An advertisement bidding adjustment method, device, equipment and storage medium

Abstract

Embodiments of the application disclose an advertisement bid adjustment method and device, equipment and a storage medium, wherein the method comprises: determining a first deep target achievement rate and a second deep target achievement rate corresponding to a target advertisement; the first deep target achievement rate is a proportion of objects achieving a deep target among objects achieving a shallow target in a reference historical period, and the second deep target achievement rate is a proportion of objects achieving a deep target among objects achieving a shallow target in a reference time interval of the reference historical period; determining a shallow price adjustment coefficient corresponding to the target advertisement according to a conversion price corresponding to the target advertisement; determining a deep price adjustment coefficient corresponding to the target advertisement according to the first deep target achievement rate and the second deep target achievement rate; and adjusting the conversion price according to the shallow price adjustment coefficient and the deep price adjustment coefficient to obtain a target conversion price corresponding to the target advertisement. The method can accurately adjust the conversion price corresponding to the advertisement.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to an advertising bid adjustment method, apparatus, device and storage medium. Background Technology

[0002] Cost Per Action (CPA) advertising is a common form of advertising in the advertising industry. This type of advertising is typically priced based on the actual performance of the ad. When advertisers run CPA ads on an ad platform, they need to set campaign objectives (such as orders, activations, or payments) and conversion prices (the amount expected to be spent to achieve a campaign objective). When the ad platform determines which ads to expose, it estimates the click-through rate (CTR) and conversion rate of the ads stored in its ad library. Then, based on the conversion price and the corresponding CTR and conversion rate, it calculates the effective cost per mille (ECPM). The higher the ECPM, the greater the likelihood that the ad will be exposed.

[0003] When advertising platforms calculate ECPM, they need to use models to predict the click-through rate (CTR) and conversion rate (CTR) of ads. However, the accuracy of these predicted CTRs and CTRs is often not high. Consequently, calculating ECPM based on inaccurate CTRs and CTRs, and then using this ECPM to decide whether to expose ads, often leads to a mismatch between the actual cost of advertising and the actual conversion results. This can result in losses for advertisers or advertising platforms.

[0004] To mitigate losses caused by inaccurate click-through rate and conversion rate predictions, advertising platforms typically need to adjust the conversion prices set by advertisers to ensure that the actual advertising results match the actual costs. This process of adjusting conversion prices is known in the industry as ad bid adjustment. However, accurately adjusting conversion prices to achieve this desired effect remains a pressing issue for the industry. Summary of the Invention

[0005] This application provides an advertising bid adjustment method, apparatus, device, and storage medium that can accurately adjust the conversion price corresponding to the advertisement, thereby making the actual cost of the advertisement closer to the advertiser's expected cost.

[0006] In view of the above, the first aspect of this application provides an advertising bid adjustment method, the method comprising:

[0007] Determine the first deep-level goal achievement rate and the second deep-level goal achievement rate corresponding to the target advertisement; the first deep-level goal achievement rate is the proportion of objects that achieve the deep-level goal among those that achieve the shallow-level goal within the reference historical period, and the second deep-level goal achievement rate is the proportion of objects that achieve the deep-level goal among those that achieve the shallow-level goal within the reference time interval of the reference historical period, and achieving the shallow-level goal is a prerequisite for achieving the deep-level goal;

[0008] Based on the conversion price corresponding to the target advertisement, determine the shallow price adjustment coefficient corresponding to the target advertisement; based on the first deep target achievement rate and the second deep target achievement rate, determine the deep price adjustment coefficient corresponding to the target advertisement.

[0009] The conversion price is adjusted based on the shallow adjustment coefficient and the deep adjustment coefficient to obtain the target conversion price corresponding to the target advertisement.

[0010] A second aspect of this application provides an advertising bid adjustment device, the device comprising:

[0011] The deep goal achievement rate determination module is used to determine the first deep goal achievement rate and the second deep goal achievement rate corresponding to the target advertisement; the first deep goal achievement rate is the proportion of objects that achieve the deep goal among those that achieve the shallow goal within the reference historical period, and the second deep goal achievement rate is the proportion of objects that achieve the deep goal among those that achieve the shallow goal within the reference time interval of the reference historical period, and achieving the shallow goal is a prerequisite for achieving the deep goal;

[0012] The shallow price adjustment coefficient determination module is used to determine the shallow price adjustment coefficient corresponding to the target advertisement based on the conversion price corresponding to the target advertisement;

[0013] The deep-level price adjustment coefficient determination module is used to determine the deep-level price adjustment coefficient corresponding to the target advertisement based on the first deep-level target achievement rate and the second deep-level target achievement rate.

[0014] The conversion price adjustment module is used to adjust the conversion price according to the shallow price adjustment coefficient and the deep price adjustment coefficient to obtain the target conversion price corresponding to the target advertisement.

[0015] A third aspect of this application provides an apparatus comprising a processor and a memory:

[0016] The memory is used to store computer programs;

[0017] The processor is configured to perform the steps of the advertising bid adjustment method as described in the first aspect above, according to the computer program.

[0018] A fourth aspect of this application provides a computer-readable storage medium for storing a computer program for performing the steps of the advertising bid adjustment method described in the first aspect.

[0019] A fifth aspect of this application provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps of the advertising bid adjustment method described in the first aspect.

[0020] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0021] This application provides an advertising bid adjustment method. This method innovatively utilizes a first deep target achievement rate and a second deep target achievement rate, which reflect the deep target achievement status of the target advertisement from different dimensions, to determine a deep price adjustment coefficient. Then, using this deep price adjustment coefficient and a shallow price adjustment coefficient determined according to the conversion price corresponding to the target advertisement, the conversion price corresponding to the target advertisement is adjusted in a coordinated manner to achieve advertising bid adjustment for the target advertisement. The aforementioned first deep-level goal achievement rate is the proportion of objects that achieve shallow-level goals within the reference historical period that also achieve deep-level goals. It comprehensively reflects the deep-level goal achievement of the target advertisement within the complete reference historical period. The aforementioned second deep-level goal achievement rate is the proportion of objects that achieve shallow-level goals within a reference time interval of the reference historical period that also achieve deep-level goals. It provides a detailed reflection of the deep-level goal achievement of the target advertisement within a specific time interval of the reference historical period. By comprehensively considering the first and second deep-level goal achievement rates, a deep-level price adjustment coefficient can be determined. This ensures that the determined deep-level price adjustment coefficient is more accurate. Adjusting the conversion price of the target advertisement using this deep-level price adjustment coefficient can better match the cost of the target advertisement with the actual achievement of its deep-level goals, thus better matching the cost of the target advertisement with its actual conversion effect, thereby minimizing losses for advertisers or advertising platforms. Attached Figure Description

[0022] Figure 1 This is a schematic diagram illustrating an application scenario of the advertising bid adjustment method provided in the embodiments of this application;

[0023] Figure 2 A flowchart illustrating the advertising bid adjustment method provided in this application embodiment;

[0024] Figure 3A flowchart illustrating the determination of the deep price adjustment coefficient provided in this application embodiment;

[0025] Figure 4 A schematic diagram illustrating the process of adjusting conversion prices provided in an embodiment of this application;

[0026] Figure 5 A schematic diagram of an advertising bid adjustment device provided in this application embodiment;

[0027] Figure 6 A schematic diagram of another advertising bid adjustment device provided in this application embodiment;

[0028] Figure 7 A schematic diagram of another advertising bid adjustment device provided in this application embodiment;

[0029] Figure 8 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application;

[0030] Figure 9 This is a schematic diagram of the server structure provided in an embodiment of this application. Detailed Implementation

[0031] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0032] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0033] To facilitate understanding of the technical solutions provided in the embodiments of this application and to understand the role of adjusting advertising bids in the advertising delivery process, the advertising delivery process will be introduced in general below.

[0034] During the advertising process, advertisers first need to create the ads they want to run on the platform. When creating the ad, they must set the corresponding campaign objectives (such as orders, activations, or payments) and conversion costs (the expected cost to achieve a campaign objective). After successfully creating the ad, it will be stored in the advertising platform's ad library. From entering the ad library to finally gaining exposure, an ad typically goes through several stages: ad recall, timeout truncation, coarse ranking, and fine ranking. The ads that perform best in each stage will ultimately receive exposure.

[0035] Whether an ad can succeed in the coarse and fine ad ranking stages usually depends on its corresponding ECPM. The higher the ECPM, the greater the likelihood of success in both stages and ultimately, the greater the chance of gaining exposure. The formula for calculating the ECPM of an ad is shown in equation (1):

[0036] (1)

[0037] in, It is the conversion price corresponding to the advertisement (that is, the conversion price set by the advertiser for the advertisement when creating the advertisement). The click-through rate (CTR) is predicted using a click-through rate prediction model. The conversion rate is estimated using a conversion rate prediction model.

[0038] Typically, the click-through rate (CTR) and conversion rate predicted by the model will deviate from the actual CTR and conversion rate. If... and / or If an ad is overestimated, it will receive a lot of exposure but little or no conversions. This exposure is considered low-quality traffic, leading to a higher actual cost-to-conversion rate than the advertiser's target, resulting in losses. Conversely, if... and / or If the click-through rate (CTR) is underestimated, the ad will receive fewer impressions, but these impressions will contribute more conversions. This traffic is considered high-quality traffic for the ad, and the actual conversion cost will be lower than the advertiser's preset conversion cost, resulting in losses for the ad platform. Furthermore, even if the model's predicted CTR and conversion rate are accurate, due to probability issues, there are still instances where the ad exceeds or falls short of its cost.

[0039] Based on equation (1), it can be seen that adjusting the ad bid (i.e., the conversion price corresponding to the ad) can correct... and / or The prediction deviation makes the actual conversion cost of the advertisement approach the conversion cost preset by the advertiser, and makes the actual conversion cost of the advertisement match its actual conversion effect. Furthermore, for and Accurate advertising forecasts, coupled with adjustments to advertising bids, can resolve over-cost and under-cost situations arising from probability issues. After introducing advertising bid adjustments, the formula for calculating the ECPM for each advertisement will be as shown in equation (2):

[0040] (2)

[0041] in, It is used to adjust the conversion price corresponding to the advertisement. The price adjustment coefficient.

[0042] In order to achieve accurate adjustment of advertising bids and ensure that the adjusted advertising bids achieve the aforementioned effects, that is, to match the actual conversion cost of the advertisement with its actual conversion effect, this application provides an advertising bid adjustment method.

[0043] In this ad bid adjustment method, the first and second deep-level goal achievement rates for the target ad are first determined. The first deep-level goal achievement rate is the proportion of targets that achieved shallow goals within a reference historical period to those that also achieved deep goals. The second deep-level goal achievement rate is the proportion of targets that achieved shallow goals within a reference time interval of the reference historical period to those that also achieved deep goals. Then, based on the conversion price of the target ad, a shallow-level price adjustment coefficient is determined; and based on the first and second deep-level goal achievement rates, a deep-level price adjustment coefficient is determined. Finally, based on the shallow-level and deep-level price adjustment coefficients, the conversion price of the target ad is adjusted to obtain the target conversion price for the target ad.

[0044] The aforementioned advertising bid adjustment method innovatively utilizes the first and second deep target achievement rates, which reflect the deep target achievement status of the target advertisement from different dimensions, to determine the deep price adjustment coefficient. Then, using this deep price adjustment coefficient and the shallow price adjustment coefficient, the conversion price corresponding to the target advertisement is adjusted in a coordinated manner to achieve the advertising bid adjustment for the target advertisement. The aforementioned first deep-level goal achievement rate is the proportion of objects that achieve shallow-level goals within the reference historical period that also achieve deep-level goals. It comprehensively reflects the deep-level goal achievement of the target advertisement within the complete reference historical period. The aforementioned second deep-level goal achievement rate is the proportion of objects that achieve shallow-level goals within a reference time interval of the reference historical period that also achieve deep-level goals. It provides a detailed reflection of the deep-level goal achievement of the target advertisement within a specific time interval of the reference historical period. By comprehensively considering the first and second deep-level goal achievement rates, a deep-level price adjustment coefficient can be determined. This ensures that the determined deep-level price adjustment coefficient is more accurate. Adjusting the conversion price of the target advertisement using this deep-level price adjustment coefficient can better match the cost of the target advertisement with the actual achievement of its deep-level goals, thus better matching the cost of the target advertisement with its actual conversion effect, thereby minimizing losses for advertisers or advertising platforms.

[0045] It should be understood that the advertising bid adjustment method provided in this application embodiment can be applied to electronic devices with data processing capabilities, which can be terminal devices or servers. Specifically, terminal devices can be computers, smartphones, tablets, personal digital assistants (PDAs), etc.; servers can be application servers or web servers, and in actual deployment, they can be standalone servers, cluster servers, or cloud servers.

[0046] To facilitate understanding of the advertising bid adjustment method provided in this application embodiment, the following example uses a server as the execution subject of the advertising bid adjustment method to illustrate the application scenarios of the advertising bid adjustment method.

[0047] See Figure 1 , Figure 1 This is a schematic diagram illustrating an application scenario of the advertising bid adjustment method provided in this application embodiment. For example... Figure 1As shown, this application scenario includes a server 110 and a database 120. The server 110 can access the database 120 via a network, or the database 120 can be integrated into the server 110. The server 110 is used to execute the advertising bid adjustment method provided in this embodiment, adjusting the conversion price of the advertisements stored in the database 120. The database 120 stores each advertisement placed by each advertiser, as well as the associated data corresponding to each advertisement. The associated data includes, but is not limited to: relevant parameters set by the advertiser for the advertisement (such as placement goals, conversion prices, reference deep goal achievement rate, targeted audience, etc.), data representing the shallow goal achievement status of the advertisement, and data representing the deep goal achievement status of the advertisement. This application does not impose any limitations on the data stored in the database 120.

[0048] In practical applications, server 110 can retrieve any advertisement from database 120 as a target advertisement, and retrieve the associated data corresponding to the target advertisement from database 120, such as the conversion price corresponding to the target advertisement (i.e., the amount that the advertiser expects to spend to complete a campaign goal by setting the target advertisement), the objects that achieved the shallow goal corresponding to the target advertisement in the reference historical period, the objects that achieved the shallow goal corresponding to the target advertisement in the reference time interval in the reference historical period, and the objects that achieved the deep goal corresponding to the target advertisement in the current period.

[0049] It should be understood that the selection of the reference historical period is usually related to the deep objective corresponding to the target ad. For example, assuming a daily data statistics period, if the deep objective corresponding to the target ad is next-day retention (i.e., there is usage behavior of the application within one day after downloading or registering the application recommended by the target ad), then the reference historical period should be the day before the current period. That is, if the current period is July 27, then the reference historical period should be July 26. If the deep objective corresponding to the target ad is three-day retention (i.e., there is usage behavior of the application within three days after downloading or registering the application recommended by the target ad), then the reference historical period should be the three days before the current period. That is, if the current period is July 27, then the reference historical period should be July 24. Furthermore, the selection of the reference time interval is usually related to the current moment of the current period. For example, the current moment can be used as the end point of the reference time interval, and then combined with the preset time interval length to determine the starting point of the reference time interval. For instance, assuming the current moment is 10:00 and the preset time interval length is 1 hour, the time interval from 9:00 to 10:00 can be selected as the reference time interval. Of course, in practical applications, the server 110 can also use other methods to determine the above-mentioned reference historical period and reference time interval. This application does not impose any limitations on the implementation method of determining the reference historical period and reference time interval.

[0050] After retrieving the associated data corresponding to the target advertisement from the database 120, the server 110 can first determine the first deep target achievement rate and the second deep target achievement rate corresponding to the target advertisement. The first deep target achievement rate is the proportion of objects that achieve the deep target among those that achieve the shallow target within the reference historical period, and the second deep target achievement rate is the proportion of objects that achieve the deep target among those that achieve the shallow target within the reference time interval of the reference historical period. For example, suppose the shallow goal of the target ad is to download the application promoted by the target ad, and the deep goal of the target ad is to have usage behavior of the application within one day after the shallow goal is achieved; then the server 110 can determine the first deep goal achievement rate in the following way: determine the objects that downloaded the application promoted by the target ad within the previous day (referred to as the first object), and then determine the objects that used the application within the current day within all the first objects (referred to as the second object), and calculate the ratio of the number of second objects to the number of first objects, which is the first deep goal achievement rate; the server 110 can determine the second deep goal achievement rate in the following way: determine the objects that downloaded the application promoted by the target ad within a reference time interval (such as 9:00-10:00) within the previous day (referred to as the third object), and then determine the objects that used the application within the current day within all the third objects (referred to as the fourth object), and calculate the ratio of the number of fourth objects to the number of third objects, which is the second deep goal achievement rate.

[0051] Then, server 110 can determine the shallow bid adjustment coefficient and deep bid adjustment coefficient corresponding to the target ad. Specifically, server 110 can determine the shallow bid adjustment coefficient corresponding to the target ad based on the conversion price corresponding to the target ad; it should be understood that, generally, the higher the conversion price corresponding to the target ad, the higher the shallow bid adjustment coefficient corresponding to the target ad. Server 110 can determine the deep bid adjustment coefficient corresponding to the target ad based on the first deep target achievement rate and the second deep target achievement rate corresponding to the target ad mentioned above.

[0052] Furthermore, server 110 can adjust the conversion price of the target ad using the shallow and deep price adjustment coefficients corresponding to the determined target ad, thus obtaining the target conversion price for the target ad. The advertising platform can then place the target ad based on this target conversion price, thereby matching the actual placement cost of the target ad with its actual performance.

[0053] It should be understood that Figure 1The application scenarios shown are merely examples. In actual applications, the advertising bid adjustment method provided in this application embodiment can also be executed by the terminal device. No limitations are made here on the application scenarios of the advertising bid adjustment method provided in this application embodiment.

[0054] The advertising bid adjustment method provided in this application will be described in detail below through method embodiments.

[0055] See Figure 2 , Figure 2 This is a flowchart illustrating the advertising bid adjustment method provided in this application embodiment. For ease of description, the following embodiments will still use a server as the executing entity of this advertising bid adjustment method. Figure 2 As shown, the method for adjusting ad bids includes the following steps:

[0056] Step 201: Determine the first deep goal achievement rate and the second deep goal achievement rate corresponding to the target advertisement; the first deep goal achievement rate is the proportion of objects that achieve the deep goal among those that achieve the shallow goal within the reference historical period, and the second deep goal achievement rate is the proportion of objects that achieve the deep goal among those that achieve the shallow goal within the reference time interval of the reference historical period, and achieving the shallow goal is a prerequisite for achieving the deep goal.

[0057] In practical applications, after an ad is displayed, the server can detect whether the ad has achieved its corresponding shallow and deep objectives based on the user's actions triggered by the ad's content. It records the number of users who achieved the shallow objective and the time they achieved it, as well as the number of users who achieved the deep objective. When the server needs to adjust the bid for a target ad, it can determine the first and second deep objective achievement rates based on the aforementioned recorded data related to the target ad.

[0058] It's important to note that after an ad is displayed, users can only achieve its deeper objective after fulfilling the ad's superficial objective. In other words, achieving the superficial objective is a prerequisite for achieving the deeper objective. In the standard process, after an ad is displayed, users can click to view the ad content based on their needs and activate it, thus achieving the superficial objective. After activating the ad content, users can perform further actions. Detecting these subsequent actions set by the advertiser indicates that the deeper objective has been achieved. The superficial and deeper objectives can be set based on the ad content and the advertiser's specific needs.

[0059] For example, suppose the advertisement promotes an application, meaning the advertiser wants to increase the number of users of that application by promoting it. The underlying objective of the advertisement could be set as either downloading the advertised application or registering as a new user of the advertised application. Specifically, after the advertisement is displayed, users can click to view the application's description and download it based on their needs. If the underlying objective is downloading the application, the user has achieved that underlying objective. If the underlying objective is registering as a user of the advertised application, the server needs to continue monitoring whether the user registers as a new user after downloading the application. If the server detects that the user has registered, the user has also achieved the underlying objective.

[0060] Building upon the aforementioned shallow objectives, the deeper objectives of the advertisement can be detecting targeted behaviors generated through the application within a preset timeframe after achieving the shallow objectives. These targeted behaviors can include at least one of the following: login behavior, information viewing behavior, multimedia content playback behavior, and payment behavior. Specifically, the deeper objectives of the advertisement can be next-day retention, three-day retention, etc. Next-day retention refers to detecting targeted behaviors generated through the application within one day after achieving the shallow objectives, and three-day retention refers to detecting targeted behaviors generated through the application within three days after achieving the shallow objectives. These targeted behaviors can be user login to the application, user browsing of information displayed on the application, user playing of audio, video, or other multimedia content through the application, or user making payments through the application. The preset timeframe and targeted behaviors can be set according to actual needs, and this application does not impose any limitations on these preset timeframes or targeted behaviors.

[0061] It should be understood that in practical applications, in addition to setting the above-mentioned shallow and deep targets for advertisements, other shallow and deep targets can be set according to actual needs. This application does not impose any limitations on the shallow and deep targets set.

[0062] When determining the achievement rates of the first and second deep goals, the server can first determine a reference historical period based on the current period and the deep goals. For example, assuming a daily period, the current period is July 27th, and the deep goal corresponding to the target ad is next-day retention, the server can determine July 26th as the reference historical period. Similarly, assuming a daily period, the current period is July 27th, and the deep goal corresponding to the target ad is three-day retention, the server can determine July 24th as the reference historical period. Furthermore, the server can determine a reference time interval based on the current moment of the current period. For example, the server can use the current moment as the end point of the reference time interval and determine the starting point of the reference time interval as the time point before the end point and at a preset time distance from the end point. For example, assuming the current moment is 10:00 and the preset time interval length is 1 hour, the time interval from 9:00 to 10:00 can be selected as the reference time interval.

[0063] It should be understood that the above implementation methods for determining the reference historical period and reference time interval are merely examples. In practical applications, the server may also use other methods to determine the reference historical period and reference time interval according to actual needs. This application does not impose any limitations on the methods for determining the reference historical period and reference time interval.

[0064] Furthermore, the server can determine the objects (i.e., the users mentioned above) that achieved the shallow goal corresponding to the target advertisement within the reference historical period, and determine the proportion of these objects that achieved the deep goal corresponding to the target advertisement. This proportion is the first deep goal achievement rate in this embodiment of the application. Taking the current period as July 27, the reference historical period as July 26, and the deep goal corresponding to the target advertisement as next-day retention as an example, the server can determine the objects that achieved the shallow goal corresponding to the target advertisement on July 26 (referred to as the first objects), and determine the objects that achieved the deep goal corresponding to the target advertisement on July 27 among all the first objects (referred to as the second objects). Then, the ratio of the number of second objects to the number of first objects is calculated as the first deep goal achievement rate.

[0065] The server can identify objects that achieve the shallow goal corresponding to the target ad within a reference time interval of a reference historical period, and determine the proportion of these objects that achieve the deep goal corresponding to the target ad. This proportion is the second deep goal achievement rate in this embodiment. Taking the current time of the current period as 10:00 on July 27, the reference time interval of the reference historical period as 9:00-10:00 on July 26, and the deep goal corresponding to the target ad as next-day retention as an example, the server can identify objects that achieve the shallow goal corresponding to the target ad between 9:00 and 10:00 on July 26 (referred to as third objects), and identify objects that achieve the deep goal corresponding to the target ad on July 27 among all third objects (referred to as fourth objects). Then, the ratio of the number of fourth objects to the number of third objects is calculated as the second deep goal achievement rate.

[0066] It should be understood that in practical applications, other devices can also calculate the first deep target achievement rate and the second deep target achievement rate corresponding to the target advertisement. When the server needs to adjust the bid of the target advertisement, it can obtain the first deep target achievement rate and the second deep target achievement rate corresponding to the target advertisement from other devices. This application does not impose any restrictions on the way the server obtains the first deep target achievement rate and the second deep target achievement rate corresponding to the target advertisement.

[0067] Step 202: Determine the shallow price adjustment coefficient corresponding to the target advertisement based on the conversion price corresponding to the target advertisement; determine the deep price adjustment coefficient corresponding to the target advertisement based on the first deep target achievement rate and the second deep target achievement rate.

[0068] Furthermore, the server can determine the price adjustment coefficient applicable to adjusting the conversion price corresponding to the target advertisement. In this embodiment, the server determines the shallow price adjustment coefficient and the deep price adjustment coefficient from two dimensions: shallow target and deep target, so that the shallow price adjustment coefficient and the deep price adjustment coefficient can be used to adjust the conversion price corresponding to the target advertisement in a coordinated manner.

[0069] When determining the shallow price adjustment coefficient for a target ad, the server can do so based on the conversion price of that target ad. It should be understood that there should be a positive correlation between the shallow price adjustment coefficient and the conversion price; that is, the higher the conversion price of the target ad, the higher the shallow price adjustment coefficient. For example, the server can calculate the shallow price adjustment coefficient for the target ad based on the expected cost optimization principle, according to the conversion price. The expected cost optimization principle means that, assuming the current time of the current period is t, and the conversions already achieved through the target ad in the current period are CV, the accumulated cost is cost, the expected conversions by the end of the current period are CV_E, and the expected cost is cost_E, then the expected cumulative cost CPA_final for the current period should be (cost + cost_E) / (CV + CV_E). Of course, in practical applications, the server can also use other methods to determine the shallow price adjustment coefficient for the target ad. This application does not impose any limitations on the method of determining the shallow price adjustment coefficient for the target ad.

[0070] When the server determines the deep bid adjustment coefficient corresponding to the target advertisement, it can determine the deep bid adjustment coefficient corresponding to the target advertisement based on the first deep target achievement rate and the second deep target achievement rate corresponding to the target advertisement determined in step 201.

[0071] In one possible implementation, the server can follow Figure 3 The process shown determines the deep bid adjustment factor corresponding to the target advertisement. For example... Figure 3 As shown, the process by which the server determines the deep bid adjustment coefficient corresponding to the target advertisement includes the following steps:

[0072] Step 301: Based on the reference deep target achievement rate, the first deep target achievement rate, and the second deep target achievement rate corresponding to the target advertisement, determine the coefficient adjustment step size and step size adjustment rate. The reference deep target achievement rate represents the expected achievement rate of the deep target set for the target advertisement.

[0073] Specifically, the server can first obtain the reference deep goal achievement rate (tDCVR) corresponding to the target ad. This reference deep goal achievement rate can be calculated based on the shallow goal achievement price and deep goal achievement price set by the advertiser for the target ad. For example, assuming the advertiser sets the shallow goal achievement price for the target ad to targetCPA1 and the deep goal achievement price to targetCPA2, then targetCPA1 / targetCPA2 can be calculated as the reference deep goal achievement rate. Alternatively, the reference deep goal achievement rate can also be the expected achievement rate of the deep goals directly set by the advertiser for the target ad.

[0074] Then, the server can determine the calculation method for the coefficient adjustment step size and step size adjustment rate based on the relationship between the reference deep target achievement rate, the first deep target achievement rate, and the second deep target achievement rate. Furthermore, using the determined calculation method for the coefficient adjustment step size, the server calculates the coefficient adjustment step size based on one of the first and second deep target achievement rates, and the reference deep target achievement rate; and using the determined calculation method for the step size adjustment rate, the server determines the step size adjustment rate based on the first target achievement rate, the second deep target achievement rate, and preset step size adjustment parameters.

[0075] This application provides exemplary methods for determining the step size of coefficient adjustment and methods for determining the step size adjustment rate. The methods for determining the step size of coefficient adjustment and methods for determining the step size adjustment rate are described below.

[0076] As an example, the server can determine the coefficient adjustment step size in the following way:

[0077] When both the achievement rates of the first and second deep-level goals are greater than the achievement rate of the reference deep-level goal, and when both the achievement rates of the first and second deep-level goals are less than the achievement rate of the reference deep-level goal, the coefficient adjustment step size is determined based on the achievement rates of the first and reference deep-level goals. When the achievement rate of the first deep-level goal is less than the achievement rate of the reference deep-level goal, and the achievement rate of the reference deep-level goal is less than the achievement rate of the second deep-level goal, and when the achievement rate of the first deep-level goal is greater than the achievement rate of the reference deep-level goal, and the achievement rate of the reference deep-level goal is greater than the achievement rate of the second deep-level goal, the difference between the achievement rate of the first deep-level goal and the achievement rate of the reference deep-level goal is determined as the first difference, and the difference between the achievement rate of the second deep-level goal and the achievement rate of the reference deep-level goal is determined as the second difference. The coefficient adjustment step size is determined based on the first and second differences.

[0078] Specifically, the achievement rate of the first deep objective is represented as DCVR. now The achievement rate of the second deep objective is represented as DCVR. now2 Let the deep target achievement rate be represented as tDCVR. When dcvr now >dcvr now2 >tDCVR, and when dcvr now2 >dcvr now >tDCVR, and when dcvr now <dcvr now2 <tDCVR, and when dcvr now2 <dcvr now When <tDCVR, the server can directly calculate (dcvr) now / tDCVR-1) is used as the coefficient to adjust the step size.

[0079] It should be understood that when DCVR now >dcvr now2 >tDCVR time and when dcvr now2 >dcvr now When the target ad's current performance is high (>tDCVR), the exposure gained by the target ad is likely to result in a conversion. At this point, the actual conversion cost of the target ad is lower than the advertiser's expected conversion cost. If the target ad is still executed at the conversion price set by the advertiser, it will result in losses for the advertising platform. This embodiment calculates the coefficient adjustment step size using the above method, obtaining a positive coefficient adjustment step size. Adjusting the reference deep price adjustment coefficient based on this positive step size will increase the deep price adjustment coefficient corresponding to the target ad. Adjusting the conversion price corresponding to the target ad based on this increased deep price adjustment coefficient will increase the target conversion price and give the target ad more exposure opportunities. This ensures that the actual conversion cost of the target ad matches its actual conversion performance and approaches the advertiser's expected conversion cost, while reducing losses for the advertising platform.

[0080] Correspondingly, when DCVR now <dcvr now2 <tDCVR when and when dcvr now2 <dcvr now When the target ad's current performance is poor (< tDCVR), the probability of conversion from the exposure received by the target ad is very low. In this case, the actual conversion cost of the target ad is higher than the advertiser's expected conversion cost. If the target ad is still executed at the conversion price set by the advertiser, it will result in losses for the advertiser. This embodiment calculates the coefficient adjustment step size using the above method, resulting in a negative coefficient adjustment step size. Adjusting the reference deep price adjustment coefficient based on this negative step size will reduce the deep price adjustment coefficient corresponding to the target ad. Adjusting the conversion price corresponding to the target ad based on the reduced deep price adjustment coefficient will further reduce the conversion price and reduce the exposure opportunities for the target ad. This ensures that the actual conversion cost of the target ad matches its actual conversion performance and approaches the advertiser's expected conversion cost, while minimizing losses for the advertiser.

[0081] When DCVR now <tDCVR<dcvr now2 At that time, and when DCVR now2 <tDCVR<dcvr now At that time, the server can calculate DCVR.now The absolute value of the difference between tDCVR and As the first gap, and, calculating DCVR now2 The absolute value of the difference between tDCVR and This serves as the second gap. Furthermore, based on the relationship between the first and second gaps, from DCVR... now and DCVR now2 Choose one option and determine the coefficient adjustment step size based on the selected deep target achievement rate and the reference deep target achievement rate.

[0082] For example, when the first gap is greater than the second gap, the server can determine the coefficient adjustment step size based on the first deep target achievement rate and the reference deep target achievement rate; when the first gap is less than the second gap, the server can determine the coefficient adjustment step size based on the second deep target achievement rate and the reference deep target achievement rate.

[0083] Specifically, when Greater than At that time, the server can calculate (dcvr) now / tDCVR-1) is used as the coefficient adjustment step size; when Less than The server can calculate (dcvr) now2 / tDCVR-1) is used as the coefficient to adjust the step size.

[0084] It should be understood that when the first deep target achievement rate and the second deep target achievement rate are located on opposite sides of the reference deep target achievement rate, the server can determine the coefficient adjustment step size based on the deep target achievement rate that differs more from the reference deep target achievement rate. Based on this coefficient adjustment step size, a deep price adjustment coefficient is determined, and the conversion price corresponding to the target ad is adjusted using this deep price adjustment coefficient. This is more conducive to matching the actual conversion cost of the target ad with its actual conversion effect, and to making the actual conversion cost closer to the conversion cost expected by the advertiser, thereby reducing the losses of the advertiser or the advertising platform.

[0085] As an example, the server can determine the step size adjustment rate in the following way:

[0086] When both the first and second deep-level target achievement rates are greater than the reference deep-level target achievement rate, the step size adjustment rate is determined based on the ratio of the second to the first deep-level target achievement rate and a preset first-step adjustment parameter. When both the first and second deep-level target achievement rates are less than the reference deep-level target achievement rate, the step size adjustment rate is determined based on the ratio of the first to the second deep-level target achievement rates and a preset second-step adjustment parameter. When the first deep-level target achievement rate is less than the reference deep-level target achievement rate, and the reference deep-level target achievement rate is less than the second deep-level target achievement rate, and when the first deep-level target achievement rate is greater than the reference deep-level target achievement rate, and the reference deep-level target achievement rate is greater than the second deep-level target achievement rate, a preset third-step adjustment parameter is determined as the step size adjustment rate.

[0087] Specifically, the achievement rate of the first deep objective is represented as DCVR. now The achievement rate of the second deep objective is represented as DCVR. now2 Let the deep target achievement rate be represented as tDCVR. When dcvr now >dcvr now2 >tDCVR and when dcvr now2 >dcvr now When >tDCVR, the server can calculate ((dcvr) now2 / dcvr now -1)+1)*rate1 is used as the step size adjustment rate, where rate1 is the preset first step size adjustment parameter, which can be set according to actual needs, for example, to 0.05. According to the calculation formula of this step size adjustment rate, dcvr... now2 >dcvr now Step size adjustment rate compared to DCVR now >dcvr now2 The step size adjustment rate is larger during this time. The reason for this setting is that DCVR... now2 It reflects the achievement of deep objectives within a time interval close to the current moment, DCVR now It reflects the achievement of deep-level goals throughout the entire cycle, in DCVR now and DCVR now2 When both are greater than tDCVR, dcvr now2 Greater than DCVR now This indicates that the deeper goals are achieved better in time intervals closer to the current moment. In order to make the actual conversion cost of the target ad more in line with its actual conversion effect, it is necessary to adjust the deeper price adjustment coefficient corresponding to the target ad more quickly in the direction of increasing.

[0088] When DCVR now <dcvrnow2 <tDCVR and when dcvr now2 <dcvr now When <tDCVR, the server can calculate ((dcvr) now / dcvr now2 -1)+1)*rate2 is used as the step size adjustment rate, where rate2 is a preset second step size adjustment parameter, which can be set according to actual needs, for example, to 0.05. Based on the calculation formula of this step size adjustment rate, it can be seen that DCVR... now2 <dcvr now Step size adjustment rate compared to DCVR now <dcvr now2 The step size adjustment rate is larger at this time. The reason for this setting is that in DCVR... now and DCVR now2 When both are less than tDCVR, dcvr now2 Less than DCVR now This indicates that the achievement of deeper objectives is worse in time intervals closer to the current moment. In order to make the actual conversion cost of the target ad more in line with its actual conversion effect, it is necessary to adjust the deeper price adjustment coefficient corresponding to the target ad more quickly in the direction of decreasing.

[0089] When DCVR now <tDCVR<dcvr now2 When and when DCVR now2 <tDCVR<dcvr now At this time, the server does not need to adjust the deep price adjustment coefficient faster or slower in the direction of increasing or decreasing. Therefore, the server can directly use the preset third step adjustment parameter rate3 as the step adjustment rate. This third step adjustment parameter can be set according to actual needs, for example, set to 0.05.

[0090] Considering that when both the first and second deep target achievement rates are greater than the reference deep target achievement rate, and when both the first and second deep target achievement rates are less than the reference deep target achievement rate, the step adjustment rate determined based on the first and second deep target achievement rates may be too large or too small, which may lead to the deep price adjustment coefficient corresponding to the target advertisement being adjusted too quickly or too slowly, causing the final deep price adjustment coefficient to deviate from a reasonable coefficient range. Therefore, this application may also impose a threshold limit on the step adjustment rate determined under the above circumstances.

[0091] Specifically, when the achievement rates of both the first and second deep-level goals are greater than the reference deep-level goal achievement rate, and when both the first and second deep-level goal achievement rates are less than the reference deep-level goal achievement rate, it is determined whether the determined step size adjustment rate is less than the lower limit and greater than the upper limit. If the determined step size adjustment rate is less than the lower limit, then the step size adjustment rate is determined to be the lower limit. If the determined step size adjustment rate is greater than the upper limit, then the step size adjustment rate is determined to be the upper limit.

[0092] For example, the server can set a reasonable range for the step adjustment rate [0.5, 2], where 0.5 is the lower limit of the step adjustment rate and 2 is the upper limit. In cases where both the first and second deep-level target achievement rates are greater than the reference deep-level target achievement rate, or both are less than the reference deep-level target achievement rate, if the determined step adjustment rate is less than 0.5, then the subsequent step adjustment rate used to determine the deep-level price adjustment coefficient is determined to be 0.5; if the determined step adjustment rate is greater than 2, then the subsequent step adjustment rate used to determine the deep-level price adjustment coefficient is determined to be 2.

[0093] It should be understood that the methods for determining the coefficient adjustment step size and the methods for determining the step size adjustment rate described above are merely exemplary implementations provided in this application. In practical applications, the server may also use other methods to determine the coefficient adjustment step size and step size adjustment rate according to actual needs.

[0094] Step 302: Based on the coefficient, adjust the step size and step size adjustment rate, and adjust the reference deep price adjustment coefficient to obtain the deep price adjustment coefficient corresponding to the target advertisement; the reference deep price adjustment coefficient is the deep price adjustment coefficient used when adjusting the conversion price corresponding to the target advertisement last time.

[0095] After the server determines the step size and step rate of the coefficient adjustment, it can use these step size and step rate to adjust the reference deep price adjustment coefficient used when adjusting the conversion price of the target ad in the last adjustment, so as to obtain the deep price adjustment coefficient applicable to the conversion price of the target ad in this adjustment. The formula for determining the deep price adjustment coefficient is shown in equation (2):

[0096] (2)

[0097] in, This is the deep price adjustment factor used when adjusting the conversion price of the target ad in this instance; The reference deep price adjustment factor is the deep price adjustment factor used when the conversion price of the target ad was last adjusted.

[0098] Based on the information provided in step 301 above, the above equation (2) can specifically represent the transformation equations (3) to (6):

[0099] When DCVR now >dcvr now2 >tDCVR and when dcvr now2 >dcvr now When ≥ tDCVR, equation (2) is specifically expressed as equation (3):

[0100] (3)

[0101] When DCVR now <dcvr now2 <tDCVR and when dcvr now2 <dcvr now When <tDCVR, equation (2) is specifically expressed as equation (4):

[0102] (4)

[0103] When DCVR now <tDCVR<dcvr now2 When and when DCVR now2 <tDCVR<dcvr now At that time, it will be based on and The size relationship between them is specifically expressed in equation (5) and equation (6) respectively; when When, equation (2) is specifically expressed as equation (5), when When, equation (2) is specifically expressed as equation (6):

[0104] (5)

[0105] (6)

[0106] The method described above for determining the deep bid adjustment coefficient for the target ad, along with the adjustment direction (increase, decrease, or a combination thereof) and the trend of the step adjustment rate (increase, decrease, or a combination thereof), are detailed in Table 1.

[0107] Table 1

[0108]

[0109] It should be understood that in practical applications, servers may also use other methods to determine the deep bid adjustment coefficient corresponding to the target advertisement. This application does not impose any restrictions on the method of determining the deep bid adjustment coefficient corresponding to the target advertisement. However, regardless of the method used to determine the deep bid adjustment coefficient, the direction of adjustment of the deep bid adjustment coefficient and the speed of adjustment of the deep bid adjustment coefficient should conform to the rules in Table 1.

[0110] Step 203: Adjust the conversion price according to the shallow price adjustment coefficient and the deep price adjustment coefficient to obtain the target conversion price corresponding to the target advertisement.

[0111] After the server determines the shallow and deep price adjustment coefficients corresponding to the target ad, it can use these coefficients to adjust the conversion price set by the advertiser for the target ad, thus obtaining the target conversion price. Targeting the ad based on this target conversion price ensures that the actual conversion cost matches the actual conversion effect, and that the actual conversion cost approaches the advertiser's expected conversion cost.

[0112] In one possible implementation, to make more accurate adjustments to the conversion price of the target ad, the server can first adjust the deep-level price adjustment coefficient of the target ad by combining the reference deep-level goal achievement rate and the predicted deep-level goal achievement rate. Then, using the adjusted deep-level price adjustment coefficient and the shallow-level price adjustment coefficient, the conversion price of the target ad is adjusted. The specific process for adjusting the conversion price is as follows: Figure 4 As shown, it includes the following steps:

[0113] Step 401: Obtain the reference deep target achievement rate and the predicted deep target achievement rate corresponding to the target advertisement; the reference deep target achievement rate represents the expected achievement rate of the deep target set for the target advertisement, and the predicted deep target achievement rate is determined by the deep target achievement rate prediction model based at least on the basic information of the target advertisement.

[0114] Before adjusting the conversion price of a target ad, the server needs to obtain the reference deep target achievement rate and the predicted deep target achievement rate for that target ad.

[0115] The reference deep goal achievement rate can be calculated based on the shallow goal achievement price and deep goal achievement price set by the advertiser for the target ad. For example, assuming the advertiser sets the shallow goal achievement price for the target ad to targetCPA1 and the deep goal achievement price to targetCPA2, then targetCPA1 / targetCPA2 can be calculated as the reference deep goal achievement rate. Alternatively, the reference deep goal achievement rate can also be the expected achievement rate of the deep goals directly set by the advertiser for the target ad.

[0116] The prediction of deep target achievement rate can be determined by the server using a deep target achievement rate prediction model based on the basic information of the target ad; the basic information here includes, but is not limited to, the conversion price of the target ad, the shallow target and / or deep target corresponding to the target ad, the characteristics of the target ad itself (such as the display style of the target ad), etc.

[0117] Step 402: Adjust the deep target price coefficient based on the reference deep target achievement rate and the predicted deep target achievement rate to obtain the target deep target price coefficient corresponding to the target advertisement.

[0118] Then, the server can make a second adjustment to the deep target price adjustment coefficient determined in step 202 based on the reference deep target achievement rate and the predicted deep target achievement rate corresponding to the target advertisement, so as to obtain the target deep target price adjustment coefficient corresponding to the target advertisement.

[0119] For example, the server can adjust the deep pricing coefficient corresponding to the target advertisement through equation (7):

[0120] (7)

[0121] in, The target deep price adjustment coefficient, The deep price adjustment coefficient is determined through step 202, pDCVR is the predicted deep target achievement rate, and tDCVR is the reference deep target achievement rate.

[0122] Step 403: Adjust the conversion price based on the shallow price adjustment coefficient and the target deep price adjustment coefficient to obtain the target conversion price.

[0123] Furthermore, the server can adjust the shallow price adjustment coefficient determined in step 202. and the target deep price adjustment coefficient determined through step 402. The conversion price corresponding to the target advertisement is adjusted to obtain the target conversion price. The specific formula for the target conversion price is shown in equation (8):

[0124] (8)

[0125] in, To achieve the final target conversion price, The conversion price set by the advertiser for the target ad.

[0126] By comprehensively considering the actual deep target achievement rate (including the first and second deep target achievement rates), the reference deep target achievement rate, and the predicted deep target achievement rate, the deep price adjustment coefficient is adjusted, and the conversion price corresponding to the target advertisement is adjusted. The adjustment trend of the deep price adjustment coefficient and the adjustment method of the conversion price are shown in Table 2.

[0127] Table 2

[0128]

[0129] DCVR in Table 2 now Including the first deep goal achievement rate DCVR now Second deep goal achievement rate DCVR now2 DCVR now Greater than the reference deep target achievement rate tDCVR refers to DCVR now and DCVR now2 Both are greater than tDCVR, DCVR now equal to tDCVR, which refers to DCVR now and DCVR now2 Both are equal to tDCVR, DCVR now Less than tDCVR means DCVR now and DCVR now2 All are less than tDCVR.

[0130] Therefore, by comprehensively considering the actual deep target achievement rate, the reference deep target achievement rate, and the predicted deep target achievement rate through the above methods, the conversion price will be adjusted to make the adjustment trend of the conversion price more accurate and closer to matching the actual conversion cost of the target advertisement with its actual conversion effect.

[0131] It should be understood that in practical applications, servers may also use other methods to adjust conversion prices by comprehensively considering the actual deep target achievement rate, the reference deep target achievement rate, and the predicted deep target achievement rate. This application does not impose any restrictions on the method of adjusting conversion prices. However, regardless of the adjustment method used, the adjustment trend of the deep price adjustment coefficient and the adjustment method of the conversion price should conform to the rules in Table 2.

[0132] After the server adjusts the conversion price for the target ad, it can then deliver the target ad based on the adjusted target conversion price. During delivery, the server also needs to obtain the predicted click-through rate (CTR) and predicted conversion rate (PCR) for the target ad. The predicted CTR is determined using a CTR prediction model based on at least the basic information of the target ad, and the predicted PCR is determined using a conversion rate prediction model based on at least the basic information of the target ad. Therefore, the server can determine the ECPM (Economic Cost Per Mille) for the target ad based on the target conversion price, predicted CTR, and predicted PCR. When the advertising platform delivers ads, it sorts the candidate ads by their respective ECPMs from highest to lowest and then selects the candidate ad with the highest ECPM for delivery.

[0133] For example, the server can use a click-through rate prediction model to determine the predicted click-through rate (pCTR) of the target ad based on the basic information corresponding to the target ad; and use a conversion rate prediction model to determine the predicted conversion rate (pCVR) of the target ad based on the basic information corresponding to the target ad; the basic information corresponding to the target ad includes, but is not limited to, the conversion price of the target ad, the characteristics of the target ad itself (such as the display style of the target ad), etc. Furthermore, the server can calculate the ECPM corresponding to the target ad using formula (9):

[0134] (9)

[0135] in, This refers to the target conversion price corresponding to the target advertisement after adjustment.

[0136] Substituting equations (7) and (8) into equation (9), we obtain equation (10):

[0137] (10)

[0138] in, The conversion price set by the advertiser for the target ad. The shallow price adjustment coefficient is determined through step 202. The deep price adjustment coefficient is determined through step 202. pDCVR is the predicted deep target achievement rate, and tDCVR is the reference deep target achievement rate.

[0139] Calculate the ECPM corresponding to the target ad according to formula (9) or (10), that is, place the target ad based on the adjusted target conversion price, so that the actual conversion cost of the target ad matches its actual conversion effect.

[0140] The aforementioned advertising bid adjustment method innovatively utilizes the first and second deep target achievement rates, which reflect the deep target achievement status of the target advertisement from different dimensions, to determine the deep price adjustment coefficient. Then, using this deep price adjustment coefficient and the shallow price adjustment coefficient, the conversion price corresponding to the target advertisement is adjusted in a coordinated manner to achieve the advertising bid adjustment for the target advertisement. The aforementioned first deep-level goal achievement rate is the proportion of objects that achieve shallow-level goals within the reference historical period that also achieve deep-level goals. It comprehensively reflects the deep-level goal achievement of the target advertisement within the complete reference historical period. The aforementioned second deep-level goal achievement rate is the proportion of objects that achieve shallow-level goals within a reference time interval of the reference historical period that also achieve deep-level goals. It provides a detailed reflection of the deep-level goal achievement of the target advertisement within a specific time interval of the reference historical period. By comprehensively considering the first and second deep-level goal achievement rates, a deep-level price adjustment coefficient can be determined. This ensures that the determined deep-level price adjustment coefficient is more accurate. Adjusting the conversion price of the target advertisement using this deep-level price adjustment coefficient can better match the cost of the target advertisement with the actual achievement of its deep-level goals, thus better matching the cost of the target advertisement with its actual conversion effect, thereby minimizing losses for advertisers or advertising platforms.

[0141] To facilitate a further understanding of the advertising bid adjustment method provided in this application embodiment, the following example uses the advertising bid adjustment method provided in this application embodiment to adjust the bid of an advertisement promoting an application. The shallow goal of this advertisement can be that users download the application promoted by the advertisement or register as new users of the application, and the deep goal of this advertisement is next-day retention. In this scenario, the adjustment method of advertising bid is qualitatively analyzed and quantitatively calculated, and the mathematical derivation process of advertising bid adjustment is introduced.

[0142] Generally, when the basic objective of an ad is achieved, if the actual next-day retention rate is high, the conversion price of the ad should be appropriately increased based on the actual next-day retention rate. When the basic objective is not achieved, the conversion price of the ad needs to be appropriately reduced to decrease competitiveness. In this case, if the actual next-day retention rate is high, whether to accept the ad should consider both activation cost and next-day retention. If the activation cost is not met, a lower price should be offered; if the next-day retention rate is high, a higher price should be offered. The final decision on whether to accept the ad should consider the competitive situation at the time to achieve system optimization.

[0143] When adjusting the deep pricing coefficient corresponding to the advertisement in this embodiment, the first deep target achievement rate (DCVR) is comprehensively considered. nowSecond deep goal achievement rate DCVR now2 Among them, the first deep goal achievement rate DCVR now The first reflects the achievement of deeper goals for objects that have already achieved shallow goals within a complete historical cycle. The second deeper goal achievement rate reflects the achievement of deeper goals for objects that have achieved shallow goals within a specific period of the historical cycle. The specific quantitative adjustment method for the deeper price adjustment coefficient is as follows:

[0144] When DCVR now >dcvr now2 >tDCVR and when dcvr now2 >dcvr now When >tDCVR,

[0145]

[0146] When DCVR now <dcvr now2 <tDCVR and when dcvr now2 <dcvr now When <tDCVR,

[0147]

[0148] When DCVR now <tDCVR<dcvr now2 When and when DCVR now2 <tDCVR<dcvr now At that time, if ,

[0149]

[0150] like ,

[0151]

[0152] in, This refers to the deep bid adjustment factor that can be used when adjusting ad bids in this instance. This is the deep bid adjustment factor used during the last ad bid adjustment; The target deep goal achievement rate set by advertisers for their ads, i.e., the reference deep goal achievement rate; It's DCVR now and The gap between them It's DCVR now2 and The difference between them; rate1, rate2 and rate3 are the preset first step length adjustment parameters, second step length adjustment parameters and third step length adjustment parameters, respectively.

[0153] Furthermore, the deep price adjustment coefficient mentioned above, as well as the shallow price adjustment coefficient determined based on the conversion price of the advertisement, can be used to adjust the conversion price of the advertisement, and the advertisement can be run using the adjusted conversion price. That is, the ECPM of the advertisement can be calculated based on the adjusted conversion price, as follows:

[0154]

[0155] in, The conversion price corresponding to the advertisement; and These are the predicted click-through rate and predicted conversion rate, respectively, obtained through model prediction. This is the shallow price adjustment coefficient; tDCVR is the deep price adjustment coefficient; pDCVR is the deep target achievement rate predicted by the model; tDCVR is the reference deep target achievement rate.

[0156] The following is a mathematical derivation of the process of adjusting advertising bids based on retention rates.

[0157] Assuming activation cost does not exceed ;

[0158] The corresponding total consumption is ;

[0159] The corresponding total number of activations is ;in, The conversion rate for shallow targets;

[0160] The activation cost should meet the following conditions: ;

[0161] By transforming the above conditions, we can obtain: .

[0162] Assuming the retention rate is not less than ;

[0163] Total residues are ;in, Conversion rate for deeper targets;

[0164] The retention rate should meet the following conditions: ;

[0165] By transforming the above conditions, we can obtain: .

[0166] Combining the two conditions above, we can conclude that:

[0167] max

[0168] subject to

[0169]

[0170]

[0171] The second-retention optimization problem can be abstracted into an integer programming problem:

[0172] max

[0173] subject to

[0174]

[0175]

[0176] Assumption

[0177]

[0178]

[0179]

[0180] At this point, we can find a linear programming problem that is close to the original integer programming problem:

[0181] max

[0182] subject to

[0183]

[0184]

[0185] It can be proven that the optimal solutions to the two problems above are very close. Assume the optimal solution to the original integer programming problem is... The optimal solution to the linear programming problem is ,but .

[0186] From a practical business perspective, substituting the original definitions of r, p, and q, we can see that... , and Multiply by the price per exposure and the estimated number of activations respectively. And it's related to the estimated number of impressions. Since a typical ad can get anywhere from a few thousand to millions or tens of millions of impressions, the metrics for a single impression are very small and almost negligible compared to the overall data.

[0187] One optimal solution takes the form of:

[0188] ,in, ;

[0189] Substituting the values ​​of x and y, we get the following representation:

[0190]

[0191]

[0192]

[0193] Perform an equivalent transformation on the above expression, multiplying both sides by... This will result in the following representation:

[0194]

[0195] Based on the above formula, the conditions that the cost should satisfy are calculated, and then extracted. This will result in the following representation:

[0196]

[0197] because This can be represented as follows:

[0198]

[0199]

[0200]

[0201] The following new parameters are introduced:

[0202] u=1+

[0203]

[0204] The exposure selection criteria are as follows:

[0205]

[0206] The specific formula for bidding is as follows:

[0207]

[0208] Another optimal solution takes the form of:

[0209]

[0210] Substituting the y-value and transforming the form of the above optimal solution, we obtain the following representation:

[0211]

[0212]

[0213]

[0214]

[0215] The exposure selection criteria are as follows:

[0216] set up , will get ;

[0217] The specific formula for bidding is as follows:

[0218]

[0219] The significance of the above bidding formula lies in the fact that when the retention rate is... If the price exceeds a certain threshold, a very high price must be paid to obtain the exposure; otherwise, the exposure will be abandoned.

[0220] In response to the advertising bid adjustment method described above, this application also provides a corresponding advertising bid adjustment device so that the above advertising bid adjustment method can be applied and implemented in practice.

[0221] See Figure 5 , Figure 5 This is consistent with the above text Figure 2 The diagram shows a structural schematic of an advertising bid adjustment device 500 corresponding to the advertising bid adjustment method illustrated. (See attached diagram.) Figure 5 As shown, the advertising bid adjustment device 500 includes:

[0222] The deep target achievement rate determination module 501 is used to determine the first deep target achievement rate and the second deep target achievement rate corresponding to the target advertisement; the first deep target achievement rate is the proportion of objects that achieve the deep target among those that achieve the shallow target within the reference historical period, and the second deep target achievement rate is the proportion of objects that achieve the deep target among those that achieve the shallow target within the reference time interval of the reference historical period, and achieving the shallow target is a prerequisite for achieving the deep target;

[0223] The shallow price adjustment coefficient determination module 502 is used to determine the shallow price adjustment coefficient corresponding to the target advertisement based on the conversion price corresponding to the target advertisement;

[0224] The deep pricing coefficient determination module 503 is used to determine the deep pricing coefficient corresponding to the target advertisement based on the first deep target achievement rate and the second deep target achievement rate.

[0225] The conversion price adjustment module 504 is used to adjust the conversion price according to the shallow price adjustment coefficient and the deep price adjustment coefficient to obtain the target conversion price corresponding to the target advertisement.

[0226] Optional, in Figure 5 Based on the advertising bid adjustment device shown, see [link / reference]. Figure 6 , Figure 6 This is a schematic diagram of another advertising bid adjustment device 600 provided in an embodiment of this application. Figure 6 As shown, the deep price adjustment coefficient determination module 503 includes:

[0227] The coefficient adjustment parameter determination unit 601 is used to determine the coefficient adjustment step size and step size adjustment rate based on the reference deep target achievement rate corresponding to the target advertisement, the first deep target achievement rate, and the second deep target achievement rate; the reference deep target achievement rate is the expected achievement rate of the deep target set for the target advertisement;

[0228] The price adjustment coefficient adjustment unit 602 is used to adjust the reference deep price adjustment coefficient based on the coefficient adjustment step size and the step size adjustment rate to obtain the deep price adjustment coefficient corresponding to the target advertisement; the reference deep price adjustment coefficient is the deep price adjustment coefficient used when the conversion price corresponding to the target advertisement was adjusted last time.

[0229] Optional, in Figure 6 Based on the advertising bid adjustment device shown, the coefficient adjustment parameter determination unit 601 specifically determines the coefficient adjustment step size in the following manner:

[0230] When both the first deep target achievement rate and the second deep target achievement rate are greater than the reference deep target achievement rate, and when both the first deep target achievement rate and the second deep target achievement rate are less than the reference deep target achievement rate, the coefficient adjustment step size is determined based on the first deep target achievement rate and the reference deep target achievement rate.

[0231] When the first deep target achievement rate is less than the reference deep target achievement rate and the reference deep target achievement rate is less than the second deep target achievement rate, and when the first deep target achievement rate is greater than the reference deep target achievement rate and the reference deep target achievement rate is greater than the second deep target achievement rate, the difference between the first deep target achievement rate and the reference deep target achievement rate is determined as the first difference, and the difference between the second deep target achievement rate and the reference deep target achievement rate is determined as the second difference. Based on the first difference and the second difference, the coefficient adjustment step size is determined.

[0232] Optional, in Figure 6 Based on the advertising bid adjustment device shown, the coefficient adjustment parameter determination unit 601 specifically determines the coefficient adjustment step size in the following manner:

[0233] When the first gap is greater than the second gap, the coefficient adjustment step size is determined based on the first deep target achievement rate and the reference deep target achievement rate;

[0234] When the first gap is smaller than the second gap, the coefficient adjustment step size is determined based on the second deep target achievement rate and the reference deep target achievement rate.

[0235] Optional, in Figure 6 Based on the advertising bid adjustment device shown, the coefficient adjustment parameter determination unit 601 specifically determines the step adjustment rate in the following manner:

[0236] When both the first deep target achievement rate and the second deep target achievement rate are greater than the reference deep target achievement rate, the step size adjustment rate is determined based on the ratio of the second deep target achievement rate to the first deep target achievement rate and the preset first step length adjustment parameter.

[0237] When both the first deep target achievement rate and the second deep target achievement rate are less than the reference deep target achievement rate, the step size adjustment rate is determined based on the ratio of the first deep target achievement rate to the second deep target achievement rate and the preset second step size adjustment parameter.

[0238] When the first deep target achievement rate is less than the reference deep target achievement rate and the reference deep target achievement rate is less than the second deep target achievement rate, and when the first deep target achievement rate is greater than the reference deep target achievement rate and the reference deep target achievement rate is greater than the second deep target achievement rate, a preset third step size adjustment parameter is determined as the step size adjustment rate.

[0239] Optional, in Figure 6Based on the advertising bid adjustment device shown, the coefficient adjustment parameter determination unit 601 is further used for:

[0240] When both the first deep target achievement rate and the second deep target achievement rate are greater than the reference deep target achievement rate, and when both the first deep target achievement rate and the second deep target achievement rate are less than the reference deep target achievement rate, it is determined whether the step size adjustment rate is less than the lower limit of the step size adjustment rate and whether it is greater than the upper limit of the step size adjustment rate.

[0241] If the step size adjustment rate is less than the lower limit of the step size adjustment rate, then the step size adjustment rate is determined to be the lower limit of the step size adjustment rate.

[0242] If the step size adjustment rate is greater than the upper limit of the step size adjustment rate, then the step size adjustment rate is determined to be the upper limit of the step size adjustment rate.

[0243] Optional, in Figure 5 Based on the advertising bid adjustment device shown, the conversion price adjustment module 504 is specifically used for:

[0244] Obtain the reference deep target achievement rate and the predicted deep target achievement rate corresponding to the target advertisement; the reference deep target achievement rate represents the expected achievement rate of the deep target set for the target advertisement, and the predicted deep target achievement rate is determined by the deep target achievement rate prediction model based at least on the basic information of the target advertisement;

[0245] Based on the reference deep target achievement rate and the predicted deep target achievement rate, the deep price adjustment coefficient is adjusted to obtain the target deep price adjustment coefficient corresponding to the target advertisement;

[0246] The conversion price is adjusted based on the shallow price adjustment coefficient and the target deep price adjustment coefficient to obtain the target conversion price.

[0247] Optional, in Figure 5 Based on the advertising bid adjustment device shown, see [link / reference]. Figure 7 , Figure 7 A schematic diagram of another advertising bid adjustment device 700 provided in an embodiment of this application. (See attached diagram.) Figure 7 As shown, the device further includes: an ECPM determination module 701; the ECPM determination module 701 is used for:

[0248] Obtain the predicted click-through rate (CTR) and predicted conversion rate corresponding to the target advertisement; the predicted CTR is determined by a CTR prediction model based at least on the basic information of the target advertisement, and the predicted conversion rate is determined by a conversion rate prediction model based at least on the basic information of the target advertisement.

[0249] Based on the target conversion price, the predicted click-through rate, and the predicted conversion rate, the revenue per thousand impressions (ECPM) for the target ad is determined.

[0250] Optional, in Figure 5 Based on the advertising bid adjustment device shown, the shallow target includes any of the following:

[0251] Download the application that is being promoted by the targeted advertisement;

[0252] Register as a new user of the application being promoted by the targeted advertisement.

[0253] Optional, in Figure 5 Based on the advertising bid adjustment device shown, the deep target includes:

[0254] Within a preset time period after the shallow goal is achieved, a target behavior is detected through the application; the target behavior includes at least one of the following: login behavior, information viewing behavior, multimedia content playback behavior, and payment behavior.

[0255] The aforementioned advertising bid adjustment device innovatively utilizes a first deep target achievement rate and a second deep target achievement rate, which reflect the deep target achievement status of the target advertisement from different dimensions, to determine a deep price adjustment coefficient. Then, using this deep price adjustment coefficient and the shallow price adjustment coefficient, the conversion price corresponding to the target advertisement is adjusted in a coordinated manner to achieve advertising bid adjustment for the target advertisement. The aforementioned first deep-level goal achievement rate is the proportion of objects that achieve shallow-level goals within the reference historical period that also achieve deep-level goals. It comprehensively reflects the deep-level goal achievement of the target advertisement within the complete reference historical period. The aforementioned second deep-level goal achievement rate is the proportion of objects that achieve shallow-level goals within a reference time interval of the reference historical period that also achieve deep-level goals. It provides a detailed reflection of the deep-level goal achievement of the target advertisement within a specific time interval of the reference historical period. By comprehensively considering the first and second deep-level goal achievement rates, a deep-level price adjustment coefficient can be determined. This ensures that the determined deep-level price adjustment coefficient is more accurate. Adjusting the conversion price of the target advertisement using this deep-level price adjustment coefficient can better match the cost of the target advertisement with the actual achievement of its deep-level goals, thus better matching the cost of the target advertisement with its actual conversion effect, thereby minimizing losses for advertisers or advertising platforms.

[0256] This application also provides a device for adjusting advertising bids. This device may be a terminal device or a server. The terminal device and server provided in this application will be described below from the perspective of hardware implementation.

[0257] See Figure 8 , Figure 8 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. For example... Figure 8 As shown, for ease of explanation, only the parts related to the embodiments of this application are shown. For specific technical details not disclosed, please refer to the method section of the embodiments of this application. The terminal can be any terminal device including mobile phones, tablets, personal digital assistants, point-of-sales (POS) terminals, in-vehicle computers, etc. Taking a computer as an example:

[0258] Figure 8 This is a block diagram illustrating a portion of the structure of a computer associated with the terminal provided in an embodiment of this application. (Reference) Figure 8 The computer includes: a radio frequency (RF) circuit 810, a memory 820, an input unit 830 (including a touch panel 831 and other input devices 832), a display unit 840 (including a display panel 841), a sensor 850, an audio circuit 860 (which can connect to a speaker 861 and a microphone 862), a wireless fidelity (WiFi) module 870, a processor 880, and a power supply 890, etc. Those skilled in the art will understand that... Figure 8 The computer architecture shown does not constitute a limitation on the computer and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0259] The memory 820 can be used to store software programs and modules. The processor 880 executes various computer functions and data processing by running the software programs and modules stored in the memory 820. The memory 820 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one 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 computer (such as audio data, telephone directory, etc.). In addition, the memory 820 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.

[0260] The processor 880 is the control center of the computer, connecting various parts of the computer through various interfaces and lines. It performs various computer functions and processes data by running or executing software programs and / or modules stored in the memory 820, and by calling data stored in the memory 820. Optionally, the processor 880 may include one or more processing units; preferably, the processor 880 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may also not be integrated into the processor 880.

[0261] In this embodiment of the application, the processor 880 included in the terminal also has the following functions:

[0262] Determine the first deep-level goal achievement rate and the second deep-level goal achievement rate corresponding to the target advertisement; the first deep-level goal achievement rate is the proportion of objects that achieve the deep-level goal among those that achieve the shallow-level goal within the reference historical period, and the second deep-level goal achievement rate is the proportion of objects that achieve the deep-level goal among those that achieve the shallow-level goal within the reference time interval of the reference historical period, and achieving the shallow-level goal is a prerequisite for achieving the deep-level goal;

[0263] Based on the conversion price corresponding to the target advertisement, determine the shallow price adjustment coefficient corresponding to the target advertisement; based on the first deep target achievement rate and the second deep target achievement rate, determine the deep price adjustment coefficient corresponding to the target advertisement.

[0264] The conversion price is adjusted based on the shallow adjustment coefficient and the deep adjustment coefficient to obtain the target conversion price corresponding to the target advertisement.

[0265] Optionally, the processor 880 is further configured to execute steps of any implementation of the advertising bid adjustment method provided in the embodiments of this application.

[0266] See Figure 9 , Figure 9This is a schematic diagram of the structure of a server 900 provided in an embodiment of this application. The server 900 can vary significantly due to different configurations or performance, and may include one or more central processing units (CPUs) 922 (e.g., one or more processors) and memory 932, and one or more storage media 930 (e.g., one or more mass storage devices) for storing application programs 942 or data 944. The memory 932 and storage media 930 can be temporary or persistent storage. The program stored in the storage media 930 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 922 may be configured to communicate with the storage media 930 and execute the series of instruction operations in the storage media 930 on the server 900.

[0267] Server 900 may also include one or more power supplies 926, one or more wired or wireless network interfaces 950, one or more input / output interfaces 958, and / or one or more operating systems, such as Windows Server. TM Mac OS X TM Unix TM Linux TM FreeBSD TM etc.

[0268] The steps performed by the server in the above embodiments can be based on this Figure 9 The server structure shown.

[0269] CPU 922 is used to perform the following steps:

[0270] Determine the first deep-level goal achievement rate and the second deep-level goal achievement rate corresponding to the target advertisement; the first deep-level goal achievement rate is the proportion of objects that achieve the deep-level goal among those that achieve the shallow-level goal within the reference historical period, and the second deep-level goal achievement rate is the proportion of objects that achieve the deep-level goal among those that achieve the shallow-level goal within the reference time interval of the reference historical period, and achieving the shallow-level goal is a prerequisite for achieving the deep-level goal;

[0271] Based on the conversion price corresponding to the target advertisement, determine the shallow price adjustment coefficient corresponding to the target advertisement; based on the first deep target achievement rate and the second deep target achievement rate, determine the deep price adjustment coefficient corresponding to the target advertisement.

[0272] The conversion price is adjusted based on the shallow adjustment coefficient and the deep adjustment coefficient to obtain the target conversion price corresponding to the target advertisement.

[0273] Optionally, CPU 922 can also be used to execute the steps of any implementation of the advertising bid adjustment method provided in the embodiments of this application.

[0274] This application also provides a computer-readable storage medium for storing a computer program that executes any one of the implementation methods of the advertising bid adjustment method described in the foregoing embodiments.

[0275] This application also provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform any of the implementation methods for adjusting advertising bids described in the foregoing embodiments.

[0276] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0277] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0278] The units described as separate components may or may not be physically separate. 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0279] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0280] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing computer programs.

[0281] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0282] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for adjusting advertising bids, characterized in that, The method includes: Determine the first deep-level goal achievement rate and the second deep-level goal achievement rate corresponding to the target advertisement; the first deep-level goal achievement rate is the proportion of objects that achieve the deep-level goal among those that achieve the shallow-level goal within the reference historical period, and the second deep-level goal achievement rate is the proportion of objects that achieve the deep-level goal among those that achieve the shallow-level goal within the reference time interval of the reference historical period, and achieving the shallow-level goal is a prerequisite for achieving the deep-level goal; Based on the conversion price corresponding to the target advertisement, determine the shallow price adjustment coefficient corresponding to the target advertisement; based on the first deep target achievement rate and the second deep target achievement rate, determine the deep price adjustment coefficient corresponding to the target advertisement. The conversion price is adjusted based on the shallow adjustment coefficient and the deep adjustment coefficient to obtain the target conversion price corresponding to the target advertisement; The step of determining the deep-level price adjustment coefficient corresponding to the target advertisement based on the first deep-level target achievement rate and the second deep-level target achievement rate includes: Based on the reference deep target achievement rate, the first deep target achievement rate, and the second deep target achievement rate corresponding to the target advertisement, the coefficient adjustment step size and step size adjustment rate are determined; the reference deep target achievement rate represents the expected achievement rate of the deep target set for the target advertisement; Based on the coefficient adjustment step size and the step size adjustment rate, the reference deep price adjustment coefficient is adjusted to obtain the deep price adjustment coefficient corresponding to the target advertisement; the reference deep price adjustment coefficient is the deep price adjustment coefficient used when the conversion price corresponding to the target advertisement was adjusted last time.

2. The method according to claim 1, characterized in that, The step of determining the coefficient adjustment step size based on the reference deep target achievement rate, the first deep target achievement rate, and the second deep target achievement rate corresponding to the target advertisement includes: When both the first deep target achievement rate and the second deep target achievement rate are greater than the reference deep target achievement rate, and when both the first deep target achievement rate and the second deep target achievement rate are less than the reference deep target achievement rate, the coefficient adjustment step size is determined based on the first deep target achievement rate and the reference deep target achievement rate. When the first deep target achievement rate is less than the reference deep target achievement rate and the reference deep target achievement rate is less than the second deep target achievement rate, and when the first deep target achievement rate is greater than the reference deep target achievement rate and the reference deep target achievement rate is greater than the second deep target achievement rate, the difference between the first deep target achievement rate and the reference deep target achievement rate is determined as the first difference, and the difference between the second deep target achievement rate and the reference deep target achievement rate is determined as the second difference. Based on the first difference and the second difference, the coefficient adjustment step size is determined.

3. The method according to claim 2, characterized in that, Determining the coefficient adjustment step size based on the first gap and the second gap includes: When the first gap is greater than the second gap, the coefficient adjustment step size is determined based on the first deep target achievement rate and the reference deep target achievement rate; When the first gap is smaller than the second gap, the coefficient adjustment step size is determined based on the second deep target achievement rate and the reference deep target achievement rate.

4. The method according to claim 1, characterized in that, The step size adjustment rate is determined based on the reference deep target achievement rate, the first deep target achievement rate, and the second deep target achievement rate corresponding to the target advertisement, including: When both the first deep target achievement rate and the second deep target achievement rate are greater than the reference deep target achievement rate, the step size adjustment rate is determined based on the ratio of the second deep target achievement rate to the first deep target achievement rate and the preset first step length adjustment parameter. When both the first deep target achievement rate and the second deep target achievement rate are less than the reference deep target achievement rate, the step size adjustment rate is determined based on the ratio of the first deep target achievement rate to the second deep target achievement rate and the preset second step size adjustment parameter. When the first deep target achievement rate is less than the reference deep target achievement rate and the reference deep target achievement rate is less than the second deep target achievement rate, and when the first deep target achievement rate is greater than the reference deep target achievement rate and the reference deep target achievement rate is greater than the second deep target achievement rate, a preset third step size adjustment parameter is determined as the step size adjustment rate.

5. The method according to claim 1 or 4, characterized in that, The method further includes: When both the first deep target achievement rate and the second deep target achievement rate are greater than the reference deep target achievement rate, and when both the first deep target achievement rate and the second deep target achievement rate are less than the reference deep target achievement rate, it is determined whether the step size adjustment rate is less than the lower limit of the step size adjustment rate and whether it is greater than the upper limit of the step size adjustment rate. If the step size adjustment rate is less than the lower limit of the step size adjustment rate, then the step size adjustment rate is determined to be the lower limit of the step size adjustment rate. If the step size adjustment rate is greater than the upper limit of the step size adjustment rate, then the step size adjustment rate is determined to be the upper limit of the step size adjustment rate.

6. The method according to claim 1, characterized in that, The step of adjusting the conversion price based on the shallow price adjustment coefficient and the deep price adjustment coefficient to obtain the target conversion price corresponding to the target advertisement includes: Obtain the reference deep target achievement rate and the predicted deep target achievement rate corresponding to the target advertisement; the reference deep target achievement rate represents the expected achievement rate of the deep target set for the target advertisement, and the predicted deep target achievement rate is determined by the deep target achievement rate prediction model based at least on the basic information of the target advertisement; Based on the reference deep target achievement rate and the predicted deep target achievement rate, the deep price adjustment coefficient is adjusted to obtain the target deep price adjustment coefficient corresponding to the target advertisement; The conversion price is adjusted based on the shallow price adjustment coefficient and the target deep price adjustment coefficient to obtain the target conversion price.

7. The method according to claim 1 or 6, characterized in that, The method further includes: Obtain the predicted click-through rate (CTR) and predicted conversion rate corresponding to the target advertisement; the predicted CTR is determined by a CTR prediction model based at least on the basic information of the target advertisement, and the predicted conversion rate is determined by a conversion rate prediction model based at least on the basic information of the target advertisement. Based on the target conversion price, the predicted click-through rate, and the predicted conversion rate, the revenue per thousand impressions (ECPM) for the target ad is determined.

8. The method according to claim 1, characterized in that, The shallow target includes any of the following: Download the application that is being promoted by the targeted advertisement; Register as a new user of the application being promoted by the targeted advertisement.

9. The method according to claim 8, characterized in that, The deep targets include: Within a preset time period after the shallow goal is achieved, a target behavior is detected through the application; the target behavior includes at least one of the following: login behavior, information viewing behavior, multimedia content playback behavior, and payment behavior.

10. An advertising bid adjustment device, characterized in that, The device includes: The deep goal achievement rate determination module is used to determine the first deep goal achievement rate and the second deep goal achievement rate corresponding to the target advertisement; the first deep goal achievement rate is the proportion of objects that achieve the deep goal among those that achieve the shallow goal within the reference historical period, and the second deep goal achievement rate is the proportion of objects that achieve the deep goal among those that achieve the shallow goal within the reference time interval of the reference historical period, and achieving the shallow goal is a prerequisite for achieving the deep goal; The shallow price adjustment coefficient determination module is used to determine the shallow price adjustment coefficient corresponding to the target advertisement based on the conversion price corresponding to the target advertisement; The deep-level price adjustment coefficient determination module is used to determine the deep-level price adjustment coefficient corresponding to the target advertisement based on the first deep-level target achievement rate and the second deep-level target achievement rate. The conversion price adjustment module is used to adjust the conversion price according to the shallow price adjustment coefficient and the deep price adjustment coefficient to obtain the target conversion price corresponding to the target advertisement; The deep price adjustment coefficient determination module includes: The coefficient adjustment parameter determination unit is used to determine the coefficient adjustment step size and step size adjustment rate based on the reference deep target achievement rate, the first deep target achievement rate, and the second deep target achievement rate corresponding to the target advertisement; the reference deep target achievement rate is the expected achievement rate of the deep target set for the target advertisement; The price adjustment coefficient adjustment unit is used to adjust the reference deep price adjustment coefficient based on the coefficient adjustment step size and the step size adjustment rate to obtain the deep price adjustment coefficient corresponding to the target advertisement; the reference deep price adjustment coefficient is the deep price adjustment coefficient used when the conversion price corresponding to the target advertisement was adjusted last time.

11. The apparatus according to claim 10, characterized in that, The conversion price adjustment module is specifically used for: Obtain the reference deep target achievement rate and the predicted deep target achievement rate corresponding to the target advertisement; the reference deep target achievement rate represents the expected achievement rate of the deep target set for the target advertisement, and the predicted deep target achievement rate is determined by the deep target achievement rate prediction model based at least on the basic information of the target advertisement; Based on the reference deep target achievement rate and the predicted deep target achievement rate, the deep price adjustment coefficient is adjusted to obtain the target deep price adjustment coefficient corresponding to the target advertisement; The conversion price is adjusted based on the shallow price adjustment coefficient and the target deep price adjustment coefficient to obtain the target conversion price.

12. The apparatus according to claim 10, characterized in that, The coefficient adjustment parameter determination unit is specifically used for: When both the first deep target achievement rate and the second deep target achievement rate are greater than the reference deep target achievement rate, and when both the first deep target achievement rate and the second deep target achievement rate are less than the reference deep target achievement rate, the coefficient adjustment step size is determined based on the first deep target achievement rate and the reference deep target achievement rate. When the first deep target achievement rate is less than the reference deep target achievement rate and the reference deep target achievement rate is less than the second deep target achievement rate, and when the first deep target achievement rate is greater than the reference deep target achievement rate and the reference deep target achievement rate is greater than the second deep target achievement rate, the difference between the first deep target achievement rate and the reference deep target achievement rate is determined as the first difference, and the difference between the second deep target achievement rate and the reference deep target achievement rate is determined as the second difference. Based on the first difference and the second difference, the coefficient adjustment step size is determined.

13. A device, characterized in that, The device includes a processor and a memory; The memory is used to store computer programs; The processor is configured to execute the advertising bid adjustment method according to any one of claims 1 to 9 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 for performing the advertising bid adjustment method according to any one of claims 1 to 9.

15. A computer program product, characterized in that, The computer program product includes computer instructions, which are executed by the processor of the computer device to cause the computer device to perform the advertising bid adjustment method according to any one of claims 1 to 9.

Images