A method and system for pushing advertisement investment information for attracting investment

Abstract

The application discloses a kind of investment promotion advertisement delivery information push method and system, belong to advertisement delivery technical field, including: time period value quantization, a week is divided into each unit time period, through the active degree coefficient and decision conversion coefficient double-dimension weighting, calculate each time period comprehensive value coefficient;Time budget allocation, based on comprehensive value screening high, low value time period, combined with total budget to complete time precision budget allocation;Reach frequency control, fitting conversion utility and negative utility function, to maximize net utility to solve optimal reach upper limit, generate delivery hard constraint rule;Differentiated reach adjustment, based on user full-link behavior calculation intention score and stratification, generate differentiated adaptive reach strategy, the application effectively improves advertisement budget utilization and user conversion efficiency, balance conversion effect and brand risk, adapt to the fine delivery demand of investment promotion scene.

Description

Technical Field

[0001] This invention relates to the field of advertising technology, and more specifically, to a method and system for pushing advertising information to attract investment and customers. Background Technology

[0002] With the rapid development of the digital marketing industry, online advertising has become a core customer acquisition channel for various scenarios such as investment promotion, attracting customers to physical stores, and expanding business services. Compared with brand exposure advertising, investment promotion advertising has the characteristics of a long conversion chain, high average order value, long user decision-making cycle, and high sensitivity of conversion effect to the accuracy of the advertising. Its core goal is to achieve efficient use of the advertising budget, improve user conversion efficiency, and controllability of negative brand effects, which places extremely high demands on the ability to finely manage the entire advertising process.

[0003] Currently, industry optimization technologies for online advertising still have the following technical shortcomings in the specific scenario of attracting investment and customers: The level of refinement in time-segment value assessment and budget allocation is insufficient, resulting in low budget utilization efficiency. Existing technologies often use a uniform daily budget allocation in the time-segment budget allocation stage of advertising campaigns, failing to fully identify the differentiated characteristics of user activity scale and conversion decision-making ability in different time periods. The lack of a quantitative control mechanism for reach frequency makes it impossible to balance conversion effect and negative brand effect. Existing technologies in the advertising reach stage mostly focus on front-end indicators such as exposure and click volume as the core objectives, lacking a refined control and quantitative evaluation system for the reach frequency of individual users. On the one hand, no fitting model has been established between reach frequency and user conversion effect, making it impossible to identify the marginal benefit inflection point of reach frequency on conversion. This can easily lead to problems such as insufficient reach frequency causing failure in user decision guidance, or excessive reach causing diminishing marginal conversion benefits. On the other hand, the negative feedback effects of excessive outreach, such as user blocking, reporting, and brand aversion, are not taken into account. There is a lack of quantitative evaluation and constraint mechanisms for negative effects. It is impossible to solve the optimal reach limit for a single user throughout the entire cycle, making it difficult to maximize the net utility of the campaign. Especially in the long decision-making cycle of attracting merchants and customers, unrestrained repeated outreach is very likely to cause user resistance. Not only will it fail to improve conversion rates, but it will also cause irreversible damage to brand reputation and further increase customer acquisition costs.

[0004] In summary, a method and system for pushing advertising information to attract investment and customers is proposed. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a method and system for pushing information on investment promotion and customer attraction advertising.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method for pushing out advertising information to attract investment and customers includes: Time Period Value Quantification: The past week is divided into 168 consecutive time periods, each with a unique number t. For each time period t, the activity coefficient and decision conversion coefficient of the advertising platform are calculated. Based on the dual-dimensional coefficients and preset weights, the comprehensive value coefficient of the advertising platform for each time period in the past week is output. Time-based budget allocation: Based on the comprehensive value coefficient of the advertising platform in each time period over the past week, high-value and low-value time periods are filtered using pre-edited filtering rules. The budget for each time period in the coming week is then allocated by combining the results of the filtering rules with the total pre-invested advertising budget within the period. Reach frequency control: Based on the full-link data of advertising on the advertising platform over the past week, fit the conversion utility function of reach frequency and conversion effect, and the negative utility function of reach frequency and brand negative feedback. By maximizing net utility, solve for the optimal reach limit for a single user throughout the entire cycle and the reach limit for a single day, and generate hard constraint rules for advertising. Differentiated outreach adjustment: Based on the user intention level corresponding to different events, and combined with real-time user behavior data, calculate the intention score of a single user. Based on the intention score, complete the standardized stratification of the user decision-making stage. Under the hard constraint rules of outreach, generate differentiated adaptive outreach strategies for different user strata for the next week.

[0007] Specifically, the logic for calculating the two-dimensional coefficients; Activity coefficient of time period t ; in The number of unique active users of the target customer group within time period t; Of the 168 time periods in the entire cycle, The maximum value; Decision conversion coefficient ; In the formula, The return on investment (ROI) for advertising within time period t; The user conversion rate within time period t; This represents the average production ratio over the entire cycle of 168 time periods. This represents the average conversion rate over the entire cycle of 168 time periods.

[0008] Specifically, the logic for calculating the comprehensive value coefficient; The comprehensive value coefficient for each time period t is obtained by comprehensively weighting the activity coefficient and decision conversion coefficient of time period t with preset weights.

[0009] Specifically, the selection logic for high-value and low-value time periods; The 168 time periods were sorted in descending order of comprehensive value coefficient; then, the high-value time periods were selected. Calculate the percentage of cumulative value ; Let be the comprehensive value coefficient for the i-th time period; when the cumulative value ratio is higher than the preset reference value ratio, the corresponding first k time periods are the set of high-value time periods; and the remaining time periods are the set of low-value time periods.

[0010] Specifically, the logic behind determining the budget allocation for each period in the coming week; Multiply the total pre-invested advertising budget within the period by the cumulative value percentage to obtain the total budget for high-value periods; remove the total budget for high-value periods and use the remainder as the total budget for low-value periods; Precise budget allocation for a single time period; Budget allocation within high-value periods is based on the percentage of value scores; budget allocation for each period within the high-value period set. Through formula Calculated; Total budget for high-value periods, For high-value time periods; S This represents the comprehensive value coefficient of each time period within the set of high-value time periods; Distribute evenly within low-value periods.

[0011] Specifically, the fitting logic for the transformation utility function and the negative utility function; Define the cumulative effective reach frequency of a single user over the past week as the independent variable f; Conversion utility function fitting: Conversion utility U(f) is the cumulative conversion probability of a user after f effective triggers. A Logistic growth model is used for fitting, and the function expression is: ; Where K is the conversion probability saturation value, a is the conversion growth rate coefficient, and b is the conversion growth inflection point; Negative utility function fitting: Negative utility N(f) is the probability of negative feedback after reaching f users effectively. An exponential growth model is used for fitting, and the function expression is: ; in Based on the negative feedback rate, This represents the negative feedback growth coefficient, and is greater than 0. The frequency of negative feedback growth threshold.

[0012] Specifically, the logic for generating hard constraint rules for ad placement; Optimal trigger limit solution: Construct the net utility function NU(f), and calculate it using the following formula: ;in The negative utility weight is preset; Based on the first-order condition for maximizing net utility, the cumulative reach frequency corresponding to the maximum net utility is obtained. Simultaneously calculate the maximum trigger limit for a single user per day, then... The result is obtained by multiplying by a preset proportional coefficient; The maximum reach limit for the entire period and the maximum reach limit for a single day will be used as mandatory constraints for ad placement.

[0013] Specifically, the logic for calculating a single user's intention score; Based on the user intent level corresponding to different events, construct a full-link behavioral event set E={c1,c2,c3,c4} from low to high. Where c1 is a valid reach event, c2 is a valid click event, c3 is a deep stay event where the continuous stay time on the landing page exceeds the preset deep stay time, and c4 is a lead generation behavior event. Construct an m×n order user behavior matrix Where m is the number of pre-screened target customers, and n=4 is the number of behavioral events. Let be the number of times the j-th action event of the i-th user occurs; The user's intention score is obtained by combining the number of occurrences of different behavioral events for each user with the preset intention weights corresponding to different behavioral events.

[0014] Specifically, the logic for generating differentiated adaptive outreach strategies; Users are segmented into stages using pre-established intention score-intention type mapping rules; Intent types include no-intent users, low-intent users, medium-intent users, and high-intent users; Pre-editing different initial adaptive outreach strategies for different user target types; After adjusting the hard constraints of ad delivery by inputting them into the initial adaptive outreach strategy, this becomes the differentiated adaptive outreach strategy for each user in the target customer group for the next week.

[0015] A system for pushing information on investment promotion and customer attraction advertising includes: The Time Period Value Quantification Module is used to divide the campaign period into several consecutive time periods, calculate the activity coefficient and decision conversion coefficient for each time period, and calculate the comprehensive value coefficient of each time period based on the dual-dimensional coefficient and preset weights. The time-based budget allocation module is used to filter high-value and low-value time periods based on the comprehensive value coefficient of each time period, and to complete the allocation of advertising budget for each time period in combination with the total budget of the campaign period. The frequency control module is used to fit the conversion utility function of frequency of reach and conversion effect, and the negative utility function of frequency of reach and negative feedback of brand based on standardized datasets. Based on the maximization of net utility, it solves the optimal reach limit for a single user throughout the whole cycle and the reach limit for a single day, and generates hard constraint rules for ad delivery reach. The differentiated outreach adjustment module is used to calculate the intention score of a single user based on the user's full-link behavioral events, complete the user decision-making stage stratification, and generate differentiated adaptive outreach strategies for different user strata under the hard constraint rules of outreach.

[0016] The technical effects and advantages of this invention are as follows: This invention constructs a dual-dimensional evaluation model of activity coefficient and decision conversion coefficient through the time period value quantification process. It accurately quantifies the comprehensive value of each time period within a week, and then selects high-value time periods by cumulative value ratio. The total budget is tilted towards high-value time periods, and the budget for each time period is accurately allocated according to the value ratio of high-value time periods. Low-value time periods are evenly allocated as a safety net. This completely solves the drawbacks of traditional extensive budget allocation, avoids the budget waste of high-traffic, low-conversion time periods, and adapts to the characteristics of mismatch between user activity and conversion peaks in merchant recruitment and customer acquisition scenarios. It achieves accurate matching of budget and time period value and maximizes the conversion value of each budget. Based on historical full-link data, this invention constructs a net utility function and solves for the optimal reach limit for a single user throughout the entire cycle and on a single day, forming a hard constraint rule for ad placement. This not only solves the problem of user decision-making guidance failure caused by insufficient reach frequency, but also prevents diminishing marginal conversion benefits, user resistance, and damage to brand reputation caused by excessive reach. It is suitable for the long decision-making cycle of merchant recruitment and customer acquisition, and achieves the global optimization of the overall net utility of ad placement while maximizing the conversion pull effect. This invention constructs a user behavior matrix based on behavioral events throughout the entire customer acquisition and marketing process. It calculates individual user intent scores by combining intent weights, achieving standardized stratification of the user decision-making stage. Under hard-constraint rules for reaching users, it develops differentiated adaptive outreach strategies for users at different intent levels. This breaks the limitations of traditional homogeneous outreach, preventing ineffective campaigns to users with no intent, providing gradual brand and product awareness guidance to low- to medium-intent users, and tackling in-depth conversion for high-intent users. Simultaneously, it forms a closed-loop, end-to-end campaign system that accurately adapts to the needs of users throughout the entire decision-making process, significantly reducing customer acquisition costs and improving final conversion efficiency. Attached Figure Description

[0017] Figure 1 This is a flowchart of a method for pushing advertising information to attract investment and customers according to the present invention; Figure 2 This is a schematic diagram of an advertising information push system for attracting investment and customers, as described in this invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1 like Figure 1 As shown, a method for pushing advertising information to attract investment and customers is as follows: Metadata definition: Collect full-link data of advertising on the target platform and platform behavior data of the target customer group within the week prior to the current time. The collected data is deduplicated, de-cheated, and missing values ​​are filled in to form a standardized dataset, providing a valid data foundation for subsequent calculations.

[0020] The entire advertising delivery data includes effective reach events, effective click events, lead generation behavior events, final conversion events, and negative feedback events; Effective reach events: Exposure events in which the ad creative is fully displayed in the user's visible area and the continuous display duration is longer than the preset reference display duration; the reference display duration is, for example, 3 seconds; excluding invalid exposures such as cheating, duplication, background loading, and those that are not noticed by the user; Valid click event: A user's active click on the ad creative, and the duration of the user's stay on the landing page after clicking is longer than the preset reference duration; the reference duration can be set to 1 second; erroneous clicks, fraudulent clicks, and duplicate clicks are excluded; Lead generation behavior event: The user completes a business recruitment or consumption form on the landing page and submits it, and the form contains valid contact information; Final conversion event: The user completes the franchise agreement and pays the deposit, or the user completes an in-store purchase and pays for the order; Negative feedback events: negative behaviors such as user disinterest in the ad, blocking, reporting, and leaving negative reviews; The platform behavior data of the target customer group includes basic customer attribute data and platform active behavior data at all times; Customer basic attribute data includes user anonymous unique identifier, age, gender, permanent residence, occupation, income level, education level, marital status, device type, and network environment; The platform's real-time activity data includes user anonymous unique identifiers, daily platform login time, online duration by time period, number of APP launches, activity status by time period, weekly / monthly activity frequency, and frequently active periods.

[0021] Time-based value quantification: The past week is divided into 168 consecutive time units; that is, each time unit is 1 hour; each time unit is assigned a unique number t. For each unit time period t, calculate the activity coefficient and decision conversion coefficient of the advertising platform. Based on the dual-dimensional coefficient and preset weights, output the comprehensive value coefficient of the advertising platform for each unit time period over the past week. Specifically: The activity coefficient represents the scale of platform activity of the target customer group within time period t; The system counts the number of unique active users of the target customer group within the statistical period t, identifies the highest value of the number of unique active users in the 168 unit time periods of the entire period, calculates the ratio between the number of unique active users in period t and the highest value, and outputs the activity coefficient of the target customer group in period t. The calculation formula is as follows: Activity coefficient of time period t ; in The number of unique active users of the target customer group within time period t; Of the 168 time periods in the entire cycle, The maximum value; A higher activity coefficient indicates higher user activity within the target time period.

[0022] The decision conversion coefficient represents the advertising conversion capability of time period t, distinguishing between ineffective activity and effective conversion; Decision conversion coefficient ; In the formula, The return on investment (ROI) for advertising within time period t is obtained by dividing the total revenue generated from conversions within time period t by the total cost of advertising. The user conversion rate within time period t is obtained by dividing the number of users who ultimately converted within time period t by the number of unique users effectively reached. This represents the average production ratio over the entire cycle of 168 time periods. This represents the average conversion rate over the entire cycle of 168 time periods. To clarify, the criteria for determining effective reach to independent users are statistically determined based on the aforementioned criteria for effective reach events; the criteria for determining final conversion users are statistically determined based on the aforementioned criteria for final conversion events.

[0023] The higher the decision conversion coefficient, the higher the conversion rate of advertising within the target time period.

[0024] The comprehensive value coefficient for each time period t is obtained by combining the activity coefficient and decision conversion coefficient of time period t with preset weights. The formula is expressed as follows: ;in and These represent the preset weight coefficients, and their sum is one; Supplementary Explanation and The setting reference logic; Based on the Analytic Hierarchy Process (AHP), a consistency judgment matrix is ​​constructed. In the current scenario (business attraction), the priority of decision conversion capability is set higher than that of platform activity capability, and a second-order judgment matrix is ​​constructed: ; Matrix row / column correspondence: activity level Decision Transformation A value of 1 / 3 indicates that activity level is significantly less important than decision conversion. After normalizing the feature vectors, the activity weights are obtained. =0.25, decision transformation weight =0.75.

[0025] Time-based budget allocation: Based on the comprehensive value coefficient of the advertising platform in each time period over the past week, high-value and low-value time periods are filtered using pre-edited filtering rules. The budget for each time period in the coming week is then allocated by combining the results of the filtering rules with the total pre-invested advertising budget within the period. Specifically: The 168 time periods were arranged in descending order of comprehensive value coefficient; satisfying... .

[0026] Then, high-value time periods are selected by... Calculate the percentage of cumulative value ; This is the comprehensive value coefficient for the i-th time period; when the cumulative value ratio is higher than the preset reference value ratio; the reference value ratio can be preset to 80% and dynamically adjusted by technical personnel; the corresponding first k time periods are the set of high-value time periods; the remaining time periods are the set of low-value time periods; Additional notes: The mandatory constraint k / 168 ≤ the preset constraint ratio (which can be set to 20%).

[0027] Multiply the total pre-investment advertising budget within the period by the cumulative value percentage to obtain the total budget for the high-value period; The remaining budget after deducting the total budget for high-value periods is used as the total budget for low-value periods. After determining the total budget for high-value periods and low-value periods, precise budget allocation is carried out for each period. The budget allocation within high-value periods is based on the percentage of value scores, and the budget allocation for each period within the set of high-value periods is as follows: Through formula Calculated; Total budget for high-value periods, For high-value time periods; S This represents the comprehensive value coefficient of each time period within the set of high-value time periods; The budget is evenly distributed within low-value periods, which is obtained by dividing the total budget for low-value periods by the total number of periods within the low-value period set.

[0028] Reach frequency control: Based on the full-link data of advertising on the advertising platform over the past week, fit the conversion utility function of reach frequency and conversion effect, and the negative utility function of reach frequency and brand negative feedback. By maximizing net utility, solve for the optimal reach limit for a single user throughout the entire cycle and the reach limit for a single day, and generate hard constraint rules for advertising. Specifically: The cumulative effective reach frequency of a single user in the past week is defined as the independent variable f, where f is a positive integer and ≥1. The statistical rule is the cumulative number of effective reach events of a single user in the past week, and the statistical granularity is the anonymous unique identifier of the user.

[0029] Conversion utility function fitting: Conversion utility U(f) is the cumulative conversion probability of a user after f effective triggers. A Logistic growth model is used for fitting, and the function expression is: ; Where K is the conversion probability saturation value, a is the conversion growth rate coefficient, and b is the conversion growth inflection point; Additional explanation, The conversion probability saturation value is the theoretical upper limit of user conversion rate in the scenario of attracting and retaining customers. That is, the limit that the conversion rate cannot be broken when the number of touches increases infinitely. Calculate the user conversion rate corresponding to each frequency f. The conversion is based on the user completing the final conversion event. The user conversion rate is obtained by dividing the number of converted users reached f times by the total number of users reached f times. Monitor the trend of conversion rate changes. When f continues to increase and the user conversion rate no longer increases and is within a preset fixed value range, the conversion rate at this time is taken as K. The final value of K can be determined by fitting and optimizing using the least squares method. The conversion rate increase coefficient is the efficiency of user conversion rate growth for each additional effective reach. The larger the value, the stronger the pull effect of reach on conversion. By using f as the independent variable and conversion rate as the dependent variable, a fitting dataset is constructed, and the optimal value of a is obtained by sampling least squares fitting. The constraint condition is: a > 0.

[0030] The inflection point of transformation growth is the frequency at which the marginal transformation utility reaches its maximum value. It is the critical point for dividing the intervals of increasing / decreasing marginal utility. It is solved synchronously with a by least squares fitting. The fitting result verification: when f=b, the first derivative (marginal utility) of the transformation utility function reaches its maximum value. The constraint condition is that b is a positive integer.

[0031] Negative utility function fitting: Negative utility N(f) is the probability of negative feedback after reaching f users effectively. An exponential growth model is used for fitting, and the function expression is: ; in Based on the negative feedback rate, This represents the negative feedback growth coefficient, and is greater than 0. The frequency of negative feedback increases; Additional explanation, The baseline negative feedback rate is the negative feedback rate of users who are effectively reached only once throughout the entire period. It is the benchmark value for negative feedback. The baseline negative feedback rate is obtained by counting the total number of users who are effectively reached only once and the number of users who trigger negative feedback events in this group. To determine the critical point at which the user negative feedback rate enters an exponentially rapid growth range after accumulating more than a certain frequency, the frequency-specific negative feedback rate corresponding to each frequency f is calculated. The changing trend of the frequency-specific negative feedback rate is observed to locate the critical frequency at which the growth slope changes from gentle to steep (using the second-order difference, i.e., the rate of change of the slope; when the second-order difference reaches the preset reference rate of change threshold, it is determined that the growth slope changes from gentle to steep). This is used as the negative feedback growth threshold frequency, with the constraint condition being a positive integer. The negative feedback growth coefficient represents the rate of increase in the negative feedback rate after overexposure. The larger the value, the faster the brand damage from overexposure increases. By taking the natural logarithm of both sides of the negative utility function, the exponential model is transformed into a linear model. With f as the independent variable, As the dependent variable, the least squares method is used for linear fitting, and the slope obtained is used as the negative feedback growth coefficient. Constraints: >0.

[0032] Additional notes: The least squares method was used to fit the data based on the standardized dataset, and a goodness of fit ≥ 0.9 was considered acceptable; e is the natural constant (Euler number), a fixed irrational constant in mathematics.

[0033] The optimal trigger limit is then determined by constructing a net utility function NU(f), which represents the overall delivery benefit when the cumulative number of touches is f. The calculation formula is: ;in The default negative utility weight is 2. Based on the first-order condition for maximizing net utility The cumulative reach frequency corresponding to the maximum net utility is obtained by solving the problem. This is the maximum effective trigger limit within a single user's cycle; If the maximum trigger limit for a single user per day is calculated synchronously, then... The result is obtained by multiplying by a preset proportional coefficient; The preset ratio coefficient is set to 0.3, meaning that the number of touches per day will not exceed 30% of the upper limit of the entire cycle, and the calculation result will be rounded down.

[0034] The maximum reach limit for the entire period and the maximum reach limit for a single day will be used as mandatory constraints for ad placement. When the number of times a single user is reached, all channels will immediately stop reaching them.

[0035] Differentiated outreach adjustment: Based on the user intention level corresponding to different events, and combined with real-time user behavior data, calculate the intention score of a single user. Based on the intention score, complete the standardized stratification of the user decision-making stage. Under the hard constraint rules of outreach, generate differentiated adaptive outreach strategies for different user strata for the next week. Specifically: Based on the user intention level corresponding to different events, which can be divided into 1-5 levels, the higher the level, the higher the degree of user intention corresponding to the event; construct the full-link behavioral event set E={c1,c2,c3,c4} from low to high. Where c1 is a valid reach event, c2 is a valid click event, c3 is a deep stay event where the continuous stay time on the landing page exceeds the preset deep stay time (30s), and c4 is a lead generation behavior event; Additional notes: When a user triggers a negative feedback event, they are directly marked as an unintentional user.

[0036] Construct an m×n order user behavior matrix Where m is the number of pre-screened target customers, and n=4 is the number of behavioral events. Let be the number of times the j-th action event of the i-th user occurs; Perform min-max normalization on the behavior matrix to obtain the normalized matrix. The normalization formula is as follows: ,in and These are the minimum and maximum values ​​of the total number of occurrences of the j-th behavioral event across the entire sample, respectively.

[0037] The intention score of a single user is obtained by combining the number of occurrences of different behavioral events for each user with the preset intention weights corresponding to different behavioral events. That is, using formulas Calculate the single user intention score ;in This is the standardized value of the number of occurrences of the j-th behavioral event for a single user. The preset intention weights for each behavioral event are summed to one.

[0038] Users are segmented into stages using pre-established intention score-intention type mapping rules; Intent types include no-intent users, low-intent users, medium-intent users, and high-intent users; This means setting four sets of intention score intervals, with each set of intention score intervals corresponding to an intention type; the higher the intention score, the higher the probability of matching a high intention.

[0039] Pre-editing different initial adaptive outreach strategies for different user target types; For example, Uninterested users: Immediately stop all-channel advertising outreach and permanently remove them from the target customer group package for the current campaign period; Low-intent / Medium-intent users: Set the optimal reach interval to 72 hours for low-intent users and push brand awareness content; set the optimal reach interval to 48 hours for medium-intent users and push product / franchise advantages and store case studies content; the weekly reach frequency shall not exceed the set ratio of the total reach limit for the entire cycle. High-intent users: Preset the frequency of contact increase coefficient, shorten the contact interval to 24 hours, and push high-conversion content such as franchise policies, profit models, and exclusive offers until the user completes the final conversion or reaches the contact limit.

[0040] After adjusting the hard constraints of the ad delivery strategy by inputting them into the initial adaptive outreach strategy, it becomes the differentiated adaptive outreach strategy for the target customer group per user for the next week. Example

[0041] Please see Figure 2 As shown, based on Embodiment 1 of this application, a method for pushing information on investment promotion and customer attraction advertising is provided. Embodiment 2 of this application proposes a system for pushing information on investment promotion and customer attraction advertising. Embodiment 2 is merely a preferred embodiment of Embodiment 1, and its implementation will not affect the individual implementation of Embodiment 1.

[0042] Specifically, Embodiment 2 of this application provides a business attraction and customer acquisition advertising information push system, comprising: The Time Period Value Quantification Module is used to divide the campaign period into several consecutive time periods, calculate the activity coefficient and decision conversion coefficient for each time period, and calculate the comprehensive value coefficient of each time period based on the dual-dimensional coefficient and preset weights. The time-based budget allocation module is used to filter high-value and low-value time periods based on the comprehensive value coefficient of each time period, and to complete the allocation of advertising budget for each time period in combination with the total budget of the campaign period. The frequency control module is used to fit the conversion utility function of frequency of reach and conversion effect, and the negative utility function of frequency of reach and negative feedback of brand based on standardized datasets. Based on the maximization of net utility, it solves the optimal reach limit for a single user throughout the whole cycle and the reach limit for a single day, and generates hard constraint rules for ad delivery reach. The differentiated outreach adjustment module is used to calculate the intention score of a single user based on the user's full-link behavioral events, complete the user decision-making stage stratification, and generate differentiated adaptive outreach strategies for different user strata under the hard constraint rules of outreach.

[0043] The above formulas are all dimensionless calculations. Dimensionless calculations can be performed using various methods such as standardization, which will not be elaborated here. The formulas are derived from software simulations based on a large amount of collected data, and the preset parameters in the formulas can be set by those skilled in the art according to the actual situation.

[0044] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, ATA hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state ATA hard disk.

[0045] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0046] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0047] 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 through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0048] 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; 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, depending on actual needs.

[0049] In addition, 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.

[0050] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they 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 a portion 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 various media capable of storing program code, such as USB flash drives, portable ATA hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0051] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for pushing advertising information to attract investment and customers, characterized in that, include: Time period value quantification: The past week is divided into 168 consecutive time periods, each of which is assigned a unique number t; For each unit time period t, calculate the activity coefficient and decision conversion coefficient of the advertising platform. Based on the dual-dimensional coefficient and preset weights, output the comprehensive value coefficient of the advertising platform for each unit time period in the past week. Time-based budget allocation: Based on the comprehensive value coefficient of the advertising platform in each time period over the past week, high-value and low-value time periods are filtered using pre-edited filtering rules. The budget for each time period in the coming week is then allocated by combining the results of the filtering rules with the total pre-invested advertising budget within the period. Reach frequency control: Based on the full-link data of advertising on the advertising platform over the past week, fit the conversion utility function of reach frequency and conversion effect, and the negative utility function of reach frequency and brand negative feedback. By maximizing net utility, solve for the optimal reach limit for a single user throughout the entire cycle and the reach limit for a single day, and generate hard constraint rules for advertising. Differentiated outreach adjustment: Based on the user intention level corresponding to different events, and combined with real-time user behavior data, calculate the intention score of a single user. Based on the intention score, complete the standardized stratification of the user decision-making stage. Under the hard constraint rules of outreach, generate differentiated adaptive outreach strategies for different user strata for the next week.

2. The method for pushing advertising information to attract investment and customers according to claim 1, characterized in that: Two-dimensional coefficient calculation logic; Activity coefficient of time period t ; in The number of unique active users of the target customer group within time period t; Of the 168 time periods in the entire cycle, The maximum value; Decision conversion coefficient ; In the formula, The return on investment (ROI) for advertising within time period t; The user conversion rate within time period t; This represents the average production ratio over the entire cycle of 168 time periods. This represents the average conversion rate over the entire cycle of 168 time periods.

3. The method for pushing advertising information to attract investment and customers according to claim 2, characterized in that: The logic for calculating the comprehensive value coefficient; The comprehensive value coefficient for each time period t is obtained by comprehensively weighting the activity coefficient and decision conversion coefficient of time period t with preset weights.

4. The method for pushing advertising information to attract investment and customers according to claim 3, characterized in that: The logic behind filtering high-value and low-value time periods; The 168 time periods were sorted in descending order of comprehensive value coefficient; then, the high-value time periods were selected. Calculate the percentage of cumulative value ; Let be the comprehensive value coefficient for the i-th time period; when the cumulative value ratio is higher than the preset reference value ratio, the corresponding first k time periods are the set of high-value time periods; and the remaining time periods are the set of low-value time periods.

5. The method for pushing advertising information for attracting investment and customers according to claim 4, characterized in that: The logic behind budget allocation for different time periods in the coming week; Multiply the total pre-invested advertising budget within the period by the cumulative value percentage to obtain the total budget for high-value periods; remove the total budget for high-value periods and use the remainder as the total budget for low-value periods; Precise budget allocation for a single time period; Budget allocation within high-value periods is based on the percentage of value scores; budget allocation for each period within the high-value period set. Through formula Calculated; Total budget for high-value periods, For high-value time periods; S This represents the comprehensive value coefficient of each time period within the set of high-value time periods; Distribute evenly within low-value periods.

6. The method for pushing advertising information for attracting investment and customers according to claim 1, characterized in that: Fitting logic for transformation utility function and negative utility function; Define the cumulative effective reach frequency of a single user over the past week as the independent variable f; Conversion utility function fitting: Conversion utility U(f) is the cumulative conversion probability of a user after f effective triggers. A Logistic growth model is used for fitting, and the function expression is: ； Where K is the conversion probability saturation value, a is the conversion growth rate coefficient, and b is the conversion growth inflection point; Negative utility function fitting: Negative utility N(f) is the probability of negative feedback after reaching f users effectively. An exponential growth model is used for fitting, and the function expression is: ； in Based on the negative feedback rate, This represents the negative feedback growth coefficient, and is greater than 0. The frequency of negative feedback growth threshold.

7. The method for pushing advertising information for attracting investment and customers according to claim 6, characterized in that: The logic for generating hard constraint rules for ad placement; Optimal trigger limit solution: Construct the net utility function NU(f), and calculate it using the following formula: ;in The negative utility weight is preset; Based on the first-order condition for maximizing net utility, the cumulative reach frequency corresponding to the maximum net utility is obtained. ; If the maximum trigger limit for a single user per day is calculated synchronously, then... The result is obtained by multiplying by a preset proportional coefficient; The maximum reach limit for the entire period and the maximum reach limit for a single day will be used as mandatory constraints for ad placement.

8. The method for pushing advertising information for attracting investment and customers according to claim 7, characterized in that: Logic for calculating single-user intention scores; Based on the user intent level corresponding to different events, construct a full-link behavioral event set E={c1,c2,c3,c4} from low to high. Where c1 is a valid reach event, c2 is a valid click event, c3 is a deep stay event where the continuous stay time on the landing page exceeds the preset deep stay time, and c4 is a lead generation behavior event. Construct an m×n order user behavior matrix Where m is the number of pre-screened target customers, and n=4 is the number of behavioral events. Let be the number of times the j-th action event of the i-th user occurs; The user's intention score is obtained by combining the number of occurrences of different behavioral events for each user with the preset intention weights corresponding to different behavioral events.

9. The method for pushing advertising information for attracting investment and customers according to claim 7, characterized in that: Differentiated adaptive outreach strategy generation logic; Users are segmented into stages using pre-established intention score-intention type mapping rules; Intent types include no-intent users, low-intent users, medium-intent users, and high-intent users; Pre-editing different initial adaptive outreach strategies for different user target types; After adjusting the hard constraints of ad delivery by inputting them into the initial adaptive outreach strategy, this becomes the differentiated adaptive outreach strategy for each user in the target customer group for the next week.

10. A system for pushing information on investment promotion and customer attraction advertising, applied to the method for pushing information on investment promotion and customer attraction advertising as described in any one of claims 1-9, characterized in that: The Time Period Value Quantification Module is used to divide the campaign period into several consecutive time periods, calculate the activity coefficient and decision conversion coefficient for each time period, and calculate the comprehensive value coefficient of each time period based on the dual-dimensional coefficient and preset weights. The time-based budget allocation module is used to filter high-value and low-value time periods based on the comprehensive value coefficient of each time period, and to complete the allocation of advertising budget for each time period in combination with the total budget of the campaign period. The frequency control module is used to fit the conversion utility function of frequency of reach and conversion effect, and the negative utility function of frequency of reach and negative feedback of brand based on standardized datasets. Based on the maximization of net utility, it solves the optimal reach limit for a single user throughout the whole cycle and the reach limit for a single day, and generates hard constraint rules for ad delivery reach. The differentiated outreach adjustment module is used to calculate the intention score of a single user based on the user's full-link behavioral events, complete the user decision-making stage stratification, and generate differentiated adaptive outreach strategies for different user strata under the hard constraint rules of outreach.

Images