Advertisement frame effect evaluating method, program and information processing device

Abstract

To provide a technique that evaluates a contribution level of an advertisement frame to an advertisement effect.SOLUTION: According to an advertisement frame effect evaluating method of the present invention, a base model is set which is obtained by making a model of an impact of, to an outcome index, an advertisement makeup index of an advertisement including multiple advertisement frames in an interest area. A regression model is set by a frame factor of each of the multiple advertisement frames based on an assumption such that a ratio between the advertisement makeup index and the outcome index expressed by the base model is unchanged. A partial regression coefficient of the frame factor for each of the multiple advertisement frames is obtained using the regression model. The effect of the advertisement frame is evaluated with the partial regression coefficient of the frame factor for each of the multiple advertisement frames or a four-arithmetic operation result based on the partial regression coefficient as an advertisement frame effect index for each of the multiple advertisement frames.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a technique for evaluating the degree to which an advertising space contributes to advertising effectiveness. [Background technology]

[0002] Companies spend a huge amount of budget and resources on advertising campaigns. Therefore, accurately estimating the effectiveness of advertising is an essential business challenge in maximizing the results of corporate activities. Therefore, when placing an advertisement, it is important to accurately estimate the impact that the advertisement will have on viewers, etc. In order to maximize the results of an advertising campaign, it is necessary to grasp indicators according to the advertising objectives and analyze the data.

[0003] Against this background, Patent Publication No. 2023-86659 (Patent Document 1) discloses an invention of an information processing device that can evaluate the value of the material used in a commercial and the value of the program linked to that commercial.

[0004] The invention disclosed in Patent Document 1 employs a method of acquiring data related to key performance indicators (KPIs) related to a product or service, predicting future trends in the KPIs using reference information indicating the relationship between past time trends and the corresponding time trends in the KPIs, calculating the difference between the ongoing KPIs and the predicted KPIs from now on as an estimated commercial effect, and calculating the material commercial effect of the material used in a commercial for the product or service and the program commercial effect of the program linked to the commercial based on the estimated commercial effect. It is believed that employing such a method makes it possible to evaluate the value of the material used in a commercial and the value of the program linked to the commercial.

[0005] However, while the method disclosed in Patent Document 1 makes it possible to determine whether a conversion occurred due to factors such as the material used in the commercial or the program linked to that commercial, it estimates the commercial effect using data from areas where the commercial is not aired (non-airing areas), and then regresses the commercial material and program slots from the estimated overall effect to calculate the material commercial effect and the program commercial effect, thereby evaluating the impact of each. This requires a huge amount of data for evaluation, which inevitably increases the evaluation costs. Furthermore, if the amount of data related to key performance indicators is small, evaluation is not possible, and even if evaluation is performed, there is a problem in that the reliability of the results cannot be guaranteed. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 2023-86659 Summary of the Invention [Problem to be solved by the invention]

[0007] The present invention has been made in consideration of the above-mentioned problems, and its purpose is to provide a method that enables a simpler, more convenient, and lower-cost evaluation of the impact on advertising effectiveness of advertising slots themselves, such as the day of the week, time slot, broadcasting station, and program genre on which television and digital advertisements are distributed, as well as a program and information processing device for executing the method. [Means for solving the problem]

[0008] In order to solve the above problem, the method for evaluating the effectiveness of advertising spaces according to the present invention sets a base model that models the influence of advertising investment indicators for advertisements including a plurality of advertising spaces in an area of ​​interest on outcome indicators, A regression model is set on the assumption that the ratio of the advertising investment index and the outcome index expressed in the base model does not change depending on each of the frame factors of the plurality of advertising frames; Using the regression model, a partial regression coefficient of a frame factor for each of the plurality of advertising frames is calculated; Evaluating the effectiveness of the advertising spaces using partial regression coefficients of the frame factors for each of the plurality of advertising spaces or results of arithmetic operations based on the partial regression coefficients as advertising space effectiveness indicators for each of the plurality of advertising spaces. It is characterized by:

[0009] In one embodiment, the outcome measure is session entry count.

[0010] In one embodiment, the advertisement is a digital advertisement, and the advertising investment metric is the number of impressions.

[0011] In one embodiment, the advertisement is a television advertisement, and the advertising investment index includes GRP.

[0012] In one embodiment, the advertising investment metrics for the television advertisement further include a TRP.

[0013] In one embodiment, the base model is set for each of the GRP and TRP included in the advertising investment indicators for the television advertisement, and the effectiveness of the advertising space is evaluated for each GRP and TRP as an advertising space effectiveness indicator for each of the multiple advertising spaces.

[0014] In one embodiment, the TRP includes a plurality of segment TRPs, which are individual viewer ratings for each segment, and the effectiveness of the advertising space is evaluated for each of the segment TRPs as an advertising space effectiveness index for each of the plurality of advertising spaces.

[0015] In one embodiment, the TRP includes a plurality of core TRPs, which are individual viewer ratings of a core demographic, and the effectiveness of the advertising space is evaluated for each of the core TRPs as an advertising space effectiveness index for each of the plurality of advertising spaces.

[0016] In one embodiment, the base model is set by using a Hill function and fitting accumulated data on advertising investment indicators and outcome indicators for advertisements including multiple advertising spaces in the area of ​​interest to determine the parameters of the Hill function.

[0017] In one aspect, an outcome index for each of the plurality of advertising spaces is estimated based on the advertising space effectiveness index, and the average value of the estimated outcome index for each advertising space divided by the average value of the advertising investment index is set as an average cost-effectiveness index for the advertising space; Similarly, an outcome index of a specific advertising space among the plurality of advertising spaces is estimated based on the advertising space effectiveness index, and the estimated outcome index of the specific advertising space is divided by the average value of the advertising investment indexes to obtain an estimated cost-effectiveness index of the specific advertising space; The cost-effectiveness of the specific advertising space is evaluated using the comparison between the value of the estimated cost-effectiveness index of the specific advertising space and the value of the average cost-effectiveness index of the advertising space as an index.

[0018] In one embodiment, the amount of accumulated data n of advertising investment indicators and outcome indicators of advertisements including a plurality of advertising spaces in the area of ​​interest is P is the given amount n that gives total confidence F If it does not reach (n P <n F ), and the partial reliability is z=(n P / n F ) 1 / 2 The advertising space effectiveness index of the advertising space is corrected by the partial reliability and re-evaluated as the advertising space effectiveness index of the advertising space.

[0019] A program of the present invention is a program for causing an information processing device to execute a method for evaluating the effectiveness of an advertisement space, the method being the above-described method for evaluating the effectiveness of an advertisement space.

[0020] The information processing device of the present invention is an information processing device that includes a data acquisition unit, a data storage unit, a data processing unit, and a data output unit, executes the program of the present invention described above, and outputs the advertising space effectiveness index. [Effects of the Invention]

[0021] The present invention provides a method that enables simpler, more convenient, and lower-cost evaluation of the impact of advertising slots themselves, such as the day of the week, time slot, broadcasting station, and program genre on the effectiveness of television and digital advertisements, as well as a program and information processing device for executing the method. [Brief explanation of the drawings]

[0022] [Figure 1] 1 is a flowchart for illustrating a series of processes in a method for evaluating the effectiveness of an advertising space according to the present invention. [Figure 2] 1 is a block diagram illustrating an example of the configuration of an information processing device for executing an advertising space effectiveness evaluation method according to the present invention. [Figure 3] This is an example of fitting using accumulated data on advertising investment indicators and outcome indicators for advertisements including multiple advertising spaces in an area of ​​interest to determine the parameters b, k, and s of the Hill function. [Figure 4] FIG. 10 is a diagram showing an example of partial regression coefficient β (horizontal axis) for each frame factor (vertical axis) estimated by a regression model. [Figure 5] 1 is a flowchart for illustrating a process of estimating cost performance in a method for evaluating the effectiveness of an advertising space according to the present invention. [Figure 6] 10 is a flowchart for explaining, by way of example, processing in a case where the advertising investment index for a television advertisement includes TRP. DETAILED DESCRIPTION OF THE INVENTION

[0023] Hereinafter, a method for evaluating the effectiveness of an advertising space according to the present invention will be described with reference to the drawings.

[0024] [Method of evaluating the effectiveness of advertising space] FIG. 1 is a flowchart for illustrating an example of a series of processes in the method for evaluating the effectiveness of an advertising space according to the present invention.

[0025] Also, Figure 2 is a block diagram for explaining an example configuration of an information processing device for executing the above-mentioned method for evaluating the effectiveness of advertising spaces. This information processing device 100 includes a data acquisition unit 101, a data storage unit 102, a data processing unit 103, and a data output unit 104, and executes a program for executing the above-mentioned method for evaluating the effectiveness of advertising spaces, and outputs advertising space effectiveness indicators.

[0026] As shown in Figure 1, in the method for evaluating the effectiveness of advertising spaces according to the present invention, first, a base model is set (S101) that models the influence of advertising investment indices for advertisements including multiple advertising spaces in an area of ​​interest on an outcome indicator. Next, a regression model is set (S102) assuming that the ratio between the advertising investment index and the outcome indicator expressed in the base model set in S101 does not change depending on the frame factors of the multiple advertising spaces. Then, using this regression model, partial regression coefficients of the frame factors for each of the multiple advertising spaces are calculated (S103), and the partial regression coefficients of the frame factors for each of the multiple advertising spaces, or the results of arithmetic operations based on the partial regression coefficients, are used as advertising space effectiveness indicators for each of the multiple advertising spaces to evaluate the effectiveness of the advertising spaces (S104).

[0027] When performing this process, the time difference of any outcome in the area of ​​interest of the advertisement to be evaluated, the actual advertising investment index of the advertisement to be evaluated, and the slot information of the advertising slot are required.

[0028] Here, the outcome indicator is, for example, the number of session inflows (also called the number of sessions, inflows, or visits). When the outcome indicator is the number of session inflows, the time difference r of any outcome of an advertisement aired at time t is t is calculated as the difference between the mean outcome for the period from time t to time (t+δ) and the mean outcome for the period from time (t-δ) to time t ({mean outcome for the period (t~t+δ)} - {mean outcome for the period (t-δ~t)}), where δ is an arbitrary time interval.

[0029] Furthermore, the advertising investment index may be GRP if the advertisement to be evaluated is a television advertisement, or may be the number of impressions if the advertisement to be evaluated is a digital advertisement.

[0030] The base model used in the advertising space effectiveness evaluation method of the present invention can be any model, but the Hill function can be used as an example. The Hill function is a function that models the "shape effect." Here, the shape effect refers to an effect that saturates or, conversely, continues to grow depending on the amount of advertising investment. In other words, it can be said to be a function that expresses diminishing returns to advertising effectiveness. Parameter estimation of the Hill function is a method used in the field of pharmacology, such as Bayesian estimation, which is a method for estimating pharmacokinetic parameter values ​​in subjects. The Hill function is expressed, for example, by the following formula. In this example, it is assumed that the advertisement being evaluated is a television advertisement, and that the advertising investment index that most influences the outcome index r of the advertisement is GRP (Gross Rating Point), i.e., the total viewership rating of television advertisements broadcast over a certain period of time. In the following formula, b, k, and s are parameters.

[0031]

number

[0032] These parameters b, k, and s are determined by fitting accumulated data on advertising investment indicators and outcome indicators for advertisements including multiple advertising spaces in the area of ​​interest, and a base model is set.

[0033] Figure 3 shows an example of fitting using accumulated data on advertising investment indicators and outcome indicators for advertisements including multiple advertising spaces in an area of ​​interest to determine the parameters b, k, and s.

[0034] By using such a base model, it is possible to model how much the outcome indicator r increases when GRP is increased.

[0035] As described above, in the method for evaluating the effectiveness of advertising spaces according to the present invention, a regression model is set up under the assumption that the ratio between the advertising investment index and the outcome index expressed in the base model does not change depending on the space factors of a plurality of advertising spaces. By setting up a regression model in this way, the present invention makes it possible to more simply, easily, and at low cost evaluate the influence of advertising spaces themselves, such as the day of the week, time slot, broadcast station, and program genre on which television and digital advertisements are distributed, on advertising effectiveness.

[0036] The method for evaluating the effectiveness of advertising spaces according to the present invention evaluates the effectiveness of advertising spaces as an advertising space effectiveness index for each of a plurality of advertising spaces, and therefore the parameter of interest is the space factor X. In the present invention, proportionality between this space factor X and the base model (r0(grp) in the above example) is assumed. In other words, it is assumed that the effect of the space factor X is constant regardless of increases or decreases in GRP. Therefore, the effect of the space factor X can be expressed as a value obtained by multiplying the value calculated by the base model (r0(grp) in the above example) by a predetermined coefficient.

[0037] For example, if the above-mentioned Hill function (r0(grp)) is used as the base model, the outcome index r of the advertisement to be evaluated is calculated by multiplying the advertising space x i Partial regression coefficient β for each i can be expressed as the following equation using as a parameter:

[0038]

number

[0039] Figure 4 shows an example of the partial regression coefficient β for each frame factor estimated by the regression model. In this figure, the vertical axis shows each frame factor (drama, news / reports, sports, etc.), and the horizontal axis shows the partial regression coefficient for each frame factor.

[0040] For example, if the ad slot x1 is at a certain broadcasting station (Station A) and its partial regression coefficient β1 is 0.2, the outcome index r can be evaluated as exp(0.2) times (approximately 1.22 times) of the base model.The effectiveness of the ad slot is then evaluated using these evaluation results as the ad slot effectiveness index for each ad slot.

[0041] Therefore, the information processing device 100 that executes the program of the present invention having the configuration shown in Figure 2 is connected to a network, and the data acquisition unit 101 acquires data such as the time difference of any outcome in the area of ​​interest of the advertisement to be evaluated, the actual advertising investment index of the advertisement to be evaluated, and the space information of the advertising space, and the data storage unit 102 stores this data.

[0042] Based on the above data, the data processing unit 103 sets a base model that models the influence of advertising investment indexes for advertisements including multiple advertising spaces in an area of ​​interest on outcome indexes (S101), sets a regression model assuming that the ratio of the advertising investment index and outcome index expressed in the base model does not change depending on the space factors of the multiple advertising spaces (S102), calculates partial regression coefficients of the space factors for each of the multiple advertising spaces using the regression model (S103), and evaluates the effectiveness of the advertising spaces using the partial regression coefficients of the space factors for each of the multiple advertising spaces or the results of arithmetic operations based on the partial regression coefficients as advertising space effectiveness indexes for each of the multiple advertising spaces (S104).The data output unit 104 then outputs this evaluation result (advertising space evaluation index).

[0043] In one aspect of the present invention, the cost performance of an advertisement space is estimated in order to evaluate the quality of the advertisement space.

[0044] 5 is a flowchart illustrating an exemplary process for estimating cost-effectiveness in the method for evaluating the effectiveness of advertising spaces according to the present invention. As shown in this figure, the estimation is performed by estimating outcome indicators for each of a plurality of advertising spaces based on the advertising space effectiveness indicators described above (S201), dividing the estimated outcome indicators for each advertising space by the average advertising investment indicators to obtain an average value, which is used as the average cost-effectiveness indicator for the advertising spaces (S202), estimating an outcome indicator for a specific advertising space among the plurality of advertising spaces based on the advertising space effectiveness indicators (S203), dividing the estimated outcome indicator for the specific advertising space by the average advertising investment indicators to obtain an estimated cost-effectiveness indicator for the specific advertising space (S204), and evaluating the cost-effectiveness of the specific advertising space using the comparison between the estimated cost-effectiveness indicator for the specific advertising space and the average cost-effectiveness indicator for the advertising spaces as an indicator (S205). This allows for an estimation of the cost-effectiveness of the advertising space, enabling the quality of the advertising space to be evaluated.

[0045] More specifically, if the advertising investment index is GRP, its average value is 3, and the average number of session inflows is 27, the average cost-effectiveness index estimated in step S202 above is 9 (= 27 / 3). Here, if the partial regression coefficient of a certain broadcasting station (Station A) is 0.2, this can be interpreted as approximately 1.22 (= exp(0.2)) times on average. In other words, the estimated cost-effectiveness of Station A is 10.98.

[0046] However, there are naturally some ad slots with biased GRPs. For example, it is known that "weekday morning" slots have relatively low GRPs. For such ad slots, it is not appropriate to calculate the average cost-effectiveness index by dividing the average number of visitors by the average GRP.

[0047] In such cases, the following procedure can be used. First, the GRP is discretized (e.g., divided into three parts), and then the probability of the GRP being realized when the "weekday morning" time slot is selected is calculated. This can be done by simple aggregation or using a machine learning model such as Naive Bayes. (Example: Small GRP range: 74%, Medium GRP range: 21%, Large GRP range: 4%)

[0048] Next, calculate the cost-effectiveness for each of the three GRPs. The numerator is estimated using the regression formula shown above, and the denominator uses the average GRP for that range. (Example: Small GRP range: 5.11, Medium GRP range: 4.20, Large GRP range: 3.46)

[0049] Then, the estimated cost-effectiveness index for the "weekday morning" slot is calculated by taking into account the probability of realization of the three intervals, and calculating the average cost-effectiveness index. In this case, for the example above, it is (5.11 x 0.74) + (4.20 x 0.21) + (3.46 x 0.04) = 4.8. Since the average cost-effectiveness is 9, this advertising slot with an index value of 4.8 can be considered a "slot that is not recommended."

[0050] In the present invention, when the amount of data used as the basis for calculating the evaluation index is small, that is, when the amount of accumulated data n of the advertising investment index and the outcome index of the advertisement including a plurality of advertising spaces in the area of ​​interest is small, P is the given amount n that gives total confidence F If it does not reach (n P <n F ) is set to z=(n P / n F ) 1 / 2 The advertising space effectiveness index of the advertising space is corrected by the partial reliability and re-evaluated as the advertising space effectiveness index of the advertising space. This makes it possible to obtain a highly reliable advertising space effectiveness index even when the amount of data used as a base for calculating the evaluation index is small.

[0051] In conventionally known methods, if the amount of data used as the basis for calculating the evaluation index is not sufficiently large, the reliability of the statistics is low, and therefore, even if evaluation is performed under such conditions, a highly reliable ad space effectiveness index cannot be obtained. However, such data should still contain information about the ad space evaluation index. Therefore, as one aspect of the present invention, when the amount of data used as the basis for calculating the evaluation index is small, reliability is ensured to a degree commensurate with the statistical amount, and such data is used to evaluate the ad space evaluation index.

[0052] Specifically, as an example, the advertising space effectiveness index of the advertising space is re-evaluated using the following formula.

[0053]

number

[0054]

number

[0055]

number

[0056] [Method of evaluating the effectiveness of advertising space when including TRP] The following describes an example of an embodiment in which the advertising investment index for television advertising further includes TRP. Here, the target demographic in TRP is as follows: C layer: Boys and girls aged 4 to 12 (Child) T demographic: Men and women aged 13 to 19 (Teen-age) M1 group: Males aged 20-34 M2: Men aged 35-49 M3: Men over 50 years old F1 group: Females aged 20 to 34 F2 group: Women aged 35 to 49 F3 group: Women over 50 years old In addition, the demographic that can be expected to have a higher advertising effect is called the core demographic.

[0057] If the advertising investment index for television advertising further includes TRP, in addition to the above-mentioned processing, the following processing is performed.

[0058] First, the following TRP basic formula (1) and TRP basic formula (2) are formulated. TRP basic formula (1): JPEG2025127450000007.jpg14102TRP basic formula (2): JPEG2025127450000008.jpg13102

[0059] In the above formula, the meanings of the symbols are as follows: JPEG2025127450000009.jpg4896

[0060] The purpose of basic formula (1) is to estimate which TRP corresponds to the core layer. For example, if there are TRPs trp_m1 and trp_f1, it estimates which TRP corresponds to the core layer, and it estimates that trp_f1 is more likely to be in the core layer than trp_m1. In basic formula (1), the variable of interest is θ d , which is TRP d is the sensitivity to the outcome of the study, and ε is the estimation error.

[0061] In addition, in the basic formula (2), the above sensitivity θ d is a function of ad space X.

[0062] Using these basic equations (1) and (2), the following procedure is performed for all TRPs (trp_c, trp_t, trp_m1, ..., trp_f3). This procedure is shown in Figure 6. For ease of explanation, step 303 (S303) is placed after step 302 (S302), but in actual processing, both processes are performed in parallel. In addition, step 303 may be omitted in some cases.

[0063] First, in step 301 (S301), two arbitrary supervised learning methods are prepared as follows. JPEG2025127450000010.jpg2848

[0064] The above learning a is learning in which the target TRP is regressed on a TRP other than the target TRP, and the above learning b is learning in which the outcome is regressed on a TRP other than the target TRP.

[0065] Next, in step 302 (S302), the above two models are used to calculate the residual.

[0066] Specifically, based on the results of learning the model in step 301 (S301) above, the residuals for each are calculated as follows: JPEG2025127450000011.jpg2646

[0067] The above d tilde is d with the influence of TRPs other than the target TRP removed, and the above r tilde is r with the influence of TRPs other than the target TRP removed.

[0068] Then, in step 303 (S303), cross-fitting is performed to reduce estimation errors due to overfitting of both models. This process is preferably performed if there is a sufficient amount of data, but can be omitted if there is not enough data to observe.

[0069] In this cross-fitting, we first randomly divide the dataset into K folds (subsets), each of which has approximately equal size.

[0070] Next, the residuals are estimated by cross-fitting.

[0071] Specifically, for each subset k, the residual is calculated as follows: -khat and f -k hat has been trained outside the subset. JPEG2025127450000012.jpg2652

[0072] Then, we combine all the results for each field to obtain the residual as follows: JPEG2025127450000013.jpg2636

[0073] Following the above process, in step 304 (S304), the calculation of the above-mentioned basic formula (1) is performed. By this calculation, the degree of importance of the TRP of interest to the client is evaluated.

[0074] In this process, the following linear regression problem is solved and the sensitivity θ d The same process is performed for each TRP in turn to obtain θd for each TRP. Table 1 below shows an output example, and the coefficients in this table correspond to this. Also, p-values ​​are calculated as needed. JPEG2025127450000014.jpg1336

[0075] [Table 1]

[0076] From the results shown in Table 1, we can see that, for example, M1 (men aged 20-34) and M2 (men aged 35-49) were highly sensitive, while M3 (men aged 50 and over) and all F demographics (women aged 20 and over) were less sensitive and barely responded at all. In other words, we can see that men aged roughly 20 to 49 should generally be targeted, while older men and women should not be targeted.

[0077] Following the above process, in step 305 (S305), the above-mentioned basic formula (2) is calculated. This calculation determines the compatibility of the focused TRP (attribute) with the ad slot X. This makes it possible to evaluate which ad slot is particularly good for that attribute (trp).

[0078] JPEG2025127450000016.jpg17169JPEG2025127450000017.jpg1271

[0079] JPEG2025127450000018.jpg17169

[0080] Using the basic equation (2) described above, Θ is grasped as a regularized linear regression model. The regularization parameters are determined by cross-validation, as in normal machine learning. An example output is shown in Table 2.

[0081] [Table 2]

[0082] In this table, the "Overall Trend" column shows the effect of each ad slot obtained from the GRP described above. In the "Overall Trend," "TV Station E" has a value of -0.05, making it an ad slot that should be avoided, but if the target audience is older men (=M3 demographic), the value is 0.20, suggesting that "TV Station E" could be an effective ad slot.

[0083] As described above, the method for evaluating the effectiveness of advertising spaces according to the present invention includes: setting a base model that models the influence of advertising investment indicators for advertisements including a plurality of advertising spaces in an area of ​​interest on outcome indicators; A regression model is set on the assumption that the ratio of the advertising investment index and the outcome index expressed in the base model does not change depending on each of the frame factors of the plurality of advertising frames; Using the regression model, a partial regression coefficient of a frame factor for each of the plurality of advertising frames is calculated; Evaluating the effectiveness of the advertising spaces using partial regression coefficients of the frame factors for each of the plurality of advertising spaces or results of arithmetic operations based on the partial regression coefficients as advertising space effectiveness indicators for each of the plurality of advertising spaces. It is characterized by:

[0084] In one embodiment, the outcome measure is session entry count.

[0085] In one embodiment, the advertisement is a digital advertisement, and the advertising investment metric is the number of impressions.

[0086] In one embodiment, the advertisement is a television advertisement, and the advertising investment index includes GRP.

[0087] In one embodiment, the advertising investment metrics for the television advertisement further include a TRP.

[0088] In one embodiment, the base model is set for each of the GRP and TRP included in the advertising investment indicators for the television advertisement, and the effectiveness of the advertising space is evaluated for each GRP and TRP as an advertising space effectiveness indicator for each of the multiple advertising spaces.

[0089] In one embodiment, the TRP includes a plurality of segment TRPs, which are individual viewer ratings for each segment, and the effectiveness of the advertising space is evaluated for each of the segment TRPs as an advertising space effectiveness index for each of the plurality of advertising spaces.

[0090] In one embodiment, the TRP includes a plurality of core TRPs, which are individual viewer ratings of a core demographic, and the effectiveness of the advertising space is evaluated for each of the core TRPs as an advertising space effectiveness index for each of the plurality of advertising spaces.

[0091] In one embodiment, the base model is set by using a Hill function and fitting accumulated data on advertising investment indicators and outcome indicators for advertisements including multiple advertising spaces in the area of ​​interest to determine the parameters of the Hill function.

[0092] In one aspect, an outcome index for each of the plurality of advertising spaces is estimated based on the advertising space effectiveness index, and the average value of the estimated outcome index for each advertising space divided by the average value of the advertising investment index is set as an average cost-effectiveness index for the advertising space; Similarly, an outcome index of a specific advertising space among the plurality of advertising spaces is estimated based on the advertising space effectiveness index, and the estimated outcome index of the specific advertising space is divided by the average value of the advertising investment indexes to obtain an estimated cost-effectiveness index of the specific advertising space; The cost-effectiveness of the specific advertising space is evaluated using the comparison between the value of the estimated cost-effectiveness index of the specific advertising space and the value of the average cost-effectiveness index of the advertising space as an index.

[0093] In one embodiment, the amount of accumulated data n of advertising investment indicators and outcome indicators of advertisements including a plurality of advertising spaces in the area of ​​interest is P is the given amount n that gives total confidence F If it does not reach (n P <n F ), and the partial reliability is z=(n P / n F ) 1 / 2 The advertising space effectiveness index of the advertising space is corrected by the partial reliability and re-evaluated as the advertising space effectiveness index of the advertising space.

[0094] [Advertising space effectiveness evaluation program and information processing device] The program of the present invention is a program for causing an information processing device having a configuration as shown in Figure 2 to execute the advertising space effectiveness evaluation method of the present invention, and the information processing device of the present invention executes the program to output the above-mentioned advertising space effectiveness indicator.

[0095] As described above, in the method for evaluating the effectiveness of advertising spaces according to the present invention, a regression model is set up under the assumption that the ratio between the advertising investment index and the outcome index expressed in the base model does not change depending on the space factors of a plurality of advertising spaces. By setting up a regression model in this way, the present invention makes it possible to more simply, easily, and at low cost evaluate the influence of advertising spaces themselves, such as the day of the week, time slot, broadcast station, and program genre on which television and digital advertisements are distributed, on advertising effectiveness. [Industrial Applicability]

[0096] The present invention provides a method that enables simpler, more convenient, and lower-cost evaluation of the impact on advertising effectiveness of advertising slots themselves, such as the day of the week, time slot, broadcasting station, and program genre on which television and digital advertisements are distributed, as well as a program and information processing device for executing the method. [Explanation of symbols]

[0097] 100 Information processing device 101 Data Acquisition Unit 102 Data storage unit 103 Data Processing Unit 104 Data output section

Claims

1. A base model is established that models the influence of advertising spending indicators for advertisements including multiple advertising spaces in an area of ​​interest on outcome indicators. A regression model is set on the assumption that the ratio of the advertising investment index and the outcome index expressed in the base model does not change depending on each of the frame factors of the plurality of advertising frames; Using the regression model, a partial regression coefficient of a frame factor for each of the plurality of advertising frames is calculated; Evaluating the effectiveness of the advertising spaces using partial regression coefficients of the frame factors for each of the plurality of advertising spaces or results of arithmetic operations based on the partial regression coefficients as advertising space effectiveness indicators for each of the plurality of advertising spaces. A method for evaluating the effectiveness of an advertising space.

2. The method for evaluating the effectiveness of an advertising space according to claim 1 , wherein the outcome indicator is the number of sessions.

3. The method for evaluating the effectiveness of an advertising space according to claim 1 or 2, wherein the advertisement is a digital advertisement, and the advertising investment index is the number of impressions.

4. 3. The method for evaluating the effectiveness of an advertising space according to claim 1, wherein the advertisement is a television advertisement, and the advertising investment index is an index including GRP.

5. The method for evaluating the effectiveness of an advertising space according to claim 4 , wherein the advertising investment index for the television advertisement further includes a TRP.

6. 6. The method for evaluating the effectiveness of advertising spaces described in claim 5, wherein the base model is set for each of the GRP and TRP included in the advertising investment index of the television advertisement, and the effectiveness of the advertising space is evaluated for each GRP and TRP as an advertising space effectiveness index for each of the plurality of advertising spaces.

7. The method for evaluating the effectiveness of advertising spaces described in claim 6, wherein the TRP includes a plurality of segment TRPs, which are individual viewership ratings for each segment, and the effectiveness of the advertising space is evaluated for each of the segment TRPs as an advertising space effectiveness indicator for each of the plurality of advertising spaces.

8. The method for evaluating the effectiveness of advertising spaces described in claim 6, wherein the TRP includes a plurality of core TRPs, which are individual viewership ratings of a core demographic, and the effectiveness of the advertising spaces is evaluated for each of the core TRPs as an advertising space effectiveness indicator for each of the plurality of advertising spaces.

9. 3. A method for evaluating the effectiveness of advertising spaces as described in claim 1 or 2, wherein a Hill function is used as the base model, and the parameters of the Hill function are determined by fitting using accumulated data on advertising investment indicators and outcome indicators for advertisements including multiple advertising spaces in the area of ​​interest, thereby setting the base model.

10. An outcome index for each of the plurality of advertising spaces is estimated based on the advertising space effectiveness index, and the average value of the estimated outcome index for each advertising space divided by the average value of the advertising investment index is set as an average cost-effectiveness index for the advertising space; Similarly, an outcome index of a specific advertising space among the plurality of advertising spaces is estimated based on the advertising space effectiveness index, and the estimated outcome index of the specific advertising space is divided by the average value of the advertising investment indexes to obtain an estimated cost-effectiveness index of the specific advertising space; evaluating the cost-effectiveness of the specific advertising space using a comparison between the value of the estimated cost-effectiveness index of the specific advertising space and the value of the average cost-effectiveness index of the advertising space as an index; The method for evaluating the effectiveness of an advertising space according to claim 1 or 2.

11. The amount of accumulated data n of advertising investment indicators and outcome indicators for advertisements including a plurality of advertising spaces in the area of ​​interest Ｐ is a given quantity n that gives total confidence Ｆ If it does not reach (n Ｐ <n Ｆ ), and the partial reliability is z = (n Ｐ / n Ｆ ) １／２ and correcting the advertising space effectiveness index of the advertising space with the partial reliability and re-evaluating the advertising space effectiveness index of the advertising space.

12. 3. A program for causing an information processing device to execute a method for evaluating the effectiveness of an advertising space, the method being the method for evaluating the effectiveness of an advertising space according to claim 1 or 2.

13. An information processing device comprising: a data acquisition unit, a data storage unit, a data processing unit, and a data output unit; and executing the program according to claim 12 to output the advertising space effectiveness index.

Images