Information processing device, information processing method, and program

Abstract

To provide a technique capable of efficiently delivering advertisements to web pages highly relevant to commercial materials while excluding web pages with poor user experience in advance.SOLUTION: An information processing device comprises: an advertisement delivery unit 51 which delivers advertisements to web pages; a score calculation unit 52 which calculates an attention score on the basis of a plurality of indicators for evaluating the visibility of the web pages; a delivery destination determination unit 53 which excludes web pages with an attention score less than a predetermined reference value from advertising delivery targets: an EAR calculation unit 54 which calculates an EAR (Engaged Audience Rate) which is a ratio of the number of responses to the number of displays of questionnaire advertisements about web pages with an attention score higher than or equal to the reference value; a weighting calculation unit 55 which weighs the web pages on the basis of the calculated EAR; and a delivery volume determination unit 56 which determines a delivery volume of advertisements to the web pages according to the weighting.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] Traditionally, in the field of online advertising, it has been required to reach (display advertisements to) those who are interested in a product, and the click-through rate (CTR) has been used as an indicator of success. The click-through rate is calculated by dividing the number of clicks by the number of impressions (number of times the advertisement is displayed). Click-through rates are disclosed, for example, in Patent Document 1 below. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-40386 Summary of the Invention [Problem to be solved by the invention]

[0004] When observing web pages with unusually high click-through rates, it is very common to see them using ad formats designed to generate unintended clicks. This is most evident in overlay ads that appear on top of content and interstitial ads that appear full-screen during page transitions, and web pages that use these ad formats commonly experience a significantly poor user experience (referring to the quality of the user's entire experience when browsing a web page, such as whether or not they were able to take the intended action, such as reading an article).

[0005] The present invention has been made in consideration of such circumstances, and aims to provide a technology that can efficiently deliver advertisements to web pages that are highly relevant to a product while excluding web pages that provide a poor user experience in advance. [Means for solving the problem]

[0006] In order to achieve the above object, an information processing device according to one aspect of the present invention comprises: An information processing device that distributes online advertisements, an advertisement distribution means for distributing advertisements to a web page; a score calculation means for calculating an attention score based on a plurality of indices for evaluating the visibility of the web page; a destination determination means for excluding web pages having an attention score less than a predetermined reference value from advertisement delivery targets; an EAR calculation means for calculating an EAR (Engaged Audience Rate), which is the ratio of the number of responses to the number of times a survey advertisement is displayed, for web pages whose attention scores are equal to or greater than the reference value; a weight calculation means for weighting the web page based on the calculated EAR; a distribution amount determination means for determining an advertisement distribution amount to the web page in accordance with the weighting; Equipped with.

[0007] An information processing method and a program according to one aspect of the present invention are a method and a program corresponding to an information processing device according to one aspect of the present invention. [Effects of the Invention]

[0008] According to the present invention, it is possible to efficiently deliver advertisements to web pages that are highly relevant to a product while excluding web pages that provide a poor user experience in advance. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram showing an overview of a service that can be realized by an information processing system to which a server according to an embodiment of the information processing device of the present invention is applied; [Figure 2] 1 is a diagram illustrating an example of a configuration of an information processing system to which a server according to an embodiment of the information processing device of the present invention is applied. [Figure 3] 3 is a block diagram showing an example of a hardware configuration of a server in the information processing system of FIG. 2. FIG. [Figure 4] 4 is a functional block diagram showing an example of a functional configuration of the server of FIG. 3 that constitutes the information processing system of FIG. 2. [Figure 5] 1A and 1B are diagrams illustrating examples of a campaign advertisement and a questionnaire advertisement according to the present embodiment. [Figure 6] FIG. 10 is a diagram showing the difference in clickable areas (areas that can be clicked) between a normal advertisement and a survey advertisement. [Figure 7] FIG. 10 is a diagram showing distribution groups in the EAR evaluation experiment. [Figure 8] FIG. 10 is a diagram showing the layout and question contents of a questionnaire advertisement for the EAR evaluation experiment. [Figure 9] FIG. 10 is a diagram showing the results of an EAR evaluation experiment. [Figure 10] 10A and 10B are diagrams relating to the EAR evaluation experiment (smartphone), in which FIG. 10A shows a group of distributions, and FIG. 10B shows changes to a survey advertisement. [Figure 11] FIG. 10 shows the layout of a questionnaire advertisement and the content of questions in the EAR evaluation experiment (smartphone). [Figure 12] 10(a) and 10(b) are diagrams showing the results of an EAR evaluation experiment (smartphone), respectively. [Figure 13] FIG. 1 is a diagram illustrating a first example of an advertisement distribution method using an EAR. [Figure 14] FIG. 10 is a diagram showing an example of an optimal delivery amount (number of advertisement impressions) of a web page in the first example. [Figure 15] FIG. 10 is a diagram showing the format of EAR Ads. [Figure 16] FIG. 10 is a diagram illustrating a second example of an advertisement distribution method using an EAR. [Figure 17] FIG. 10 is a diagram illustrating an example of weighted optimization distribution using EAR. [Figure 18]FIG. 10 is a diagram showing an example of the operation flow of EAR Ads on the advertisement distribution platform, including the behavior up to advertisement distribution. [Figure 19] FIG. 10 is a diagram showing an operational flow of EAR Ads on an advertisement distribution platform, illustrating an example of behavior after advertisement distribution (behavior when answering a questionnaire). [Figure 20] FIG. 1 illustrates an example of a mechanism for attention scoring (assessing the visibility of a web page). [Figure 21] FIG. 10 is a diagram showing an example of numerical values ​​of attention scores. [Figure 22] FIG. 10 is a diagram illustrating an example of EAR advertisement delivery utilizing attention scores. [Figure 23] FIG. 10 is a diagram showing an example of exclusion due to a low attention score. [Figure 24] FIG. 10 is a diagram showing an example of the operational flow of EAR Ads on the advertisement distribution platform, from advertisement distribution to attention score measurement. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0011] First, referring to Figure 1, we will explain an overview of a service (hereinafter referred to as "this service") that can be realized by an information processing system (see Figure 2 described later) to which a server according to one embodiment of the information processing device of the present invention is applied. FIG. 1 is a diagram showing an outline of the present service that can be realized by an information processing system to which a server according to an embodiment of the information processing device of the present invention is applied.

[0012] This service is an online advertising distribution service that calculates an attention score based on multiple indicators that evaluate the visibility of a web page, excludes web pages with an attention score below a specified standard value from advertising distribution targets, and calculates an Engaged Audience Rate (EAR), which is the ratio of the number of responses to the number of times a survey advertisement is displayed, for web pages that are above the standard value, weights the web pages based on this EAR, and determines the amount of advertising distribution based on that weighting.

[0013] This service first uses attention scores to filter out web pages with poor user experiences, and then uses EAR to efficiently deliver advertisements to web pages that are highly relevant to the product. This service has the effect of being able to efficiently deliver advertisements to web pages that are highly relevant to the product while excluding web pages that provide a poor user experience in advance.

[0014] For example, the service in the example of FIG. 1 is provided by a service manager SK to users U1 to Un (n is an integer value of 1 or more). Therefore, the present service in the example of FIG. 1 uses a server 1 managed by a service manager SK and user terminals 2-1 to 2-n managed by users U1 to Un, respectively. The server 1 is an information processing device that distributes online advertisements.

[0015] The example of this service in FIG. 1 will be described in detail below with reference to other figures. In step S1 of the example of this service in Fig. 1, the server 1 executes advertisement distribution. Specifically, the server 1 distributes advertisements to web pages. The advertisements here refer to, for example, campaign advertisements and survey advertisements related to commercial products (described later). In step S2, the server 1 executes score calculation. Specifically, the server 1 calculates an attention score, which will be described later, based on a plurality of indices for evaluating the visibility of a web page. In step S3, the server 1 executes a distribution destination determination. Specifically, the server 1 excludes web pages with attention scores below a predetermined reference value from advertisement distribution targets.

[0016] In step S4, the server 1 calculates the EAR. Specifically, the server 1 calculates the EAR (Engaged Audience Rate) (described later), which is the ratio of the number of responses to the number of times a survey advertisement is displayed, for web pages whose attention scores are equal to or greater than a reference value. In step S5, the server 1 performs weighting calculation. Specifically, the server 1 weights the web pages based on the calculated EAR. In step S6, the server 1 determines the amount of advertisement to be delivered to the web page in accordance with the weighting.

[0017] As described above, by combining the attention score and the EAR, the server 1 can efficiently deliver advertisements to web pages that are highly relevant to the product while excluding web pages that provide poor user experience in advance.

[0018] In addition, the attention score calculated by server 1 is calculated based on at least one of the following: a time-in-view score indicating the viewing time of the advertisement; an in-view rate score indicating the percentage of the advertisement that reached the appropriate viewing time; a click timing score indicating the time from the advertisement display to the click; an advertisement space number score indicating the number of advertisement spaces within the web page; and an advertisement space occupancy rate score indicating the display ratio of advertisement spaces to the web page display area. This allows for more flexible visibility evaluation, as appropriate indicators can be selectively used depending on the characteristics of each web page and the measurement environment.

[0019] In addition, the advertisement delivery by the server 1 will be in a two-stage display format, including a first screen containing questions and answer options regarding the product, and a second screen containing an advertisement for the product that is displayed after the answer option is selected. This allows for effective brand awareness while gradually drawing out user interest.

[0020] Furthermore, it is preferable to set the distribution rate of the questionnaire advertisement within a predetermined range for calculating the EAR by the server 1. By specifying an appropriate distribution rate, it is possible to ensure the accuracy of EAR measurement while minimizing the impact on normal advertisement distribution.

[0021] Next, with reference to FIG. 2, a description will be given of the configuration of an information processing system that realizes the provision of the above-described service, that is, an information processing system to which a server according to an embodiment of the information processing device of the present invention is applied. FIG. 2 is a diagram showing an example of the configuration of an information processing system to which a server according to an embodiment of the information processing device of the present invention is applied.

[0022] The information processing system shown in FIG. 2 is configured to include a server 1 and user terminals 2-1 to 2-n (n is an integer value of 1 or more). The server 1 and the user terminals 2-1 to 2-n are connected to each other via a network NW such as the Internet.

[0023] The server 1 is an information processing device managed by a service manager SK (see FIG. 1) of this service. The server 1 executes various processes for realizing this service while appropriately communicating with user terminals 2-1 to 2-n.

[0024] The user terminals 2-1 to 2-n are information processing devices operated by users U1 to Un (see FIG. 1), respectively, and are configured as smartphones, tablets, personal computers, or the like. When there is no need to distinguish between them, they will be referred to as user terminals 2.

[0025] FIG. 3 is a block diagram showing an example of a hardware configuration of a server in the information processing system shown in FIG.

[0026] The server 1 is configured to include a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a bus 14, an input / output interface 15, an input unit 16, an output unit 17, a memory unit 18, a communication unit 19, and a drive 20.

[0027] The CPU 11 executes various processes according to a program recorded in the ROM 12 or a program loaded from the storage unit 18 into the RAM 13 . The RAM 13 also stores data and the like necessary for the CPU 11 to execute various processes.

[0028] The CPU 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output interface 15 is also connected to this bus 14. To the input / output interface 15, an input unit 16, an output unit 17, a storage unit 18, a communication unit 19, and a drive 20 are connected.

[0029] The input unit 16 is configured by, for example, a keyboard, and accepts input of various information. The output unit 17 is configured with a display such as a liquid crystal display, a speaker, etc., and outputs various information as images and sounds. The storage unit 18 is configured with a DRAM (Dynamic Random Access Memory) or the like, and stores various data. The communication unit 19 communicates with other devices (for example, the user terminal 2 in FIG. 2) via a network NW including the Internet.

[0030] Removable media 30, such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, is appropriately attached to the drive 20. A program read from the removable media 30 by the drive 20 is installed in the storage unit 18 as needed. Furthermore, the removable medium 30 can also store various data stored in the storage unit 18 in the same manner as the storage unit 18.

[0031] Although not shown, the user terminal 2 in Fig. 2 can also have a configuration that is basically the same as the hardware configuration shown in Fig. 3. Therefore, a description of the hardware configuration of the user terminal 2 will be omitted.

[0032] The various hardware and software components constituting the information processing system of FIG. 2, including the server 1 of FIG. 3, work together to execute various processes for providing the present service of FIG.

[0033] FIG. 4 is a functional block diagram showing an example of the functional configuration of the server of FIG. 3 in the information processing system of FIG.

[0034] As shown in FIG. 4, in the CPU 11 of the server 1, an advertisement distribution unit 51, a score calculation unit 52, a distribution destination determination unit 53, an EAR calculation unit 54, a weight calculation unit 55, and a distribution amount determination unit 56 function. In addition, an area of ​​the storage unit 18 of the server 1 stores an advertisement distribution information DB 71, a score information DB 72, a distribution destination determination information DB 73, an EAR information DB 74, a weighting information DB 75, and a distribution amount determination information DB 76.

[0035] The advertisement distribution unit 51 executes control to distribute advertisements to web pages. Specifically, the advertisement distribution unit 51 distributes regular campaign advertisements and survey advertisements related to merchandise to web pages at a predetermined rate (a predetermined range distribution rate (e.g., 2-4%)). The survey advertisements include questions and options related to the merchandise, and are configured to allow the user to consciously make a selection. Furthermore, in the survey advertisements, only the radio buttons of the options can be made clickable to prevent erroneous operation. The advertisement distribution unit 51 stores and manages the URL information of the web page and information related to advertisement distribution in the advertisement distribution information DB 71.

[0036] The advertisement distribution unit 51 can distribute the questionnaire advertisement in a two-stage display format including a first screen including questions and answer options related to the product, and a second screen including an advertisement for the product that is displayed after the answer option is selected. The advertisement distribution unit 51 can achieve effective brand recognition while gradually drawing out the user's interest. Furthermore, the advertisement distribution unit 51 can display the brand logo or product image of the merchandise larger on the second screen after the answer to the questionnaire advertisement is selected than on the first screen. By impressing the brand at a point when the user's interest is heightened, more effective brand recall can be achieved.

[0037] The score calculation unit 52 executes control to calculate the attention score based on a plurality of indices for evaluating the visibility of a web page. Specifically, the score calculation unit 52 uses at least one of the time-in-view score, in-view rate score, click timing score, number of ad spaces score, and ad space occupancy rate score, weights these, and calculates the attention score so that the sum of these is 1. The score calculation unit 52 stores and manages information relating to each score in the score information DB 72. The score calculation unit 52 can selectively use appropriate indices depending on the characteristics of each web page and the measurement environment, thereby enabling more flexible visibility evaluation.

[0038] The above-mentioned time-in-view score is a score that determines that a web page has a poor layout if the time between when the survey advertisement is displayed and when the user leaves the page is less than a first predetermined number of seconds (e.g., less than 3 seconds). Specific time criteria allow for objective and quantitative determination of low-quality content.

[0039] The above-mentioned click timing score is a score for determining a click as an unintentional click when the click occurs within a second predetermined second (for example, within one second) that is shorter than the above-mentioned first predetermined second from the advertisement display. Since unintentional clicks can be quantitatively determined from a time perspective, the accuracy of measuring advertising effectiveness can be improved.

[0040] The attention score described above is a score calculated so that the sum of the weights of at least one selected score is 1. By making it possible to adjust the importance of each index depending on the situation, it is possible to achieve a more appropriate visibility evaluation.

[0041] The delivery destination determination unit 53 executes control to exclude web pages with attention scores below a predetermined reference value (or equal to or less than the reference value) from advertisement delivery targets. Specifically, for example, if a predetermined reference value for the attention score is 40 points, the delivery destination determination unit 53 can determine that a web page with an attention score of less than 40 points (40 points or less, as described below, in this embodiment) has an extremely poor user experience and exclude it. This makes it possible to avoid delivery of advertisements to web pages that induce unintended clicks. The delivery destination determination unit 53 stores and manages information relating to advertisement delivery targets in the delivery destination determination information DB 73 .

[0042] The EAR calculation unit 54 executes control to calculate the EAR (described later), which is the ratio of the number of responses to the number of times a survey advertisement is displayed, for web pages whose attention scores are equal to or greater than a reference value (exceed the reference value). Specifically, the EAR calculation unit 54 can quantify the degree of interest in the product among users viewing the web page by dividing the number of responses to a survey advertisement related to the product by the number of times it is displayed. The EAR calculation unit 54 stores and manages information related to the EAR in the EAR information DB 74.

[0043] The weight calculation unit 55 executes control to weight the web pages based on the calculated EAR. Specifically, the weighting calculation unit 55 can calculate the weighting of the amount of distribution to each web page by dividing the EAR of each web page by the total EAR of all target web pages. This allows more advertisements to be distributed to web pages that are highly relevant to the product. The weight calculation unit 55 stores and manages information relating to the weight in the distribution amount determination information DB 76.

[0044] The server 1 as described above corresponds to an advertisement distribution server (on-premise or cloud) which will be described later with reference to Figures 18, 19 and 24. The storage unit 18 also corresponds to a database (on-premise or cloud) which will be described later with reference to Figures 18, 19 and 24.

[0045] From here on, we will use specific examples to explain this service, including its background and issues.

[0046] First, we will explain how general online advertising metrics work. The ultimate goal of an online advertising campaign is primarily to promote product sales and increase brand awareness. To achieve this, it is necessary to "reach" (display ads) people who are "interested" in the product or item. The most commonly used indicator of success in this method (reaching users who are interested) is the click-through rate (also known as CTR [Click Through Rate]). Click-through rate is calculated using the following formula:

[0047] Clicks / impressions (number of times an ad is displayed)

[0048] An impression is when an ad appears on a web page, and a click is when a user clicks on the ad with their mouse (or taps on a mobile device), which takes them to the advertiser's website promoting the product.

[0049] Click-through rate is an indicator created on the assumption that users (viewers) who are interested in the product will likely click on the ad because they have a strong desire to purchase or gather information. Theoretically, factors that affect click-through rates are as follows: (1) the response rate of the advertising materials used, (2) the match between the promoted product and the web page content on which the advertisement is displayed (assuming that users viewing that content are also interested in related products), and (3) the timing of the advertisement display (day of the week and time of day).

[0050] As mentioned above, (1) can contribute to the click-through rate by designing the advertising material itself to encourage users to respond positively, but in the end, there is little point in arbitrarily making users who are not very interested click, so the key point to actually evaluate in terms of click-through rate is the match between the product (mentioned above in (2)) and the web page on which the advertisement is displayed.

[0051] In modern online advertising campaigns, it is common to deliver ads to hundreds or even thousands of web pages at once, but web media (domains) and web pages are evaluated by click-through rate, and ads are delivered actively to media and web pages with high click-through rates, while delivery to media and pages with low click-through rates is weakened or stopped, thereby optimizing delivery toward the ultimate goal of "contributing to sales" and "increasing brand awareness." Modern ad delivery platforms primarily automate this type of click-through rate optimization.

[0052] Next, the problem of click-through rate will be explained. As mentioned above, click-through rates are based on the assumption that people interested in a particular product will click on the ad. However, with the evolution of the online advertising market and changes in the environment surrounding users, major problems have come to light.

[0053] This applies not only to advertisers who want to efficiently deliver ads to interested users, but also to the web media that delivers those ads. This is because media revenues tend to increase with a higher click-through rate. In many cases, pay-per-click systems only pay media when a click actually occurs, and even in pay-per-impression systems, where payment is made when an ad is actually displayed, there is a tendency to allocate budgets to media that efficiently generate clicks (optimizing ad delivery). In other words, a high click-through rate for the media itself leads to profitability.

[0054] The most effective methods that media companies can use to increase click-through rates include optimizing the position of advertising space and optimizing the timing of advertising display. Simply put, optimizing the position of an ad space means placing an ad space on a web page in a location that catches the user's eye, or changing the web page layout. However, depending on the medium that wants to maximize revenue, there may be cases where ad space is placed in a location that induces accidental ad clicks or the layout is changed. For example, there are cases where ad space is inserted into a list of recommended articles. Also, when placing ads on mobile devices, ads that are always displayed at the bottom of the screen (anchor ads) are often prone to being clicked by mistake.

[0055] A more aggressive ad layout on mobile devices is overlay ads that cover content, which also induces a lot of unwanted clicks. As a method for media companies to efficiently optimize revenue from clicks, they often use layout changes and timing of ad display. For example, a common method is to display an ad at the same time as clicking the "Read the main text" button in the content, in order to encourage clicks.

[0056] Additionally, interstitial ads, which are full-screen ads that are displayed when clicking a link or button to navigate to a new page, often result in unintended clicks due to forced viewing. In the worst cases, the button to close the full-screen ad is difficult to find, further encouraging accidental clicks. In addition to interstitial ads, some websites now display full-screen or overlay ads when you press the back button on your browser.

[0057] When we look at the methods for increasing click-through rates on the media side, it becomes clear that it becomes extremely difficult to "assess the match between the promoted product and the web page content displayed in the advertisement," which is what advertisers originally wanted to achieve through click-through rates. This is because the optimization techniques used by media outlets often end up being mechanisms that induce users to make the wrong clicks and have nothing to do with the user's interests. To take an extreme example, if the ad being served is for women's cosmetic products, it may still be clicked on even if it is served alongside martial arts content, which is likely to have a predominantly male audience.

[0058] In fact, a survey conducted by a US company found that there is absolutely no correlation between click-through rates and the ultimate goals of advertisers' campaigns: "purchase intent (sales)," "brand awareness," and "ad awareness."

[0059] Next, we will explain the mechanism behind the new indicator, EAR. It is highly likely that the click-through rate cannot accurately evaluate the important objective of online advertising distribution, "the match between the promoted product and the web page content on which the advertisement is displayed (assuming that users viewing that content are also interested in related products)." Therefore, we will use EAR (Engaged Audience Rate) as a new indicator to replace it.

[0060] First, a survey advertisement related to the product is used to evaluate the match between the promoted product and the web page content on which the advertisement is displayed. FIG. 5 is a diagram showing an example of a campaign advertisement and a questionnaire advertisement in this embodiment.

[0061] Survey ads are used to assess whether users viewing a particular web page are interested in the product. The left side of Figure 5 shows a regular campaign advertisement, while the right side shows a survey advertisement, which includes questions and options about the product. The main purposes of survey ads are twofold: (1) to see if users viewing a specific web page are interested in the product, and (2) to prevent users from answering the survey by making a mistake (unlike clicking, the probability of making a mistake is extremely low). Regarding (1), the evaluation is based on the assumption that if people are interested, they are likely to answer the survey. On the other hand, regarding (2), not all advertisements are clickable, and people must view the survey content and consciously choose their answer.

[0062] Based on these two objectives, survey ads are structured so that questions and options are clearly displayed, allowing users to consciously select their answers, as shown in Figure 5. This makes it possible to properly measure the EAR (Engaged Audience Rate), which is the ratio of the number of responses to the number of times the survey ad is displayed, as described below. This configuration allows for a more accurate evaluation of the match between the product and the web page content than the click-through rate, and in particular, eliminates errors caused by accidental or unintended clicks, enabling a more accurate measurement of interest and attention.

[0063] FIG. 6 is a diagram showing the difference in clickable areas (areas that can be clicked) between a normal advertisement and a questionnaire advertisement.

[0064] Figure 6(a) shows an example of a regular advertisement, where the entire advertisement display area is clickable, which makes it easy for users to click by mistake. In contrast, Figure 6(b) shows an example of a survey advertisement, where only the radio buttons are clickable, so the viewer must look at the survey content and consciously select the answer.

[0065] Specifically, in the normal advertisement in Figure 6(a), the entire advertisement banner is a clickable area, so unintentional clicks may occur depending on the display position of the advertisement and the timing of the user's operation. In particular, if a click occurs within a short time-in-view period of less than the first predetermined second, or if a click occurs within the second predetermined second from the time of display (however, the second predetermined second is shorter than the first predetermined second), it is likely to be an unintentional click. In contrast, the survey ad in Figure 6(b) requires the user to consciously select an option by making only the limited area of ​​the radio button clickable. This prevents unintentional clicks and allows only responses from users who are genuinely interested in the product to be collected.

[0066] This configuration makes it possible to more accurately measure the Engaged Audience Rate (EAR), which is the ratio of the number of responses to the number of times a survey ad is displayed, and to appropriately evaluate the match between the product and the web page content based on the measurement results.

[0067] In today's online advertising, the number of web pages to be delivered is extremely large (hundreds to thousands, and sometimes even tens of thousands), and new content (web pages) is constantly being added, so survey ads are also delivered randomly (at a rate of about 2-4%) to the same web pages as the actual campaign ads. Also, since the content of the survey advertisements is solely intended to gauge interest, they do not ask complex questions about the product (as difficult questions may scare people away from the survey advertisement in the first place).

[0068] Here we will show the formula for calculating EAR. The formula for calculating EAR is as follows: EAR = Number of survey responses / Number of survey ad impressions If a survey ad is displayed 100 times and there is one response, the EAR is 1%. By using this EAR to evaluate multiple web pages, it is possible to limit the amount of ads delivered to web pages with low EAR and deliver more ads to web pages with high EAR.

[0069] Here, we formulate a hypothesis: "When a survey ad is delivered to content (webpage) related to a product, users viewing that webpage are more likely to be interested in the survey content and to respond to it; conversely, when a survey ad is delivered to content (webpage) that is completely unrelated to the product, users will not participate in the survey because the content is not of interest to them."

[0070] FIG. 7 is a diagram showing distribution groups in the EAR evaluation experiment. FIG. 8 shows the layout of the questionnaire advertisement and the content of the questions in the EAR evaluation experiment. Note that "web page" may be written as "web page" in the figures from Figure 7 onwards.

[0071] To prove the validity of EAR as a proposed index, we conducted the following evaluation experiment. Decide on the product and distribute it to the following three groups. That is, the distribution is performed to three groups: (a) a group of related web pages, (b) a group of web pages that are assumed to be largely unrelated, and (c) a group of web pages with a high click-through rate.

[0072] Here, we assume that (a) has a high EAR and (b) has a low EAR. (c) is a group of web pages with a high click-through rate, and is assumed to contain many false clicks, so we hypothesize that it also has a low EAR. In addition, because the degree to which advertising banner images occupy the screen and the user's behavior and operations differ between the PC and smartphone environments, separate experiments will be conducted for each.

[0073] As shown in Figure 7, in this experiment, the product was a "protein shake," and six types of protein-related survey advertisements were delivered to the following three groups: Group A is a group of web pages related to protein (number of URLs: 2007), Group B is a group of web pages related to gourmet food (number of URLs: 200), and Group C is a group of web pages with a high click-through rate (number of URLs: 125).

[0074] The reason for the difference in the number of URLs is that in the case of niche content, the number of users who view it is small, so in order to achieve roughly the same number of ad displays within the same period, it was necessary to increase the number of target URLs. Also, the group of webpages with high click-through rates in Group C are webpages that have been proven to have high click-through rates based on the actual delivery results of protein products delivered in December 2023.

[0075] As shown in Figure 8, the survey advertisements distributed were six in total, each with two different layouts and three questions each. This experiment makes it possible to evaluate the match between the product and the web page, as well as the user's true level of interest.

[0076] 9 shows the results of the EAR evaluation experiment, specifically the results of questionnaire advertisement distribution on PCs. The distribution summary and EAR results are shown below.

[0077] In Group A (protein-related), the number of survey ad impressions was 41,446, while the number of survey responses was 92, resulting in an EAR of 0.22%. In Group B (food-related), the number of survey ad impressions was 46,322, while the number of survey responses was 44, resulting in an EAR of 0.09%, which is 2.34 times higher than Group A. In group c (high click-through rate), the number of survey ad impressions was 43,588, while the number of survey responses was 52, resulting in an EAR of 0.12%, which is 1.86 times higher than group a.

[0078] As can be seen in Figure 9, when the survey advertisement is related to protein, users viewing a web page containing protein content (group a) are 2.34 times more likely to respond to the survey than users viewing a completely unrelated gourmet-related content (group b). Additionally, click-through rate is an indicator that is currently considered in the advertising industry to measure the match between advertisements and content, but viewers of the group of web pages with high click-through rates (group c) had a low response rate, just like group b. The web pages in Group C were actually selected from web pages that had a high click-through rate at the time when a protein shake campaign was being distributed, but it can be seen that the viewers of these web pages were not particularly interested in the protein shake product.

[0079] In this way, the hypothesis that the closer the question (product) in the survey ad is to the content on the web page, the more likely users who view that content are to respond to the survey has been proven. EAR is therefore considered to be superior to click-through rate in terms of "evaluating the match between the promoted product and the web page content displayed in the advertisement," which is one of the important purposes of online advertising campaigns.

[0080] Next, we will explain the experimental results on smartphones. FIG. 10 is a diagram relating to the EAR evaluation experiment (smartphone), in which (a) is a diagram showing distribution groups, and (b) is a diagram showing changes to a survey advertisement. FIG. 11 is a diagram relating to the EAR evaluation experiment (smartphone), in which (a) is a diagram showing distribution groups, and (b) is a diagram showing changes to survey advertisements. FIG. 12 is a diagram showing the layout of a questionnaire advertisement and the content of questions in the EAR evaluation experiment (smartphone). 13(a) and 13(b) are diagrams showing the results of the EAR evaluation experiment (smartphone).

[0081] Similar to the PC experiment described above, six types of protein-related survey advertisements (see Figure 12) were distributed to three groups (see Figure 10) from December 18, 2023 to January 17, 2024, and displayed approximately 40,000 times to test the effectiveness of EAR.

[0082] As shown in Figure 10, the experiment was conducted on the following three groups: That is, the survey was conducted on three groups: Group A, a group of protein-related web pages (number of URLs: 2007); Group B, a group of gourmet-related web pages (number of URLs: 200); and Group C, a group of web pages with high click-through rates (number of URLs: 125). However, for smartphones, the method for responding to the survey advertisements was changed as shown in Figure 11 in order to minimize the possibility of users responding to the survey incorrectly due to the high screen occupancy rate of the advertisements and special formats such as interstitial advertisements.

[0083] As shown in Figure 11(a), when a survey ad is displayed in an interstitial ad space, there is a possibility that users may mistakenly believe that the ad will disappear once they answer the survey, which could lead to them answering surveys that they are not interested in. Therefore, as shown in Figure 11(b), in order to eliminate such erroneous operations, we changed the survey response method as follows: (1) Select answers using radio buttons, and (2) Click explicitly to answer or not answer (see also Figure 12).

[0084] The distribution summary and EAR results on the smartphone are shown in Figure 13. In Group A (protein-related), the number of survey ad impressions was 40,293, but the number of survey responses (including no responses) was 32, resulting in an EAR of 0.079%. On the other hand, the number of responses without responses was 20, resulting in an EAR of 0.050%. In Group B (food-related), the number of survey ad impressions was 40,265, but the number of survey responses (including no responses) was 18, resulting in an EAR of 0.045%. On the other hand, the number of responses that did not include a response was 7, resulting in an EAR of 0.017%. In group c (high click-through rate), the number of survey ad impressions was 47,063, but the number of survey responses (including no responses) was 18, resulting in an EAR of 0.038%. On the other hand, the number of responses without responses was 7, resulting in an EAR of 0.015%.

[0085] As a result, when we looked further at the strictness of the responses, in the case of "not including no response", the difference in EAR was even greater than that in PC. When the survey ad was protein-related, users viewing a web page containing protein content (group a) were 2.85 times more likely to complete the survey than users viewing completely unrelated gourmet content (group b), and 3.33 times more likely to complete the survey than users viewing a highly click-through ad (group c). In addition, the intention behind the comparison of EAR "including not answering" was that although people chose not to answer, they consciously paid attention to protein-related questions, and so can be defined as having a certain degree of interest or concern. It is interesting to note that when compared with group c, which is a group of sites where clicking due to incorrect operation is induced, the EAR is two-fold different from that of the protein-related group (a).

[0086] A first embodiment of an advertisement distribution method using an EAR will be described. FIG. 13 is a diagram showing a first example of an advertisement distribution method using an EAR. FIG. 14 is a diagram showing an example of the optimum delivery amount (number of times advertisements are displayed) of the web page in the first example.

[0087] EAR is considered to be a more effective indicator than click-through rate for measuring the "match between the promoted product and the web page content on which the advertisement is displayed," and EAR is used to efficiently reach (display advertisements to) "those with an interest or concern." As an example, let's say an advertising distribution company (advertising distribution platform company) receives an order from an advertiser for an advertising campaign budget of 1 million yen at a cost of 1 yen per impression. A typical online advertising platform distributes ads to hundreds or thousands of web pages, but for simplicity's sake, we will distribute ads to three different web pages (A, B, and C), and the display rate of the survey ad will be 4%.

[0088] When the number of impressions of the survey exceeds a certain threshold (for example, 4,000 impressions), the EAR of each webpage is evaluated.As shown in Figure 13, assume that, with a budget of 1 million yen, 100,000 impressions are sent to each webpage (a budget of 300,000 yen has been used), and the EAR results are: Webpage A = EAR 0.2%, Webpage B = EAR 0.4%, and Webpage C = EAR 0.1%.

[0089] In this case, we can create an optimization logic that weights and distributes the remaining 700,000 yen. The weighting can be expressed as follows: Weighting of Web Page A = EAR of Web Page A / (EAR of Web Page A + EAR of Web Page B + EAR of Web Page C) Weighting of web page A = 0.2 / (0.2 + 0.4 + 0.1) = 0.286 ≒ 0.29 The optimized ad delivery volume is calculated using the following formula: Optimal delivery volume for web page A = remaining budget x weighting for web page A Optimal delivery volume for web page A = 700,000 yen x 0.29 = 203,000 Therefore, the budget for advertising will be 203,000 yen. Note that the calculations for web pages B and C are omitted as they are calculated in the same way as web page A.

[0090] FIG. 15 shows the optimal delivery volume (number of advertisement impressions) for each web page. This allows for efficient display of advertisements to those who are interested in the product by utilizing EAR. As an advertising distribution platform, it is required to automate this process. Furthermore, as EAR is measured for various advertising campaigns, general EAR figures (average and median) will be determined, and a new logic could be envisaged in which no advertisements are delivered to web pages where EAR measurements deviate below a certain EAR.

[0091] A second embodiment of the advertisement distribution method using the EAR will be described. FIG. 15 is a diagram showing the format of EAR Ads. FIG. 16 is a diagram showing a second example of an advertisement distribution method using an EAR. FIG. 17 is a diagram showing an example of weighted optimization distribution using EAR.

[0092] In the first embodiment of the advertising distribution method described above, the EAR of each web page is learned using survey ads, and actual ads are distributed in an optimized form based on the results. However, it is also possible to decorate survey ads with the advertiser's logo and product images and distribute them as actual commercial ads. We will call these "EAR Ads." This has the following advantages:

[0093] That is, (1) since the survey advertisement itself can quickly generate ad impressions, the period for evaluating the EAR of a web page can be shortened. Another advantage is that (2) when the attention of a user who is interested in content (surveys) related to the product is captured, showing the brand logo or product image can contribute to "brand recall (the association of seeing a specific item with a specific brand)." Another advantage is that (3) when the "Thank you for your cooperation" message is displayed after selecting an answer (called the second interaction page), the brand logo and product image can be displayed in larger size, further increasing brand recall. In addition, (4) when developing automatic delivery optimization logic, the EAR can be obtained from the advertisement itself, which has the advantage of enabling optimized delivery to a large number of web pages in near real time. In addition, (5) the revenue model has the advantage of allowing for easy pricing, such as "XX yen per 1 EAR Ad display."

[0094] Figure 15 shows the format of "EAR Ads." It consists of two screens: the first interaction screen (questionnaire screen) and the second interaction screen (thank you screen). After answering the question, a second interaction page will be displayed, and clicking a button will take you to the advertiser's web page. As shown in FIG. 16, after answering the first interaction (questionnaire screen), the second interaction (thank you screen) can be displayed. FIG. 17 shows the optimal delivery volume (number of advertisement impressions) for each web page. This allows for efficient display of advertisements to "those who are interested in the product" by utilizing EAR, as in Figure 14.

[0095] The behavior up to ad delivery will be explained. FIG. 18 is a diagram showing an example of the operation flow of EAR Ads on the advertisement distribution platform, and the behavior up to advertisement distribution.

[0096] In FIG. 18, assuming that multiple EAR Ads advertising campaigns are registered on the advertising distribution platform, the decision as to which advertisement to distribute is generally made through the following steps. First, (1) place the ad space tag (consisting of Javascript and HTML) on the web page. Next, (2) the ad space tag is loaded at the same time as the web page is loaded. Next, (3) when the ad space tag is loaded, a request is sent to the ad distribution server (at that time, URL information is also passed on). Next, (4) the EAR of the web page (URL) is evaluated using data stored in the database on the ad distribution server. The data can be anything from (4-a) the past EAR for each product category for the same URL, (4-b) the EAR for the advertising products currently registered on the ad distribution platform, and (4-c) the average EAR for the same URL. Next, (5) after evaluating the URL based on the EAR, the advertising campaign that is considered to have the highest degree of matching is selected. Next, (6) the ad distribution tag of the selected ad is returned to the ad space. Finally, (7) this will cause the advertisement for the campaign to be displayed on the device.

[0097] This section explains the behavior after ad delivery (behavior when responding to a questionnaire). FIG. 19 is a diagram showing an operational flow of EAR Ads on the advertisement distribution platform, and is a diagram showing an example of behavior after advertisement distribution (behavior when answering a questionnaire).

[0098] In Figure 19, the flow for answering the survey after the survey advertisement is distributed is roughly as follows: First, (1) when you answer the survey questions, the content of the first interaction page that has already been delivered will be dynamically changed to the content of the second interaction page via the ad delivery tag. Next, (2) at the same time, the answer result is returned to the advertisement distribution server. Next, (3) record and update the response results for the relevant advertising campaign. Finally, (4) at the same time, the EAR of the corresponding advertising material for the URL is recorded and updated.

[0099] We explain the increase in EAR response rates and the detection of poor web pages. EAR is extremely effective at "evaluating the match between the promoted product and the web page content on which the advertisement is displayed," but the demonstration experiment suggested that it may also be useful in detecting poor quality web pages that induce unintended clicks. When observing web pages with unusually high click-through rates, it is common to see them using ad formats designed to generate unintended clicks. This is the most obvious example of the problem mentioned above regarding click-through rates, such as overlay ads that appear on top of content or full-screen interstitial ads that appear between pages. Web pages that use these ad formats often have a significantly poor user experience, as they often force unintended clicks and then redirect the browser to the advertiser's landing page. If users try to read the article on the web page or continue reading the article, they are redirected to a completely unrelated advertiser's page, which is a very frustrating flow for users, and they are unlikely to respond to EAR's survey advertisements. Because they are not paying attention to the advertisement in the first place, they do not understand the content of the survey, and they are not likely to respond to the survey in the midst of such an unwanted series of actions. This is reflected in the results.

[0100] From EAR's perspective, it is a waste to deliver a survey to an ad space where no one will ever read it, so by detecting ad format spaces that induce such accidental clicks and avoiding ad delivery, it will be possible to "more efficiently evaluate the compatibility between advertising materials and web content."

[0101] The detection of poor quality web pages where survey responses are not expected will be carried out using the following indicators, and measurement JavaScript will be distributed at the same time as the EAR advertisement.

[0102] The indexes shown in FIG. 20 will be explained. FIG. 20 is a diagram illustrating an example of a mechanism for an attention score (evaluation of the visibility of a web page).

[0103] First, time-in-view (ad viewing time) measures the time from when the survey ad is displayed until the user leaves the page. The meaning of this metric is that if the viewing time is abnormally short (for example, less than 3 seconds), it is considered that the content is deemed to have no value, including poor layout. The in-view rate (the percentage of times the page is viewed for a certain number of seconds) is calculated by determining an appropriate viewing time (for example, 3 seconds or more) and measuring the percentage of times the page is viewed for that number of seconds. As an indicator, web pages with a high rate of not being viewed are considered to have low quality content or layout. Click timing measures the time between when an ad is displayed and when a click occurs. As an indicator, if the timing of a click is extremely short (for example, within one second), it can be assumed that the click was unintentional. The number of ad slots measures the number of ad slots per web page. The metric determines that having too many ad slots leads to a poor experience. The ad space occupancy rate measures the ratio of the ad space displayed to the web page display area. As an indicator, if the ad space display ratio is abnormally high, it is determined that the user experience is significantly degraded.

[0104] We will call the scoring system that measures and scores these things the attention score (degree of attention). The scoring formula is generally based on the concept below, and will be fine-tuned as data is collected.

[0105] Attention score = (Time-in-view score x weighting) + (In-view rate score x weighting) + (Click timing score x weighting) + (Number of ad slots score x weighting) + (Ad slot occupancy rate score x weighting)

[0106] FIG. 21 is a diagram showing an example of numerical values ​​of attention scores. As shown in Figure 21, the score of each index is weighted, and the attention score is calculated by summing the weighted scores. For example, if the time-in-view score is 70 and the weighting is 0.25, the weighted score is 17.5. In this way, the weighted score is calculated for each indicator, and the sum of these 51.1 is the attention score. The strength of the attention score is that it can detect sites that are likely to have a poor user experience with a small number of impressions, without relying on the survey response rate or number of responses. This allows more advertising budget to be distributed to highly relevant web pages.

[0107] We will explain EAR advertising delivery using attention scores. FIG. 22 is a diagram showing an example of EAR advertisement delivery using attention scores. FIG. 23 is a diagram showing an example of exclusion due to a low attention score.

[0108] As shown in Figure 22, by starting with measuring the attention score, sites with poor visibility can be eliminated at an early stage. Distribution begins with a budget of 1 million yen and 1 yen per impression, and the attention score is measured once each web page has been displayed 1,000 times. The attention score for web page A is calculated as 62, for web page B as 80, and for web page C as 21. At this time, web page c, which is judged to be a poor web page with an attention score of 40 or less, is excluded. Then, for web page A and web page B, the remaining budget can be weighted and distributed based on the EAR. In this way, early screening using attention scores allows web pages with low visibility to be excluded early on, enabling more efficient ad delivery.

[0109] We will explain the behavior from ad delivery to attention score measurement. FIG. 24 is a diagram showing an example of the operation flow of EAR Ads on the advertisement distribution platform, from advertisement distribution to attention score measurement.

[0110] In FIG. 24, first, (1) an advertising space tag (made up of Javascript and HTML) is placed on a web page. Next, (2) the ad space tag is loaded at the same time as the web page is loaded. Next, (3) when the ad space tag is loaded, a request is sent to the ad distribution server (at that time, URL information is also passed on). Next, (4) the EAR of the web page (URL) is evaluated using data stored in the database on the ad distribution server. The data can be anything from (4-a) the past EAR for each product category for the same URL, (4-b) the EAR for the advertising products currently registered on the ad distribution platform, and (4-c) the average EAR for the same URL. Next, (5) after evaluating the URL based on the EAR, the advertising campaign that is considered to have the highest degree of matching is selected. Next, (6) the ad delivery tag of the selected ad is sent back to the ad space along with JavaScript for measuring the attention score. Finally, (7) when the advertisement for the campaign is displayed on the device, measurement of each indicator of the attention score begins.

[0111] To summarize what we have discussed so far, one of the most important ultimate goals in the online advertising industry is to reach (display ads to) people who are interested in a product, and in order to do this efficiently, click-through rate has been the most widely used indicator, but with the evolution of devices and changes in the industry, it has become a less effective indicator. To overcome this, this service provides a new index concept called EAR. The effectiveness of EAR is clear from the above experiments. The market itself is extremely fixated on click metrics, resulting in a significant decline in the visibility of websites and user experience. Sites with poor visibility have extremely low survey response rates, so excluding these sites from ad delivery, as with this service, greatly improves EAR ad delivery efficiency. A unique evaluation axis called attention score is effective in assessing visibility. This allows the system to first identify poor quality web pages using attention scores and exclude them from delivery, and then automatically optimize ad delivery based on EAR, which evaluates the degree of match between the product and content. It is possible to offer survey ads that obtain EAR as a new advertising format, and at the same time, by establishing attention scores, it can also serve as a warning to website owners who are degrading the user experience.

[0112] As explained with reference to Figures 1 to 24, by combining attention score and EAR, this service can efficiently deliver advertisements to web pages that are highly relevant to the product while excluding web pages with poor user experience in advance.

[0113] Although one embodiment of the present invention has been described above, the present invention is not limited to the above-described embodiment, and modifications, improvements, etc. within the scope of achieving the object of the present invention are considered to be included in the present invention.

[0114] For example, the system configuration shown in FIG. 2 and the hardware configuration of the server 1 shown in FIG. 3 are merely examples for achieving the object of the present invention, and are not particularly limited.

[0115] Furthermore, the functional block diagram shown in Fig. 4 is merely an example and is not particularly limited. That is, it is sufficient if the information processing system in Fig. 2 is provided with functions that can execute the various processes described above as a whole, and the functional blocks and databases used to realize these functions are not particularly limited to the example in Fig. 4.

[0116] Furthermore, the locations of the functional blocks and databases are not limited to those shown in FIG. 4 and may be arbitrary. For example, at least some of the functional blocks and databases arranged on the server 1 side may be provided on the user terminal 2 side or on another information processing device (not shown).

[0117] The above-described series of processes can be executed by hardware or software. Furthermore, one functional block may be configured as a single piece of hardware, a single piece of software, or a combination thereof.

[0118] When a series of processes is executed by software, the programs that make up the software are installed into a computer or the like from a network or a recording medium. The computer may be a computer built on dedicated hardware. The computer may also be a computer capable of executing various functions by installing various programs, such as a server, a general-purpose smartphone, or a personal computer.

[0119] The recording medium containing such a program may be composed not only of a removable medium (not shown) that is distributed separately from the device main body in order to provide the program to the user, but also of a recording medium that is provided to the user in a state that is pre-installed in the device main body.

[0120] In this specification, the steps describing the program to be recorded on the recording medium include not only processes that are performed in chronological order, but also processes that are not necessarily performed in chronological order but are performed in parallel or individually.

[0121] To sum up, the information processing device to which the present invention is applied is sufficient as long as it has the following configuration, and can take on a variety of different embodiments. That is, an information processing device to which the present invention is applied (for example, the server 1 in FIGS. 1 to 4) An information processing device that distributes online advertisements, Advertisement distribution means (e.g., advertisement distribution unit 51 in FIG. 4) for distributing advertisements (e.g., EAR Ads in FIG. 22 and FIG. 23) to web pages (e.g., web pages in FIG. 22 and FIG. 23), score calculation means (e.g., score calculation unit 52 in FIG. 4 ) for calculating an attention score (e.g., the attention score in the above-mentioned formula) based on a plurality of indices for evaluating the visibility of the web page (e.g., the indices shown in FIGS. 20 and 21 ); a destination determination unit (e.g., the destination determination unit 53 in FIG. 4) that excludes web pages (e.g., web page C in FIG. 22) having an attention score less than a predetermined reference value from advertisement delivery targets; EAR calculation means (e.g., EAR calculation unit 54 in FIG. 4 ) for calculating EAR (Engaged Audience Rate) (e.g., EAR in the above-mentioned formula) which is the ratio of the number of responses to the number of times a survey advertisement is displayed for web pages (e.g., Web pages A and B in FIG. 22 ) whose attention scores are equal to or greater than the reference value; a weighting calculation means (such as the weighting calculation unit 55 in FIG. 4) for weighting the web page (such as the weighting in FIG. 23) based on the calculated EAR; a distribution amount determination means (for example, the distribution amount determination unit 56 in FIG. 4 ) for determining the advertisement distribution amount (for example, the number of advertisements displayed in FIG. 23 ) to the web page in accordance with the weighting; It is enough to have this.

[0122] According to such an information processing device (for example, server 1 in Figures 1 to 4), by combining attention score and EAR, it is possible to efficiently deliver advertisements to web pages that are highly relevant to the product while excluding web pages with poor user experience in advance.

[0123] In an information processing device to which the present invention is applied (for example, the server 1 in FIGS. 1 to 4), The score calculation means A time-in-view score indicating the viewing time of the advertisement (for example, the time-in-view score in Figure 21), An in-view rate score indicating the percentage of times the appropriate viewing time was reached (e.g., the in-view rate score in Figure 21), A click timing score (e.g., Figure 21) that indicates the time from ad display to click, An ad space count score indicating the number of ad spaces in a web page (for example, the ad space count score in Figure 21), An ad space occupancy score (e.g., the ad space occupancy score in Figure 21) indicating the display ratio of the ad space to the web page display area, Calculating the attention score based on at least one of the following: It is possible. This allows for more flexible visibility evaluation, as appropriate indicators can be selectively used depending on the characteristics of each web page and the measurement environment.

[0124] In addition, in an information processing device to which the present invention is applied (for example, the server 1 in FIGS. 1 to 4), The time-in-view score is a score for determining that a web page has a poor layout when the time from when the survey advertisement is displayed until the user leaves the page is less than a first predetermined second (e.g., less than 3 seconds). It can be said that: This allows low-quality content to be objectively and quantitatively determined based on specific time criteria.

[0125] In addition, in an information processing device to which the present invention is applied (for example, the server 1 in FIGS. 1 to 4), The click timing score is a score for determining a click as an unintentional click when the click occurs within a second predetermined second (for example, within one second) that is shorter than the first predetermined second from the advertisement display. It can be said that: This allows unintended clicks to be quantitatively determined from a time perspective, improving the accuracy of measuring advertising effectiveness.

[0126] In addition, in an information processing device to which the present invention is applied (for example, the server 1 in FIGS. 1 to 4), The attention score is At least one score selected from the time-in-view score, the in-view rate score, the click timing score, the number of ad slots score, and the ad slot occupancy rate score is weighted, and the weighted score is calculated so that the sum of the weighted scores of the selected at least one score is 1 (for example, the sum of 1.0 in FIG. 21). It can be said that: This allows the importance of each index to be adjusted depending on the situation, thereby achieving a more appropriate visibility evaluation.

[0127] In addition, in an information processing device to which the present invention is applied (for example, the server 1 in FIGS. 1 to 4), The advertisement distribution means The survey advertisement is displayed on a first screen (for example, the first interaction (survey screen) in FIG. 15) including questions and answer options related to the product; a second screen (e.g., the second interaction (thank you screen) in FIG. 15 ) including an advertisement for the product that is displayed after the answer option is selected; It will be delivered in a two-stage display format, including It is possible. This allows for effective brand awareness while gradually drawing out user interest.

[0128] In addition, in an information processing device to which the present invention is applied (for example, the server 1 in FIGS. 1 to 4), On a second screen after the answer to the questionnaire advertisement is selected, a brand logo or a product image of the merchandise is displayed larger than on the first screen (for example, as shown in FIG. 15 ). It is possible. This allows the brand to be more effectively remembered by making an impression on users at a time when their interest is heightened.

[0129] In addition, in an information processing device to which the present invention is applied (for example, the server 1 in FIGS. 1 to 4), The EAR calculation means sets the distribution rate of the survey advertisement to a predetermined range (e.g., 2-4%). It is possible. This allows for the determination of an appropriate delivery ratio, ensuring the accuracy of EAR measurement while minimizing the impact on normal ad delivery. It is possible. [Explanation of symbols]

[0130] 1 Server, 2 User terminal, 11 CPU, 12 ROM, 13 RAM, 14 Bus, 15 Input / output interface, 16 Input unit, 17 Output unit, 18 Storage unit, 19 Communication unit, 20 Drive, 30 Removable media, 51 Advertisement delivery unit, 52 Score calculation unit, 53 Delivery destination determination unit, 54 EAR calculation unit, 55 Weighting calculation unit, 56 Delivery amount determination unit, 71 Advertisement delivery information DB, 72 Score information DB, 73 Delivery destination determination information DB, 74 EAR information DB, 75 Weighting information DB, 76 Delivery amount determination information DB

Claims

1. An information processing device that distributes online advertisements, an advertisement distribution means for distributing advertisements to a web page; a score calculation means for calculating an attention score based on a plurality of indices for evaluating the visibility of the web page; a destination determination means for excluding web pages having an attention score less than a predetermined reference value from advertisement delivery targets; an EAR calculation means for calculating an EAR (Engaged Audience Rate), which is the ratio of the number of responses to the number of impressions of a survey advertisement, for web pages whose attention scores are equal to or greater than the reference value; a weight calculation means for weighting the web page based on the calculated EAR; a distribution amount determination means for determining an advertisement distribution amount to the web page in accordance with the weighting; An information processing device comprising:

2. The score calculation means a time-in-view score indicating the time spent viewing the advertisement; an in-view rate score indicating the percentage of people who achieved the appropriate viewing time; A click timing score, which indicates the time between the ad being displayed and the click, and an ad slot count score indicating the number of ad slots in the webpage; an ad space occupancy score indicating the proportion of ad space displayed relative to the web page display area; Calculating the attention score based on at least one of the following:

2. The information processing apparatus according to claim 1, wherein:

3. The time-in-view score is a score for determining a web page as having a poor layout when the time from when a survey advertisement is displayed until the user leaves the page is less than a first predetermined second.

3. The information processing apparatus according to claim 2, wherein:

4. the click timing score is a score for determining a click as an unintentional click when the click occurs within a second predetermined second, which is shorter than the first predetermined second, from the advertisement display; 4. The information processing apparatus according to claim 3,

5. The attention score is a score calculated by weighting at least one score selected from the time-in-view score, the in-view rate score, the click timing score, the number of ad slots score, and the ad slot occupancy rate score so that the sum of the weights of the selected at least one score is 1; 3. The information processing apparatus according to claim 2, wherein:

6. The advertisement distribution means a first screen including questions and answer options related to the survey advertisement; a second screen including an advertisement for the product, which is displayed after the answer option is selected; It will be delivered in a two-stage display format, including 2. The information processing apparatus according to claim 1, wherein:

7. a brand logo or a product image of the merchandise is displayed on a second screen after the answer to the questionnaire advertisement is selected, the brand logo or the product image being displayed larger than that on the first screen; 7. The information processing apparatus according to claim 6,

8. the EAR calculation means sets the distribution rate of the survey advertisement to a predetermined range distribution rate; 2. The information processing apparatus according to claim 1, wherein:

9. An information processing method executed by an information processing device that distributes online advertisements, comprising: an advertisement delivery step of delivering advertisements to the web page; a score calculation step of calculating an attention score based on a plurality of indices for evaluating the visibility of the web page; a destination determination step of excluding web pages having an attention score less than a predetermined reference value from advertisement delivery targets; an EAR (Engaged Audience Rate) calculation step of calculating an EAR, which is the ratio of the number of responses to the number of impressions of a survey advertisement, for web pages whose attention scores are equal to or greater than the reference value; a weight calculation step of weighting the web page based on the calculated EAR; a delivery amount determination step of determining an advertisement delivery amount to the web page in accordance with the weighting; An information processing method including:

10. On the computer, an advertisement delivery step of delivering advertisements to the web page; a score calculation step of calculating an attention score based on a plurality of indices for evaluating the visibility of the web page; a destination determination step of excluding web pages having an attention score less than a predetermined reference value from advertisement delivery targets; an EAR (Engaged Audience Rate) calculation step of calculating an EAR, which is the ratio of the number of responses to the number of impressions of a survey advertisement, for web pages whose attention scores are equal to or greater than the reference value; a weight calculation step of weighting the web page based on the calculated EAR; a delivery amount determination step of determining an advertisement delivery amount to the web page in accordance with the weighting; A program that executes control processing including:

Images