Search advertising target-specific weighting automation operation system

Abstract

The present invention relates to an automated operation system for weighting by search advertising target, and more specifically, to an automated operation system for weighting by search advertising target that maximizes the efficiency of search advertising by providing data analysis, automated weight adjustment, and user convenience. According to the present invention, advertising performance data by gender, age, time of day, and region is collected and analyzed to enable scientific targeting based on objective data, thereby allowing advertisers to establish efficient targeting strategies based on data without relying on intuition or experience.

Description

Technology Field

[0001] The present invention relates to an automated operation system for weighting by search advertising target, and more specifically, to an automated operation system for weighting by search advertising target that maximizes the efficiency of search advertising by providing data analysis, automated weight adjustment, and user convenience. Background Technology

[0002] In existing search advertising systems, advertisers or marketers had to directly set and adjust weights for each ad target, which required significant time and effort, and it was difficult to set efficient weights if they lacked professional knowledge.

[0003] Furthermore, in the past, the analysis of advertising performance data was limited, and it was difficult to make scientific decisions using it.

[0004] Therefore, weights were often adjusted based on intuition or experience, which became a factor hindering advertising efficiency.

[0005] Furthermore, while advertising performance constantly changes depending on time, situation, and trends, existing systems found it difficult to adjust weights in real-time in response to these changes.

[0006] Therefore, optimized advertising operations were difficult, and advertising effectiveness could decrease.

[0007] In addition, manual weight adjustments, a lack of data analysis, and difficulties in real-time response led to reduced efficiency in advertising operations and potential waste of advertising budgets.

[0008] To solve the aforementioned problems, the present invention proposes an automated operation system for search ad target-specific weights equipped with data analysis and automated weight adjustment functions. Prior art literature

[0009] Republic of Korea Registered Patent No. 10-2620128 The problem to be solved

[0010] Accordingly, the present invention has been devised to resolve the aforementioned conventional problems, and the objective of the present invention is to collect and analyze advertising performance data by gender, age, time of day, and region to search for the optimal target capable of increasing advertising efficiency.

[0011] Another objective of the present invention is to analyze advertising performance in real time and automatically adjust advertising weights based on this to maximize the return on investment (ROI).

[0012] Another objective of the present invention is to improve advertising operational efficiency by automatically adjusting weights according to key objectives (number of conversions, conversion rate, sales revenue, ROAS, etc.) set by the advertiser.

[0013] Another objective of the present invention is to automatically apply weights through API integration and to enable easy identification and management of advertising performance through a user-friendly interface. means of solving the problem

[0014] In order to achieve the problem that the present invention aims to solve,

[0015] An automated operation system for weighting by search ad target according to one embodiment of the present invention,

[0016] A data collection unit (100) that collects advertising performance data by gender, age group, time of day, and region;

[0017] A performance analysis unit (200) that analyzes the collected data above to calculate at least two of the following: number of conversions, conversion rate, conversion revenue, advertising return rate, number of impressions, number of clicks, cost per click, and total cost;

[0018] A weight calculation unit (300) that calculates the optimal weight by applying the weight ratio for each key indicator by referring to the weight table for each key indicator;

[0019] The problem of the present invention is solved by including an API integration unit (400) that automatically applies the weights calculated above through the API. Effects of the invention

[0020] The automated operation system for weighting by search advertisement target of the present invention provides the following significant effects.

[0021] First, it provides accurate, data-driven scientific targeting.

[0022] In other words, by collecting and analyzing advertising performance data by gender, age, time of day, and region, it enables scientific targeting based on objective data. Through this, advertisers can establish efficient targeting strategies based on data, rather than relying on intuition or experience.

[0023] Second, it provides ROI improvement through real-time performance analysis.

[0024] In other words, by analyzing advertising performance in real time and automatically adjusting ad weights based on this, it enables a rapid response to the changing advertising environment, maximizing advertising effectiveness and improving the return on investment (ROI).

[0025] Third, it provides increased efficiency in advertising operations.

[0026] In other words, by automatically applying optimal weights based on the key objectives set by advertisers through automated weight adjustment functions, the efficiency of advertising operations is increased, saving time and costs while improving advertising performance.

[0027] Fourth, it provides improved user convenience.

[0028] In other words, since weights are automatically applied through API integration, users can conveniently use the system without a complex setup process.

[0029] In addition, user convenience is enhanced by enabling easy identification and management of advertising performance through a user-friendly interface. Brief explanation of the drawing

[0030] FIG. 1 is an overall configuration diagram of an automated operation system for weighting by search ad target according to an embodiment of the present invention. FIG. 2 is an example of a full screen displayed by the operation of an automated operating system for weighting by search advertisement target according to an embodiment of the present invention. FIG. 3 is a conceptual diagram of the configuration of a data collection unit (100) of an automated operation system for weighting by search advertisement target according to an embodiment of the present invention. Figure 4 is a graph illustrating the ratio of each of the three fields of male, female, and unknown relative to the average value of the automated operation system for weighting by search ad target according to an embodiment of the present invention. Figure 5 is a graph showing the sum calculated by age group and displayed as a ratio in an automated operation system for weighting by search advertisement target according to an embodiment of the present invention. FIG. 6 is an example of a CRITERION report for an automated operation system for weighting by search ad target according to an embodiment of the present invention. FIG. 7 is an example of an Ad Performance report of an automated operation system for weighting by search ad target according to an embodiment of the present invention. FIG. 8 is a conceptual diagram of the configuration of the performance analysis unit (200) of the automated operation system for weighting by search advertisement target according to one embodiment of the present invention. FIG. 9 is a conceptual diagram of the configuration of the weight calculation unit (300) of the automated operation system for weighting by search ad target according to an embodiment of the present invention. FIG. 10 is an example diagram of a weighting table by key indicator of an automated operation system for weighting by search ad target according to an embodiment of the present invention. FIG. 11 is an example diagram showing the application of a correction weight to an age group in the case of equal emphasis in an automated operation system for weighting by search advertisement target according to an embodiment of the present invention. FIG. 12 is an example diagram showing the application of a correction weight to an age group when the conversion target is a weighting point of an automated operation system for search advertising targeting according to an embodiment of the present invention. FIG. 13 is a conceptual diagram of the configuration of the API linkage unit (400) of the automated operation system for weighting by search advertisement target according to an embodiment of the present invention. FIG. 14 is an example diagram showing the performance of a target-specific weighting automation operation system for search advertisements according to an embodiment of the present invention. FIGS. 15 and 16 are exemplary diagrams showing conversion count, conversion rate, conversion revenue, and advertising return rate layers on the gender layer, time-of-day layer, and age-group layer through the display of an automated operation system for weighting by search ad target according to an embodiment of the present invention. FIG. 17 is an example diagram showing how an administrator of an automated operation system for weighting by search ad target according to an embodiment of the present invention sets key indicators and weighting application targeting to change weights. Specific details for implementing the invention

[0031] The following description merely illustrates the principles of the present invention. Therefore, those skilled in the art may invent various devices that embody the principles of the present invention and are included within the concept and scope of the present invention, even though they are not explicitly described or illustrated in this specification.

[0032] Furthermore, all conditional terms and embodiments listed in this specification are, in principle, explicitly intended only for the purpose of enabling an understanding of the concept of the invention and should be understood not as being limited to the embodiments and conditions specifically listed as such.

[0033] An automated operation system (1000) for weighting by search advertisement target according to one embodiment of the present invention,

[0034] A data collection unit (100) that collects advertising performance data by gender, age group, time of day, and region;

[0035] A performance analysis unit (200) that analyzes the collected data above to calculate at least two of the following: number of conversions, conversion rate, conversion revenue, advertising return rate, number of impressions, number of clicks, cost per click, and total cost;

[0036] A weight calculation unit (300) that calculates the optimal weight by applying the weight ratio for each key indicator by referring to the weight table for each key indicator;

[0037] It is characterized by being configured to include an API integration unit (400) that automatically applies the weights calculated above through the API.

[0038] At this time, the data collection unit (100) is,

[0039] Gender data module (110) that collects data classified into male, female, and unclassified based on the user's login information;

[0040] An age data module (120) that collects data by dividing it into 12 intervals from under 14 years of age to 60 years of age or older;

[0041] Time zone data module (130) that collects data by dividing 24 hours into 1-hour units;

[0042] It is characterized by including a regional data module (140) that collects data by classifying it into metropolitan / provincial and city / county / district units.

[0043] At this time, the above performance analysis unit (200) is,

[0044] A conversion analysis module (210) that aggregates the number of conversions occurring by target attribute;

[0045] A conversion rate analysis module (220) that calculates the ratio of conversions to clicks;

[0046] A conversion sales analysis module (230) that calculates the total sum of sales amounts per conversion;

[0047] It is characterized by including an advertising return rate analysis module (240) that calculates the ratio of conversion sales to advertising costs.

[0048] At this time, the weight calculation unit (300) is,

[0049] Key indicator setting module (310) for setting the type of key indicator;

[0050] A performance analysis module (320) that applies weight ratios differently according to the set key indicators;

[0051] A weight calculation module (330) that adjusts the calculated weight within the range of 50 to 500;

[0052] It is characterized by including a verification module (340) that verifies the validity of the calculated weights.

[0053] At this time, the API integration unit (400) is,

[0054] A request generation module (410) that generates an API request URL and parameters;

[0055] Response processing module (420) that validates and processes API responses;

[0056] It is characterized by including a history management module (430) that manages the history of weight changes.

[0057] Hereinafter, an example of the automated operation system for weighting by search advertisement target according to the present invention will be described in detail.

[0058] FIG. 1 is an overall configuration diagram of an automated operation system for weighting by search ad target according to an embodiment of the present invention.

[0059] As illustrated in FIG. 1, the automated operation system (1000) for search advertising target-specific weighting according to one embodiment of the present invention is,

[0060] A data collection unit (100) that collects advertising performance data by gender, age group, time of day, and region;

[0061] A performance analysis unit (200) that analyzes the collected data above to calculate the number of conversions, conversion rate, conversion revenue, and advertising return rate;

[0062] A weight calculation unit (300) that calculates the optimal weight by applying the weight ratio for each key indicator by referring to the weight table for each key indicator;

[0063] It is characterized by being configured to include an API integration unit (400) that automatically applies the weights calculated above through the API.

[0064] Through the above configuration, a screen such as that shown in FIG. 2 can be provided, for example, by displaying layers of API authentication, performance by target, and weighted recommendation, and is characterized by providing exposures, clicks, advertising costs, conversions, sales, etc. on the performance by target layer.

[0065] The number of conversions described in this invention is an indicator that shows how many valid results an advertisement actually produces, such as when 5 people purchase a product through an advertisement, the number of conversions becomes 5.

[0066] In addition, the conversion rate is the ratio of people who actually convert among those who clicked on an advertisement; for example, if 100 people clicked on an advertisement and 10 of them purchased a product, the conversion rate is 10%, and this is an indicator that measures the persuasiveness and effectiveness of an advertisement.

[0067] In addition, conversion revenue is the total revenue generated through conversion; for example, if 10 people purchased a product and the price of each product was 20,000 won, the conversion revenue would be 200,000 won, which is an indicator showing how much revenue was actually generated through advertising.

[0068] In addition, the return on advertising is an indicator that shows how much revenue is generated relative to the cost spent on advertising. If 100,000 won is spent on advertising and the converted revenue is 200,000 won, the ROAS is 200%. This is a key indicator for evaluating the efficiency of advertising investment, and a higher ROAS means that the advertising is being operated more efficiently.

[0069] Through the above configuration, the present invention enables the realization of accurate data-based scientific targeting, thereby providing an improvement in ROI through real-time performance analysis.

[0070] The components of the present invention will be described in detail below with reference to the drawings.

[0071] FIG. 3 is a conceptual diagram of the configuration of a data collection unit (100) of an automated operation system for weighting by search advertisement target according to an embodiment of the present invention.

[0072] As illustrated in FIG. 3, the data collection unit (100) performs the function of collecting advertising performance data by gender, age group, time of day, and region.

[0073] In order to perform the above function, the data collection unit (100) is,

[0074] Gender data module (110) that collects data classified into male, female, and unclassified based on the user's login information;

[0075] An age data module (120) that collects data by dividing it into 12 intervals from under 14 years of age to 60 years of age or older;

[0076] Time zone data module (130) that collects data by dividing 24 hours into 1-hour units;

[0077] It is characterized by including a regional data module (140) that collects data by classifying it into metropolitan / provincial and city / county / district units.

[0078] To explain in detail, the gender data module (110) performs the function of collecting data by classifying it into male (M), female (F), and unclassified (unknown) based on the user's login information.

[0079] At this time, clicks, impressions, and conversions are collected separately for male (M), female (F), and unclassified (unknown).

[0080] For example, when a male user clicks on a specific ad, the corresponding click data is aggregated in the 'Male' category; when a female user purchases a specific product, the corresponding conversion data is aggregated in the 'Female' category; and ad impression data from users without gender information is aggregated in the 'Uncategorized' category.

[0081] In addition, Figure 4 is a graph illustrating the proportion of each field relative to the average value of male, female, and unknown, showing the average values ​​of male, female, and unknown for conversions, conversion rate, conversion revenue, and advertising return rate, and then expressing them as percentages by separating them into male and female.

[0082] The above age data module (120) performs the function of collecting data by dividing it into 12 segments from under 14 years of age to over 60 years of age.

[0083] For example, data is collected by dividing it into 12 intervals, such as under 14 years old, 14 to 18 years old, 19 to 24 years old, ..., 55 to 59 years old, 60 to 99, unknown, etc.

[0084] At this time, when a 25-year-old user visits the website through an advertisement, the data is aggregated in the '25-29' age range, and when a 48-year-old user sees an advertisement and purchases a product, the data is aggregated in the '45-49' age range.

[0085] In addition, in the case of Figure 5, the number of conversions, conversion rate, conversion revenue, and advertising return rate are first calculated as the proportion of each age group to the total value, and after calculating the total by age group, it is displayed as a ratio graph.

[0086] The above time zone data module (130) performs the function of collecting data by dividing 24 hours into 1-hour units.

[0087] For example, data is collected by dividing the time of occurrence of advertising-related events (impressions, clicks, conversions, etc.) into one-hour intervals, and data is collected in a total of 24 time zones by dividing 24 hours into one-hour intervals.

[0088] The above regional data module (140) performs the function of collecting data by classifying it into metropolitan / provincial and city / county / district units.

[0089] For example, it involves collecting data based on users' regional information, specifically by classifying regions into metropolitan / provincial and city / county / district units.

[0090] Meanwhile, the above data collection unit (100) is characterized by collecting indicators for gender and age as CRITERION / CRITERION_CONVERSION and collecting indicators for wide area and time as Ad Performance / Conversion Report.

[0091] For example, Figure 6 shows an example of a CRITERION report, and Figure 7 shows an example of an Ad Performance report.

[0092] That is, in the case of Fig. 6 (CRITERION report), information such as gender, age, device, impressions, clicks, and costs is included, and each row represents data for a specific criterion (e.g., male), and in the case of Fig. 7 (Ad Performance report), information such as date, time, region code, impressions, clicks, and costs is included, and each row represents data for a specific time and region.

[0093] In addition, gender and age data are collected through the CRITERION report, which is characterized by providing advertising performance indicators based on gender (male, female, unknown) and age (12 age groups including under 14, 14–18, 19–24, etc.).

[0094] The CRITERION_CONVERSION report is characterized by collecting conversion-related metrics (e.g., number of conversions, conversion rate, conversion revenue, impressions, clicks, cost per click, total cost) in addition to the gender and age data collected from the CRITERION report.

[0095] In addition, regional and time data are collected through the Ad Performance report, which is characterized by providing advertising performance indicators (e.g., impressions, clicks, advertising costs) based on regional areas (metropolitan cities and provinces) and time (in 1-hour units).

[0096] The Conversion Report is characterized by collecting conversion-related metrics (e.g., number of conversions, conversion rate) in addition to the regional and temporal data collected from the Ad Performance Report.

[0097] And, as illustrated in FIG. 8, the performance analysis unit (200) is characterized by analyzing the collected data to calculate at least two of the following: number of conversions, conversion rate, conversion revenue, advertising return rate, number of impressions, number of clicks, cost per click, and total cost.

[0098] In order to perform the above function, the performance analysis unit (200) is,

[0099] A conversion analysis module (210) that aggregates the number of conversions occurring by target attribute;

[0100] A conversion rate analysis module (220) that calculates the ratio of conversions to clicks;

[0101] A conversion sales analysis module (230) that calculates the total sum of sales amounts per conversion;

[0102] It is characterized by including an advertising return rate analysis module (240) that calculates the ratio of conversion sales to advertising costs.

[0103] To explain in detail, the conversion analysis module (210) performs the function of aggregating the number of conversions by target attribute, and the formula is as follows.

[0104] Conversions = Σ(Number of conversions)

[0105] For example, if 10 purchase conversions occurred among women in their 20s and 5 purchase conversions occurred among men in their 30s, the conversion counts are counted as 10 and 5, respectively.

[0106] As another example, if 20 purchase conversions occurred in the 'Seoul' region, the number of conversions in the 'Seoul' region is counted as 20.

[0107] The above conversion rate analysis module (220) performs the function of calculating the ratio of conversions to clicks, and the formula is as follows.

[0108] Conversion Rate (%) = (Conversions / Clicks) × 100

[0109] For example, if the number of clicks for a target audience of women in their 20s is 100 and the number of conversions is 10, the conversion rate is calculated as 10%.

[0110] As another example, if the number of clicks in the '2 PM' time slot is 50 and the number of conversions is 5, the conversion rate is calculated as 10%.

[0111] The above conversion sales analysis module (230) performs the function of calculating the total sum of sales generated through conversion for each target attribute, and the formula is as follows.

[0112] Conversion Sales = Σ(Sales per Conversion)

[0113] For example, if 10 purchase conversions occurred among women in their 20s and the purchase amounts were 20,000 won, 30,000 won, 10,000 won, etc., the total is calculated to determine the conversion revenue.

[0114] As another example, if 5 purchase conversions occurred in the 'Busan' region and the purchase amounts were 100,000 won, 50,000 won, 30,000 won, etc., the total is calculated to determine the conversion revenue.

[0115] The above advertising return rate analysis module (240) performs the function of analyzing the return on advertising (ROAS) by calculating the ratio of conversion sales to advertising costs for each target attribute, and the formula is as follows.

[0116] ROAS(%) = (Conversion Revenue / Advertising Cost) × 100

[0117] For example, if the advertising cost for a target audience of men in their 30s is 100,000 won and the conversion revenue is 200,000 won, the ROAS is calculated as 200%.

[0118] As another example, if the advertising cost for the 'Gyeonggi' region is 500,000 won and the conversion revenue is 1,000,000 won, the ROAS is calculated as 200%.

[0119] And, as illustrated in FIG. 9, the weight calculation unit (300) is characterized by calculating the optimal weight by applying the weight ratio for each key indicator by referring to the weight table for each key indicator.

[0120] To perform the above function, the weight calculation unit (300) is,

[0121] Key indicator setting module (310) for setting the type of key indicator;

[0122] A performance analysis module (320) that applies weight ratios differently according to the set key indicators;

[0123] A weight calculation module (330) that adjusts the calculated weight within the range of 50 to 500;

[0124] It is characterized by including a verification module (340) that verifies the validity of the calculated weights.

[0125] To explain in detail, as illustrated in FIG. 10, the weighting table for each key indicator above is,

[0126] The rows display conversion count, conversion rate, sales revenue, and ROAS, and

[0127] It is characterized by displaying equal weighting, conversion count focus, conversion rate focus, revenue focus, and ROAS focus in columns.

[0128] At this time, the standard ratio for each key indicator is characterized as follows.

[0129] In the case of equal focus, it is characterized by conversions:conversion rate:sales:ROAS = 1:1:1:1, and

[0130] In the case of the conversion-focused point, it is characterized by the conversions:conversion rate:sales:ROAS = 1.6:0.8:0.8:0.8, and

[0131] In the case of conversion rate focus, it is characterized by conversions:conversion rate:sales:ROAS = 0.8:1.6:0.8:0.8, and

[0132] In the case of sales focus, it is characterized by conversions:conversion rate:sales:ROAS = 0.8:0.8:1.6:0.8, and

[0133] In the case of ROAS focus, it is characterized by conversions:conversion rate:sales:ROAS = 0.8:0.8:0.8:1.6.

[0134] At this time, the above-mentioned key indicator setting module (310) performs the function of setting the type of key indicator, and sets the key indicator among the number of conversions, conversion rate, sales revenue, and ROAS according to the key goal desired by the advertiser.

[0135] The above-mentioned key indicator is characterized as being one of the same focus, conversion rate focus, conversion rate focus, sales focus, and ROAS focus.

[0136] For example, if an advertiser's goal is to increase brand awareness, they could set 'conversions' as a key metric to expose ads to a large number of people and drive website visits.

[0137] As another example, if an advertiser's goal is to increase sales volume, they can set 'Revenue' as the key metric, and if their goal is to maximize efficiency relative to advertising costs, they can set 'ROAS' as the key metric.

[0138] The above performance analysis module (320) applies weight ratios differentially according to the set key indicators, and applies weight ratios differentially by referring to the weight table of FIG. 10 according to the key indicators set in the key indicator setting module (310).

[0139] At this time, it is obvious that the above weighting ratio can be set by the administrator, and that the setting value can be changed as needed.

[0140] For example, if 'Conversions' is set as the key indicator, the 'Conversions Key' row of the weight table is referenced, so in this case, a weight of 1.6 is applied to the conversions, and a weight of 0.8 is applied to the conversion rate, sales, and ROAS respectively.

[0141] In addition, if 'ROAS' is set as the key indicator, the 'ROAS Key' row of the weight table is referenced to apply a weight of 1.6 to ROAS and 0.8 to the remaining indicators.

[0142] In addition, if set to 'Equal Weight (Equal Emphasis)', a weight of 1 is applied to all indicators.

[0143] The above weight calculation module (330) can apply the following formula to adjust the calculated weight within the range of 50 to 500.

[0144] 1) Default weight = 100

[0145] 2) Adjustment range = 50 ~ 500

[0146] 3) Weighting Formula = Base Weight × (1 + (Performance Value - Median Value) / Median Value × Key Indicator Ratio)

[0147] The above default weight is a default weight applied to all target attributes, and is set to 100 here.

[0148] In other words, if there is no significant difference in performance, all target attributes will have a weight of 100.

[0149] In addition, the adjustment range refers to the minimum and maximum values ​​that the calculated weight can have, and here it is set to 50 to 500.

[0150] In other words, the weight is not set below 50 no matter how low the performance is, and it is not set above 500 no matter how high the performance is.

[0151] In addition, the performance value refers to performance indicator values ​​such as the number of conversions, conversion rate, revenue, and ROAS for each target attribute; for example, if the conversion rate of the target audience of women in their 20s is 10%, the 'performance value' becomes 10.

[0152] In addition, the median refers to the median of the performance values ​​of all target attributes; for example, if the median is 5% when investigating the conversion rate of all age groups, the 'median' is 5.

[0153] In addition, the key indicator ratio refers to the ratio taken from the weight table of FIG. 10 according to the key indicator set in the key indicator setting module (310). For example, if it is set to 'conversion rate key', the key indicator ratio for the conversion rate becomes 1.6.

[0154] The above formula adjusts weights based on how much higher or lower the performance value of each target attribute is compared to the median value; the higher the performance value is compared to the median, the higher the weight becomes, and the lower the performance value is compared, the lower the weight becomes. Additionally, the higher the ratio of key indicators, the greater the range of weight adjustment.

[0155] For example, assuming the conversion rate for the target audience of women in their 20s is 10%, the median conversion rate for all age groups is 5%, and the key indicator is 'conversion rate focus' (key indicator ratio 1.6), the weights are calculated as follows according to the weighting formula.

[0156] 100 × (1 + (10 - 5) / 5 × 1.6) = 260

[0157] In other words, since the conversion rate of the target audience of women in their 20s is higher than the median, a weight of 260, which is higher than the basic weight of 100, can be applied.

[0158] Through this weighting formula, the system will be able to quantitatively evaluate the performance of each target attribute and, based on this, automatically calculate weights that can maximize advertising efficiency.

[0159] In addition, the reason for adjustment is that weights can be adjusted to meet the weight range required by the advertising platform or to reflect the advertiser's specific requirements.

[0160] For example, if the weight of a specific target is calculated as 10 but the minimum weight allowed by the advertising platform is 50, it will be adjusted to 50, and if the advertiser wants to target a specific age group more importantly, the weight of that age group can be increased to 500.

[0161] As another example, Figure 11 shows an example of age groups in the case of equal emphasis, where the number of conversions, conversion rate, sales revenue, and ROAS are 1:1:1:1. Therefore, weights are calculated by referring to this, and finally, all four values ​​are added together to set a sum weight, and a correction weight is applied based on this.

[0162] In addition, Figure 12 shows an example of age groups when the conversion rate is the focus point, and since the conversion rate, conversion rate, sales revenue, and ROAS are 1.6 : 0.8 : 0.8 : 0.8, weights are calculated by referring to this, and finally, all four values ​​are added together to set a sum weight, and a correction weight is applied based on this.

[0163] The verification module (340) performs the function of verifying the validity of the calculated weights.

[0164] In other words, it involves verifying whether the calculated weights meet the requirements of the advertising platform, such as validating the weight range and data format.

[0165] In addition, it is intended to utilize historical data to simulate the expected effects of applying weights and to verify whether there are any issues.

[0166] For example, if the calculated weight exceeds the range allowed by the advertising platform, it outputs an error message and allows the weight to be readjusted.

[0167] In addition, if simulation results indicate that applying a specific weight is expected to actually decrease advertising efficiency, it is designed to display a warning message and recommend weight adjustments.

[0168] Specifically, when applying a Multilayer Perceptron (MLP)-based prediction model, the input values ​​are historical advertising campaign data (e.g., target attributes, impressions, clicks, conversions, revenue, advertising costs, etc.), and the output values ​​are advertising performance indicators expected after applying adjusted weights (e.g., conversions, conversion rate, revenue, ROAS).

[0169] At this time, the multilayer perceptron model, consisting of an input layer, a hidden layer, and an output layer, learns the relationship between weights and advertising performance by training the MLP model using past data, applies adjusted weights to new advertising campaign data, and validates the validity by predicting expected performance through the MLP model.

[0170] In addition, when a simulation-based verification model is applied, performance is simulated using past advertising campaign data and adjusted weights.

[0171] For example, in the case of Monte Carlo simulation, multiple scenarios can be generated by considering various variables (e.g., click-through rates, conversion rate fluctuations), and performance can be simulated by applying adjusted weights to each scenario.

[0172] In addition, for A / B test simulations, the simulation is conducted by dividing the groups into one with adjusted weights applied and one without, and validity is verified by comparing the performance difference between the two groups.

[0173] In this case, the simulation-based verification model can predict the effect of weights without conducting actual advertising campaigns.

[0174] And, as illustrated in FIG. 13, the API integration unit (400) is characterized by automatically applying the calculated weight through the API.

[0175] In order to perform the above function, the API integration unit (400) is,

[0176] A request generation module (410) that generates an API request URL and parameters;

[0177] Response processing module (420) that validates and processes API responses;

[0178] It is characterized by including a history management module (430) that manages the history of weight changes.

[0179] Specifically, the request generation module (410) generates an API request URL and parameters, and generates the API request URL and parameters to transmit the weight calculated by the weight calculation unit (300) to the advertising platform, and is, for example, as follows.

[0180] 1) adgroupId : grp-a001-01-000000024265117 (Ad Group ID)

[0181] 2) dictionaryCode : AG (Targeting code, here meaning age group)

[0182] 3) bidWeight : 150 (Weight to change)

[0183] It is characterized by generating the following API request URL and parameters using the above information.

[0184] 1) Request URL: PUT / ncc / criterion / grp-a001-01-000000024265117 / bidWeight

[0185] 2) Parameter : {dictionaryCode: AG, bidWeight: 150}

[0186] The above request means to change the weight for age group targeting of the "grp-a001-01-000000024265117" ad group to 150.

[0187] The above response processing module (420) performs the function of verifying and processing API responses.

[0188] For example, response code verification is as follows.

[0189] 1) 200 (Success): This means that the weight change has been successfully completed, and the change history is recorded in the history management module (430).

[0190] 2) 400 (Parameter Error): This indicates that there is an error in the request parameters. For example, if the adgroupId or bidWeight value is incorrect, the error is analyzed, and if necessary, the request is modified and resent.

[0191] 3) 500 (System Error): This indicates that an error has occurred on the advertising platform server. In this case, the request will be resent after a certain period of time.

[0192] In addition, regarding response data parsing, when the advertising platform sends a response in JSON format, the response processing module (420) parses the JSON data and extracts the necessary information.

[0193] In addition, for error handling, if an API request fails, it will be retried up to 3 times, and if it fails even after 3 retries, an error message will be logged and a notification will be sent to the administrator.

[0194] The above history management module (430) manages the history of weight changes.

[0195] At this time, the stored information is characterized by including the change time, ad group ID (adgroupId), targeting code (dictionaryCode), weight before change, weight after change, API response code, and API response message.

[0196] In addition, Figure 14 is an example diagram showing performance by target. The system of the present invention provides current weight information and recommended weight information based on the history management module, and when this is applied, the recommended weight is applied by the operation of the API linkage unit (400).

[0197] Meanwhile, the automated operation system (1000) for search advertising target-specific weighting according to one embodiment of the present invention is,

[0198] It is characterized by providing a performance indicator inquiry and weight change layer.

[0199] At this time, the above performance indicator inquiry layer is,

[0200] It is characterized by including an impression layer, a click layer, an ad cost layer, a conversion layer, and a sales layer,

[0201] It is characterized by including a core target layer, a gender layer, a time-based layer, and an age-based layer.

[0202] At this time, as illustrated in FIGS. 15 to 16, the gender layer, time-based layer, and age-based layer are characterized by including layers for conversion count, conversion rate, conversion revenue, and advertising return rate.

[0203] At this time, the above weight change layer,

[0204] A key indicator layer for selecting one of the same weight, conversion weight, conversion rate weight, revenue weight, and ROAS weight, and

[0205] It is characterized by including a weighted targeting layer for selecting multi-value targeting targets by gender, age group, time of day, and region.

[0206] For example, as illustrated in FIG. 17, a focus indicator layer for selecting one of the same focus, conversion focus, conversion rate focus, sales focus, and ROAS focus is displayed at a specific location on the screen, and a weighted targeting layer for selecting multi-value targeting targets by gender, age group, time of day, and region is displayed at a different location.

[0207] At this point, the administrator sets the key indicators for changing weights and the targets for applying weights.

[0208] In addition, depending on additional aspects, the system of the present invention,

[0209] It is characterized by collectively changing target weights and setting ad exposure status for multiple ad campaigns or ad groups through API integration with an advertising system.

[0210] In other words, the weights of multiple groups are changed at once.

[0211] According to the present invention, advertising performance data by gender, age, time of day, and region is collected and analyzed to enable scientific targeting based on objective data, thereby allowing advertisers to establish efficient targeting strategies based on data without relying on intuition or experience.

[0212] Those skilled in the art to which the present invention pertains will understand that the present invention, as described above, may be implemented in other specific forms without altering the technical concept or essential features of the invention. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive.

[0213] The scope of the present invention is defined by the claims set forth below rather than by the detailed description above, and all modifications or variations derived from the meaning and scope of the claims and equivalent concepts thereof should be interpreted as being included within the scope of the present invention. Explanation of the symbols

[0214] 100 : Data Collection Unit 200 : Performance Analysis Department 300 : Weight calculation unit 400 : API Integration Section

Claims

Claim 1 In an automated operation system for weighting by search ad target, the system comprises: a data collection unit (100) that collects ad performance data by gender, age group, time of day, and region; a performance analysis unit (200) that analyzes the data collected by the data collection unit and calculates at least two of the following: number of conversions, conversion rate, conversion revenue, ad return rate, number of impressions, number of clicks, cost per click, and total cost; a weight calculation unit (300) that calculates an optimal weight by applying a weight ratio for each key indicator by referring to a weight table for each key indicator; and an API integration unit (400) that automatically applies the weights calculated by the weight calculation unit through an API. The performance analysis unit (200) includes: a conversion analysis module (210) that aggregates the number of conversions occurring by target attribute; a conversion rate analysis module (220) that calculates the ratio of conversions to clicks; a conversion revenue analysis module (230) that calculates the total sum of revenue per conversion; and an ad return rate analysis module (240) that calculates the ratio of conversion revenue to ad costs. The weight calculation unit (300) includes: a key indicator setting module (310) for setting the type of key indicator; a performance analysis module (320) for differentially applying weight ratios according to the set key indicator; a weight calculation module (330) for adjusting the calculated weight within the range of 50 to 500; and a verification module (340) for verifying the validity of the calculated weight. The key indicator is any one of identical key, conversion rate key, conversion rate key, sales revenue key, and ROAS key. The performance analysis module (320) differentially applies weight ratios by referring to a weight table according to the key indicator set in the key indicator setting module (310). The weight calculation module (330) has a basic weight of 100, an adjustment range of 50 to 500, and a weight calculation formula = basic weight x (1 + (performance value - median value) / median value x key indicator ratio). The above API integration unit (400) includes: a request generation module (410) that generates an API request URL and parameters; a response processing module (420) that verifies and processes the API response; and a history management module (430) that manages the history of weight changes.An automated operation system for weights by search ad target, comprising: a request generation module (410) that generates an API request URL and parameters to transmit weights calculated by a weight calculation unit (300) to an advertising platform, wherein the parameters include an ad group ID, a targeting code, and weights to be changed, and the stored information of the history management module (430) includes a change time, an ad group ID, a targeting code, weights before change, weights after change, an API response code, and an API response message.

Images