Advertising analysis system and advertising analysis method

Abstract

To provide an advertising analysis system capable of analyzing the effectiveness of a specific advertisement. [Solution] The advertising analysis system is an advertising analysis system that analyzes the effectiveness of advertising using an electronic information terminal 5, and the electronic information terminal 5 is equipped with an advertising-related information acquisition means 51 that acquires advertising-related information including basic information, viewing information, and performance information for each advertisement for the advertising user themselves, products, services, etc., and a correlation analysis means 52 that analyzes the correlation between elements contained in the advertising-related information, and the viewing information and performance information are paired pieces of information for each specified period of time, and the correlation analysis means 52 is a means for excluding viewing information or performance information that is noise information among the advertising-related information during a specified analysis period and has a noise basis that can be grasped from the advertising-related information or other related information during the analysis period, together with the viewing information or performance information paired with that information, from the correlation analysis and analyzing it.

Description

[Technical field]

[0001] The present invention relates to an advertising analysis system and an advertising analysis method for analyzing the effectiveness of an advertisement. [Background technology]

[0002] Traditionally, advertising has been widely used as a means to provide consumers with information about products and services and increase their willingness to purchase and awareness of them. In recent years, with the development of digital technology, in addition to traditional media such as television, radio, newspapers, magazines, flyers, and billboards, advertising via the Internet, SNS (social networking services), direct mail, etc. has also become widespread.

[0003] Regarding the content of advertisements, in addition to information about the functions and performance of the products and services handled by the advertiser (hereinafter referred to as "advertising user"), advertisements about the advertising user themselves are also increasing. This is expected to improve the recognition and brand image of the advertising user themselves. In addition, personalized advertisements tailored to the interests and concerns of consumers are being used, and advertisements that approach specific consumer segments are also on the rise.

[0004] As the advertising methods in the advertising industry become more diverse, there is an increasing need for efficient advertising. Conventionally, for example, a system for analyzing the effectiveness of web advertising is known, such as that disclosed in Patent Document 1.

[0005] Patent document 1 proposes an advertising effectiveness analysis system that includes a performance registration means for registering performance for each of a number of action items set in accordance with five processes (AIDMA), which are the basic principles of advertising activities, a compilation means for compiling performance status for each of the five processes based on the performance data registered by the performance registration means, and an analysis result output means for outputting the performance status compiled by the compilation means for each of the five processes in a distinct manner on a single screen. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2003-44738 A Summary of the Invention [Problem to be solved by the invention]

[0007] Nowadays, with the spread of social media and video streaming services, companies, organizations, and individuals can freely disseminate information, and fashions and trends are changing rapidly. In this modern world, where we are overwhelmed with information, even if an advertisement for a product is placed in a specific medium, there are many possible factors that led a consumer to purchase the product. For this reason, it is not easy to analyze the effectiveness of a specific advertisement.

[0008] The present invention has been made in consideration of the above circumstances, and has an object to provide an advertising analysis system and an advertising analysis method capable of analyzing the effectiveness of a specific advertisement. [Means for solving the problem]

[0009] The advertising analysis system of the present invention is an advertising analysis system that analyzes the effectiveness of an advertisement by an advertising user on an electronic information terminal when the advertising user places an advertisement, and the electronic information terminal is equipped with an advertising-related information acquisition means for acquiring advertising-related information including at least basic information, viewing information, and performance information for each advertisement for the advertising user itself, the products purchased by the advertising user, and at least one selected from the services handled by the advertising user, and a correlation analysis means for analyzing the correlation between elements included in the advertising-related information, in which the viewing information and the performance information in the advertising-related information are paired information for each specified period, and the correlation analysis means is a means for excluding viewing information or performance information that is noise information among the advertising-related information during a specified analysis period and has a noise basis that can be grasped from the advertising-related information or other related information during the analysis period, together with the paired viewing information or performance information, from the analysis of the correlation.

[0010] The predetermined unit of time is one day, and the predetermined analysis period is from three days to three months.

[0011] The noise basis determination means for determining the presence or absence of the noise basis includes a means for calculating a deviation value for each period for each period unit in the analysis target period for at least one type of information among the advertising-related information or information derived from this information, and for each period in which the deviation value is outside a predetermined range and deviates from the standard, determining the presence or absence of the noise basis from the advertising-related information or other related information that belongs to the respective period and is different from the subject at the time of calculating the deviation value.

[0012] The noise basis determination means is characterized in that it is a means for determining whether or not there is a basis for noise from temporary or periodic current affairs information, which is related information other than the advertisement-related information, belonging to each period in which the deviation value is outside a predetermined range and deviates from the standard.

[0013] The advertising-related information includes advertising costs for a specified medium, and the viewing information and performance information before and during the advertising period, and the correlation analysis means calculates the advertising costs and the actual increase or decrease in the number of specified recognitions included in the viewing information during the analysis period, and calculates the advertising costs required to obtain an increase in one recognition as the correlation.

[0014] The advertising analysis method of the present invention is an advertising analysis method that uses the advertising analysis system of the present invention to analyze the effectiveness of an advertisement when an advertising user places an advertisement, and includes an information acquisition step of acquiring advertising-related information by the advertising-related information acquisition means, the advertising-related information including at least basic information, viewing information, and performance information for each advertisement for the advertising user itself, the products purchased by the advertising user, and at least one service selected from the services handled by the advertising user, and a correlation analysis step of analyzing the correlation between elements included in the advertising-related information by the correlation analysis means, and is characterized in that the correlation analysis step includes an analysis premise step of identifying viewing information or performance information that is noise information among the advertising-related information in a specified analysis period and has a noise basis that can be grasped from the advertising-related information or other related information in the analysis period, and an actual analysis step of excluding the identified information together with the paired viewing information or performance information and analyzing the correlation.

[0015] The method is characterized in that the specified period is one day, the specified analysis period is 3 days or more and 3 months or less, and the noise basis determination step for determining whether or not there is a noise basis includes a first step of calculating a standard deviation value for each day of the analysis period for at least one type of information among the advertising-related information or information derived from this information, and a second step of determining whether or not there is a noise basis from the advertising-related information or other related information that belongs to each day and is different from the subject when the standard deviation value was calculated, for each day on which the standard deviation value is outside the specified range and deviates from the standard.

[0016] The advertising-related information includes advertising costs for a specified medium, and the viewing information and performance information before and during the advertising period, and the correlation analysis step includes a step of calculating the advertising costs and the actual increase or decrease in the number of specified recognitions included in the viewing information during the analysis period, and calculating the advertising costs required to obtain an increase in one recognition as the correlation. Effect of the Invention

[0017] The advertising analysis system of the present invention comprises an advertising-related information acquisition means for acquiring advertising-related information including at least basic information, viewing information, and performance information of an advertisement, and a correlation analysis means for analyzing the correlation between elements included in the advertising-related information, wherein the viewing information and performance information in the advertising-related information are paired pieces of information for each predetermined period of time, and the correlation analysis means is a means for excluding from the analysis of correlation viewing information or performance information that is noise information among the advertising-related information during a predetermined analysis period and has a noise basis that can be grasped from the advertising-related information or other related information during the analysis period, together with the viewing information or performance information that is paired with that information, and analyzing the same. By excluding from the analysis viewing information or performance information that is noise information among the advertising-related information during the predetermined analysis period and has a noise basis that can be grasped from the advertising-related information or other related information, the noise information is not uniformly excluded, but information without a noise basis can be included in the analysis, and the effectiveness of a specific advertisement can be appropriately analyzed.

[0018] Since the above-mentioned predetermined unit of time is one day, and the predetermined analysis period is from three days to three months, it is possible to analyze the effectiveness of advertisements over a relatively short period of time. In addition, in the case of advertisements over a relatively short period of time, the shorter the period, the greater the susceptibility to noise, but by excluding from the analysis specific information that is the above-mentioned noise information and for which noise grounds have been found, it is possible to properly analyze the effectiveness of even advertisements over a relatively short period of time.

[0019] The noise basis determination means for determining whether or not there is a noise basis includes a means for calculating a standard deviation for each period in the analysis target period for at least one type of advertising-related information or information derived from this information, and for each period in which the standard deviation is outside a specified range and deviates from the standard, determining whether or not there is a noise basis from advertising-related information or other related information that belongs to the respective period and is different from the subject at the time of calculating the standard deviation. This makes it easier to extract the noise information itself, limits the amount of information required to determine the noise basis, and makes analysis more efficient.

[0020] The noise basis determination means is a means for determining whether or not there is a basis for noise from temporary or periodic current affairs information, which is related information other than advertising-related information, belonging to each period in which the deviation value is outside the specified range and deviates from the standard, so that it is easy to determine whether or not there is a basis for noise, and the effectiveness of advertising can be properly analyzed. [Brief description of the drawings]

[0021] [Figure 1] FIG. 1 is a schematic diagram illustrating an example of a configuration of an advertisement delivery system. [Diagram 2] FIG. 1 is a schematic diagram illustrating an example of a configuration of an advertising analysis system. [Diagram 3] FIG. 13 is a diagram showing an example of a list of advertisement-related information. [Figure 4] FIG. 2 is a schematic diagram illustrating extraction of noise information from advertisement-related information. [Diagram 5] FIG. 13 is a schematic diagram showing different advertisement-related information that is the basis of noise. [Figure 6] FIG. 13 is a schematic diagram showing a determination of the presence or absence of noise grounds. [Figure 7] This is an overview of how to calculate CPA, etc. [Figure 8] FIG. 13 is a diagram showing an example of advertisement-related information when the distribution area is limited. [Figure 9] FIG. 2 is a flow chart showing an example of the procedure of an advertising analysis method of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] In the advertising analysis system of the present invention, the advertising media to be analyzed are not particularly limited. For example, various forms of advertising media, such as television commercials, radio commercials, newspaper advertisements, magazine advertisements, search engine service advertisements, SNS advertisements, video distribution service advertisements, direct mail, and outdoor advertisements such as electronic bulletin boards and billboards, are included.

[0023] Examples of search engine services include Google (registered trademark) and Yahoo! (registered trademark). Examples of SNS include Facebook (registered trademark), X, Instagram (registered trademark), TikTok, and LINE (registered trademark). Examples of video distribution services include YouTube (registered trademark).

[0024] The content of the advertisement includes information about the advertising user itself (information showing the company's business activities, corporate philosophy, corporate image, etc.), information about the products that the advertising user handles and purchases, and information about the services that the advertising user handles.

[0025] Furthermore, the analysis period for analyzing the effectiveness of the advertisement is not particularly limited, and may be set, for example, on a daily, weekly, monthly, or yearly basis. It is preferable that the analysis period is relatively short, so that the effectiveness of the advertisement can be quickly determined and the budget can be concentrated in investing in more effective advertisements. For example, it may be from one day to three months, from three days to one month, or from three days to two weeks. The advertising period (the period during which the advertisement is displayed) is not particularly limited, and may be set, for example, to the same period as the analysis period.

[0026] The configuration of the advertising analysis system of the present invention will be described below with reference to the drawings.

[0027] First, a schematic diagram of an example of the configuration of an advertisement distribution system is shown in Fig. 1. Fig. 1 shows the analysis of advertisements distributed via a network.

[0028] 1, the advertisement distribution system includes a distribution terminal 2, an advertising user terminal 3, and terminals 4a, 4b, and 4c. The distribution terminal 2, the advertising user terminal 3, and the terminals 4a, 4b, and 4c are communicatively connected via a network N. The network N is an electronic communication line capable of transmitting and receiving data, and examples of the network include a wired LAN, a wireless LAN, a WAN, and the Internet.

[0029] The distribution terminal 2 is a terminal that distributes advertisements. The distribution terminal 2 is an information processing terminal operated by an advertising medium provider (e.g., a website operator) and is, for example, a server computer or a personal computer. The distribution terminal 2 accepts bids and advertisement submissions from the advertising user terminal 3, and distributes the advertisements when the distribution conditions are satisfied.

[0030] The advertising user terminal 3 is an information processing terminal operated by an advertising user, an advertising agency, etc., and is, for example, a server computer, a personal computer, etc. The advertising user terminal 3 notifies the distribution terminal 2 of, for example, a bid amount setting, an advertisement submission setting, etc. In addition, the advertising user terminal 3 obtains various information related to the distributed advertisement from the distribution terminal 2, etc.

[0031] Terminals 4a, 4b, and 4c are personal computers, smartphones, tablet terminals, etc., and are terminals operated by consumers. The consumer here refers to a person who uses or may use advertisements, and does not necessarily have to be a consumer who ultimately purchases a product, etc. Consumers can view advertisements distributed from distribution terminal 2 through each terminal.

[0032] For example, in the case of banner ads on web pages such as SNS, when a consumer clicks on the banner ad, the consumer is taken to a link attached to the banner ad. The linked web page is a web page (a so-called landing page (LP)) that contains information to introduce products and services, etc., and promotes the products and services to encourage results such as purchases (conversions). Ads for video streaming services are shown between videos being watched, and they arouse the consumer's interest and encourage them to go to the web page intended by the advertising user.

[0033] 1, an advertisement analysis system 1 analyzes the effectiveness of advertisements distributed by the above-mentioned advertisement distribution system. The advertisement analysis system 1 includes an electronic information terminal 5. The electronic information terminal 5 is, for example, a server computer or a personal computer, and has a calculation device such as a CPU (Central Processing Unit) and a storage device. The electronic information terminal 5 can function as an advertisement analysis device by executing various programs stored in the storage device with the calculation device.

[0034] In Figure 1, the electronic information terminal 5 is communicatively connected to the distribution terminal 2 and the advertising user terminal 3, etc. via a network N, but it does not necessarily have to be connected to the network N as long as it can obtain the advertising-related information described below by any method.

[0035] The electronic information terminal 5 will be described with reference to Fig. 2. As shown in Fig. 2, the electronic information terminal 5 includes advertisement-related information acquisition means 51 for acquiring advertisement-related information, and correlation analysis means 52. Furthermore, the correlation analysis means 52 includes noise reason determination means 52a and calculation means 52b.

[0036] The advertisement-related information acquisition means 51 acquires advertisement-related information. Each piece of advertisement-related information may be acquired, for example, from the distribution terminal 2 or the advertising user terminal 3 via the network N, or may be acquired via any storage device, or may be acquired by manual input, etc. The advertisement-related information includes at least basic information, viewing information, and performance information of the advertisement. Note that the detailed information acquired from the advertisement-related information differs strictly depending on the individual advertising media through which the advertisement is distributed, etc.

[0037] The basic information is information about the advertising user, such as the company name of the advertising user, an identifier for identifying the advertising user, the advertisement name, the campaign name, the consumer target (age, gender, etc.), and the area (prefecture, etc.).

[0038] Viewing information is information that indicates how much a consumer viewed an advertisement. Examples of viewing information include advertising costs, number of impressions, number of clicks, number of sessions, total number of users, number of new users, new user rate, number of visits, etc. Advertising costs indicate the amount paid as advertising costs, impressions indicate the number of times the advertisement was displayed, clicks indicate the number of times it was clicked, and sessions indicate the number of times a consumer visits a site and leaves it, counting one session as one.

[0039] The outcome information is information that indicates the outcome of the advertisement. For example, if the objective is to purchase a product or service, it is the number of times that objective is achieved, that is, the number of purchases of the product or service (hereinafter, simply referred to as the number of conversions (CV number)).

[0040] Here, the viewing information and the performance information in the advertisement-related information are paired information for each predetermined unit of time during the advertisement period. The predetermined unit of time is, for example, one day, two days, one week, etc., and is usually set to one day.

[0041] 3 shows a table in which each piece of information is arranged in pairs for each day. Such a table is displayed on the display screen of the electronic information terminal 5, and the analyst performs operations based on this table.

[0042] In FIG. 3, the advertising period (also the period to be analyzed) is two weeks from January 1, 2024 to January 14, 2024. Here, it is assumed that the advertisement to be analyzed is related to the product "XXX", and is a banner advertisement in a specific search engine service (for example, Google). As basic information, the advertisement name "Commemorative Campaign" is described. As viewing information, advertising costs, number of impressions, and number of clicks are described on a daily basis. As performance information, the number of CVs is described on a daily basis. In this way, each element of the basic information, viewing information, and performance information is arranged in correspondence with the date. Note that FIG. 3 is merely an example, and each element acquired and displayed in each piece of basic information, viewing information, and performance information is not limited to this.

[0043] As shown in Figure 3, the numbers decrease in order of number of impressions, number of clicks, and number of conversions. This is a change in numbers that reflects the series of purchasing activities that occur when a consumer becomes aware of an ad (equivalent to number of impressions), becomes interested in it (equivalent to number of clicks), confirms it, and finally makes a purchase (equivalent to number of conversions). Note that advertising costs vary depending on factors such as the number of impressions and bidding status, and generally, the number of impressions increases as advertising costs increase. In this way, the numbers for each element change from day to day.

[0044] In Fig. 2, the correlation analysis means 52 of the electronic information terminal 5 is a means for analyzing the correlation between elements included in the advertisement-related information as shown in Fig. 3. Specifically, the correlation analysis means 52 extracts, by the noise basis determination means 52a, noise information from the advertisement-related information during the analysis period (for example, two weeks in Fig. 3). Then, among the extracted noise information, viewing information or performance information having a noise basis that can be grasped from the advertisement-related information or other related information during the analysis period is excluded from the correlation analysis together with the viewing information or performance information paired with that information and analyzed.

[0045] An example of the flow from correlation analysis to exclusion will be described with reference to Figures 4 to 6. Note that the lists shown in Figures 4 and 6 are based on the contents of the list in Figure 3.

[0046] First, noise information is extracted from the advertisement-related information shown in Fig. 4. Information considered to be noise information is extracted, for example, by calculating the standard deviation of each period for each period unit in the analysis target period.

[0047] In FIG. 4, the deviation value for each day in the two-week analysis period is calculated for the advertising cost of the viewing information. If the calculated deviation value is outside a predetermined range (for example, ±10 or more), the viewing information for that date is extracted as noise information. In FIG. 4, a deviation value within the predetermined range is indicated as "OK" and a deviation value outside the predetermined range is indicated as "NG", and the viewing information for "January 6, 2024" and "January 7, 2024" is extracted as noise information. Although not shown in FIG. 4, a deviation value for each date may be calculated for other elements of the viewing information, such as the number of views and the number of clicks, and the calculated deviation value may be used to extract the noise information. It is sufficient that a deviation value is calculated for at least one element of the viewing information, and noise information may be extracted by combining deviation values ​​for multiple elements.

[0048] In addition, in FIG. 4, the standard deviation of the number of conversions in the outcome information is calculated for each day in the two-week analysis period. If the calculated standard deviation is outside a predetermined range (for example, ±10 or more), the outcome information for that date is extracted as noise information. In FIG. 4, the outcome information for "January 9, 2024" and "January 12, 2024" is extracted as noise information.

[0049] The method of extracting noise information from viewing information and performance information is not limited to the above method. For example, instead of the deviation value, an index indicating the variation such as standard deviation or variance may be calculated as information (calculated value) derived from viewing information, and information (viewing information or performance information) to which an element showing a predetermined or greater variation belongs may be extracted as noise information.

[0050] After the noise information is extracted as described above, the presence or absence of noise grounds is determined for each period for which the noise information was acquired (four days in FIG. 4: "January 6, 2024," "January 7, 2024," "January 9, 2024," and "January 12, 2024"). Specifically, the presence or absence of noise grounds is determined from advertising-related information or other related information that is different from the subject at the time of calculation of the deviation value, etc., that belongs to each of the above periods. For example, with regard to the noise information for "January 6, 2024," the presence or absence of noise grounds is determined for the information that belongs to "January 6, 2024."

[0051] FIG. 5 shows an example of information that is the basis of noise (hereinafter, also referred to as "noise basis information"). For example, different advertising-related information related to the product "XXX" includes press releases for "XXX" on the advertising user's homepage, posts on the advertising user's SNS and blogs, web advertisements in different media, inserts, local events, etc., belonging to each period (for example, "January 7, 2024" in the case of FIG. 4). When such different advertising-related information exists, it is possible that the product is recognized by consumers through advertising-related information other than the advertisement to be analyzed (i.e., recognized through other routes), which may have resulted in noise information, that is, an increase in the number of clicks on the advertisement to be analyzed and an increase in the final number of CVs, and therefore it is determined that there is noise basis.

[0052] In Figure 5, other related information for the product "XXX" includes temporary current events such as posts by celebrities on social media, blogs, etc., and product introductions on TV, magazines, etc. Although such current events information is not based on the intended actions of advertising users, it is possible that the product will become known to consumers (i.e., will become known through other routes), and as such may have become noise information, it is determined that there is evidence of noise.

[0053] Other relevant information for the product "XXX" includes periodic current events. Depending on the characteristics of the product or service, the number of conversions may be affected by the date or time. For example, in the case of the number of mobile carrier-related contracts, the number of conversions tends to increase at the end of the month. Therefore, if information on the end of the month or a date close to that date is extracted as noise information, it may be possible to determine that there is evidence of noise based on such periodic current events.

[0054] The noise basis information is automatically collected by the noise basis determination means, or collected by an analyst's operation of an information processing device. For example, the noise basis information is collected by accessing the advertising user's homepage or blog during the period when the noise information was extracted (for example, "January 7, 2024" in the case of FIG. 4), searching for events related to the product that is the subject of the advertisement, searching for posts related to the product on each SNS, etc. Note that at least one piece of noise basis information is sufficient.

[0055] Depending on the presence or absence of such noise basis information, the presence or absence of noise basis is determined for each noise information. In the example of FIG. 6, it is determined that there is noise basis for "January 7, 2024" and "January 12, 2024", and it is determined that there is no noise basis for "January 6, 2024" and "January 9, 2024". In response to this determination, the viewing information and the result information for "January 7, 2024" and "January 12, 2024" are respectively removed (for example, the entire row is deleted), and an analysis (also called an actual analysis) is performed based on each information shown in the lower part of FIG. 6.

[0056] As shown in Figure 6, even if information is once extracted as noise information, if it is determined that there is no basis for noise, it becomes the information to be analyzed. When evaluating the effectiveness of an advertisement, for example, a method of uniformly removing viewing information or performance information with large variations and analyzing based on the information after removal is conceivable, but by including information with no basis for noise in the analysis, it is possible to make the analysis more efficient and more accurate.

[0057] FIG. 7 shows an outline of the calculation of each numerical value as the flow of the subsequent analysis. The table at the top shows the total number of conversions for the same period of days before and during distribution. Note that the average number of conversions per day may also be used. Then, the growth rate (%) due to advertising is calculated based on each total number of conversions. The calculated growth rate is used to calculate the increase or decrease in awareness. Furthermore, the CPA (advertising cost required to obtain one increase in awareness) is calculated by dividing the advertising cost by the increase or decrease in awareness. The calculation of each numerical value is performed by the calculation unit 52b (see FIG. 2) in the correlation analysis means.

[0058] In FIG. 7, the period immediately prior to the distribution period is used as the information before distribution to be compared with the information being distributed. However, this is not limited to this, and for example, information from the same period one month ago or one year ago may be used.

[0059] Depending on the form of advertisement, it may be distributed by area, time period, day of the week, etc. Even in such a case, the above-mentioned series of analysis methods can be applied.

[0060] FIG. 8(a) shows an example in which the advertisement distribution area is limited to Tokyo. In this example, basic information, viewing information, and performance information are acquired as advertisement-related information, and are acquired as paired information on a daily basis during the advertisement period. Then, as described above, noise information is extracted, and the presence or absence of noise grounds is determined for the extracted noise information, and actual analysis is performed.

[0061] In Figure 8(b), when calculating the growth rate and increase / decrease in the actual analysis, the actual growth rate (actual growth rate) and the actual increase / decrease (actual increase / decrease) are calculated by taking into account the performance of non-distribution areas. In Figure 8(b), first, the number of sessions and the number of conversions are calculated before and during distribution in the distribution area "Tokyo" and the non-distribution area "Osaka Prefecture and Aichi Prefecture". Note that here, the non-distribution area is set to "Osaka Prefecture and Aichi Prefecture," which are not neighboring prefectures and have relatively similar population size, but it may be, for example, "the whole country except Tokyo." Then, the growth rate during distribution compared to before distribution is calculated for each.

[0062] The actual growth rate is calculated using each of the growth rates obtained. For example, the actual growth rate (%) of the number of conversions in Tokyo is calculated to be 119 (= 151 / 127). This actual growth rate is then used to calculate the actual increase / decrease. For example, the actual increase / decrease in the number of conversions in Tokyo is calculated to be 3.0 (= (16 x 1.19) - 16). This actual increase / decrease is then used to calculate the CPA.

[0063] As shown in Figure 8(c), the effectiveness of advertising can be evaluated more accurately by evaluating not only the difference in the distribution area during distribution, but also by subtracting the difference in non-distribution areas. In other words, in addition to simply comparing before and during the advertising period, we predict what would have happened if the advertisement had not been displayed (counterfactual), and use the growth rate and increase / decrease compared to the predicted value as the real growth rate and real increase / decrease.

[0064] Note that the actual growth rate and actual increase / decrease can be calculated not only for advertisements by area, but also for advertisements by time period or day of the week. For example, for an advertisement that is distributed on a specific day of the week (e.g. Wednesday), in addition to comparing with the day before the advertisement (e.g. Tuesday), it is possible to grasp the trend of the specific day of the week before the advertisement (e.g. about what percentage of increase does that day have) and calculate the actual increase / decrease for that specific day by taking that trend into account.

[0065] The advertisement analysis system of the present invention is not limited to the system described in the above-mentioned Figs. 1 to 8. For example, the above describes advertisements in a form distributed via a network, but the advertisements may be, for example, TV commercials, magazine advertisements, flyer advertisements, billboard advertisements, etc. For example, for TV commercials, viewing information can be acquired based on TV viewer ratings, etc. For magazine advertisements, viewing information can be acquired based on the number of magazines sold, for flyer advertisements, the number of flyers distributed, etc. For billboard advertisements, viewing information can be acquired based on road traffic volume, the number of passengers getting on and off at stations, the number of users, etc. depending on the location where the advertisement is installed.

[0066] 9 shows a flow chart of an example of the advertising analysis method of the present invention. This advertising analysis method is an advertising analysis method in which, when an advertising user places an advertisement, the effectiveness of the advertisement is analyzed by using the above-mentioned advertising analysis system.

[0067] 9 includes an information acquisition step (corresponding to step S1) of acquiring the above-mentioned advertisement-related information, and a correlation analysis step (corresponding to steps S2 to S6) of analyzing correlations between elements included in the advertisement-related information by the above-mentioned correlation analysis means. The correlation analysis step also includes an analysis premise step (corresponding to steps S2 to S3) of identifying viewing information or performance information that is noise information among advertisement-related information during a predetermined analysis period and has a noise basis that can be grasped from the advertisement-related information during the analysis period or other related information, and an actual analysis step (corresponding to steps S4 to S6) of excluding the identified information together with its paired viewing information or performance information and analyzing the correlation.

[0068] The analysis premise step also includes a first step (corresponding to step S2) of calculating a standard deviation for each day of the analysis period for at least one type of advertising-related information or information derived from this information, and a step (corresponding to step S3) of determining whether or not there is a basis for noise for each day on which the standard deviation is outside the specified range and deviates from the standard, from advertising-related information belonging to each day that is different from the subject at the time of calculating the standard deviation or other related information.

[0069] The correlation analysis step includes a step of calculating the advertising cost and the actual increase or decrease in the number of predetermined recognitions included in the viewing information during the analysis period (corresponding to steps S4 to S5), and calculating the advertising cost required to obtain an increase of one recognition as a correlation (corresponding to step S6).

[0070] The advertising analysis method of the present invention is not limited to the method described above with reference to FIG. 9, and the details of each step may be changed as appropriate. [Explanation of symbols]

[0071] 1. Advertising analysis system 2. Distribution terminal 3. Advertising user device 4a, 4b, 4c terminals 5. Electronic information terminals 51 Means of acquiring advertising-related information 52 Correlation Analysis Methods 52a Noise basis determination means 52b Calculation method

Claims

1. An advertisement analysis system for analyzing the effectiveness of an advertisement by an advertising user on an electronic information terminal when the advertising user places an advertisement, the electronic information terminal comprises: an advertisement-related information acquisition means for acquiring advertisement-related information including at least basic information, viewing information, and result information of each advertisement for at least one selected from the advertising user itself, a product handled by the advertising user, and a service handled by the advertising user; and a correlation analysis means for analyzing correlations between elements included in the advertisement-related information; the viewing information and the performance information in the advertisement-related information are paired for each predetermined period of time, the correlation analysis means is a means for excluding viewing information or performance information that is noise information among the advertisement-related information during a predetermined analysis period and has a noise basis that can be grasped from the advertisement-related information or other related information during the analysis period, together with the paired viewing information or performance information, from the analysis of the correlation, and analyzing the information; The advertising analysis system is characterized in that it has a noise basis determination means for determining whether or not there is a basis for noise, which calculates a deviation value for each period in the analysis target period for at least one type of information among the advertising-related information or information derived from this information, and for each period in which the deviation value is outside a specified range and deviates from the standard, it has a means for determining whether or not there is a basis for noise from the advertising-related information or other related information that belongs to the respective period and is different from the subject at the time of calculating the deviation value.

2. 2. The advertising analysis system according to claim 1, wherein the predetermined unit of time is one day, and the predetermined analysis period is from three days to three months.

3. The advertising analysis system according to claim 1 or 2, characterized in that the noise basis determination means is a means for determining whether or not there is a basis for noise from temporary or periodic current affairs information which is related information other than the advertising-related information and which belongs to each period in which the deviation value is outside a specified range and deviates from the standard.

4. The advertisement-related information includes an advertisement cost for a predetermined medium, the viewing information before and during the advertisement period, and the performance information; The advertising analysis system according to claim 1 or 2, characterized in that the correlation analysis means is a means for calculating the advertising cost and the actual increase or decrease in the number of specified recognitions included in the viewing information during the analysis period, and calculating the advertising cost required to obtain an increase of one recognition as the correlation.

5. 2. An advertising analysis method for analyzing the effectiveness of an advertisement by an advertising user when the advertising user places an advertisement, using the advertising analysis system according to claim 1, an information acquisition step of acquiring, by the advertisement-related information acquisition means, advertisement-related information including at least basic information, viewing information, and performance information of each advertisement for at least one selected from the advertising user itself, a product handled by the advertising user, and a service handled by the advertising user; and a correlation analysis step of analyzing, by the correlation analysis means, a correlation between elements included in the advertisement-related information, The correlation analysis step includes an analysis premise step of identifying viewing information or performance information that is noise information among the advertisement-related information during a predetermined analysis period and has a noise basis that can be grasped from the advertisement-related information or other related information during the analysis period, and an actual analysis step of analyzing the correlation by excluding the identified information together with the paired viewing information or performance information, The predetermined period is one day, and the predetermined analysis period is from 3 days to 3 months, The advertising analysis method is characterized in that, as a noise basis determination step for determining whether or not there is a noise basis, the method includes a first step of calculating a standard deviation value for each day of the analysis period for at least one type of information among the advertising-related information or information derived from this information, and a step of determining whether or not there is a noise basis from the advertising-related information or other related information that belongs to each day and is different from the subject at the time of calculating the standard deviation value, for each day on which the standard deviation value is outside a predetermined range and deviates from the standard.

6. The advertisement-related information includes an advertisement cost for a predetermined medium, the viewing information before and during the advertisement period, and the performance information; The advertising analysis method according to claim 5, characterized in that the correlation analysis step includes a step of calculating the advertising cost and the actual increase or decrease in the number of specified recognitions included in the viewing information during the analysis period, and calculating the advertising cost required to obtain an increase of one recognition as the correlation.

Images