maximizes the cost effectiveness. For example, the control unit 11 performs calculation using a Lagrange undetermined multiplier method.
　具体的には、成果目標について条件入力を受け付けた場合、制御部１１は相関情報を参照し、当該成果目標を満たし、かつ、広告費用が最小となる各施策の構成比を算出する。例えば図１７に示すように、各施策が「Ｇ社リスティング」「Ｙ社リスティング」等である場合、制御部１１は成果（購買数）を成果目標（目標購買数）に固定して、過去の広告活動の分析結果、すなわち相関情報に基づき各施策の構成比を算出する。この場合に、制御部１１は各施策の広告単価を参照し、必要となる費用が最小となるように構成比を算出する。すなわち制御部１１は、費用対効果が最大となる構成比を算出する。上限費用について条件入力を受け付けた場合は、制御部１１は広告活動に費やす全体の費用を指定された上限費用に固定し、成果が最大となる構成比を算出する。 Specifically, when a condition input is received for the result target, the control unit 11 refers to the correlation information, and calculates the composition ratio of each measure that satisfies the result target and minimizes the advertising cost. For example, as shown in FIG. 17, when each measure is “G company listing”, “Y company listing”, etc., the control unit 11 fixes the result (the number of purchases) to the result target (the target number of purchases), The composition ratio of each measure is calculated based on the analysis result of the advertising activity, that is, the correlation information. In this case, the control unit 11 refers to the advertising unit price of each measure, and calculates the composition ratio so that the necessary cost is minimized. That is, the control unit 11 calculates a configuration ratio that maximizes cost effectiveness. When the condition input is received for the upper limit cost, the control unit 11 fixes the total cost for the advertising activity to the specified upper limit cost, and calculates the composition ratio that maximizes the result.
　制御部１１は、算出した構成比を示す予算案を出力する。具体的に制御部１１は、予算表１７４において、当該構成比を施策別の予算配分として出力する。予算表１７４は予算案の詳細を、広告活動に係る施策別に示す表である。図１７に示すように、制御部１１は予算表１７４において、各施策の名称、過去（前月）の広告費用、及び施策別の予算をそれぞれ出力する。また、制御部１１は併せて成果比較グラフ１７２、予算比較グラフ１７３を出力する。成果比較グラフ１７２及び予算比較グラフ１７３は、過去の広告活動に係る成果及び費用と、予算案に係る予想成果（予想購買数）及び予想費用との比較データを示すグラフである。予算比較グラフ１７３ではさらに、予想費用に占める施策別の予算配分が示される。 The control unit 11 outputs a budget plan indicating the calculated composition ratio. Specifically, the control unit 11 outputs the composition ratio as a budget distribution for each measure in the budget table 174. The budget table 174 is a table showing details of the budget plan for each measure related to the advertising activity. As shown in FIG. 17, in the budget table 174, the control unit 11 outputs the name of each measure, the past (previous month) advertising cost, and the measure-specific budget. The control unit 11 also outputs a result comparison graph 172 and a budget comparison graph 173. The result comparison graph 172 and the budget comparison graph 173 are graphs showing comparison data between the results and costs related to past advertising activities and the expected results (expected number of purchases) and expected costs related to the budget plan. The budget comparison graph 173 further shows the budget allocation for each measure in the expected cost.
　制御部１１はさらに、予算表１７４において、各施策の構成比に関する条件入力を受け付ける。例えば制御部１１は、予算表１７４のシミュレーション欄１７４１への操作入力に基づき予算配分の変更を受け付ける。例えば「Ｇ社リスティング」について費用の入力を受け付けた場合、制御部１１は当該施策に係る費用を固定した上で、再度予算案を算出する。すなわち制御部１１は、指定された施策以外の他の施策について予算配分を再算出し、シミュレーション欄１７１４に出力する。また、予算配分に関する条件入力を受け付けた場合、制御部１１は成果比較グラフ１７２及び予算比較グラフ１７３も自動連携させて再度算出結果を出力する。これにより、顧客企業の要望に従って予算案を自動的に再出力することができる。
　制御部１１は、プランニング画面の確定ボタン１７５への操作入力を受け付けた場合、予算案を確定させる。これにより、顧客企業は生成された予算案に基づいて広告活動を新たに開始する。制御部１１は、実施の形態１と同様に広告情報等を取得し、以下の進捗画面を顧客端末５に出力する。 In addition, the control unit 11 receives a condition input regarding the composition ratio of each measure in the budget table 174. For example, the control unit 11 receives a budget allocation change based on an operation input to the simulation field 1741 of the budget table 174. For example, when the cost input for “G company listing” is received, the control unit 11 calculates the budget plan again after fixing the cost related to the measure. That is, the control unit 11 recalculates the budget allocation for other measures other than the designated measure, and outputs it to the simulation column 1714. Further, when receiving a condition input relating to budget allocation, the control unit 11 automatically links the result comparison graph 172 and the budget comparison graph 173 and outputs the calculation result again. Thereby, the budget plan can be automatically re-output according to the request of the client company.
When the control unit 11 receives an operation input to the confirmation button 175 on the planning screen, the control unit 11 confirms the budget plan. Thereby, the customer company newly starts an advertising activity based on the generated budget plan. The control unit 11 acquires advertisement information and the like as in the first embodiment, and outputs the following progress screen to the customer terminal 5.
　図１８は、実施の形態３に係る進捗画面の一例を示す説明図である。進捗画面は、予算案に基づく広告活動について、予算の消化状況と、当該広告活動の成果、すなわち商品又はサービスの購買状況とを示す画面である。例えば進捗画面は、概算データ１８１、活動状況データ１８２、及び詳細表１８３を含む。 FIG. 18 is an explanatory diagram showing an example of the progress screen according to the third embodiment. The progress screen is a screen showing the status of the budget consumption and the result of the advertising activity, that is, the purchase status of the product or service, for the advertising activity based on the budget plan. For example, the progress screen includes approximate data 181, activity status data 182, and a detailed table 183.
　制御部１１は、取得した費用情報を参照し、広告活動を開始してから現在までに費やした施策別の費用（予算）を算出する。また、制御部１１は成果情報（購買情報）を参照し、広告活動を開始してから現在までの成果（購買数）に係る集計を行う。制御部１１は、当該費用及び成果と予算案に係る予想費用及び予想成果との比較を、概算データ１８１により出力する。概算データ１８１は、予想費用額に対する現在の費用額を示す予算消化状況、及び成果目標に対する現在の成果を示す成果目標達成度を、それぞれ棒グラフで示す。すなわち制御部１１は、概算データ１８１において予算の消化状況及び広告活動の成果（商品又はサービスの購買状況）それぞれの全体的な概算を出力する。また、制御部１１は、広告活動に係る時系列的な予算消化状況及び成果を、活動状況データ１８２に示す。活動状況データ１８２は、各日付における施策別の費用額、及び購買数を示すグラフである。 The control unit 11 refers to the acquired cost information, and calculates the cost (budget) for each measure that has been spent since the start of the advertising activity. In addition, the control unit 11 refers to the result information (purchase information), and performs aggregation related to the results (the number of purchases) from the start of the advertising activity. The control unit 11 outputs a comparison between the cost and result and the expected cost and the expected result related to the budget plan as the approximate data 181. The approximate data 181 is a bar graph showing the budget consumption status indicating the current cost amount with respect to the expected cost amount, and the achievement target achievement level indicating the current result with respect to the result target. That is, the control unit 11 outputs an overall rough estimate of the budget consumption status and the advertising activity result (purchase status of goods or services) in the approximate data 181. Moreover, the control part 11 shows the time-sequential budget consumption situation and result regarding an advertising activity in the activity status data 182. The activity status data 182 is a graph showing the cost amount for each measure and the number of purchases on each date.
　さらに制御部１１は、各日付における詳細な予算の消化状況を詳細表１８３に示す。詳細表１８３は、一日で消化すべき目標費用、現在消化している費用、及び後日に消化すべき目標費用を施策別に示す表である。顧客企業は詳細表１８３を参考にして、概算データ１８１及び活動状況データ１８２よりも詳細な予算の消化状況を確認することができる。なお、詳細表１８３において各施策名の表示欄への操作入力を受け付けた場合、図１８に示すように、制御部１１は各施策の相関関係を示すナビゲーション画面（シーケンス図）を簡易表示する。
　なお、制御部１１は詳細表１８３に、予算の消化状況だけでなく、広告活動の成果（購買数）を併せて出力するようにしてもよい。 Further, the control unit 11 shows the detailed budget consumption status on each date in the detailed table 183. The detailed table 183 is a table showing the target cost to be consumed in one day, the cost currently being consumed, and the target cost to be consumed at a later date for each measure. The customer company can confirm the digestion status of the budget in more detail than the approximate data 181 and the activity status data 182 with reference to the detailed table 183. In addition, when the operation input to the display column of each measure name is received in the detailed table 183, as illustrated in FIG. 18, the control unit 11 simply displays a navigation screen (sequence diagram) indicating the correlation between the measures.
Note that the control unit 11 may output not only the budget consumption status but also the results of advertising activities (the number of purchases) in the detailed table 183.
　図１９は、実施の形態３に係るサーバ１が実行する処理手順の一例を示すフローチャートである。図１９に基づき、本実施の形態に係るサーバ１が実行する処理内容について説明する。なお、サーバ１は実施の形態１で述べた処理により、相関情報を生成済みであるものとする。
　サーバ１の制御部１１は、過去の広告活動に関して生成済みの相関情報を参照し、広告活動に係る予算案を決定するためのプランニング画面を生成して顧客端末５に出力する（ステップＳ３０１）。例えば制御部１１は、過去の広告情報、成果情報、及び費用情報を参照し、図１７で例示したプランニング画面を生成する。プランニング画面は、例えば過去の広告活動に係る成果情報と費用情報との相関関係を示す相関グラフ１７１、過去の広告活動に係る成果（購買数）と予想成果（予想購買数）との比較を示す成果比較グラフ１７２、過去の広告活動に係る費用（予算）と予想費用との比較を示す予算比較グラフ１７３、後述するステップＳ３０４の処理により生成される予算案の詳細を示すための予算表１７４等を含む。制御部１１は生成したプランニング画面を、ネットワークＮを介して顧客端末５に出力する。 FIG. 19 is a flowchart illustrating an example of a processing procedure executed by the server 1 according to the third embodiment. Based on FIG. 19, the processing content which the server 1 which concerns on this Embodiment performs is demonstrated. It is assumed that the server 1 has already generated correlation information by the processing described in the first embodiment.
The control unit 11 of the server 1 refers to the correlation information generated with respect to the past advertising activity, generates a planning screen for determining a budget plan related to the advertising activity, and outputs the planning screen to the customer terminal 5 (step S301). For example, the control unit 11 refers to past advertisement information, result information, and cost information, and generates the planning screen illustrated in FIG. The planning screen shows, for example, a correlation graph 171 indicating the correlation between the result information related to past advertising activities and the cost information, and the comparison between the results related to past advertising activities (number of purchases) and the expected results (number of expected purchases). Result comparison graph 172, budget comparison graph 173 showing a comparison between costs (budgets) related to past advertising activities and expected costs, a budget table 174 for showing details of a budget plan generated by the processing in step S304 described later, etc. including. The control unit 11 outputs the generated planning screen to the customer terminal 5 via the network N.
　制御部１１は、成果情報又は費用情報について、予算案に関する条件入力を受け付ける（ステップＳ３０２）。具体的に制御部１１は、プランニング画面の相関グラフ１７１において、成果目標、例えば売上目標である目標購買数の設定入力を受け付ける。または、制御部１１は相関グラフ１７１において、広告活動に費やす上限費用の設定入力を受け付ける。すなわち制御部１１は、成果目標及び上限費用のいずれかについて条件入力を受け付ける。 The control part 11 receives the condition input regarding a budget plan about result information or expense information (step S302). Specifically, the control unit 11 accepts a setting input of a target purchase quantity that is a result target, for example, a sales target, in the correlation graph 171 on the planning screen. Or the control part 11 receives the setting input of the upper limit expense spent on advertising activity in the correlation graph 171. That is, the control unit 11 receives a condition input for either the achievement target or the upper limit cost.
　制御部１１は、過去の広告活動について生成済みの相関情報に基づき、ステップＳ３０２で受け付けた条件を満たす各施策の構成比を算出する（ステップＳ３０３）。具体的に制御部１１は、費用ＤＢ１４３を参照し、各施策の費用単価を算出する。また、制御部１１は、過去の広告活動について生成済みの相関情報を大容量記憶装置１４から読み出す。具体的に制御部１１は、相関情報に含まれる係数ａを読み出す。そして制御部１１は、当該費用単価、及び相関情報に係る係数ａに基づき、ステップＳ３０２で受け付けた成果目標又は上限費用の条件を満たす各施策の構成比（予算配分）を算出する。例えばステップＳ３０２で成果目標について条件入力を受け付けた場合、制御部１１は、当該成果目標を満たし、かつ、広告費用が最小となる各施策の構成比を算出する。また、ステップＳ３０２で上限費用について条件入力を受け付けた場合、制御部１１は、当該上限費用を満たし、かつ、成果が最大となる各施策の構成比を算出する。すなわち制御部１１は、成果情報（成果目標）又は費用情報（予算）の条件を満たし、かつ、費用対効果が最大となる各施策の構成比を算出する。制御部１１は、例えばラグランジュ未定乗数法を用いて当該算出処理を行う。 The control unit 11 calculates the composition ratio of each measure that satisfies the condition received in step S302 based on the correlation information generated for the past advertising activity (step S303). Specifically, the control unit 11 refers to the cost DB 143 and calculates the cost unit price of each measure. In addition, the control unit 11 reads the correlation information that has been generated for the past advertising activity from the mass storage device 14. Specifically, the control unit 11 reads the coefficient a included in the correlation information. And the control part 11 calculates the composition ratio (budget allocation) of each measure which satisfy | fills the result target received in step S302, or upper limit expense based on the said cost unit price and the coefficient a which concerns on correlation information. For example, when the condition input is received for the result target in step S302, the control unit 11 calculates the composition ratio of each measure that satisfies the result target and minimizes the advertising cost. When a condition input is received for the upper limit cost in step S302, the control unit 11 calculates the composition ratio of each measure that satisfies the upper limit cost and maximizes the result. That is, the control unit 11 calculates the composition ratio of each measure that satisfies the condition of the result information (result target) or the cost information (budget) and maximizes the cost effectiveness. The control unit 11 performs the calculation process using, for example, a Lagrange undetermined multiplier method.
　制御部１１は、算出した各施策の構成比を示す予算案を出力する（ステップＳ３０４）。例えば制御部１１はプランニング画面の予算表１７４において、施策別の予算配分を示す予算案を出力する。また、制御部１１は過去の広告活動に係る成果及び費用と当該予算案に係る予想成果及び予想費用との比較データを、成果比較データ１７２及び予算比較データ１７３に出力する。 The control unit 11 outputs a budget plan indicating the calculated composition ratio of each measure (step S304). For example, the control unit 11 outputs a budget plan indicating budget allocation for each measure in the budget table 174 on the planning screen. In addition, the control unit 11 outputs comparison data between the results and costs related to past advertising activities and the predicted results and costs related to the budget plan to the result comparison data 172 and the budget comparison data 173.
　制御部１１は、各施策の構成比について条件入力を受け付ける（ステップＳ３０５）。具体的に制御部１１は、予算表１７４のシミュレーション欄１７４１において、各施策の予想費用に関し変更の入力を受け付ける。制御部１１は、予算案を確定させるか否かを判定する（ステップＳ３０６）。例えば制御部１１は、顧客端末５においてプランニング画面に表示される確定ボタン１７５への操作入力を受け付けたか否かを判定する。予算案を確定させないと判定した場合（Ｓ３０６：ＮＯ）、制御部１１は処理をステップＳ３０３に戻す。ステップＳ３０５で構成比の条件入力を受け付けた場合、制御部１１は、当該条件も含めて再度各施策の構成比を算出し、予算案を出力することになる。 The control unit 11 receives a condition input for the composition ratio of each measure (step S305). Specifically, the control unit 11 receives an input of a change regarding the expected cost of each measure in the simulation column 1741 of the budget table 174. The control unit 11 determines whether or not to finalize the budget plan (step S306). For example, the control unit 11 determines whether or not an operation input to the confirmation button 175 displayed on the planning screen at the customer terminal 5 has been received. If it is determined not to finalize the budget plan (S306: NO), the control unit 11 returns the process to step S303. When receiving the composition ratio condition input in step S305, the control unit 11 calculates the composition ratio of each measure again including the condition and outputs a budget plan.
　予算案を確定させると判定した場合（Ｓ３０６：ＹＥＳ）、制御部１１は、各装置から広告情報、成果情報、費用情報等の各情報を取得する（ステップＳ３０７）。これにより、制御部１１は予算案に基づく顧客企業の広告活動に関して情報収集を行う。制御部１１は、取得した費用情報を参照し、予算の消化状況に係る消化予算等を算出する（ステップＳ３０８）。例えば制御部１１は、予算案に基づき広告活動を開始してから現在までに費やした費用を施策別に算出する。また、制御部１１は予算案に係る予想費用と広告活動開始後の費用とを比較し、予算の消化率を施策別に算出する。 If it is determined that the budget plan is to be confirmed (S306: YES), the control unit 11 acquires each piece of information such as advertisement information, result information, and cost information from each device (step S307). Thereby, the control part 11 collects information regarding the advertising activity of the client company based on the budget plan. The control unit 11 refers to the acquired cost information, and calculates a digestion budget and the like related to the budget digest status (step S308). For example, the control unit 11 calculates the expenses spent until the present after the start of the advertising activity based on the budget plan for each measure. Further, the control unit 11 compares the expected cost related to the budget plan with the cost after the start of the advertising activity, and calculates the budget digestion rate for each measure.
　制御部１１は、広告活動の成果等を算出する（ステップＳ３０９）。例えば制御部１１は、算案に基づき広告活動を開始してから現在までの成果（購買数）を計数する。また、制御部１１は予算案に係る成果目標と広告活動開始後の成果とを比較し、広告活動の成果について目標達成率を算出する。 The control unit 11 calculates the result of the advertising activity (step S309). For example, the control unit 11 counts the results (the number of purchases) from the start of the advertising activity to the present based on the proposal. Further, the control unit 11 compares the result target related to the budget plan with the result after the start of the advertising activity, and calculates the target achievement rate for the result of the advertising activity.
　制御部１１は、予算案に基づく広告活動について、予算の消化状況と、当該広告活動の成果とを示す進捗画面を出力する（ステップＳ３１０）。具体的に制御部１１は、ステップＳ３０８で算出した費用、消化率等の情報と、ステップＳ３０９で算出した成果、目標達成率等の情報とを進捗画面に表示する。より詳しくは、制御部１１は進捗画面に、予算消化状況及び目標達成度を示す概算データ１８１、広告活動の成果（商品又はサービスの購買数）及び施策別の費用を時系列で示す活動状況データ１８２、及び予算の消化状況の詳細を示す詳細表１８３等を出力する。 The control unit 11 outputs a progress screen showing the budget consumption status and the result of the advertising activity for the advertising activity based on the budget plan (step S310). Specifically, the control unit 11 displays information such as the cost and digestion rate calculated in step S308 and information such as the result and target achievement rate calculated in step S309 on the progress screen. More specifically, the control unit 11 displays, on the progress screen, rough data 181 indicating the budget consumption status and target achievement level, activity status data indicating the results of advertising activities (the number of products or services purchased) and costs for each measure in time series. 182 and a detailed table 183 showing details of the state of budget consumption are output.
　制御部１１は、広告活動の予定期間が経過したか否かを判定する（ステップＳ３１１）。予定期間は、予め設定されている広告活動の活動期間であり、例えば一ヶ月である。予定期間が経過していないと判定した場合（Ｓ３１１：ＮＯ）、制御部１１は処理をステップＳ３０７に戻す。予定期間が経過したと判定した場合（Ｓ３１１：ＹＥＳ）、制御部１１は一連の処理を終了する。 The control unit 11 determines whether or not the scheduled period of the advertising activity has passed (step S311). The scheduled period is an activity period of advertising activity set in advance, and is, for example, one month. If it is determined that the scheduled period has not elapsed (S311: NO), the control unit 11 returns the process to step S307. When it determines with the scheduled period having passed (S311: YES), the control part 11 complete | finishes a series of processes.
　なお、上記でサーバ１がプランニング画面及び進捗画面に係る種々の処理を実行したが、本実施の形態はこれに限るものではない。例えばサーバ１は生成済みの相関情報等を顧客端末５に送信し、顧客端末５が各画面の生成、予算案に係る条件の受け付け及び予算案の算出、進捗状況に係る消化予算等の算出などを行ってもよい。すなわち本実施の形態に係る処理は、顧客端末５等の端末装置が実行してもよい。 Although the server 1 executes various processes related to the planning screen and the progress screen as described above, the present embodiment is not limited to this. For example, the server 1 transmits generated correlation information or the like to the customer terminal 5, and the customer terminal 5 generates each screen, accepts a condition related to a budget plan, calculates a budget plan, calculates a digest budget related to progress, etc. May be performed. That is, the processing according to the present embodiment may be executed by a terminal device such as the customer terminal 5.
　以上より、本実施の形態３によれば、過去の広告活動の分析結果を流用して最適な予算案を組み立て、効率的な広告活動を支援することができる。 As described above, according to the third embodiment, it is possible to assemble an optimal budget plan using the analysis results of past advertising activities and support efficient advertising activities.
　また、本実施の形態３によれば、進捗状況を顧客企業に通知することで、上記のダッシュボード、ナビゲーション、プランニング等の画面と併せて、広告活動の適切なＰＤＣＡサイクルの構築を支援することができる。 In addition, according to the third embodiment, by notifying the client company of the progress status, it is possible to support the construction of an appropriate PDCA cycle for advertising activities together with the above-mentioned dashboard, navigation, planning and other screens. Can do.
　今回開示された実施の形態はすべての点で例示であって、制限的なものではないと考えられるべきである。本発明の範囲は、上記した意味ではなく、請求の範囲によって示され、請求の範囲と均等の意味及び範囲内でのすべての変更が含まれることが意図される。 It should be considered that the embodiment disclosed this time is illustrative in all respects and not restrictive. The scope of the present invention is defined not by the above-described meaning but by the scope of claims, and is intended to include all modifications within the meaning and scope equivalent to the scope of claims.
　１　サーバ
　１１　制御部
　１２　記憶部
　Ｐ　プログラム
　１３　通信部
　１４　大容量記憶装置
　１４１　広告ＤＢ
　１４２　成果ＤＢ
　１４３　費用ＤＢ
　２　視聴端末
　３　顧客サーバ
　４　サービスセンタ
　５　顧客端末
　８１　組み合わせグラフ
　８２　棒グラフ
　８３　集計表
　１２１　ボックス
　１２２　詳細ボックス
　１２４　貢献値グラフ
　１７１　相関グラフ
　１７１１　切替ボタン
　１７２２　設定オブジェクト
　１７２　成果比較グラフ
　１７３　予算比較グラフ
　１７４　予算表
　１７４１　シミュレーション欄
　１７５　確定ボタン
　１８１　概算データ
　１８２　活動状況データ
　１８３　詳細表 1 server 11 control unit 12 storage unit P program 13 communication unit 14 mass storage device 141 advertisement DB
142 Result DB
143 Cost DB
2 Viewing Terminal 3 Customer Server 4 Service Center 5 Customer Terminal 81 Combination Graph 82 Bar Graph 83 Total Table 121 Box 122 Detail Box 124 Contribution Value Graph 171 Correlation Graph 1711 Switch Button 1722 Setting Object 172 Result Comparison Graph 173 Budget Comparison Graph 174 Budget Table 1741 Simulation field 175 Confirm button 181 Approximate data 182 Activity status data 183 Detailed table

Claims (9)

  1. 　コンピュータに、
    　商品又はサービスの広告視聴状況を示す広告情報と、広告活動の成果を示す成果情報とを取得し、
    　前記広告情報及び成果情報に基づき、前記広告視聴状況と前記成果との相関関係を示す相関情報を生成する
    　処理を実行させることを特徴とするプログラム。     On the computer,
    Get advertisement information that shows the advertisement viewing status of the product or service and result information that shows the result of the advertising activity,
    A program for executing a process of generating correlation information indicating a correlation between the advertisement viewing state and the result based on the advertisement information and the result information.
  2. 　前記広告情報及び成果情報に基づき、多段階の回帰分析を含む共分散構造分析を実行して前記相関情報を生成する
    　ことを特徴とする請求項１に記載のプログラム。     The program according to claim 1, wherein the correlation information is generated by executing a covariance structure analysis including a multi-stage regression analysis based on the advertisement information and the result information.
  3. 　前記広告情報は、消費者の広告視聴態様が受動的である場合の受動視聴情報、能動的かつ間接的である場合の間接視聴情報、能動的かつ直接的である場合の直接視聴情報、及び前記消費者による前記商品又はサービスの確認の有無を示す確認情報を含み、
    　前記成果情報は、前記商品又はサービスの購買状況を示す購買情報であり、
    　前記受動視聴情報、間接視聴情報、直接視聴情報、確認情報、及び購買情報に含まれる各要素について回帰分析を行う
    　ことを特徴とする請求項２に記載のプログラム。     The advertising information includes passive viewing information when a consumer's advertising viewing mode is passive, indirect viewing information when active and indirect, direct viewing information when active and direct, and Including confirmation information indicating whether or not the product or service is confirmed by the consumer;
    The result information is purchase information indicating a purchase status of the product or service,
    The program according to claim 2, wherein regression analysis is performed for each element included in the passive viewing information, indirect viewing information, direct viewing information, confirmation information, and purchase information.
  4. 　前記相関情報に基づき、前記広告情報及び成果情報に含まれる要素間の相関関係を示すシーケンス画面を出力する
    　ことを特徴とする請求項１～請求項３のいずれか１項に記載のプログラム。     The program according to any one of claims 1 to 3, wherein a sequence screen showing a correlation between elements included in the advertisement information and the result information is output based on the correlation information.
  5. 　前記広告活動に関する費用情報を取得し、
    　前記広告情報、成果情報及び費用情報に基づき、前記相関情報を生成する
    　ことを特徴とする請求項１～請求項４のいずれか１項に記載のプログラム。     Obtaining cost information about the advertising activity;
    The program according to any one of claims 1 to 4, wherein the correlation information is generated based on the advertisement information, result information, and cost information.
  6. 　前記成果情報又は費用情報について条件入力を受け付け、
    　前記相関情報に基づき、前記広告活動に係る各広告施策について、受け付けた条件を満たす構成比を算出し、
    　算出した前記各広告施策の構成比を示す予算案を出力する
    　ことを特徴とする請求項５に記載のプログラム。     Accepting condition input for the outcome information or cost information,
    Based on the correlation information, for each advertising measure related to the advertising activity, calculate the composition ratio that satisfies the accepted condition,
    The program according to claim 5, wherein a budget plan indicating the calculated composition ratio of each advertising measure is output.
  7. 　前記予算案に基づく前記広告活動について、予算の消化状況と、前記成果とを示す進捗画面を出力する
    　ことを特徴とする請求項６に記載のプログラム。     The program according to claim 6, wherein a progress screen indicating a budget consumption status and the result is output for the advertising activity based on the budget plan.
  8. 　商品又はサービスの広告視聴状況を示す広告情報と、広告活動の成果を示す成果情報とを取得する取得部と、
    　前記広告情報及び成果情報に基づき、前記広告視聴状況と前記成果との相関関係を示す相関情報を生成する生成部と
    　を備えることを特徴とする情報処理装置。     An acquisition unit that acquires advertisement information indicating the advertisement viewing status of the product or service and result information indicating the result of the advertising activity;
    An information processing apparatus comprising: a generation unit that generates correlation information indicating a correlation between the advertisement viewing state and the result based on the advertisement information and the result information.
  9. 　コンピュータに、
    　商品又はサービスの広告視聴状況を示す広告情報と、広告活動の成果を示す成果情報とを取得し、
    　前記広告情報及び成果情報に基づき、前記広告視聴状況と前記成果との相関関係を示す相関情報を生成する
    　処理を実行させることを特徴とする情報処理方法。     On the computer,
    Get advertisement information that shows the advertisement viewing status of the product or service and result information that shows the result of the advertising activity,
    An information processing method comprising: executing a process of generating correlation information indicating a correlation between the advertisement viewing state and the result based on the advertisement information and the result information.
PCT/JP2017/005001 2016-02-15 2017-02-10 Program, information processing device, and information processing method WO2017141837A1 (en)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US201662295335P 2016-02-15 2016-02-15
US62/295,335 2016-02-15
JP2016-111015 2016-06-02
JP2016111015A JP6364581B2 (en) 2016-02-15 2016-06-02 Program, information processing apparatus and information processing method

Publications (1)

Publication Number Publication Date
WO2017141837A1 true WO2017141837A1 (en) 2017-08-24

Family

ID=59625893

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2017/005001 WO2017141837A1 (en) 2016-02-15 2017-02-10 Program, information processing device, and information processing method

Country Status (1)

Country Link
WO (1) WO2017141837A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7009719B1 (en) 2021-03-03 2022-01-26 しるし株式会社 Purchasing analysis system, purchasing analysis method, and computer program
JP7453076B2 (en) 2020-07-03 2024-03-19 株式会社日立製作所 Generation device, generation method, and generation program

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012088994A (en) * 2010-10-21 2012-05-10 Opto:Kk Advertising effect analysis system and method of the same
JP2015111382A (en) * 2013-12-06 2015-06-18 ヤフー株式会社 Advertisement device, estimation method, and estimation program

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012088994A (en) * 2010-10-21 2012-05-10 Opto:Kk Advertising effect analysis system and method of the same
JP2015111382A (en) * 2013-12-06 2015-06-18 ヤフー株式会社 Advertisement device, estimation method, and estimation program

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7453076B2 (en) 2020-07-03 2024-03-19 株式会社日立製作所 Generation device, generation method, and generation program
JP7009719B1 (en) 2021-03-03 2022-01-26 しるし株式会社 Purchasing analysis system, purchasing analysis method, and computer program
JP2022134595A (en) * 2021-03-03 2022-09-15 しるし株式会社 Purchase analysis system, purchase analysis method and computer program

Similar Documents

Publication Publication Date Title
Malthouse et al. Opportunities for and pitfalls of using big data in advertising research
Li et al. Attributing conversions in a multichannel online marketing environment: An empirical model and a field experiment
US10185971B2 (en) Systems and methods for planning and executing an advertising campaign targeting TV viewers and digital media viewers across formats and screen types
Draganska et al. Choice set heterogeneity and the role of advertising: An analysis with micro and macro data
Olbrich et al. Modeling consumer purchasing behavior in social shopping communities with clickstream data
Huang Designing utility-based recommender systems for e-commerce: Evaluation of preference-elicitation methods
US8140380B2 (en) Creating an incentive to author useful item reviews
Du et al. Immediate responses of online brand search and price search to TV ads
KR101167139B1 (en) Survey administration system and methods
Ha et al. Manufacturer rebate competition in a supply chain with a common retailer
Lai et al. Assessing and managing e-commerce service convenience
KR101817042B1 (en) System, method and computer-readable medium recorded with program for mediation of online video advertisement contract
Ali Measurement of e-services quality: an empirical study of University of Bahrain
Hackl et al. Market structure and market performance in E-commerce
JP6364581B2 (en) Program, information processing apparatus and information processing method
Holsing et al. Effect of user-generated content on website stickiness: the case of social shopping communities
WO2017141837A1 (en) Program, information processing device, and information processing method
KR102477687B1 (en) Apparatus and method for managing aadvertisement
KR20090097701A (en) Method for providing goods information
Mariani et al. The impact of social influence and trust on customer-to-customer online shoppers’ purchase intention: An empirical study in indonesia
Jürgensmeier et al. Measuring Fair Competition on Digital Platforms
Schreiber et al. Multivariate landing page optimization using hierarchical bayes choice-based conjoint
US20220374923A1 (en) Computational methods and processor systems for predictive marketing analysis
KR101696368B1 (en) System and Method for Purchasing Agency of Advertising Media
JP6914317B2 (en) Information providing device, information providing method, and information providing program

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17753095

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17753095

Country of ref document: EP

Kind code of ref document: A1