An advertisement delivery method and related apparatus

Abstract

The application discloses an advertisement putting method and related device, which first acquires total resources for putting advertisements of a target product and N advertisements, then respectively allocates the same first putting resource to the N advertisements, and carries out advertisement putting in a first putting stage, and when advertisement putting data of the N advertisements reaches a quality estimation condition, it is determined that the first putting stage is completed. According to an advertisement putting target of the target product and advertisement putting results of the N advertisements, putting quality of the N advertisements corresponding to the advertisement putting target is determined. According to the putting quality of the N advertisements, the remaining putting resources are re-allocated to the N advertisements, and based on the allocation result, the N advertisements are respectively put in a second putting stage. The putting quality of the N advertisements can be determined in the first putting stage based on the same first putting resource, and the putting resource is allocated to the N advertisements according to the putting quality, without relying on human experience in the process, so that the allocation efficiency and the putting quality are ensured.

Description

Technical Field

[0001] This application relates to the field of data processing, and in particular to an advertising delivery method and related apparatus. Background Technology

[0002] For a product offering strategy targeting a specific product, in order to better promote that product, one can collaborate with advertising platforms to deliver ads relevant to the product to the platform's users.

[0003] Generally, the advertising provider or product provider will create multiple advertisements for the target product, and these multiple advertisements will be used as promotional materials.

[0004] To improve advertising efficiency, limited advertising resources need to be allocated reasonably to multiple ads for a single product. Currently, the allocation of advertising resources mainly relies on human experience, which makes it difficult to guarantee both allocation efficiency and advertising quality. Summary of the Invention

[0005] To address the aforementioned technical problems, this application provides an advertising delivery method and related apparatus, which can deliver N advertisements using the same delivery resources in the first delivery stage, obtaining information on the delivery quality of each advertisement. Delivery resources are then allocated to the N advertisements based on their delivery quality, without relying on human experience, thus ensuring both allocation efficiency and delivery quality.

[0006] The embodiments of this application disclose the following technical solutions:

[0007] On one hand, embodiments of this application provide an advertising delivery method, the method comprising:

[0008] Obtain the total advertising resources for the target product and N advertisements. The total advertising resources are used to identify the total amount of advertising resources consumed during the advertising process for the target product. The advertising resources are used for advertising. N>1.

[0009] Allocate the same first delivery resource to each of the N advertisements, and perform the first delivery stage of the advertisement delivery for each of the N advertisements based on the first delivery resource;

[0010] In response to the advertising delivery data of N advertisements reaching the quality prediction conditions, the first delivery stage is determined to be completed, and the delivery quality of each of the N advertisements corresponding to the advertising delivery target is determined based on the advertising delivery target of the target product and the advertising delivery results of the N advertisements.

[0011] Based on the delivery quality of the N advertisements, the remaining delivery resources are redistributed to the N advertisements. The remaining delivery resources are the total delivery resources of the advertisements minus the delivery resources that have been consumed. The delivery resources that have been consumed are the delivery resources that were consumed when the quality prediction conditions were met.

[0012] Based on the allocation results of the remaining advertising resources, the second advertising stage is carried out for each of the N advertisements.

[0013] On the other hand, embodiments of this application provide an advertising delivery device, the device comprising: an acquisition module, a first delivery module, a determination module, an allocation module, and a second delivery module;

[0014] The acquisition module is used to acquire the total advertising resources for the target product and N advertisements. The total advertising resources are used to identify the total amount of advertising resources consumed during the advertising process of the target product. The advertising resources are used for advertising, and N>1.

[0015] The first delivery module is used to allocate the same first delivery resource to each of the N advertisements, and to perform a first delivery stage of advertising delivery for each of the N advertisements based on the first delivery resource.

[0016] The determining module is used to determine that the first placement stage is completed in response to the advertising placement data of N advertisements reaching the quality prediction condition, and to determine the placement quality of the N advertisements corresponding to the advertising placement target based on the advertising placement target of the target product and the advertising placement results of the N advertisements.

[0017] The allocation module is used to reallocate the remaining advertising resources to the N advertisements based on the quality of the N advertisements. The remaining advertising resources are the total advertising resources minus the consumed advertising resources. The consumed advertising resources are the advertising resources consumed when the quality prediction conditions are met.

[0018] The second delivery module is used to deliver advertisements in the second delivery stage to each of the N advertisements according to the allocation result of the remaining delivery resources.

[0019] In another aspect, embodiments of this application provide a computer device, which includes a processor and a memory:

[0020] Memory is used to store computer programs;

[0021] The processor is used to execute the methods described above according to a computer program.

[0022] On another front, embodiments of this application provide a computer-readable storage medium for storing a computer program that performs the methods described above.

[0023] In another aspect, embodiments of this application provide a computer program product including a computer program, which, when run on a computer device, causes the computer device to perform the methods described above.

[0024] As can be seen from the above technical solution, when targeting a product with specific advertising goals, the total advertising resources consumed during the campaign are obtained, along with N ads used for the campaign. To objectively predict the quality of these N ads relative to the advertising goals, the campaign is divided into two phases. In the first phase, the same initial advertising resources are allocated to each of the N ads. Since all ads are based on the same resources, the differences in advertising strategies are minimal, eliminating the impact of different resources and strategies on the quality assessment. This is equivalent to putting all N ads on an equal footing, resulting in advertising data that more fairly reflects the quality of each ad. When the advertising data reaches the quality prediction criteria, the first phase ends, and the remaining resources are reallocated based on the determined quality. This allows high-quality ads to consume more resources in the second phase. The entire process does not rely on human experience, and the final advertising results better meet the advertising goals, ensuring both allocation efficiency and campaign quality. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 A schematic diagram illustrating a scenario for an advertising delivery method provided in an embodiment of this application;

[0027] Figure 2 A flowchart illustrating an advertising delivery method provided in this application embodiment;

[0028] Figure 3 This is a schematic diagram illustrating resource allocation during the second deployment phase, as provided in an embodiment of this application.

[0029] Figure 4This is a schematic diagram illustrating the first stage of ad delivery as provided in an embodiment of this application.

[0030] Figure 5 A schematic diagram illustrating an advertisement placement method provided in an embodiment of this application;

[0031] Figure 6 A schematic diagram illustrating the determination of delivery quality provided in an embodiment of this application;

[0032] Figure 7 A schematic diagram illustrating an advertising delivery process provided in an embodiment of this application;

[0033] Figure 8 A schematic diagram of an advertising delivery device provided in an embodiment of this application;

[0034] Figure 9 A structural diagram of a terminal device provided in an embodiment of this application;

[0035] Figure 10 This is a structural diagram of a server provided in an embodiment of this application. Detailed Implementation

[0036] The embodiments of this application will now be described with reference to the accompanying drawings.

[0037] As society develops, product providers, in order to better promote their target products, choose to cooperate with advertising platforms to place advertisements for their target products among the platform's users. For the promotional materials used in these advertisements, the advertising provider or the product provider typically creates multiple ads as promotional materials for the target product. These ads differ in various ways, such as script differences, image composition, and duration, and each ad targets different interested users and levels of interest.

[0038] The product offering strategy for a target product often involves limited advertising resources. To achieve high advertising quality within these limited resources, it is necessary to efficiently and accurately manage the allocation of advertising resources across multiple advertisements.

[0039] In related technologies, the allocation of advertising resources for multiple ads mainly relies on human experience, which makes it difficult to guarantee the allocation efficiency and the quality of ad delivery.

[0040] Therefore, this application provides an advertising delivery method and related apparatus. In a first delivery stage, N advertisements are delivered using the same delivery resources to obtain the delivery quality of each advertisement. Based on the delivery quality, the remaining delivery resources are allocated to the N advertisements in a second delivery stage. This process does not rely on human experience, ensuring both allocation efficiency and delivery quality.

[0041] The advertising delivery method provided in this application can be implemented using computer equipment, which can be a terminal device or a server. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. Terminal devices include, but are not limited to, mobile phones, computers, smart voice interaction devices, smart home appliances, vehicle terminals, and aircraft. Terminal devices and servers can be directly or indirectly connected via wired or wireless communication, and this application does not impose any limitations on this connection.

[0042] Figure 1 This is a schematic diagram of an advertising delivery method provided in an embodiment of this application, wherein the aforementioned computer device is a server.

[0043] When advertising a target product, the server needs to be accessed to obtain the total amount of resources consumed for advertising (i.e., total advertising resources) and N ads. In the first campaign phase, the same initial advertising resources are allocated to each of the N ads, and the ads are then deployed for the first campaign phase based on these initial resources. When the advertising data meets the quality prediction criteria, the first campaign phase is considered complete. At this point, based on the advertising objectives of the target product and the results of the N ads, the campaign quality for each of the N ads can be determined. The consumed advertising resources for the N ads in the first campaign phase can also be determined. By allocating the same initial advertising resources to the N ads, the influence of different advertising resources on the evaluation of campaign quality is eliminated, ensuring that the obtained advertising data fairly and accurately reflects the campaign quality of the N ads.

[0044] Based on the total advertising resources mentioned above and the identified consumed resources, the remaining advertising resources can be determined. In the second campaign phase, considering the performance of the N ads in the first campaign phase, the remaining resources are reallocated to the N ads. Based on the allocation of the remaining resources, the N ads are then deployed. It can be seen that the reallocation of remaining resources for the N ads in the second campaign phase is based on the performance of each ad in the first campaign phase. This process is automated and does not involve human intervention, eliminating subjective human interference and making the allocation of remaining resources more efficient and reasonable. Figure 2 This is a flowchart illustrating an advertising delivery method provided in an embodiment of this application. The method can be executed by a computer device. In this embodiment, the computer device is a server as an example for explanation.

[0045] S201: Obtain the total advertising resources and N ads for the target product.

[0046] The target product refers to the product that the product provider intends to promote through advertising on an advertising platform. The type of target product is not limited here. Once the target product is identified, promotional materials (i.e., advertisements) for advertising placement need to be determined for that target product. Generally, multiple (i.e., N, N>1) advertisements will be created for a single target product. Although the main content of these advertisements is the target product, there will be some differences in their presentation. The purpose of creating N advertisements for the same target product is to cater to the needs of different audiences and avoid advertising homogenization. In this embodiment, it is expected that based on the placement quality of the N determined advertisements, placement resources will be allocated reasonably and efficiently to the N advertisements, improving the efficiency of advertising placement and maximizing the placement quality for the target product with limited advertising resources, thereby achieving conversion of the target product. It should be noted that the creator of the advertisements can be the advertising provider or the product provider; this is not limited here.

[0047] Total advertising resources are used to identify the total amount of advertising resources consumed during the advertising campaign for the target product. These resources are used for advertising. These resources can be understood as a real-world general equivalent or as virtual resources with a certain value, which can be obtained by exchanging real-world general equivalents for advertising resources through a recharge transaction with the advertising platform. In this embodiment, the advertising resources can be understood as a real-world general equivalent provided by the product provider to the advertising platform, or virtual resources obtained from the advertising platform based on a general equivalent. Total advertising resources refer to the total amount of advertising resources consumed within a certain period of advertising the target product. If the advertising for the target product is divided into periods, the total advertising resources can refer to the total amount of advertising resources consumed in a single advertising period, or the total amount of advertising resources consumed in all advertising periods; no limitation is made here.

[0048] For example, suppose the advertising campaign for a target product is divided into three advertising periods: the first period, the second period, and the third period. In this case, the total advertising resources for the target product can refer to the total amount of resources provided by the product provider to the advertising platform for any one of the three periods. Alternatively, it can refer to the total amount of resources provided by the product provider to the advertising platform for the three periods.

[0049] S202: Allocate the same first delivery resource to each of the N advertisements, and perform the first delivery stage of the advertisement delivery for each of the N advertisements based on the first delivery resource.

[0050] In the first phase of the campaign, the goal is to initially determine the quality of each of the N ads, so that the campaign resources can be redistributed based on the determined quality, thereby improving the accuracy of the resource allocation.

[0051] The first placement resource refers to the placement resources allocated to N ads in the first placement phase. For example, this first placement resource can be allocated from the total advertising placement resources and is consumed during this phase. To ensure the placement quality of the N ads in the first placement phase is supported by fair placement resources, the same first placement resource is allocated to all N ads. Since the ads are all placed based on the same resources, the differences in placement strategies are minimal, thus eliminating the impact of different placement resources and strategies on placement quality assessment. Generally, it is desirable to allocate a larger first placement resource to the N ads in the first placement phase to accelerate the process, i.e., to speed up the process of reaching the quality prediction conditions for the ad placement data.

[0052] Advertising refers to the process of displaying N advertisements corresponding to a target product to users. In this application embodiment, the product provider expects to improve the conversion rate of the target product on the user side by advertising N advertisements of the target product to users.

[0053] Once the initial ad placement resources are allocated, N ads will be deployed based on these resources for the first phase of ad delivery. During this process, the initial ad placement resources can be fully or partially consumed. The rate at which the initial ad placement resources are consumed may differ between different ads, and this rate of consumption is related to the ad's own delivery quality.

[0054] The quality of ad placement can be determined by both ad performance and the resources consumed. In the first placement phase, all N ads are allocated the same initial resources. Therefore, the quality of placement for these N ads can be determined by their ad performance (the number of conversions corresponding to the ad's conversion goal). In the second placement phase, the remaining resources are redistributed to the N ads based on their performance determined in the first phase. Therefore, the resources allocated to the N ads may differ. Thus, when measuring ad performance in the second placement phase, both ad performance and resources consumed must be considered.

[0055] There is a correlation between ad placement quality and ad placement objectives. Ad placement quality is used to measure the effectiveness of achieving ad placement objectives. When N ads are placed and the available placement resources are the same (i.e., in the first placement phase), if an ad gets more conversions corresponding to the ad placement objectives during the ad placement process, it proves that the ad placement quality is better; conversely, it proves that the ad placement quality is worse.

[0056] There is a correlation between the rate at which initial ad spend is consumed and the quality of ad placement. The rate at which initial ad spend is consumed can be understood as the cumulative rate of ad spending costs. High-quality ads can achieve the desired conversion with lower costs, while low-quality ads require higher costs to achieve the same conversion. Therefore, to achieve the same conversion rate within the same timeframe, high-quality ads will consume their initial ad spend more slowly, while low-quality ads will consume their initial ad spend more quickly.

[0057] S203: In response to the advertising delivery data of N advertisements reaching the quality prediction condition, determine that the first delivery stage is completed, and determine the delivery quality of each of the N advertisements corresponding to the advertising delivery target based on the advertising delivery target of the target product and the advertising delivery results of the N advertisements.

[0058] Advertising delivery data refers to the data generated during the first phase of advertising delivery for N ads. For example, this data may include, but is not limited to, the following: impressions, downloads, clicks, feature cost, and consumed delivery resources.

[0059] To avoid inaccurate quality predictions based on isolated cases (i.e., a small amount of ad delivery data) during the initial campaign phase, it's necessary to set quality prediction criteria in the first phase that ensure a sufficient volume of data from N ads for quality estimation. This guarantees a certain level of data volume, preventing significant discrepancies between the determined ad quality and the actual situation due to limited data.

[0060] Quality prediction conditions refer to the conditions that enable the prediction of the quality of ad placement. In this embodiment, the quality prediction conditions can refer to the data volume of N ad placement data, which is sufficient to meet the conditions for predicting the quality of ad placement. The "data volume" here can be determined by the amount of placement resources consumed, the number of conversions, and the amount of different types of data in the aforementioned ad placement data (such as the number of impressions), etc., and is not limited here.

[0061] When the ad delivery data for N ads reaches the quality prediction criteria, it means that the amount of ad delivery data obtained has reached the level required to predict the ad delivery quality. At this point, the data foundation corresponding to the initial determination of the delivery quality of each of the N ads in the first delivery phase is available, and the first delivery phase can be considered complete.

[0062] Advertising objectives refer to the conversion results that a product provider hopes to achieve when placing ads targeting a specific product. The quality of these conversion results is determined by the number of conversions generated by the ads for the target product. A conversion refers to the process by which a user performs a specific action related to the target product. For example, a conversion may include clicking a product link, downloading the software, registering on the product page, or transferring user information to the product. Generally, advertising objectives are set by product providers who have a need to promote the target product. For example, the advertising objective might be set as downloading the target product.

[0063] In other words, by defining the advertising objectives, we can identify the types of operations that lead to conversions. The number of conversions achieved through advertising measures the quality of the advertising campaign. Conversion is a crucial dimension for evaluating advertising quality. One or more operation types can be identified as conversions within the advertising objectives, such as: clicking a product link for the target product, downloading the software for the target product, completing registration on the target product, or transferring feature values ​​to the target product.

[0064] When determining the quality of ad placement, since all ads are allocated the same resources in the first placement phase, the relative quality of placement among the N ads can be determined based on the number of conversions achieved by the N ads. Ads with more conversions have higher placement quality, while ads with fewer conversions have lower placement quality.

[0065] The advertising campaign results refer to the data related to the advertising campaign objectives obtained at the end of the first campaign phase for N ads. Specifically, the advertising campaign results include data corresponding to the conversion action types determined by the advertising campaign objectives. The data included in the advertising campaign results can be determined based on the advertising campaign objectives. For example, if the advertising campaign objective determines that the conversion action type is clicking the target product's product link, then the corresponding advertising campaign results will be the data on clicks of the target product's product link obtained by the N ads in the first campaign phase. Depending on the advertising campaign objective, the advertising campaign results may include data corresponding to one or more action types. For example, if the advertising campaign objective determines that the conversion action types are: clicking the target product's product link and downloading the target product's software, then the corresponding advertising campaign results will include data on clicks of the target product's product link and downloads of the target product's software. Here, "data" can refer to the quantity of the above action types for the N ads in the first campaign phase; for example, when the action type is clicking the target product's product link, the corresponding "data" is the number of clicks.

[0066] Once the first phase of campaign deployment is complete, it's necessary to determine the campaign quality (KQ) of each of the N ads, based on the target product's advertising objectives and the campaign results of the N ads. Campaign quality can be determined by the advertising objectives. For example, if the advertising objective is to convert clicks on the target product's link into a conversion, then measuring the KQ of the N ads requires analyzing the campaign results corresponding to those objectives. Campaign quality can be understood as the ad's ability to achieve its conversion objective; this ability can be measured by the number of conversions.

[0067] Suppose there are three ads: Ad A, Ad B, and Ad C. The ad campaign objective is set to convert clicks on the target product's link into a conversion. The ad campaign result refers to the number of clicks on the target product's link. Ad A has 100 clicks, Ad B has 200 clicks, and Ad C has 70 clicks. Based on the ad campaign results, we can determine that Ad B has the best ad quality, followed by Ad A, and then Ad C. This example is a relatively rough way to determine ad quality. Alternatively, we can more accurately reflect the ad quality by determining the proportion of each ad's ad result within the total ad results.

[0068] It should be noted that the advertising target in this embodiment is variable, but the relative stability of the advertising target needs to be ensured within an advertising campaign period. One advertising campaign period includes a first campaign phase and a second campaign phase, with the end of the second campaign phase marking the end of the advertising campaign period. The second campaign phase ends when all advertising resources are exhausted.

[0069] S204: Based on the delivery quality of the N advertisements, the remaining delivery resources are redistributed to the N advertisements.

[0070] The remaining ad placement resources are the total ad placement resources minus the resources already consumed. The consumed ad placement resources are those resources that were consumed when the quality prediction conditions were met. For example, suppose that after the first placement phase, the total ad placement resources consumed by N ads (i.e., consumed resources) is 5000, and the total ad placement resources are 100000, then the remaining ad placement resources are 95000.

[0071] Based on the identified quality of the N ads, the remaining ad placement resources can be redistributed in a targeted manner. For example, more resources can be allocated to ads with higher placement quality, and fewer resources to ads with lower placement quality. Alternatively, the remaining ad placement resources can be redistributed according to the proportion of each ad's placement quality within the total placement quality of all ads.

[0072] When redistributing remaining resources using a percentage-based approach, the following formula can be used:

[0073]

[0074] Among them budget i It is the budget allocated to the i-th ad. group It represents the remaining deployment resources, convCap. i It is the quality of the i-th advertisement, ∑ N convCap i This represents the total quality of N ads. The process of determining ad placement resources based on the above formula can be viewed as normalizing the quality of each ad to determine its proportion, and then using this proportion to redistribute the remaining placement resources.

[0075] Campaign quality refers to the quality of the advertising campaign's effectiveness, which can be quantified by the number of conversions achieved. Campaign quality can be expressed as a quality score for each ad based on its campaign performance. By determining the quality score of each ad within the overall campaign quality score, the percentage of each ad's score in the total campaign quality score can be calculated. The remaining campaign resources are then allocated based on this percentage. For example, suppose ads A, B, and C have quality scores that represent 20%, 50%, and 30% of the total campaign quality score, respectively. With 10,000 remaining campaign resources, the corresponding allocations for ads A, B, and C would be 2,000, 5,000, and 3,000, respectively.

[0076] S205: Based on the allocation results of the remaining advertising resources, the second advertising stage is carried out for each of the N advertisements.

[0077] After the remaining advertising resources are redistributed to N ads, the allocation results of the remaining advertising resources can be obtained. Then, based on the allocated advertising resources, the second stage of advertising is carried out for each of the N ads.

[0078] As can be seen from the above technical solution, when targeting a product with specific advertising goals, the total advertising resources consumed during the campaign are obtained, along with N ads used for the campaign. To objectively predict the quality of these N ads relative to the advertising goals, the campaign is divided into two phases. In the first phase, the same initial advertising resources are allocated to each of the N ads. Since all ads are based on the same resources, the differences in advertising strategies are minimal, eliminating the impact of different resources and strategies on the quality assessment. This is equivalent to putting all N ads on an equal footing, resulting in advertising data that more fairly reflects the quality of each ad. When the advertising data reaches the quality prediction criteria, the first phase ends, and the remaining resources are reallocated based on the determined quality. This allows high-quality ads to consume more resources in the second phase. The entire process does not rely on human experience, and the final advertising results better meet the advertising goals, ensuring both allocation efficiency and campaign quality.

[0079] S202 mentioned above states that "the same first placement resource is allocated to each of the N advertisements," and explains that this first placement resource is a portion of the total advertising placement resources. If the number of first placement resources allocated to the N advertisements in the first placement phase is small, the speed of advertising placement will be affected due to budget constraints (i.e., the first placement resource). Therefore, in order to ensure the completion speed of the first placement phase, the first placement resource needs to be reasonably determined. In one possible implementation, the first placement resource is determined as follows: the total advertising placement resources are determined relative to the average advertising placement resources of the N advertisements, and the first placement resource is determined based on the average advertising placement resources.

[0080] Average ad placement resources can be understood as the value of ad placement resources obtained by evenly distributing the total ad placement resources across N ads. Therefore, determining the average ad placement resources relative to the total ad placement resources across N ads is equivalent to averaging the total ad placement resources across the N ads. For example, if the total ad placement resources are 10,000 and there are 5 ads targeting the target product, then the average ad placement resources would be 2,000.

[0081] In the first campaign phase, the goal is to determine the quality of N ads. When allocating initial campaign resources to these N ads, the same initial resources are allocated to each ad. This approach ensures all N ads have a level playing field for exposure, resulting in more equitable data reflecting the quality of each ad and thus determining an objective quality for the N ads. However, this undifferentiated allocation of initial resources in the first phase is not a optimal strategy. Differentiated resource allocation in the second campaign phase is more efficient. Therefore, throughout the campaign, the aim is to shorten the campaign time in the first phase and extend the campaign time in the second phase to improve efficiency. To shorten the campaign time in the first phase, a higher initial campaign resource allocation (i.e., greater than the average campaign resource allocation mentioned in this embodiment) can be considered for the N ads.

[0082] Once the average ad allocation is determined, the first allocation resource needs to be determined based on the average ad allocation resource. The first allocation resource is greater than the average ad allocation resource. The purpose of this is to provide a higher first allocation resource for N ads in the first allocation phase, so as to avoid the allocation strategy determined based on a lower first allocation resource, which would affect the speed of completing the first allocation phase.

[0083] The first campaign phase can be understood as the exploration phase. In this phase, N ads are allocated the same initial campaign resources to determine their conversion rate. The goal is to initially assess the conversion quality of these ads. Based on the determined campaign quality, the ability of each ad to achieve its campaign objectives and the corresponding conversion type can be clearly defined. The second campaign phase can be understood as the formal campaign phase. In this phase, based on the campaign quality determined in the first phase, the remaining campaign resources are redistributed. A differentiated allocation method allows for higher-quality ads to receive more resources, improving campaign efficiency. Therefore, within an advertising campaign cycle, the campaign time in the first phase should be shortened to allocate more time to the second phase.

[0084] In the first phase of ad delivery, a certain volume of ad data is needed to objectively predict ad quality and avoid significant discrepancies between predictions based on isolated cases and actual results. For example, suppose there are 5 ads with a total ad resource of 10,000. The first ad resource A is determined to be 5,000 (greater than the average ad resource), and the first ad resource B is 2,000 (equal to the average ad resource). Compared to the first ad resource A, the first ad resource B, being less abundant, is consumed more slowly in the first phase. This results in a longer timeframe for the ad data to reach the required volume, thus lengthening the first phase and reducing the time available for the second phase, ultimately impacting ad delivery efficiency.

[0085] Therefore, in this embodiment, the average advertising resource is used as a reference, and the first advertising resource is set to a value greater than the average advertising resource, so that all N advertisements can obtain a relatively high budget. During the advertising campaign in the first campaign phase, budget constraints will not affect the completion time of the first campaign phase, thereby improving the efficiency of advertising campaigns.

[0086] It is understood that, in this embodiment of the application, the average advertising resources can also be used as a reference, and the first advertising resource can be set to the value of the average advertising resources, so that all N advertisements receive the same advertising resources. Under this allocation method, the total first advertising resources of the N advertisements are exactly the same as the total advertising resources, which can control the cost of advertising resources, and the allocation method is simple and easy to implement.

[0087] The method described above for determining the first advertising resource, which is set to be greater than the average advertising resource, ensures that each advertisement has a higher budget in the first stage. This allows for more flexible and controlled resource consumption during the campaign, thus accelerating the completion of the first stage. With a fixed total advertising time, shortening the advertising time in the first stage (determining the quality of N campaigns) and extending the advertising time in the second stage (the stage where differentiated advertising is actually conducted) allows for a longer period of effective advertising in the second stage, improving the efficiency of advertising for the target product.

[0088] S203 mentioned above states that "in response to the advertising delivery data of N advertisements reaching the quality prediction condition, it is determined that the first delivery stage is completed." The aforementioned quality prediction condition refers to the amount of data from the advertising delivery of the N advertisements being sufficient to predict the quality of the advertising delivery. In one possible implementation, "data volume" can refer to the proportion of consumed delivery resources in the total advertising delivery resources. This proportion tends to be a specified proportion, which can be determined by those skilled in the art based on the actual situation and application scenario, and is not limited here. When the quality prediction condition is a specified proportion of consumed delivery resources in the total advertising delivery resources, the corresponding advertising delivery data is the delivery resources consumed by the advertising delivery. In this case, the corresponding method for determining that the advertising delivery data has reached the quality prediction condition is: first, determine the sum of the delivery resources consumed by the advertising delivery of the N advertisements in the first delivery stage; then, when the sum of delivery resources accounts for a specified proportion of the total advertising delivery resources, determine that the advertising delivery data of the N advertisements has reached the quality prediction condition.

[0089] Once the first campaign phase is complete, the sum of the resources consumed by the N ads in that phase can be determined. It's important to note that due to varying ad quality, the amount of resources consumed by different ads during the first phase may differ. Therefore, to determine the sum of the resources consumed by the N ads in the first phase, you can simply add the resources consumed by each ad together. There might be a special case where different ads consume the same amount of resources during the first phase. In this case, to determine the sum of the resources consumed by the N ads in the first phase, you can multiply the resources consumed by one ad by N.

[0090] The determination of the specified percentage is related to the total advertising resources, and factors such as the number of ad placements can also be considered. Based on these factors, the specified percentage should be determined by those skilled in the art according to the actual situation and application scenario. For example, the specified percentage can be set to 20% of the total advertising resources.

[0091] In the embodiments of this application, quality assessment conditions can be identified by the following formula:

[0092]

[0093] Among them, cost i The resources consumed in delivering the i-th ad can be obtained from the ad platform's logs, ∑ N cost iIt is the sum of the advertising resources consumed by N ads, where p is the specified percentage.

[0094] When the sum of the resources consumed by N ads in a campaign reaches a specified percentage of the total advertising resources, it can be determined that the advertising data of the N ads has met the quality prediction criteria. For example, assuming the total advertising resources are 10,000 and the specified percentage is 20%, then when the sum of the resources consumed by the N ads exceeds 2,000, it can be determined that the advertising data of the N ads has met the quality prediction criteria.

[0095] This embodiment uses the cost of advertising (i.e., advertising resources) as a guide to determine the quality prediction conditions. In other words, the purpose of the first stage of advertising is to preliminarily determine the quality of each of the N ads. During this process, excessive advertising resources should not be consumed; more resources need to be allocated to the second stage of advertising. Therefore, defining the quality prediction condition as a specified percentage of consumed advertising resources out of the total advertising resources allows for reasonable control over the amount of resources consumed during the advertising process.

[0096] The method described above for determining whether advertising data meets the quality prediction criteria sets the criteria as a specified percentage of consumed advertising resources out of the total advertising resources. The corresponding advertising data is defined as the resources consumed by each advertisement. In the first campaign phase, when the sum of the resources consumed by N advertisements reaches the specified percentage, the quality prediction criteria are considered met. By defining the specified percentage of consumed advertising resources out of the total advertising resources as the quality assessment criterion, the goal of controlling the resources consumed in the first campaign phase can be achieved. Based on this specified percentage, the maximum value of the resources that N advertisements can consume in the first campaign phase is limited, thus achieving cost control and enabling a reasonable allocation of advertising resources between the first and second campaign phases.

[0097] S203 mentioned above states that "in response to the advertising delivery data of N advertisements reaching the quality prediction condition, it is determined that the first delivery stage is completed." This mentions that the determination of the quality prediction condition is guided by the cost of advertising delivery (i.e., delivery resources). In one possible implementation, the quality assessment condition can be determined based on conversion results. In this case, "data volume" can refer to the specified conversion number corresponding to the advertising delivery target. When the quality prediction condition is the specified conversion number corresponding to the advertising delivery target, the corresponding advertising delivery data is the conversion number achieved by the N advertisements in the first delivery stage. In this case, the corresponding method for determining that the advertising delivery data has reached the quality prediction condition is as follows: First, count the conversion numbers corresponding to the N advertisements in the first delivery stage. Then, when the conversion numbers corresponding to the N advertisements all reach the specified conversion number, it is determined that the advertising delivery data of the N advertisements has reached the quality prediction condition.

[0098] The aforementioned conversion refers to the type of operation determined by the advertising campaign objective. Depending on the advertising campaign objective, conversions can include various types, such as clicking a product link of the target product, downloading the software of the target product, completing a registration operation on the target product, and transferring feature values ​​to the target product. The specified conversion quantity refers to the number of conversions that, in the first campaign phase, correspond to N ads that meet the requirements for evaluating campaign quality, from one or more of the conversion types determined by the advertising campaign objective. This specified conversion quantity generally cannot be set to a small value, because if the specified conversion quantity is too small, the last ad to reach the specified conversion quantity will have less campaign data, leading to an inaccurate prediction of the campaign quality. It should be noted that the specified conversion quantity can also be determined by those skilled in the art based on the actual situation and application scenario, and is not limited here.

[0099] During the first phase of campaign deployment, it is necessary to statistically analyze the conversion counts for each of the N ads. When each ad in the first phase has multiple conversion types, the statistical results will include the conversion counts for each conversion type. When it is determined that the conversion counts for each of the N ads in the statistical results have all reached the specified conversion counts, the ad delivery data for the N ads is considered to have met the quality prediction conditions.

[0100] It should be noted that the specified conversion quantity mentioned above refers to the minimum conversion quantity corresponding to N ads in the first campaign phase, which is one or more of the aforementioned conversion types. When the specified conversion quantity refers to the minimum conversion quantity corresponding to N ads in the first campaign phase for a single conversion type, when calculating the conversion quantity corresponding to each of the N ads in the first campaign phase, only the conversion quantity generated by that conversion type for the N ads needs to be counted. This simplifies the statistical process, reduces the amount of data, and improves the efficiency of the statistical process.

[0101] When the specified conversion count refers to the minimum conversion count corresponding to N ads in the first campaign phase across multiple conversion types, when calculating the conversion count for each of the N ads in the first campaign phase, you can simplify the calculation process by only counting the conversion count generated by the N ads for each of the multiple conversion types. This reduces the amount of data and improves the efficiency of the calculation. Alternatively, you can choose to count the conversion count generated by the N ads for all the conversion types involved, and then filter out the conversion count generated by the N ads corresponding to the specified multiple conversion types.

[0102] For example, suppose the specified conversion count refers to the number of clicks on the target product's link, and this count is set to 100. There are three ads: Ad A, Ad B, and Ad C. Statistics show that Ad A has 100 conversions from clicks on the target product's link, Ad B has 90 conversions, and Ad C has 400 conversions. Since the conversion count for Ad B does not reach the quality conversion count, it can be determined that the ad performance data for all N ads has not yet met the quality prediction criteria.

[0103] This embodiment determines the quality assessment conditions based on conversion results. In this way, it can be ensured that each advertisement has reached a basic number of conversions, which means that the amount of advertising placed for each advertisement is relatively sufficient. The corresponding conversion data of each advertisement in the first stage of placement has certain data support and can ensure high reliability.

[0104] The foregoing introduction mentioned two quality prediction conditions: (1) the specified percentage of the total advertising resources consumed; and (2) the specified number of conversions corresponding to the advertising target. It should be noted that in this application embodiment, the determination of the completion of the first advertising phase using these two quality prediction conditions can be done individually or simultaneously, and is not limited here. For situations where the total advertising resources are small and the conversion rate is slow, meeting any one of the above quality prediction conditions is sufficient to be considered as completion of the first advertising phase.

[0105] The method described above for determining whether advertising data meets the quality prediction criteria involves setting the quality prediction criteria as a specified number of conversions corresponding to the advertising campaign objective, and the corresponding advertising data as the number of conversions achieved by N ads in the first campaign phase. The conversion numbers for each of the N ads in the first campaign phase are statistically analyzed. When the conversion numbers for each ad reach the specified number, the advertising data is considered to have met the quality prediction criteria. By using the specified number of conversions corresponding to the advertising campaign objective as the quality evaluation criterion, it is ensured that the conversion numbers for each ad in the first campaign phase reach a certain level. This means that the advertising campaign's volume is relatively sufficient, and the conversion results shown in the statistical results have substantial data support and high reliability, facilitating accurate evaluation of the quality of each ad campaign.

[0106] The aforementioned S205 states that "based on the allocation results of the remaining placement resources, the N advertisements are placed in the second placement phase." In fact, during the second placement phase, to further improve the quality of ad placement, the placement resources for the N advertisements can be dynamically allocated to adapt to changes in the placement quality of each advertisement. Therefore, in one possible implementation, the method for allocating placement resources in the second placement phase is as follows:

[0107] A1: At time j in the advertising delivery process of the second delivery stage, based on the advertising delivery target and the advertising delivery results of N ads up to time j, determine the delivery quality of the N ads corresponding to the advertising delivery target at time j.

[0108] The aforementioned j-th time point refers to any time point within the second campaign phase. At any time during the advertising campaign in the second campaign phase, the campaign quality of each of the N ads at that time point can be determined based on the advertising campaign objective and the campaign results of the N ads up to that time point. In other words, a j-th time point can be randomly determined within the second campaign phase, and the campaign results of the N ads prior to that time point can be determined based on that j-th time point. Then, combining the advertising campaign objective and the campaign results of the N ads, the campaign quality of the N ads at time point j can be determined.

[0109] For example, suppose the second campaign phase runs from 12:00 to 16:00, with the j-th moment being 15:15. The advertising objective is to reach 100 clicks on the target product's product link, and there are two ads, Ad A and Ad B. At this point, based on the advertising objective and the campaign results of Ad A and Ad B up to 15:15 (i.e., from 12:00 to 15:15), we need to consider that the campaign results for Ad A and Ad B may include more than just the number of clicks on the target product's product link. When determining the campaign quality of Ad A and Ad B, we can only consider the campaign results for the campaign type corresponding to the advertising objective. Then, based on the campaign results, we can determine the campaign quality of Ad A and Ad B at 15:15. Assuming that Ad A has 200 clicks on the target product's product link and Ad B has 70 clicks, we can conclude that Ad A has a better campaign quality than Ad B.

[0110] As mentioned earlier, when determining the performance quality of N ads in the second delivery phase, it is necessary to combine the ad delivery results and the resources consumed (i.e., the resources consumed up to time j). That is, the ad delivery results and the resources consumed are used as two factors to determine the performance quality, and these two factors are combined to determine the ad delivery quality. For example, assuming the ad delivery objective indicates that the conversion operation type is click, and the ad delivery results (i.e., the number of clicks) for both ad A and ad B up to time j are 100, we can then obtain the resources consumed by ad A and ad B up to time j. Combining the resource consumption, we can further determine the performance quality of ad A and ad B, where ad A consumed 1000 resources and ad B consumed 3000 resources. Since ad A consumed fewer resources within the same time period and achieved the same ad delivery result as ad B, it can be determined that the performance quality of ad A is better than that of ad B.

[0111] Continuing with the example above, if we find that both Ad A and Ad B have consumed 1000 resources up to time j, we can then obtain the ad delivery results for Ad A and Ad B up to time j. Combining these results, we can further determine the delivery quality of Ad A and Ad B. Ad A's delivery result (i.e., number of clicks) is 100, and Ad B's delivery result (i.e., number of clicks) is 300. Since Ad A consumed the same number of resources as Ad B within the same timeframe but achieved fewer clicks, we can conclude that Ad A's delivery quality is inferior to Ad B's.

[0112] When the results and resources consumed by different ads vary, the quality of the ad campaign can be determined by the number of ad results obtained per unit of resources, or the amount of resources consumed per unit of ad results. Generally speaking, the more ad results obtained per unit of resources, the better the ad campaign quality; conversely, the more resources consumed per unit of ad results, the worse the ad campaign quality.

[0113] A2: Based on the delivery quality of the N advertisements corresponding to the advertising delivery target at time j, the remaining delivery resources of the total advertising delivery as of time j are reallocated to the N advertisements.

[0114] In section A1, the delivery quality of each of the N ads at time j is determined according to its corresponding advertising target. Based on time j, the delivery resources consumed up to that time can also be determined. The remaining delivery resources can be calculated by combining the total delivery resources with the consumed resources. Then, combined with the determined delivery quality of the N ads, the remaining delivery resources are reallocated among the N ads.

[0115] When redistributing the remaining advertising resources among the N ads based on their performance, more resources can be allocated to ads with higher performance and less resources to ads with lower performance. Alternatively, the remaining resources can be redistributed according to the proportion of each ad's performance in the total performance of all ads. The specific redistribution method is not limited here.

[0116] It should be noted that, in this embodiment, the advertising delivery results of the N advertisements determined at different times are not the same. When redistributing, you can choose to redistribute the remaining delivery resources according to the delivery quality of the N advertisements at each time in the second delivery stage, or you can choose to redistribute the remaining delivery resources according to the delivery quality of the N advertisements at a certain time interval.

[0117] Figure 3 This application provides a schematic diagram illustrating resource allocation during the second deployment phase, as shown in the embodiment. Figure 3As shown, in the second delivery phase, based on the delivery quality of the N ads determined in the first delivery phase, the remaining delivery resources 1 are allocated to each ad. Then, based on the allocation results, the N ads (i.e., ad 1, ad 2... ad N in the diagram) are delivered in the second delivery phase. At time j in the second delivery phase, based on the ad delivery results up to time j and the ad delivery objectives, the delivery quality of the N ads (ad 1 delivery quality, ad 2 delivery quality... ad N delivery quality in the diagram) is determined. Based on the delivery quality, the remaining delivery resources (i.e., remaining delivery resources 2 in the diagram) up to time j are reallocated to the N ads.

[0118] The method described above for reallocating remaining advertising resources in the second campaign phase allows for the reallocation of remaining resources across the N ads at any point during the second campaign phase, ensuring the quality of ad delivery for each ad's target audience. This enables dynamic adjustment of resources for the N ads during the second campaign phase, facilitating real-time monitoring of changes in ad delivery quality and timely adjustments to resource allocation. This improves the accuracy of remaining resource allocation during the second campaign phase, ultimately enhancing the quality and effectiveness of ad delivery.

[0119] The aforementioned method dynamically adjusts the allocation of remaining advertising resources for N ads in the second campaign phase. In one possible implementation, the second campaign phase may span a long period. Due to this extended duration, the cumulative advertising results for each of the N ads will be large. Even if some ads experience a surge or drop in advertising results, it won't be significantly reflected in the overall large-scale advertising results. Therefore, this embodiment proposes a method of periodically conducting the first and second campaign phases to avoid the problem of inaccurately reflecting differences in advertising results caused by excessively long campaign durations. In one possible implementation, assume that the total advertising resources are used to identify the total amount of advertising resources consumed within one advertising campaign cycle for the target product, and N ads are used for advertising campaigns over M campaign cycles, where M>1. The corresponding implementation method for the first campaign phase is as follows: at the beginning of the k-th advertising campaign cycle out of the M campaign cycles, the same first advertising resources are allocated to each of the N ads, and the first campaign phase is then performed on each of the N ads based on these first advertising resources.

[0120] As mentioned above, in this embodiment of the application, an advertising campaign cycle includes a first campaign stage and a second campaign stage, with the end of the second campaign stage marking the end of the advertising campaign cycle. The second campaign stage ends when all advertising resources are exhausted.

[0121] As mentioned in the preceding introduction to total advertising resources, total advertising resources refer to the total amount of resources consumed during a certain period of advertising for a target product. If advertising for a target product is divided into periods, then total advertising resources can refer to the total amount of resources consumed during a single advertising period. N ads can be used for advertising across multiple advertising periods. An advertising period refers to the time span of an advertising campaign; for example, in this embodiment, the advertising period can be one day. The total advertising resources can be different or the same in different advertising periods.

[0122] At the start of the k-th advertising campaign period, the same initial advertising resources need to be allocated to N ads. The reasons for and process of allocating these initial resources have been detailed in the preceding discussion and will not be repeated here. Once the initial advertising resources are allocated, the first advertising campaign phase is then performed on each of the N ads based on these resources.

[0123] For example, suppose N ads are used for 5 ad campaigns, with each campaign lasting one day. At the start of the second ad campaign (day 2), the same initial ad placement resource needs to be allocated to all N ads. Then, based on this initial resource, the first ad placement phase is performed on each of the N ads. This first placement phase is repeated for each ad campaign.

[0124] The process for the second campaign phase after the first phase is similar to the previous description. Generally, the quality of N ads is determined, and then the remaining resources are redistributed among the N ads based on their quality. Finally, based on the allocation of the remaining resources, the second campaign phase is conducted for each of the N ads. More detailed information can be found in the preceding discussion and will not be repeated here.

[0125] The method in this embodiment is equivalent to resetting the allocation of resources for N ads in each advertising campaign period, and also resetting the advertising results for the N ads. This avoids the aforementioned situation where, due to the large base number of accumulated advertising results for the N ads over time, sudden changes in advertising quality cannot be significantly reflected. It can redetermine the advertising quality of each ad within each advertising campaign period, display the changes in advertising quality, and adjust the allocation strategy of advertising resources accordingly.

[0126] Figure 4This application provides an embodiment of an advertisement placement process for the first placement stage, as shown in the diagram. Figure 4 As shown, N ads (i.e., ad 1, ad 2, ..., ad N in the diagram) are used for ad placement over M ad placement cycles (i.e., ad placement cycle 1, ad placement cycle 2, ..., ad placement cycle M in the diagram). At the beginning of each ad placement cycle, the same first placement resource is allocated to all N ads, and then the ad placement operation for the first placement phase is performed on the N ads based on the first placement resource.

[0127] Figure 5 This is a schematic diagram of an advertising placement method provided in an embodiment of this application, such as... Figure 5 As shown, there are two advertising campaign periods: Campaign Period 1 and Campaign Period 2. The target product includes two advertisements, Ad 1 and Ad 2. At the beginning of each advertising campaign period, both advertisements are allocated the same initial ad placement resources. In Campaign Period 1, the placement quality 1 for Ad 1 is determined to be 100, and the placement quality 2 for Ad 2 is determined to be 300. The remaining ad placement resources allocated based on placement quality are: Ad 1 is allocated 200 ad placement resources, and Ad 2 is allocated 600 ad placement resources. In Campaign Period 2, the placement quality 1 for Ad 1 is determined to be 300, and the placement quality 2 for Ad 2 is determined to be 100. The remaining ad placement resources allocated based on placement quality are: Ad 1 is allocated 600 ad placement resources, and Ad 2 is allocated 200 ad placement resources. This demonstrates that the ad placement quality changes in different advertising campaign periods, and the corresponding allocation of ad placement resources is adjusted accordingly.

[0128] It should be noted that the allocation of advertising resources in the embodiments of this application is all done automatically, without the need for manual intervention or adjustment.

[0129] The method described above for implementing the first delivery phase involves allocating the same first delivery resources to N ads at the beginning of each advertising delivery cycle. Then, based on these first delivery resources, the N ads are delivered for the first delivery phase. This allows for a re-evaluation of the delivery quality of the N ads in each advertising delivery cycle. Therefore, based on the changes in the delivery quality of each ad in different advertising delivery cycles, the allocation method of delivery resources for the N ads within a cycle can be redefined. This enables timely and significant reflection of changes in ad delivery quality, improving the accuracy of delivery resource allocation.

[0130] S203 mentioned above states that "based on the advertising campaign objective of the target product and the advertising results of the N ads, the campaign quality of each of the N ads corresponding to the advertising campaign objective is determined." In determining the campaign quality, it can be based on the actual conversion count. For the i-th ad among the N ads, in one possible implementation, the method for determining the campaign quality is as follows: First, based on the advertising campaign objective of the target product and the advertising results of the i-th ad, determine the actual conversion count achieved by the i-th ad corresponding to the advertising campaign objective. Then, determine the campaign quality of the i-th ad corresponding to the advertising campaign objective based on the actual conversion count.

[0131] Advertising performance refers to the conversion results achieved by users based on N ads, corresponding to the advertising objectives. Depending on the advertising objectives, conversions can include various action types, such as clicks, downloads, registrations, and feature transfers. The advertising performance can be measured by the number of conversions, such as the number of clicks, downloads, registrations, and feature transfers.

[0132] The aforementioned advertising objective refers to the desired outcome of advertising targeting a specific product. This outcome is determined by the type of user action the product provider expects to take in response to the advertisement for the target product. Advertising objectives may include requirements for one or more action types. For example, product provider A may expect advertising to result in users clicking on the target product's link and downloading the target product; the corresponding conversion refers to clicking on the target product's link and downloading the target product.

[0133] Actual conversions refer to the number of action types that actually occur for the target product during the first phase of advertising campaigns for N ads, corresponding to the advertising campaign's objective. The actual conversion count can also include one or more action types, determined by the specific advertising objectives. For example, assuming the advertising objective uses clicks as the conversion action type, and the first phase of advertising campaigns uses ads A and B, where ad A achieves conversions including clicks and downloads, while ad B achieves conversions only including clicks, then the actual conversion count for ads A and B includes the number of clicks.

[0134] For the i-th ad out of N ads, based on the target product's advertising objective and the ad's performance, the actual number of conversions achieved by the i-th ad corresponding to that objective can be determined. Here, the actual conversion number refers to the number of actual conversions as the action type corresponding to the advertising objective. For example, suppose the advertising objective is to achieve clicks, and the ad data for the i-th ad includes both clicks and downloads. Then, to determine the actual conversion number corresponding to the advertising objective, only the click count needs to be extracted to determine the ad quality.

[0135] Once the actual conversion rate achieved for the ad campaign objective corresponding to the i-th ad is determined, the campaign quality for the i-th ad campaign objective is determined based on the actual conversion rate. Campaign quality can be determined in quantifiable ways, such as determining the percentage of the actual conversion rate of the i-th ad in the total actual conversion rate of the N ads, and then determining the campaign quality based on this percentage. Alternatively, the campaign quality can be determined by measuring the relative size of the actual conversion rate of the i-th ad to the actual conversion rates of other ads.

[0136] Figure 6 This is a schematic diagram illustrating the determination of delivery quality provided in an embodiment of this application, such as... Figure 6 As shown, the advertising objective is to achieve conversion type 1. In the advertising data of the i-th ad, the number of conversion type 1 is 200, and the number of conversion type 2 is 50. It can be determined that the actual number of conversions achieved by the advertising objective corresponding to the i-th ad is 200. Based on this actual number of conversions, the advertising quality of the advertising objective corresponding to the i-th ad can be determined.

[0137] The method described above for determining ad delivery quality determines the actual number of conversions achieved by the ad delivery objective corresponding to the i-th ad, and then determines the ad delivery quality of the ad delivery objective corresponding to the i-th ad based on the actual number of conversions. This method enables targeted determination of ad delivery quality based on ad delivery objectives, providing a clear standard for quality determination. Furthermore, since the determination of ad delivery quality is based on the actual number of conversions corresponding to the ad delivery objective, it does not require consideration of data from other conversion types not involved in the ad delivery objective, thus simplifying the quality determination process and improving its efficiency.

[0138] The aforementioned methods for determining campaign quality rely on actual conversions. However, in some scenarios, it's possible that no conversions are generated in the initial stages of advertising campaigns due to the difficulty in obtaining the required conversions. To ensure accurate campaign quality determination, it's advisable to combine the expected and actual conversion counts. Therefore, in one possible implementation, the expected conversion count must first be determined. This is achieved by calculating the expected conversion count for the i-th ad based on the cumulative resources consumed in running the i-th ad and the expected conversion resources for the target product per campaign.

[0139] Expected conversion resource per campaign (EPS) is used to identify the expected value of the total advertising resources a provider expects to consume to complete a single conversion through advertising. The provider of total advertising resources can refer to the product provider or other third parties. EPS can be determined based on the average amount of advertising resources consumed during the advertising campaign period.

[0140] Cumulative resources refer to the total resources consumed by the i-th ad from the start of ad campaigning until the expected conversion rate is determined. For example, assuming ad campaigning begins at 12:00 and the expected conversion rate is determined at 12:12, cumulative resources refer to the total resources consumed from 12:00 to 12:12. Based on the cumulative resources consumed by the i-th ad campaign and the expected conversion rate for a single campaign of the target product, the expected conversion rate for the i-th ad can be determined.

[0141] In the embodiments of this application, the expected conversion number of the i-th advertisement is expectedConv. i It can be determined using the following formula:

[0142]

[0143] Among them, cost i It is the cumulative spending resources consumed by the i-th ad placement, and averageTargetCpa is the expected conversion resources per ad placement.

[0144] When determining the quality of ad placement, the first stage of placement has been completed. At this point, all N ads have actual conversion numbers corresponding to their ad placement goals, but these actual conversion numbers may be relatively low. For slow conversion rates, determining the quality of N ads solely based on actual conversion numbers, while providing an objective estimate, may still differ from the actual quality due to the limited number of conversions. Therefore, this embodiment introduces the expected conversion number. The expected conversion number is combined with the actual conversion number to determine the quality of ad placement for the i-th ad. This increased conversion number ensures a closer alignment with the actual quality of ad placement for the i-th ad.

[0145] Since the expected conversion count is determined based on the expected conversion resources for a single campaign, and these resources are provided by the provider of the total advertising resources, the expected value of the resources consumed to complete a conversion through advertising is objectively determined based on market conditions and is inherently objective. Therefore, the corresponding expected conversion count is also objective. Combining the expected conversion count with the actual conversion count to determine the quality of the i-th ad campaign ensures objectivity while improving the accuracy of the quality determination.

[0146] The above describes the method for determining the expected conversion number of the i-th advertisement. When determining the delivery quality of the i-th advertisement, the delivery quality can be determined based on both the actual conversion number and the expected conversion number. This can be achieved by summing the actual and expected conversion numbers to obtain the total conversion number of the i-th advertisement, and then using this total conversion number to determine the delivery quality of the i-th advertisement.

[0147] By introducing the expected conversion count, the campaign quality can be evaluated even if the i-th ad only generates a small number of actual conversions at the end of the first campaign phase, taking into account the expected conversion count. Furthermore, since the expected conversion count is introduced for all N ads, it avoids situations where the expected conversion count is only introduced for ads that generate a small number of actual conversions, thus preventing inaccurate or unfair evaluations of conversion quality.

[0148] The method for determining ad delivery quality described above introduces the expected conversion count. This expected conversion count is combined with the actual conversion count to jointly determine the ad delivery quality for the i-th ad's target audience. By introducing the expected conversion count, when the actual conversion count for the i-th ad is less than ideal after the first delivery phase, the discrepancy between the expected and actual conversion counts can be reduced while maintaining objectivity, thus improving the accuracy of ad delivery quality prediction.

[0149] As mentioned earlier, when the first campaign phase ends and the actual conversion rate of the i-th ad is not ideal due to slow conversion speed, resulting in a deviation from the determined actual campaign quality, the expected conversion number is introduced. This can reduce the discrepancy with the actual campaign quality while ensuring data objectivity. In fact, the actual conversion number gradually increases over time, and it better reflects the actual campaign quality. Therefore, when the actual conversion number increases, it is necessary to highlight its proportion. This can be achieved by reducing the expected conversion number, which can be done by adjusting the expected weight of the expected conversion number. Thus, in one possible implementation, as mentioned above, "the campaign quality of the i-th ad is determined based on both the actual and expected conversion numbers." The method for determining this campaign quality is: determine the campaign quality of the i-th ad based on the actual conversion number, the expected weight, and the expected conversion number.

[0150] Expected weight refers to the weight value assigned to the expected number of conversions. This expected weight decreases as the ad campaign duration increases. This is because the actual number of conversions gradually increases with the ad campaign duration, and the quality assessment for the i-th ad tends to be based on the actual number of conversions. Therefore, when the actual number of conversions increases, the value of the expected number of conversions needs to be reduced, thus decreasing the weight corresponding to the expected number of conversions and increasing the share of the actual number of conversions in the overall campaign quality assessment. As mentioned earlier, campaign quality can be a quality score assigned to each ad based on its campaign performance, and this quality score is correlated with both the actual number of conversions and the expected number of conversions.

[0151] Therefore, when determining the ad delivery quality for the i-th ad's target audience by combining actual and expected conversions, an expected conversion weight can be introduced to adjust the proportion of expected conversions in the ad delivery quality during the ad delivery process. As the actual conversions increase, the expected conversion weight is reduced, meaning the proportion of expected conversions in the ad delivery quality is decreased, thus highlighting the actual conversions in the ad delivery quality assessment. Determining the ad delivery quality for the i-th ad's target audience based on a higher proportion of actual conversions more closely reflects the actual conversion rate of the i-th ad, making the determined ad delivery quality more realistic and reliable.

[0152] In this application embodiment, the delivery quality of the i-th advertisement is convCap i It can be determined using the following formula:

[0153] convCap i =conv i +smooth t*expectConv i

[0154] Where, conv i This represents the actual number of conversions generated after the i-th ad campaign. The meaning of this actual conversion number varies depending on the ad campaign objective. For example, if the ad campaign objective identifies the corresponding action type as "activation," then conv... i This represents the number of activations generated by the i-th ad. expectConv i It represents the expected conversion quantity, smoothness. t This is the expected weight, a weighted smoothing coefficient that decays over time. When a calendar day is considered as an advertising campaign period, the formula for calculating the expected weight is as follows:

[0155]

[0156] Where t is the number of minutes that have elapsed in a calendar day (i.e., 24 hours) (0≤t≤1440). λ is a set hyperparameter, generally set to λ=2.5, but can also be adjusted according to the actual campaign situation. When the advertising campaign period is one day, the above formula for calculating the expected weight can be used; if the advertising campaign period is not one day, the formula can be adjusted accordingly.

[0157] For example, suppose the advertising campaign period is 1 day, and the quality of ad placement for the i-th ad is determined at 9:00 AM, with an expected conversion rate of 100. If the actual conversion rate for the i-th ad at 9:00 AM is 50, then the determined quality of ad placement for the i-th ad is 50 + (1 - 540 / 1440). 2.5 *100≈81. At 16:00, the actual conversion rate of the i-th ad was 200, and the expected conversion rate was also 200. Therefore, the quality of the i-th ad at this point is determined to be 200 + (1 - 960 / 1440). 2.5 *200≈213.

[0158] The method described above for determining ad delivery quality, based on the premise of determining the ad delivery quality for the i-th ad's target audience by combining actual and expected conversion numbers, introduces an expected conversion weight to adjust the expected conversion number. This allows the expected conversion number to decrease as the actual conversion number increases, thus highlighting the role of actual conversion numbers in the ad delivery quality prediction process. This more closely reflects the actual conversion situation of the i-th ad, making the determined ad delivery quality more realistic and reliable.

[0159] In the second campaign phase, based on the campaign quality of the N ads determined in the first phase, the remaining campaign resources were redistributed to the N ads. It should be noted that, in one possible scenario, the reason why some ads with suboptimal campaign quality in the first phase might be that a suitable campaign strategy was not set for them in that phase, thus failing to fully reflect their true campaign quality. Therefore, in one possible implementation, the campaign strategies for the N ads could be adjusted in the second campaign phase based on the campaign quality of each ad's corresponding campaign objective.

[0160] An advertising placement strategy refers to the plan developed during the process of placing advertisements. An advertising placement strategy may include, but is not limited to, the following: advertising time period, advertising target audience, and advertising format.

[0161] In the second campaign phase, the remaining campaign resources were redistributed to the N ads based on their performance. However, considering that some ads were identified as having poor performance, possibly due to inappropriate ad delivery strategies implemented in the first campaign phase, adjustments to the ad delivery strategies for those ads identified as having poor performance in the first phase (i.e., those allocated fewer campaign resources) should be considered in the second campaign phase.

[0162] For example, suppose there is an ad A whose content is about fitness, and the target audience A is fitness enthusiasts, while the target audience B is wellness enthusiasts. In the first campaign phase, the ad delivery strategy assigned to ad A corresponds to target audience B, resulting in a lower quality of delivery for ad A in that phase. In the second campaign phase, the ad delivery strategy for ad A can be adjusted, changing the target audience to target audience A, and ad delivery can be based on the new strategy to re-evaluate the quality of delivery for ad A.

[0163] By adjusting the advertising strategy in the second campaign phase, it's possible to determine whether the poor performance of ads identified as having low campaign quality in the first phase was due to an inappropriate advertising strategy. If the ad's performance improves after adjusting the strategy, this confirms the initial assessment and prevents the misallocation of resources for high-quality ads due to improper campaign strategies.

[0164] Meanwhile, in the second campaign phase, the advertising strategy can be adjusted for ads that performed well in the first phase. For example, suppose there is ad B, which was scheduled to run from 5:00 to 6:00 in the first phase. In the second phase, considering that users are more likely to view ads during the morning rush hour, the ad period for ad B can be adjusted to 8:00 to 9:00 to improve the effectiveness of ad B and, to some extent, improve the quality of ad delivery.

[0165] In this embodiment, to avoid the potential arbitrariness of relying solely on the advertising quality determined in the first campaign phase to allocate resources for each advertisement, the second campaign phase considers adjusting the advertising strategy. Ads with poor performance in the first campaign phase are reassessed after the strategy is changed. This helps prevent high-quality ads from being overlooked in the second campaign phase due to inadequate strategies in the earlier (first) phase. Simultaneously, for ads with good performance in the first campaign phase, appropriate adjustments to the advertising strategy can be considered to better stimulate their performance and improve campaign efficiency.

[0166] It should be noted that, in the embodiments of this application, the advertising strategy in the second delivery stage can be adjusted once or multiple times, and the specific adjustment time is not limited, that is, it can be dynamically adjusted in the second delivery stage.

[0167] The aforementioned method for adjusting advertising strategies involves adjusting the strategies for each of the N ads in the second campaign phase based on the quality of their respective campaign objectives. This approach allows for better optimization of the advertising results for each ad through strategy adjustments. Furthermore, the dynamic adjustment of advertising strategies enables more accurate assessment of ad quality, facilitating the precise allocation of resources and improving overall advertising efficiency.

[0168] S203 mentioned above states, "Based on the advertising placement objective of the target product and the advertising placement results of N advertisements, determine the placement quality of each of the N advertisements corresponding to the advertising placement objective." This shows that the advertising placement objective is a dimension for measuring the quality of advertising placement. Within the advertising placement objective, the operation type that constitutes a conversion can be determined, and different operation types correspond to different interaction depths. In one possible implementation, the advertising placement objective includes at least one of click, download, product acquisition, and feature value transfer. Click, download, product acquisition, and feature value transfer are different operation types, and one or more operation types can be determined as conversions within the advertising placement objective.

[0169] In this embodiment, the advertising target can be predetermined or dynamically determined during the advertising process. The predetermined target can be the product provider of the target product, while the dynamically determined target can be the product provider of the target product or the advertising platform conducting the advertising. The aforementioned clicks, downloads, product acquisitions, and feature value transfers reflect the different levels of user interaction with the target product based on the advertisement. For example, the interaction depth of downloads is greater than that of clicks, and the interaction depth of feature value transfers is greater than that of downloads. The determination of the advertising target can be based on the required interaction depth. Furthermore, if the operation type used for conversion within the advertising target is not determined during the advertising process, multiple different operation types can be used simultaneously as the conversions corresponding to the advertising target to jointly measure the quality of N advertisements.

[0170] In one possible implementation, the conversion operation type in the advertising campaign objective can be predetermined. If, when determining the quality of N ads based on the advertising campaign objective, it is found that the quality of the N ads determined based on the objective is not ideal, then the advertising campaign objective can be changed, and the quality of the N ads can be determined based on the newly determined advertising campaign objective.

[0171] It's important to note that this assumes multiple advertising campaigns. Within a single campaign cycle, the advertising objectives need to remain relatively stable; that is, the objectives generally do not change within that cycle. However, the advertising objectives can differ between different campaign cycles.

[0172] By defining advertising objectives involving different depths of interaction, the selection of conversion action types can be based on the depth of interaction when determining the quality of N ads. Furthermore, during ad campaigns, advertising objectives can be adjusted according to needs and the quality of the determined N ads. Different ad campaign periods may correspond to different determined advertising objectives, allowing for adaptive adjustments during ad campaigns. Moreover, different advertising objectives involve different action types corresponding to different evaluation dimensions of campaign quality.

[0173] Figure 7 This is a schematic diagram of an advertising delivery process provided in an embodiment of this application, such as... Figure 7 As shown in (a), the three ads targeting the product can be considered as an ad group, ad group = {ad A, ad B, ad C}, and the total ad spending can be represented as budget. The total ad spending shown in the figure is 600. For example... Figure 7 As shown in (b), when allocating initial campaign resources for each ad in the first campaign phase, an automatic budget allocation method can be used. In the diagram, ads A, B, and C are allocated the same initial campaign resource of 600. It should be noted that transitioning from the first campaign phase to the second campaign phase requires meeting a condition: the ad's campaign data must meet a quality prediction condition. This quality prediction condition can be that the consumed campaign resources reach a specified percentage of the total ad campaign resources. In this case, the ad campaign data represents the campaign resources consumed in the first campaign phase. The quality prediction condition can then be expressed as... ∑ i cost i This represents the total cost of the entire ad group. As shown in the diagram, at the end of the first campaign phase, 500 ad resources remain. Once the first campaign phase is complete, the ad delivery quality for each ad targeting the respective ad objective is determined based on the target product's advertising goals and the ad delivery results. For example... Figure 7 As shown in (c), the determined ad delivery quality is represented by percentages, where ad A has a delivery quality of 10%, ad B has a delivery quality of 33%, and ad C has a delivery quality of 57%. Based on the delivery quality of each ad, the remaining delivery resources are redistributed among ad A, ad B, and ad C, with ad A allocated 50 resources, ad B allocated 165 resources, and ad C allocated 285 resources.

[0174] In the foregoing Figure 1-7 Based on the corresponding embodiments, Figure 8This is a schematic diagram of an advertising delivery device provided in an embodiment of the present application. The advertising delivery device 800 includes: an acquisition module 801, a first delivery module 802, a determination module 803, an allocation module 804, and a second delivery module 805.

[0175] The acquisition module 801 is used to acquire the total advertising resources for the target product and N advertisements. The total advertising resources are used to identify the total amount of advertising resources consumed during the advertising process of the target product. The advertising resources are used for advertising, and N>1.

[0176] The first delivery module 802 is used to allocate the same first delivery resource to each of the N advertisements, and to perform a first delivery stage of advertisement delivery for each of the N advertisements based on the first delivery resource;

[0177] The determining module 803 is used to determine that the first placement stage is completed in response to the advertising placement data of N advertisements reaching the quality prediction condition, and to determine the placement quality of the N advertisements corresponding to the advertising placement target based on the advertising placement target of the target product and the advertising placement results of the N advertisements.

[0178] The allocation module 804 is used to reallocate the remaining placement resources to the N advertisements according to the placement quality of the N advertisements. The remaining placement resources are the total placement resources of the advertisements minus the placement resources that have been consumed. The placement resources that have been consumed are the placement resources that were consumed when the quality prediction conditions were met.

[0179] The second delivery module 805 is used to deliver advertisements in the second delivery stage to the N advertisements according to the allocation result of the remaining delivery resources.

[0180] In one possible implementation, the apparatus further includes a resource determination module, which determines a first deployment resource, the first deployment resource being determined in the following manner:

[0181] Determine the total advertising resources relative to the average advertising resources of the N ads;

[0182] The first advertising resource is determined based on the average advertising resource, and the first advertising resource is greater than the average advertising resource.

[0183] In one possible implementation, the quality prediction condition includes a specified percentage of the consumed advertising resources in the total advertising resources, the advertising data includes the advertising resources consumed by the advertising campaign, and the device further includes a condition determination module, which is used to:

[0184] Determine the sum of the advertising resources consumed by the N advertisements in the first advertising stage;

[0185] When the sum of the resources allocated to advertising reaches the specified percentage as a percentage of the total resources allocated to advertising, the advertising data of N of the advertisements are determined to meet the quality prediction conditions.

[0186] In one possible implementation, the quality prediction condition includes a specified number of conversions corresponding to the advertising target, the advertising data includes the conversion numbers achieved by N ads in the first advertising phase, and the device further includes a condition determination module, which is used to:

[0187] Calculate the conversion count for each of the N ads in the first campaign phase;

[0188] When the conversion counts corresponding to each of the N ads reach the specified conversion count, it is determined that the ad delivery data of the N ads has met the quality prediction condition.

[0189] In one possible implementation, the apparatus further includes a reallocation module, the reallocation module being used for:

[0190] At the j-th moment in the advertising delivery process of the second delivery phase, based on the advertising delivery target and the advertising delivery results of N ads up to the j-th moment, the delivery quality of the N ads at the j-th moment corresponding to the advertising delivery target is determined respectively, where the j-th moment is any moment in the second delivery phase;

[0191] Based on the delivery quality of the N advertisements corresponding to the advertising delivery target at time j, the remaining delivery resources of the total advertising delivery as of time j are reallocated to the N advertisements.

[0192] In one possible implementation, the total advertising resources are used to identify the total amount of advertising resources consumed within one advertising cycle of the target product, and N advertisements are used for advertising for M advertising cycles, where M>1. The device further includes a third advertising module, which is used for:

[0193] At the beginning of the k-th advertising campaign period among the M advertising campaign periods, the operation of allocating the same first advertising resource to each of the N advertisements and performing the first advertising campaign stage for each of the N advertisements based on the first advertising resource is executed.

[0194] In one possible implementation, for the i-th advertisement among the N advertisements, the determining module 803 is used to:

[0195] Based on the advertising campaign objectives of the target product and the advertising results of the i-th advertisement, determine the actual number of conversions achieved by the i-th advertisement corresponding to the advertising campaign objectives;

[0196] The delivery quality of the i-th advertisement corresponding to the advertising delivery target is determined based on the actual conversion number.

[0197] In one possible implementation, the apparatus further includes a quantity determination module, the quantity determination module being used to:

[0198] Based on the cumulative resources consumed in placing the i-th advertisement and the expected conversion resources for a single instance of the target product, the expected conversion quantity of the i-th advertisement is determined. The expected conversion resources for a single instance are used to identify the expected value of the resources consumed by the provider of the total advertising resources to complete one conversion through advertising.

[0199] The determining module 803 is used for:

[0200] The delivery quality of the i-th advertisement corresponding to the advertising delivery target is determined based on the actual conversion number and the expected conversion number.

[0201] In one possible implementation, the determining module 803 is used to:

[0202] The delivery quality of the i-th advertisement corresponding to the advertising delivery target is determined based on the actual conversion quantity, the expected weight, and the expected conversion quantity, wherein the expected weight decreases as the advertising delivery time increases.

[0203] In one possible implementation, the device further includes an adjustment module, the adjustment module being used for:

[0204] In the second delivery phase, the advertising delivery strategy for the N ads is adjusted based on the delivery quality of the ads corresponding to the advertising delivery targets.

[0205] In one possible implementation, the advertising delivery objective includes at least one of clicks, downloads, product acquisition, and feature transfer.

[0206] The aforementioned advertising delivery device, when delivering ads to a target product, acquires the total advertising resources consumed during the delivery process, and N ads used for delivery. To objectively predict the delivery quality of these N ads relative to the advertising objective, the advertising delivery is divided into two stages. In the first stage, the same initial delivery resources are allocated to each of the N ads. Since all ads are delivered using the same resources, the differences in delivery strategies are minimal, eliminating the impact of different delivery resources and strategies on the delivery quality assessment. This is equivalent to putting all N ads on an equal footing for exposure, resulting in advertising delivery data that more fairly reflects the delivery quality of each ad. When the advertising delivery data reaches the quality prediction criteria, the first stage ends, and the remaining delivery resources are reallocated based on the determined delivery quality. This allows high-quality ads to consume more delivery resources in the second stage. The entire process does not rely on human experience, and the final advertising delivery result better meets the advertising objective, ensuring both allocation efficiency and delivery quality.

[0207] This application also provides a computer device, including a terminal device or a server, in which the aforementioned advertising delivery device can be configured. The computer device will now be described in conjunction with the accompanying drawings.

[0208] If the computer device is a terminal device, please refer to Figure 9 As shown, this application provides a terminal device, taking a mobile phone as an example:

[0209] Figure 9 The diagram shown is a block diagram of a portion of the structure of a mobile phone provided in an embodiment of this application. (Reference) Figure 9 The mobile phone includes components such as a radio frequency (RF) circuit 1410, a memory 1420, an input unit 1430, a display unit 1440, a sensor 1450, an audio circuit 1460, a Wi-Fi module 1470, a processor 1480, and a power supply 1490. Those skilled in the art will understand that... Figure 9 The mobile phone structure shown does not constitute a limitation on the mobile phone and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0210] The following is combined with Figure 9 A detailed introduction to each component of a mobile phone:

[0211] The RF circuit 1410 can be used to receive and transmit signals during information transmission or calls. In particular, it receives downlink information from the base station and processes it with the processor 1480; in addition, it transmits uplink data to the base station.

[0212] The memory 1420 can be used to store software programs and modules. The processor 1480 executes various mobile phone functions and data processing by running the software programs and modules stored in the memory 1420. The memory 1420 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory 1420 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0213] The input unit 1430 can be used to receive input numeric or character information, and to generate key signal inputs related to user settings and function control of the mobile phone. Specifically, the input unit 1430 may include a touch panel 1431 and other input devices 1432.

[0214] The display unit 1440 can be used to display information input by the user or information provided to the user, as well as various menus of the mobile phone. The display unit 1440 may include a display panel 1441.

[0215] The mobile phone may also include at least one sensor 1450, such as a light sensor, a motion sensor, and other sensors.

[0216] Audio circuitry 1460, speaker 1461, and microphone 1462 provide an audio interface between the user and the mobile phone.

[0217] WiFi is a short-range wireless transmission technology. Through the WiFi module 1470, mobile phones can help users send and receive emails, browse web pages, and access streaming media, providing users with wireless broadband internet access.

[0218] The processor 1480 is the control center of the mobile phone. It connects to various parts of the mobile phone through various interfaces and lines. It performs various functions of the mobile phone and processes data by running or executing software programs and / or modules stored in the memory 1420 and calling data stored in the memory 1420.

[0219] The mobile phone also includes a power supply 1490 (such as a battery) that powers the various components.

[0220] In this embodiment, the processor 1480 included in the terminal device is also used to execute the steps in the methods of the various embodiments of this application.

[0221] If the computer device is a server, this application embodiment also provides a server; please refer to [link to relevant documentation]. Figure 10 As shown, Figure 10 This is a structural diagram of a server 1500 provided in an embodiment of this application. The server 1500 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 1522 (e.g., one or more processors) and a memory 1532, and one or more storage media 1530 (e.g., one or more mass storage devices) for storing application programs 1542 or data 1544. The memory 1532 and storage media 1530 can be temporary or persistent storage. The program stored in the storage media 1530 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the server. Furthermore, the CPU 1522 may be configured to communicate with the storage media 1530 and execute the series of instruction operations in the storage media 1530 on the server 1500.

[0222] Server 1500 may also include one or more power supplies 1526, one or more wired or wireless network interfaces 1550, one or more input / output interfaces 1558, and / or one or more operating systems 1541, such as Windows Server. TM Mac OS X TM Unix TM Linux TM FreeBSD TM etc.

[0223] The steps performed by the server in the above embodiments can be based on Figure 10 The server structure shown.

[0224] In addition, this application embodiment also provides a storage medium for storing a computer program for executing the method provided in the above embodiment.

[0225] This application also provides a computer program product including a computer program, which, when run on a computer device, causes the computer device to perform the method provided in the above embodiments.

[0226] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium can be at least one of the following media: read-only memory (ROM), RAM, magnetic disk or optical disk, and other media that can store computer programs.

[0227] In this application embodiment, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0228] It should be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for the device and system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments. The device and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0229] The above description is merely one specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Moreover, based on the implementation methods provided in the above aspects, this application can be further combined to provide more implementation methods. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An advertising placement method, characterized in that, The method includes: Obtain the total advertising resources for the target product and N advertisements. The total advertising resources are used to identify the total amount of advertising resources consumed during the advertising process for the target product. The advertising resources are used for advertising. N>1. Allocate the same first delivery resource to each of the N advertisements, and perform the first delivery stage of the advertisement delivery for each of the N advertisements based on the first delivery resource; In response to the advertising delivery data of N advertisements reaching the quality prediction conditions, the first delivery stage is determined to be completed, and the delivery quality of each of the N advertisements corresponding to the advertising delivery target is determined based on the advertising delivery target of the target product and the advertising delivery results of the N advertisements. Based on the delivery quality of the N advertisements, the remaining delivery resources are redistributed to the N advertisements. The remaining delivery resources are the total delivery resources of the advertisements minus the delivery resources that have been consumed. The delivery resources that have been consumed are the delivery resources that were consumed when the quality prediction conditions were met. Based on the allocation results of the remaining advertising resources, the second advertising stage is carried out for each of the N advertisements.

2. The method according to claim 1, characterized in that, The first resource allocation is determined in the following way: Determine the total advertising resources relative to the average advertising resources of the N ads; The first advertising resource is determined based on the average advertising resource, and the first advertising resource is greater than the average advertising resource.

3. The method according to claim 1, characterized in that, The quality prediction criteria include the specified percentage of the consumed advertising resources in the total advertising resources, the advertising data includes the advertising resources consumed by the advertising campaign, and the method further includes: Determine the sum of the advertising resources consumed by the N advertisements in the first advertising stage; When the sum of the resources allocated to advertising reaches the specified percentage as a percentage of the total advertising resources allocated to advertising, the advertising data of N of the advertisements are determined to meet the quality prediction conditions.

4. The method according to claim 1, characterized in that, The quality prediction conditions include a specified number of conversions corresponding to the advertising delivery target, and the advertising delivery data includes the number of conversions achieved by N ads in the first delivery phase. The method further includes: Calculate the conversion count for each of the N ads during the first campaign phase; When the conversion counts corresponding to each of the N ads reach the specified conversion count, it is determined that the ad delivery data of the N ads has met the quality prediction condition.

5. The method according to claim 1, characterized in that, The method further includes: At the j-th moment in the advertising delivery process of the second delivery phase, based on the advertising delivery target and the advertising delivery results of N ads up to the j-th moment, the delivery quality of the N ads at the j-th moment corresponding to the advertising delivery target is determined respectively, where the j-th moment is any moment in the second delivery phase; Based on the delivery quality of the N advertisements corresponding to the advertising delivery target at time j, the remaining delivery resources of the total advertising delivery as of time j are reallocated to the N advertisements.

6. The method according to claim 1, characterized in that, The total advertising resources are used to identify the total amount of advertising resources consumed within one advertising cycle of the target product. N advertisements are used for advertising across M advertising cycles, where M > 1. The method further includes: At the beginning of the k-th advertising campaign period among the M advertising campaign periods, the operation of allocating the same first advertising resource to each of the N advertisements and performing the first advertising campaign stage for each of the N advertisements based on the first advertising resource is executed.

7. The method according to any one of claims 1 to 6, characterized in that, For the i-th advertisement among the N advertisements, determining the delivery quality of each of the N advertisements corresponding to the advertising delivery target based on the advertising delivery target of the target product and the advertising delivery results of the N advertisements includes: Based on the advertising campaign objectives of the target product and the advertising results of the i-th advertisement, determine the actual number of conversions achieved by the i-th advertisement corresponding to the advertising campaign objectives; The delivery quality of the i-th advertisement corresponding to the advertising delivery target is determined based on the actual conversion number.

8. The method according to claim 7, characterized in that, The method further includes: Based on the cumulative resources consumed in placing the i-th advertisement and the expected conversion resources for a single instance of the target product, the expected conversion quantity of the i-th advertisement is determined. The expected conversion resources for a single instance are used to identify the expected value of the resources consumed by the provider of the total advertising resources to complete one conversion through advertising. Determining the delivery quality of the i-th advertisement corresponding to the advertising delivery target based on the actual conversion quantity includes: The delivery quality of the i-th advertisement corresponding to the advertising delivery target is determined based on the actual conversion number and the expected conversion number.

9. The method according to claim 8, characterized in that, Determining the delivery quality of the i-th advertisement corresponding to the advertising delivery target based on the actual conversion number and the expected conversion number includes: The delivery quality of the i-th advertisement corresponding to the advertising delivery target is determined based on the actual conversion quantity, the expected weight, and the expected conversion quantity, wherein the expected weight decreases as the advertising delivery time increases.

10. The method according to any one of claims 1 to 6, characterized in that, The method further includes: In the second delivery phase, the advertising delivery strategy for the N ads is adjusted based on the delivery quality of the ads corresponding to the advertising delivery targets.

11. The method according to any one of claims 1 to 6, characterized in that, The advertising delivery objectives include at least one of clicks, downloads, product acquisition, and feature transfer.

12. An advertising delivery device, characterized in that, The device includes: an acquisition module, a first delivery module, a determination module, an allocation module, and a second delivery module; The acquisition module is used to acquire the total advertising resources for the target product and N advertisements. The total advertising resources are used to identify the total amount of advertising resources consumed during the advertising process of the target product. The advertising resources are used for advertising, and N>1. The first delivery module is used to allocate the same first delivery resource to each of the N advertisements, and to perform a first delivery stage of advertising delivery for each of the N advertisements based on the first delivery resource. The determining module is used to determine that the first placement stage is completed in response to the advertising placement data of N advertisements reaching the quality prediction condition, and to determine the placement quality of the N advertisements corresponding to the advertising placement target based on the advertising placement target of the target product and the advertising placement results of the N advertisements. The allocation module is used to reallocate the remaining advertising resources to the N advertisements based on the quality of the N advertisements. The remaining advertising resources are the total advertising resources minus the consumed advertising resources. The consumed advertising resources are the advertising resources consumed when the quality prediction conditions are met. The second delivery module is used to deliver advertisements in the second delivery stage to each of the N advertisements according to the allocation result of the remaining delivery resources.

13. A computer device, characterized in that, The computer device includes a processor and memory: The memory is used to store computer programs; The processor is configured to perform the method according to any one of claims 1-11 according to the computer program.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program that, when executed by a computer device, performs the method described in any one of claims 1-11.

15. A computer program product comprising a computer program, which, when run on a computer device, causes the computer device to perform the method of any one of claims 1-11.

Images