Digital Placement Adjustment Method and System Based on Multi-Platform Data Analysis

Abstract

The present invention discloses a digital placement adjustment method and system based on multi-platform data analysis, which relates to the technical field of advertising placement, and includes obtaining platform advertising placement information, classifying the placed advertisements according to the platform advertising placement information to obtain classified advertisement placement information, and obtaining advertisement effect data according to the platform advertising placement information. By dividing the placed advertisements with the same product type into the same group, the present invention facilitates the subsequent classification of the placed advertisements, improves the advertisement classification efficiency, further divides the advertisements in the placed advertisement group based on the classified reference advertisement, ensuring the accuracy and similarity of the advertisement classification. By analyzing the advertisement effect through the advertisement effect index while considering the user overlap of different platforms, the accuracy of the advertisement effect evaluation is ensured. By adjusting the digital placement of the advertisement according to the advertisement order information, the stability of the advertisement effect is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of advertising placement, and specifically relates to a digital placement adjustment method and system based on multi-platform data analysis. Background Art

[0002] With the rapid development of Internet technology, the advertising industry has rapidly shifted from the traditional offline mode to the online mode. Digital advertising has the advantages of wide dissemination range, relatively low cost, high accuracy, etc., and has gradually become the main choice of advertisers. For corresponding advertisements, the advertising placement strategy greatly affects the advertising effect. Therefore, advertising placement has become a crucial step in digital advertising.

[0003] Currently, for digital advertising placement, there are still problems such as being unable to classify the placed advertisements according to the advertisement content, unable to centrally analyze similar advertisements, reducing the efficiency of advertising placement adjustment, often directly evaluating the advertising effect through the click-through rate and conversion rate of the advertisement, without considering the influence of the association between different platforms on advertising placement, often directly replacing the advertisements with poor advertising effects without further analyzing the reasons for abnormal advertising effects, and being unable to timely adjust the advertising placement. Summary of the Invention

[0004] To solve the above technical problems, a digital placement adjustment method and system based on multi-platform data analysis are provided. This technical solution solves the problems raised in the above background art, such as being unable to classify the placed advertisements according to the advertisement content, unable to centrally analyze similar advertisements, reducing the efficiency of advertising placement adjustment, often directly evaluating the advertising effect through the click-through rate and conversion rate of the advertisement, without considering the influence of the association between different platforms on advertising placement, often directly replacing the advertisements with poor advertising effects without further analyzing the reasons for abnormal advertising effects, and being unable to timely adjust the advertising placement.

[0005] To achieve the above purposes, the technical solution adopted by the present invention is as follows:

[0006] A digital placement adjustment method based on multi-platform data analysis, including:

[0007] Obtain platform advertisement placement information, where the platform advertisement placement information includes placed advertisement information, platform information, and advertisement effect data;

[0008] Classify the placed advertisements according to the platform advertisement placement information to obtain placed advertisement classification information;

[0009] Obtain advertisement effect data according to the platform advertisement placement information, where the advertisement effect data includes click-through rate data and conversion rate data corresponding to each platform when each advertisement is placed;

[0010] Based on the advertising effect analysis requirements, different weights are set for the click-through rate and conversion rate;

[0011] According to the classified information of the placed advertisements, obtain the placement platform information corresponding to each advertisement;

[0012] According to the advertising effect data, the weights of the click-through rate and conversion rate, and the placement platform information, obtain the advertising effect index;

[0013] Based on the classified information of the placed advertisements, sort the advertisements of the same category in descending order according to the advertising effect index to obtain the advertisement order information;

[0014] Adjust the digital placement of the advertisements according to the advertisement order information.

[0015] Preferably, the step of classifying the placed advertisements according to the platform advertising placement information to obtain the classified information of the placed advertisements specifically includes:

[0016] According to the platform advertising placement information, obtain the product category information corresponding to the placed advertisements;

[0017] Taking the product category as the basis, group the placed advertisements with the same product category into the same group to obtain multiple groups of placed advertisements;

[0018] Based on the groups of placed advertisements, obtain the target user information corresponding to each advertisement in the same group of advertisements, and the target user information includes target user age range information;

[0019] Take any one advertisement in each group of placed advertisements as the characteristic advertisement of the group of placed advertisements;

[0020] According to the target user information, select the advertisements in the same group of advertisements that have no overlap with the target user age range corresponding to the characteristic advertisement;

[0021] Take the selected advertisements and the characteristic advertisement as the classified reference advertisements of the group of placed advertisements;

[0022] Based on the classified reference advertisements, further divide the advertisements in the group of placed advertisements to obtain the classified information of the placed advertisements.

[0023] Preferably, the step of further dividing the advertisements in the group of placed advertisements based on the classified reference advertisements to obtain the classified information of the placed advertisements specifically includes:

[0024] According to the platform advertising placement information, obtain the platform information, and the platform information includes multiple platform name information and the user data corresponding to each platform;

[0025] According to the platform information, obtain the platform user age ratio information corresponding to each platform, and the platform user age ratio information represents the proportion of users of different ages in the platform;

[0026] Based on the platform information, obtain the daily active user quantity information corresponding to each platform;

[0027] According to the platform information, the user age proportion information corresponding to each platform, and the daily active user quantity information, obtain the user age proportion information, where the user age proportion information represents the proportion information of the daily active user quantities of users of different ages in the total sum of the daily active user quantities of all platforms;

[0028] Taking the classified benchmark advertisement as a reference, obtain the overlapping age range of the target users of any advertisement in the placed advertisement group and the classified benchmark advertisement;

[0029] According to the user age proportion information, the overlapping age range of the target users, and the classified benchmark advertisement, determine whether this advertisement and the classified benchmark advertisement belong to the same classification. If not, use this advertisement as the classified benchmark advertisement. If so, classify this advertisement and the classified benchmark advertisement into the same category of advertisements to obtain the classified information of the placed advertisements;

[0030] Among them, if the sum of the user age proportions of all age users in the overlapping age range of the target users exceeds half of the user age proportion of the target user age range corresponding to the classified benchmark advertisement, that is , then this advertisement and the classified benchmark advertisement belong to the same classification:

[0031] ;

[0032] In the formula, represents the sum of the user age proportions of all age users in the overlapping age range of the target users, represents the sum of the user age proportions of the target user age range corresponding to the classified benchmark advertisement, represents the user age proportion of the i-th user age in the overlapping age range of the target users, represents the user age proportion of the j-th user age in the target user age range corresponding to the classified benchmark advertisement.

[0033] Preferably, the obtaining of the advertisement effect index according to the advertisement effect data, the weight of the click-through rate, the weight of the conversion rate, and the placement platform information specifically includes:

[0034] Obtain the target user age range information corresponding to each advertisement and the user age proportion information corresponding to each platform;

[0035] According to the target user age range information corresponding to each advertisement and the placement platform information, obtain the proportion information of the target user age range in each placement platform;

[0036] Use the placement platform with the largest proportion of the target user age range of each advertisement in the user age proportion corresponding to the platform as the benchmark placement platform of this advertisement;

[0037] Conduct a user overlap survey for the placement platforms corresponding to each advertisement to obtain the user overlap between the benchmark placement platform and the placement platforms of each advertisement;

[0038] Obtain the advertisement effect index based on the advertisement effect data, the weights of the click-through rate and the conversion rate, and the user overlap;

[0039] Among them, the calculation formula of the advertisement effect index is:

[0040] ;

[0041] In the formula, Q is the advertisement effect index, is the advertisement click-through rate of the benchmark placement platform, is the advertisement conversion rate of the benchmark placement platform, is the advertisement click-through rate of the sth placement platform, is the advertisement conversion rate of the sth placement platform, is the user overlap between the sth placement platform and the benchmark placement platform, h is the number of placement platforms corresponding to the advertisement, and are the weights of the click-through rate and the conversion rate respectively.

[0042] Preferably, the digital placement of the advertisement is adjusted according to the advertisement order information, which specifically includes:

[0043] Set the advertisement click-through rate threshold and the advertisement conversion rate threshold based on the advertisement placement cost analysis;

[0044] Based on the advertisement effect data, the advertisement click-through rate threshold, and the advertisement conversion rate threshold, obtain the abnormal advertisement information according to the advertisement order information. The abnormal advertisement information represents the digital advertisement in which the click-through rate corresponding to each platform during the placement of each advertisement in the advertisement order information is lower than the advertisement click-through rate threshold or the conversion rate is lower than the advertisement conversion rate threshold;

[0045] Adjust the abnormal advertisement placement according to the abnormal advertisement information;

[0046] Adjust the advertisement placement budget according to the advertisement order information based on the ratio of the advertisement effect index;

[0047] Among them, the ratio of the advertisement placement budget in the same category of advertisements with the advertisement classification information is the same as the ratio of the advertisement effect index.

[0048] Preferably, obtain the advertisement effect data corresponding to the abnormal advertisement according to the abnormal advertisement information;

[0049] According to the advertisement effect data, take the average value of the advertisement conversion rates of the placement platforms corresponding to the abnormal advertisement as the advertisement conversion rate of the abnormal advertisement;

[0050] Based on the advertisement conversion rate and the advertisement conversion rate threshold, it is determined whether to adjust the advertisement content of the abnormal advertisement. If the advertisement conversion rate is lower than the advertisement conversion rate threshold, the advertisement content of the abnormal advertisement is creatively adjusted;

[0051] If the advertisement conversion rate is higher than the advertisement conversion rate threshold, the placement platform of the abnormal advertisement is adjusted until the click-through rate of the abnormal advertisement is higher than the advertisement click-through rate threshold. The adjustment of the placement platform of the abnormal advertisement includes re-selecting the placement platform of the abnormal advertisement and adjusting the advertisement placement time of each placement platform corresponding to the abnormal advertisement.

[0052] Furthermore, a digital placement adjustment system based on multi-platform data analysis is proposed to implement the adjustment method as described above, including:

[0053] A main control module, which is used to determine whether the advertisement belongs to the same category as the classification reference advertisement according to the user age ratio information, the overlapping age range of target users and the classification reference advertisement, determine whether to adjust the advertisement content of the abnormal advertisement according to the advertisement conversion rate and the advertisement conversion rate threshold, divide the placement advertisements with the same product type into the same group, obtain multiple placement advertisement groups, further divide the advertisements in the placement advertisement groups based on the classification reference advertisement, obtain the placement advertisement classification information, and adjust the digital placement of the advertisement according to the advertisement order information;

[0054] An information acquisition module, which is used to acquire platform advertisement placement information, placement advertisement information, platform information and advertisement effect data, acquire advertisement effect data according to the platform advertisement placement information, the advertisement effect data includes the click-through rate data and conversion rate data corresponding to each platform when each advertisement is placed, acquire the product type information corresponding to the placement advertisement according to the platform advertisement placement information, acquire platform information according to the platform advertisement placement information, and the platform information includes multiple platform name information and user data corresponding to each platform;

[0055] An evaluation module, which is used to acquire the user age ratio information according to the platform information, the user age ratio information corresponding to each platform and the daily active user quantity information, conduct a user overlap survey on the placement platform corresponding to each advertisement, acquire the user overlap between the reference placement platform and the placement platform of each advertisement, and acquire the advertisement effect index according to the advertisement effect data, the weights of the click-through rate and the conversion rate, and the user overlap;

[0056] A display module, which interacts with the main control module and is used to output and display the placement advertisement classification information, advertisement effect data, advertisement effect index and advertisement order information.

[0057] Optionally, the main control module specifically includes:

[0058] A control unit, which is used to divide the advertising placements with the same product type into the same group, obtain multiple advertising placement groups, further divide the advertisements in the advertising placement groups based on the classified reference advertisement, obtain the advertising placement classification information, and adjust the digital placement of the advertisements according to the advertisement order information;

[0059] An information receiving unit, which interacts with the information acquisition module and the evaluation module, and is used to receive data and transmit it to the judgment unit;

[0060] A judgment unit, which is used to judge whether the advertisement belongs to the same classification as the classified reference advertisement according to the user age ratio information, the overlapping age range of the target users, and the classified reference advertisement, and judge whether to adjust the advertisement content of the abnormal advertisement according to the advertisement conversion rate and the advertisement conversion rate threshold.

[0061] Optionally, the information acquisition module specifically includes:

[0062] A first acquisition unit, which is used to acquire the platform advertisement placement information, the advertisement placement information, the platform information, and the advertisement effect data, and acquire the advertisement effect data according to the platform advertisement placement information. The advertisement effect data includes the click-through rate data and the conversion rate data corresponding to each platform when each advertisement is placed;

[0063] A second acquisition unit, which is used to acquire the product type information corresponding to the advertisement placement according to the platform advertisement placement information, and acquire the platform information according to the platform advertisement placement information. The platform information includes multiple platform name information and the user data corresponding to each platform.

[0064] Optionally, the evaluation module specifically includes:

[0065] A first evaluation unit, which is used to acquire the user age ratio information according to the platform information, the user age ratio information corresponding to each platform, and the daily active user quantity information;

[0066] A first evaluation unit, which is used to conduct a user overlap survey on the placement platform corresponding to each advertisement, obtain the user overlap degree between the reference placement platform and the placement platform of each advertisement, and obtain the advertisement effect index according to the advertisement effect data, the weight of the click-through rate, the weight of the conversion rate, and the user overlap degree.

[0067] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0068] The present invention provides a digital placement adjustment method and system based on multi-platform data analysis. By dividing the placed advertisements with the same product type into the same group, it is convenient to classify the placed advertisements subsequently, improving the advertisement classification efficiency. By further dividing the advertisements in the placed advertisement group based on the classified benchmark advertisement, the accuracy and similarity of advertisement classification are ensured. By analyzing the advertisement effect based on the advertisement effect index considering the user overlap on different platforms, the accuracy of advertisement effect evaluation is ensured. By the advertisement order information, the digital placement of advertisements is adjusted to ensure the stability of advertisement effect. Description of the Drawings

[0069] Figure 1 It is a flowchart of the digital placement adjustment method based on multi-platform data analysis proposed by the present invention;

[0070] Figure 2 It is a flowchart for obtaining classified information of placed advertisements in the present invention;

[0071] Figure 3 It is a flowchart for obtaining the advertisement effect index in the present invention;

[0072] Figure 4 It is a flowchart for obtaining abnormal advertisement information in the present invention;

[0073] Figure 5 It is a block diagram of the structure of the digital placement adjustment system based on multi-platform data analysis proposed by the present invention. Detailed Embodiment

[0074] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations.

[0075] Referring to Figure 1 - Figure 4 As shown, the digital placement adjustment method based on multi-platform data analysis in the embodiment of the present invention includes:

[0076] Obtain platform advertisement placement information, where the platform advertisement placement information includes placed advertisement information, platform information, and advertisement effect data;

[0077] Classify the placed advertisements according to the platform advertisement placement information to obtain classified information of the placed advertisements;

[0078] Specifically, classifying the placed advertisements according to the platform advertisement placement information to obtain classified information of the placed advertisements specifically includes:

[0079] Obtain the product type information corresponding to the placed advertisements according to the platform advertisement placement information;

[0080] Based on the product categories, group the advertisements with the same product category into the same group, and obtain multiple advertisement groups;

[0081] Based on the advertisement groups, obtain the target user information corresponding to each advertisement in the same group of advertisements, and the target user information includes target user age group information;

[0082] Take any one advertisement in each advertisement group as the characteristic advertisement of the advertisement group;

[0083] According to the target user information, select the advertisements in the same group of advertisements that have no overlap with the target user age group corresponding to the characteristic advertisement;

[0084] Take the selected advertisements and the characteristic advertisement as the classification benchmark advertisements of the advertisement group;

[0085] Based on the classification benchmark advertisements, further divide the advertisements in the advertisement group to obtain advertisement classification information.

[0086] In this solution, grouping the advertisements according to the product categories and dividing the advertisements with the same product category into the same group can quickly conduct preliminary classification and sorting of a large number of advertisements. This helps advertisers understand the placement situation of advertisements of different product categories from a macro level, facilitating subsequent management and analysis. For example, it is possible to count information such as the placement quantity and budget of advertisements of a certain product category (such as food, electronic products, daily necessities, etc.), providing basic data support for the overall advertisement placement strategy. By obtaining the target user information corresponding to each advertisement in the same group of advertisements, especially the target user age group information, it helps to deeply understand the audience of each advertisement. By taking any one advertisement in each advertisement group as the characteristic advertisement and selecting the advertisements that have no overlap with the target user age group of the characteristic advertisement as the classification benchmark advertisements, it is convenient to classify the remaining advertisements and divide the advertisements more clearly. The classification efficiency of the advertisements is improved, ensuring that key indicators such as the click-through rate and conversion rate of the advertisements are more targeted and representative when analyzing the advertisement effects subsequently.

[0087] Specifically, based on the classification benchmark advertisements, further divide the advertisements in the advertisement group to obtain advertisement classification information, which specifically includes:

[0088] According to the platform advertisement placement information, obtain platform information, and the platform information includes multiple platform name information and user data corresponding to each platform;

[0089] According to the platform information, obtain the platform user age proportion information corresponding to each platform, and the platform user age proportion information represents the proportion of users of different ages in the platform;

[0090] Based on the platform information, obtain the daily active user quantity information corresponding to each platform;

[0091] According to the platform information, the user age ratio information corresponding to each platform, and the daily active user quantity information, obtain the user age ratio information, where the user age ratio information represents the proportion information of the daily active user quantities of users of different ages in the total daily active user quantity of all platforms;

[0092] Taking the classified benchmark advertisement as a reference, obtain the overlapping age range of the target users of any advertisement in the placed advertisement group and the classified benchmark advertisement;

[0093] According to the user age ratio information, the overlapping age range of the target users, and the classified benchmark advertisement, determine whether the advertisement and the classified benchmark advertisement belong to the same category. If not, use the advertisement as the classified benchmark advertisement. If so, classify the advertisement and the classified benchmark advertisement into the same category of advertisements to obtain the classified information of the placed advertisements;

[0094] Among them, if the sum of the user age ratios of all age users in the overlapping age range of the target users exceeds half of the user age ratio of the target user age range corresponding to the classified benchmark advertisement, that is , then the advertisement and the classified benchmark advertisement belong to the same category:

[0095] ;

[0096] In the formula, represents the sum of the user age ratios of all age users in the overlapping age range of the target users, represents the sum of the user age ratios of the target user age range corresponding to the classified benchmark advertisement, represents the user age ratio of the i-th user age in the overlapping age range of the target users, represents the user age ratio of the j-th user age in the target user age range corresponding to the classified benchmark advertisement.

[0097] In this solution, by obtaining the platform information (including platform name, user data, etc.), the platform user age ratio information, and the daily active user quantity information, it is possible to understand the age distribution and activity status of users from a macroscopic perspective of multiple platforms. Calculate the proportion information of the daily active user quantities of users of different ages in the total daily active user quantity of all platforms (i.e., the user age ratio information), which provides a more comprehensive and objective basis for advertisement classification. Based on the classified benchmark advertisement, combining these platform data to determine whether advertisements belong to the same category can classify advertisements more accurately, avoid inaccurate classification caused by simply considering the target user age range, make advertisement classification more scientific and reasonable. Based on the accurate advertisement classification information, advertisement publishers can make more targeted placement adjustments on different platforms according to the characteristics and target user groups of different classified advertisements.

[0098] Obtain advertisement effect data according to the advertisement placement information of the platform. The advertisement effect data includes click-through rate data and conversion rate data corresponding to each platform when each advertisement is placed.

[0099] Set different weights for the click-through rate and the conversion rate based on the advertisement effect analysis requirements.

[0100] Obtain the placement platform information corresponding to each advertisement according to the advertisement classification information.

[0101] Obtain the advertisement effect index according to the advertisement effect data, the weight of the click-through rate, the weight of the conversion rate, and the placement platform information.

[0102] Specifically, obtaining the advertisement effect index according to the advertisement effect data, the weight of the click-through rate, the weight of the conversion rate, and the placement platform information specifically includes:

[0103] Obtain the target user age group information corresponding to each advertisement and the user age ratio information corresponding to each platform.

[0104] Obtain the proportion information of the target user age group on each placement platform according to the target user age group information corresponding to each advertisement and the placement platform information.

[0105] Take the placement platform with the largest proportion of the target user age group of each advertisement in the user age ratio corresponding to the platform as the benchmark placement platform for the advertisement.

[0106] Conduct a user overlap survey on the placement platforms corresponding to each advertisement to obtain the user overlap between the benchmark placement platform and the placement platform of each advertisement.

[0107] Obtain the advertisement effect index according to the advertisement effect data, the weight of the click-through rate, the weight of the conversion rate, and the user overlap.

[0108] Among them, the calculation formula of the advertisement effect index is:

[0109] ;

[0110] In the formula, Q is the advertisement effect index, is the advertisement click-through rate of the benchmark placement platform, is the advertisement conversion rate of the benchmark placement platform, is the advertisement click-through rate of the s-th placement platform, is the advertisement conversion rate of the s-th placement platform, is the user overlap between the s-th placement platform and the benchmark placement platform, h is the number of placement platforms corresponding to the advertisement, and are the weights of the click-through rate and the conversion rate respectively.

[0111] In this solution, by taking the placement platform with the largest proportion in the corresponding user age ratio of each advertising target user age group on the platform as the benchmark placement platform for the advertisement, a core reference platform can be provided for subsequent advertisement effect evaluation and placement adjustment. The benchmark placement platform is usually the platform with the highest degree of fit with the advertising target audience. Based on this for analysis and decision-making, the advertisement placement strategy can be made more focused and effective. Conduct a user overlap survey for the placement platforms corresponding to each advertisement to obtain the user overlap between the benchmark placement platform and the placement platform of each advertisement, and accurately analyze the advertisement effect through the user overlap.

[0112] It can be understood that if the user overlap of multiple placement platforms is high, it means that advertising on these platforms may lead to repeated exposure of the same group of users, which will not only cause waste of advertising resources but also introduce certain errors in the advertisement effect data. By considering the user overlap to adjust the advertisement effect evaluation results, it can remind advertisers to pay attention to this situation of repeated placement when evaluating the advertisement effect and avoid overestimating the advertisement effect due to false high exposure. For example, if the user overlap of two platforms is extremely high, even if the click-through rate and conversion rate of the advertisement on these two platforms are both high, in fact, it may just be multiple interactions of the same group of users, not really expanding the audience range and influence of the advertisement. At this time, adjusting the advertisement effect index through the user overlap can more accurately reflect the true effect of the advertisement.

[0113] Based on the classified information of the placed advertisements, sort the advertisements of the same category in descending order according to the advertisement effect index to obtain the advertisement order information;

[0114] Adjust the digital placement of the advertisements according to the advertisement order information.

[0115] Specifically, adjusting the digital placement of the advertisements according to the advertisement order information specifically includes:

[0116] Based on the analysis of the advertisement placement cost, set the advertisement click-through rate threshold and the conversion rate threshold;

[0117] Based on the advertisement effect data, the advertisement click-through rate threshold, and the conversion rate threshold, and based on the advertisement order information, obtain the abnormal advertisement information, where the abnormal advertisement information represents the digital advertisement in which the click-through rate corresponding to each platform is lower than the advertisement click-through rate threshold or the conversion rate is lower than the advertisement conversion rate threshold when each advertisement in the advertisement order information is placed;

[0118] Adjust the abnormal advertisement placement according to the abnormal advertisement information;

[0119] Adjust the advertisement placement budget according to the advertisement order information based on the ratio of the advertisement effect index;

[0120] Among them, the ratio of the advertising investment budget ratio to the advertising effect index is the same for advertisements of the same category in the classified information of the placed advertisements.

[0121] Specifically, according to the abnormal advertisement information, the adjustment of the abnormal advertisement placement is carried out, specifically including:

[0122] According to the abnormal advertisement information, obtain the advertisement effect data corresponding to the abnormal advertisement;

[0123] According to the advertisement effect data, take the average advertisement conversion rate of the placement platform corresponding to the abnormal advertisement as the advertisement conversion rate of the abnormal advertisement;

[0124] According to the advertisement conversion rate and the advertisement conversion rate threshold, determine whether to adjust the advertisement content of the abnormal advertisement. If the advertisement conversion rate is lower than the advertisement conversion rate threshold, perform creative adjustment on the advertisement content of the abnormal advertisement;

[0125] If the advertisement conversion rate is higher than the advertisement conversion rate threshold, adjust the placement platform of the abnormal advertisement until the click-through rate of the abnormal advertisement is higher than the advertisement click-through rate threshold. The adjustment of the placement platform of the abnormal advertisement includes reselecting the placement platform of the abnormal advertisement and adjusting the advertisement placement time of each placement platform corresponding to the abnormal advertisement.

[0126] In this solution, through the advertisement effect data, the advertisement click-through rate threshold, the conversion rate threshold, and the advertisement order information, the abnormal advertisement information is obtained, and it is possible to accurately find out which advertisements have unqualified effects (the click-through rate is lower than the threshold or the conversion rate is lower than the threshold) on which platforms in the advertisement placement sequence. This enables advertisers to clearly understand the problem, focus the key points of the optimization work on these abnormal advertisements, avoid blind adjustment, improve the optimization efficiency, and adjust the advertising investment budget based on the ratio of the advertising effect index according to the advertisement order information, and the ratio of the advertising investment budget ratio to the advertising effect index is the same for advertisements of the same category. Ensure the scientificity and rationality of the advertising investment budget allocation. The advertising effect index is obtained by comprehensively considering various factors such as the click-through rate, the conversion rate, and the platform user overlap degree, and can comprehensively reflect the actual effect of the advertisement. Linking the budget with the effect index enables advertisers to allocate more resources to advertisements with good effects, improve the use efficiency of the budget, and maximize the advertising placement benefits.

[0127] It should be noted that the click-through rate thresholds and conversion rate thresholds for advertisements of different product types are also very different and need to be set according to specific situations. For example, the click-through rate threshold in the e-commerce industry is generally between 1% and 3%. For some popular categories such as clothing and beauty, where user demand is relatively strong and the click-through rate is relatively high, it may reach 2% - 3%; while for some relatively niche or highly specialized categories such as industrial equipment and special chemicals, the click-through rate is 1%. The conversion rate threshold is approximately between 0.5% and 3%. Well-known e-commerce brands, products with large discount margins or unique selling points have a higher conversion rate of 2% - 3%; new brands or products with weaker competitiveness may have a conversion rate between 0.5% and 1%.

[0128] For the game industry, the click-through rate threshold is usually between 2% and 5%. For popular game types such as role-playing games (RPGs) and action games, due to the wide audience and the often strong attractiveness of game promotion materials, the click-through rate is relatively high, reaching 3% - 5%; while for some relatively niche game types such as strategy simulation management games, the click-through rate is 2%. The conversion rate threshold is between 0.3% and 2%. New works launched by large well-known game manufacturers, or games with unique gameplay and powerful marketing promotions, have a relatively high conversion rate, approaching 1% - 2%; products of small game studios or games with insufficient competitiveness have a conversion rate between 0.3% and 0.5%.

[0129] It can be seen that the thresholds corresponding to different industries and different product types are also different. Therefore, specific settings need to be made according to the product categories.

[0130] Refer to Figure 5 As shown, further, in combination with the above digital placement adjustment method based on multi-platform data analysis, a digital placement adjustment system based on multi-platform data analysis is proposed, including:

[0131] A main control module, which is used to determine whether the advertisement belongs to the same category as the classified reference advertisement according to the user age ratio information, the overlapping age range of the target users and the classified reference advertisement, judge whether to adjust the advertisement content of the abnormal advertisement according to the advertisement conversion rate and the advertisement conversion rate threshold, divide the placement advertisements with the same product type into the same group, obtain multiple placement advertisement groups, further divide the advertisements in the placement advertisement group with the classified reference advertisement as the benchmark to obtain the placement advertisement classification information, and adjust the digital placement of the advertisement according to the advertisement order information;

[0132] An information acquisition module, which is used to acquire platform advertisement placement information, placed advertisement information, platform information, and advertisement effect data. According to the platform advertisement placement information, it acquires the advertisement effect data. The advertisement effect data includes click-through rate data and conversion rate data corresponding to each platform when each advertisement is placed. According to the platform advertisement placement information, it acquires the product category information corresponding to the placed advertisement. According to the platform advertisement placement information, it acquires the platform information. The platform information includes multiple platform name information and user data corresponding to each platform;

[0133] An evaluation module, which is used to acquire the user age ratio information according to the platform information, the user age ratio information corresponding to each platform, and the daily active user quantity information, conduct a user overlap survey on the placement platforms corresponding to each advertisement, acquire the user overlap between the benchmark placement platform and the placement platform of each advertisement, and acquire the advertisement effect index according to the advertisement effect data, the weight of the click-through rate, the weight of the conversion rate, and the user overlap;

[0134] A display module, which interacts with the main control module and is used to output and display the classified information of the placed advertisement, the advertisement effect data, the advertisement effect index, and the advertisement order information.

[0135] The main control module specifically includes:

[0136] A control unit, which is used to divide the placed advertisements with the same product category into the same group, acquire multiple groups of placed advertisements, and further divide the advertisements in the placed advertisement group based on the classified benchmark advertisement to acquire the classified information of the placed advertisement, and adjust the digital placement of the advertisement according to the advertisement order information;

[0137] An information receiving unit, which interacts with the information acquisition module and the evaluation module and is used to receive data and transmit it to the judgment unit;

[0138] A judgment unit, which is used to judge whether the advertisement and the classified benchmark advertisement belong to the same category according to the user age ratio information, the target user overlap age range, and the classified benchmark advertisement, and judge whether to adjust the advertisement content of the abnormal advertisement according to the advertisement conversion rate and the advertisement conversion rate threshold.

[0139] The information acquisition module specifically includes:

[0140] A first acquisition unit, which is used to acquire platform advertisement placement information, placed advertisement information, platform information, and advertisement effect data. According to the platform advertisement placement information, it acquires the advertisement effect data. The advertisement effect data includes click-through rate data and conversion rate data corresponding to each platform when each advertisement is placed;

[0141] A second acquisition unit, which is configured to acquire product category information corresponding to the placed advertisement according to the platform advertisement placement information, and acquire platform information according to the platform advertisement placement information, where the platform information includes multiple platform name information and user data corresponding to each platform.

[0142] An evaluation module, specifically including:

[0143] A first evaluation unit, which is configured to acquire user age ratio information according to the platform information, user age ratio information corresponding to each platform, and daily active user quantity information;

[0144] A first evaluation unit, which is configured to conduct a user overlap survey on the placement platform corresponding to each advertisement, acquire the user overlap between the benchmark placement platform and the placement platform of each advertisement, and acquire an advertisement effect index according to the advertisement effect data, the weights of the click-through rate and the conversion rate, and the user overlap.

[0145] In summary, the advantages of the present invention are as follows: By dividing the placed advertisements with the same product category into the same group and acquiring multiple groups of placed advertisements, it is convenient for subsequent classification of the placed advertisements, improving the advertisement classification efficiency. By further dividing the advertisements in the placed advertisement group based on the classified benchmark advertisement, the accuracy and similarity of the advertisement classification are ensured. By using the advertisement effect data, the weights of the click-through rate and the conversion rate, and the platform information, an advertisement effect index is obtained. By analyzing the advertisement effect based on the advertisement effect index considering the user overlap on different platforms, the accuracy of the advertisement effect evaluation is ensured. By using the advertisement order information, the digital placement of the advertisement is adjusted to ensure the stability of the advertisement effect.

[0146] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A digital delivery adjustment method based on multi-platform data analysis, characterized in that: include: Obtaining platform advertising information, wherein the platform advertising information includes advertising information, platform information, and advertising effect data; According to the platform advertising information, the advertisements are classified and the classified information of the advertisements is obtained; Acquire advertising effect data based on platform advertising delivery information, wherein the advertising effect data includes click-through rate data and conversion rate data corresponding to each platform when each advertisement is delivered; Based on the needs of advertising effect analysis, set different weights for click-through rate and conversion rate; According to the classified information of the advertisements, the information of the delivery platform corresponding to each advertisement is obtained; Obtain the advertising effect index based on advertising effect data, click-through rate weight, conversion rate weight and delivery platform information; Based on the classified information of the advertisements placed, the advertisements of the same category are sorted in descending order according to the advertisement effect index to obtain the advertisement sequence information; Adjust the digital delivery of advertisements based on the advertisement sequence information; The step of classifying the placed advertisements according to the platform advertisement placement information and obtaining the placed advertisement classification information specifically includes: According to the platform advertising information, obtain the product category information corresponding to the advertising; Based on product categories, advertisements with the same product categories are divided into the same group to obtain multiple advertisement groups; Based on the delivered advertisement group, obtaining target user information corresponding to each advertisement in the same advertisement group, wherein the target user information includes target user age group information; Use any advertisement in each delivery ad group as the featured advertisement of the delivery ad group; According to the target user information, select the advertisements in the same group of advertisements that have no overlap in the target user age group corresponding to the feature advertisements; The selected advertisements and characteristic advertisements are used as classified benchmark advertisements of the delivery advertisement group; Based on the classified benchmark advertisements, the advertisements in the delivered advertisement group are further divided to obtain the delivered advertisement classification information; The step of further dividing the advertisements in the delivery advertisement group based on the classified benchmark advertisements to obtain the delivery advertisement classification information specifically includes: According to the platform advertisement delivery information, platform information is obtained, wherein the platform information includes multiple platform name information and user data corresponding to each platform; According to the platform information, the platform user age ratio information corresponding to each platform is obtained, wherein the platform user age ratio information indicates the proportion of users of different ages in the platform; Based on platform information, obtain the number of daily active users corresponding to each platform; According to the platform information, the user age ratio information and the daily active user number information corresponding to each platform, the user age ratio information is obtained, where the user age ratio information indicates the proportion of the daily active user number of users of different ages in the total number of daily active users of all platforms; Taking the classified benchmark advertisement as a benchmark, obtain the age group of target users that overlaps with the classified benchmark advertisement for any advertisement in the delivery advertisement group; According to the user age ratio information, the overlapping age group of the target users and the classified benchmark advertisement, it is determined whether the advertisement and the classified benchmark advertisement belong to the same category. If not, the advertisement is used as the classified benchmark advertisement. If so, the advertisement and the classified benchmark advertisement are classified as the same category of advertisements, and the classification information of the delivered advertisement is obtained; Among them, if the target user overlaps the user age ratio of all age users in the age group and exceeds half of the user age ratio of the target user age group corresponding to the classified benchmark advertisement, that is, , then the ad belongs to the same category as the classified benchmark ad: ； In the formula, Indicates the age ratio of all users in the target user's overlapping age group and, The user age ratio and the age group of the target user corresponding to the classified benchmark advertisement. Indicates the age ratio of the i-th user in the target user's overlapping age group, represents the user age ratio of the jth user age in the target user age group corresponding to the classified benchmark advertisement; The step of obtaining the advertising effect index according to the advertising effect data, the weight of the click rate, the weight of the conversion rate and the delivery platform information specifically includes: Obtain the target user age group information for each advertisement and the user age ratio information for each platform; According to the target user age group information and delivery platform information corresponding to each advertisement, obtain the proportion of the target user age group on each delivery platform; The platform with the largest proportion of the target user age group of each advertisement in the corresponding user age ratio of the platform is used as the benchmark delivery platform for the advertisement; Conduct a user overlap survey on the delivery platform corresponding to each advertisement to obtain the user overlap between the benchmark delivery platform and the delivery platform for each advertisement; Obtain the advertising effect index based on advertising effect data, click-through rate weight, conversion rate weight and user overlap; The calculation formula of the advertising effect index is: ； In the formula, Q is the advertising effect index, The click-through rate of ads on the benchmark delivery platform, The advertising conversion rate of the benchmark delivery platform, is the click-through rate of the advertisement on the sth delivery platform, is the advertising conversion rate of the sth delivery platform, is the user overlap between the sth delivery platform and the benchmark delivery platform, h is the number of delivery platforms corresponding to the advertisement, and are the weights of click-through rate and conversion rate respectively.

2. The digital delivery adjustment method based on multi-platform data analysis according to claim 1 is characterized in that: The adjusting of the digital delivery of advertisements according to the advertisement sequence information specifically includes: Based on the analysis of advertising cost, set the advertising click rate threshold and conversion rate threshold; According to the advertising effect data, the advertising click rate threshold and the conversion rate threshold, based on the advertising sequence information, abnormal advertising information is obtained, wherein the abnormal advertising information indicates a digital advertisement whose click rate corresponding to each platform when each advertisement in the advertising sequence information is delivered is lower than the advertising click rate threshold or whose conversion rate is lower than the advertising conversion rate threshold; Adjust abnormal advertising delivery based on abnormal advertising information; According to the advertisement order information and based on the ratio of the advertisement effect index, the advertisement delivery budget is adjusted; Among them, the advertising budget ratio and advertising effect index ratio in the same type of advertisements placed in the advertising classification information are the same.

3. The digital delivery adjustment method based on multi-platform data analysis according to claim 2 is characterized in that: The adjusting of abnormal advertisement delivery according to abnormal advertisement information specifically includes: According to the abnormal advertisement information, the advertisement effect data corresponding to the abnormal advertisement is obtained; According to the advertising effect data, the average advertising conversion rate of the delivery platform corresponding to the abnormal advertisement is used as the advertising conversion rate of the abnormal advertisement; According to the advertisement conversion rate and the advertisement conversion rate threshold, determine whether to adjust the advertisement content of the abnormal advertisement; if the advertisement conversion rate is lower than the advertisement conversion rate threshold, adjust the creative content of the abnormal advertisement; If the advertisement conversion rate is higher than the advertisement conversion rate threshold, the delivery platform of the abnormal advertisement is adjusted until the click-through rate of the abnormal advertisement is higher than the advertisement click-through rate threshold. The adjustment of the delivery platform of the abnormal advertisement includes reselecting the delivery platform of the abnormal advertisement and adjusting the advertisement delivery time of each delivery platform corresponding to the abnormal advertisement.

4. A digital delivery adjustment system based on multi-platform data analysis, used to implement the adjustment method according to any one of claims 1 to 3, characterized in that: include: A main control module, the main control module is used to determine whether the advertisement and the classified benchmark advertisement belong to the same category according to the user age ratio information, the overlapping age group of the target user and the classified benchmark advertisement, determine whether to adjust the advertisement content of the abnormal advertisement according to the advertisement conversion rate and the advertisement conversion rate threshold, divide the advertisements with the same product category into the same group, obtain multiple advertisement groups, further divide the advertisements in the advertisement groups based on the classified benchmark advertisement, obtain the advertisement classification information, and adjust the digital delivery of the advertisement according to the advertisement sequence information; An information acquisition module, the information acquisition module is used to acquire platform advertising information, advertising information, platform information and advertising effect data. According to the platform advertising information, the advertising effect data is acquired. The advertising effect data includes click-through rate data and conversion rate data corresponding to each platform when each advertisement is delivered. According to the platform advertising information, the product category information corresponding to the delivered advertisement is acquired. According to the platform advertising information, the platform information is acquired. The platform information includes multiple platform name information and user data corresponding to each platform; An evaluation module, the evaluation module is used to obtain user age ratio information based on platform information, user age ratio information corresponding to each platform, and daily active user number information, conduct a user overlap survey on the delivery platform corresponding to each advertisement, obtain the user overlap between the benchmark delivery platform and the delivery platform for each advertisement, and obtain an advertising effect index based on advertising effect data, click-through rate weight, conversion rate weight, and user overlap; The display module interacts with the main control module and is used to output and display advertising classification information, advertising effect data, advertising effect index and advertising sequence information.

5. The digital delivery adjustment system based on multi-platform data analysis according to claim 4 is characterized in that: The main control module specifically includes: A control unit, the control unit is used to divide the advertisements with the same product category into the same group, obtain multiple advertisement groups, further divide the advertisements in the advertisement groups based on the classified benchmark advertisements, obtain classified information of the advertisements, and adjust the digital delivery of the advertisements according to the advertisement sequence information; An information receiving unit, which interacts with the information acquisition module and the evaluation module to receive data and transmit it to the judgment unit; A judgment unit is used to judge whether the advertisement and the classification benchmark advertisement belong to the same category based on user age ratio information, overlapping age groups of target users and classification benchmark advertisements, and to judge whether to adjust the advertisement content of the abnormal advertisement based on the advertisement conversion rate and the advertisement conversion rate threshold.

6. The digital delivery adjustment system based on multi-platform data analysis according to claim 4 is characterized in that: The information acquisition module specifically includes: A first acquisition unit, the first acquisition unit is used to acquire platform advertising delivery information, delivery advertising information, platform information and advertising effect data, and acquire advertising effect data according to the platform advertising delivery information, wherein the advertising effect data includes click-through rate data and conversion rate data corresponding to each platform when each advertisement is delivered; The second acquisition unit is used to obtain product category information corresponding to the advertisement according to the platform advertisement delivery information, and to obtain platform information according to the platform advertisement delivery information, wherein the platform information includes multiple platform name information and user data corresponding to each platform.

7. The digital delivery adjustment system based on multi-platform data analysis according to claim 4 is characterized in that: The evaluation module specifically includes: A first evaluation unit, the first evaluation unit is used to obtain user age ratio information according to platform information, user age ratio information corresponding to each platform, and daily active user number information; The first evaluation unit is used to conduct a user overlap survey on the delivery platform corresponding to each advertisement, obtain the user overlap between the benchmark delivery platform and the delivery platform of each advertisement, and obtain the advertising effect index according to the advertising effect data, the weight of the click rate, the weight of the conversion rate and the user overlap.

Images