Rule-based advertisement putting method and terminal

Abstract

The invention provides a rule-based advertisement putting method and terminal, and the method comprises the steps: setting execution strategies corresponding to different conditions with the exposure amount, the click amount and the order output amount as conditions, and carrying out the associated storage of the conditions and the execution strategies as a rule; according to the advertisement updating period of the target advertisement, obtaining a current exposure amount, a current click amount and a current order output amount as matching parameters; obtaining a target condition and a target execution strategy in a target rule corresponding to the target condition according to the matching parameter; and executing the target execution strategy. According to the invention, the advertisement which is poor in current putting effect and needs to change the execution strategy can be accurately screened out, and the advertisement is adjusted according to the execution strategy under the corresponding condition; and the execution strategy is automatically adjusted according to the execution result of the execution strategy in one advertisement updating period, so that automatic adjustment of the target advertisement execution strategy according to the specific execution effect is realized.

Description

Technical Field

[0001] The present invention relates to the field of advertisement delivery, and in particular to a rule-based advertisement delivery method and terminal. Background Art

[0002] With the development of Internet technology, advertising has become an important means for companies to promote their products. However, traditional advertising requires manual logging into the advertising management backend of the corresponding platform on a regular basis to view the advertising data of the store on the platform. These data include but are not limited to key indicators such as the exposure, clicks, advertising costs, and number of orders of advertisements placed on various channels. After viewing these data, advertisers need to analyze the data in combination with their experience in product operations and conduct an in-depth analysis of these data. Based on the analysis results, advertisers need to make corresponding modifications and optimizations to the factors that affect the effectiveness of advertising. This includes but is not limited to the adjustment of advertising bidding and advertising budgets. Through these measures, the purpose of reducing the cost of advertising and increasing the exposure and number of orders of product advertisements can be achieved, thereby obtaining higher sales profits.

[0003] Moreover, in the actual implementation process, the process of obtaining relevant data on Amazon advertising and optimizing the advertising effect after analyzing the data is not a one-time process. During the sales cycle of the product, it is necessary to continuously monitor and adjust the advertising strategy to ensure the maximum advertising effect. This implementation method is not only time-consuming and labor-intensive, but the effect of the optimization plan is extremely dependent on the personal experience of each advertiser. There is no standardized process, which makes it difficult to achieve the optimal advertising effect. At the same time, different advertisements have different analysis and delivery strategies for products, and it is impossible to flexibly analyze and adjust. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an advertisement delivery strategy combination method and a terminal to realize automatic adjustment of advertisement delivery strategies.

[0005] In order to solve the above technical problems, a technical solution adopted by the present invention is: A rule-based advertising method comprises the following steps: Using exposure, clicks, and orders as conditions, set execution strategies corresponding to different conditions, and associate the conditions and execution strategies and store them as rules; According to the advertisement update cycle of the target advertisement, the current exposure, the current click volume and the current order volume are obtained as matching parameters; and according to the matching parameters, the target condition and the target execution strategy in the target rule corresponding to the target condition are obtained; The target execution policy is executed.

[0006] In order to solve the above technical problems, another technical solution adopted by the present invention is: A rule-based advertising delivery terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: Using exposure, clicks, and orders as conditions, set execution strategies corresponding to different conditions, and associate the conditions and execution strategies and store them as rules; According to the advertisement update cycle of the target advertisement, the current exposure, the current click volume and the current order volume are obtained as matching parameters; and according to the matching parameters, the target condition and the target execution strategy in the target rule corresponding to the target condition are obtained; The target execution policy is executed.

[0007] The beneficial effects of the present invention are as follows: exposure, clicks and orders are used as conditions for setting rules, and execution strategies under different conditions are set; the conditions include main tags for judging advertising effects, so that advertisements that are currently not effective and need to change their execution strategies can be accurately screened out, and the advertisements are adjusted according to the execution strategies under corresponding conditions; and an advertising update cycle is set, and the exposure, clicks and orders of the target advertisement are obtained according to the advertising update cycle, and it is determined whether there are matching rules, and the execution strategy can be automatically adjusted according to the execution results of the execution strategy within an advertising update cycle, thereby realizing automatic adjustment of the target advertisement execution strategy according to the specific execution effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A flowchart of a rule-based advertising delivery method according to an embodiment of the present invention; Figure 2 A schematic diagram of a process for obtaining advertisement data according to an embodiment of the present invention; Figure 3 A schematic diagram of a process for formulating and executing rules according to an embodiment of the present invention; Figure 4 A schematic diagram of a process of automatically adjusting rules according to execution results according to an embodiment of the present invention; Figure 5 A schematic diagram of the structure of a rule-based advertising delivery terminal according to an embodiment of the present invention; Description of labels: 1. A rule-based advertising delivery terminal; 2. A processor; 3. A memory. DETAILED DESCRIPTION

[0009] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in conjunction with the implementation modes and the accompanying drawings.

[0010] Please refer to Figure 1 , a rule-based advertising delivery method, comprising the steps of: Using exposure, clicks, and orders as conditions, set execution strategies corresponding to different conditions, and associate the conditions and execution strategies and store them as rules; According to the advertisement update cycle of the target advertisement, the current exposure, the current click volume and the current order volume are obtained as matching parameters; and according to the matching parameters, the target condition and the target execution strategy in the target rule corresponding to the target condition are obtained; The target execution policy is executed.

[0011] From the above description, it can be seen that the beneficial effects of the present invention are: taking exposure, clicks and orders as conditions for setting rules, setting execution strategies under different conditions; the conditions include the main tags for judging the advertising effect, so that advertisements that are currently not effective and need to change the execution strategy can be accurately screened out, and the advertisements are adjusted according to the execution strategies under the corresponding conditions; and an advertising update cycle is set, and the exposure, clicks and orders of the target advertisement are obtained according to the advertising update cycle, and it is determined whether there are matching rules. The execution strategy can be automatically adjusted according to the execution results of the execution strategy within an advertising update cycle, thereby realizing automatic adjustment of the target advertisement execution strategy according to the specific execution effect.

[0012] Further, obtaining the target condition according to the matching parameter and the target execution strategy in the target rule corresponding to the target condition includes: The target rule, the target advertisement, the current exposure, the current click volume, and the current order volume are associated and saved as historical execution data; The executing of the target execution strategy comprises: After a preset number of advertisement update cycles, the current new exposure, new click volume, and new order volume of the target advertisement are obtained again, associated with the target rule corresponding to the target advertisement, and saved as update execution data; The execution effect of the target advertisement under the target rule is obtained according to the comparison result between the historical execution data and the updated execution data.

[0013] From the above description, it can be seen that after matching the target execution strategy, the exposure, clicks and orders when the target advertisement and the target execution strategy are matched are retained; and after the target execution strategy is executed, the exposure, clicks and orders are saved again. The execution effect of the target advertisement under the target execution strategy can be obtained by comparing the features before and after execution, thereby judging whether the matched target execution strategy can improve the effect of advertising delivery.

[0014] Furthermore, obtaining the execution effect of the target advertisement under the target rule according to the comparison result between the historical execution data and the updated execution data includes: Obtain a first growth rate of the new exposure compared to the current exposure, a second growth rate of the new click volume equivalent to the current click volume, and a third growth rate of the new order volume equivalent to the current order volume; If the third growth rate is greater than the third preset value, the output execution effect reaches the expected result; If the third growth rate is less than the third preset value, the first growth rate is greater than the first preset value, and the second growth rate is greater than the second preset value, then the output conversion rate does not meet expectations; If the third growth rate is less than the third preset value, the first growth rate is greater than the first preset value and the second growth rate is less than the second preset value, outputting that the keyword does not match; If the third growth rate is less than the third preset value and the first growth rate is less than the first preset value, the output advertisement bidding needs to be adjusted; Otherwise, mark the execution effect as requiring verification.

[0015] From the above description, it can be seen that the ultimate goal of advertising is to promote the sales of goods. Therefore, when judging whether the target rules have achieved the expected effect, the main judgment is whether the number of orders has increased. When the growth in the number of orders does not meet expectations, the specific reasons why the growth in the number of orders has not met expectations are further summarized based on the changes in exposure and click volume. In this way, the reasons why the growth in the number of orders has not met expectations can be directly analyzed and predicted, which is convenient for the subsequent replacement of the target rules matched by the target advertisement or the modification of the target rules themselves.

[0016] Furthermore, it also includes: Obtain all execution effects corresponding to the target rule at a preset interval; The target rule is adjusted according to the expected proportion of the execution effect.

[0017] From the above description, it can be seen that each preset period is based on the target rule as the theme, and all execution effects corresponding to the target rule are exported. It can be seen that if the expected proportion of the execution effect is high, it means that there is no big problem with the setting of the target rule itself, but it is not compatible with some advertisements. There is no need to modify the target rule, and it is only necessary to block the rules with poor effects during the matching process of the corresponding target advertisements; if the expected proportion of the execution effect is low, it means that there is a problem with the setting of the rule itself, and it is necessary to consider adjusting the rule. Different rule adjustment methods can be further set according to the execution effect, so as to realize automatic adjustment of the rules.

[0018] Further, the storing of the conditions and the execution strategies as rules in association with each other includes: Acquire a preset number of combination rules whose conditions are different from each other from the rules; Combining the combination rules into an advertisement delivery strategy to obtain at least one advertisement delivery strategy; The adjusting the target rule according to the expected proportion of the execution effect comprises: If the expected proportion of the execution effect is less than a fourth preset value, obtaining the target advertisement delivery strategy where the target rule is located; A replacement rule having the same condition as the target rule is obtained, and the target rule is replaced with the replacement rule in the target advertisement delivery strategy.

[0019] From the above description, it can be seen that because different rules correspond to advertisements of different product types or different advertising platforms, when there are multiple different execution strategies for the same conditions, different rules with different conditions are obtained to form an advertising delivery strategy, which can be combined according to the execution effects of the rules matched by different types of advertisements or advertisements on different advertising platforms. In this way, when there are corresponding types of products or advertisements on advertising platforms in the future, they can be directly matched in the corresponding advertising delivery strategy, which improves the efficiency of advertising matching and increases the probability that the final execution effect reaches the expected result. At the same time, if the expected proportion of the execution effect of the target rule is low, it means that the target rule is not suitable for the current scenario or advertising type. The corresponding replacement rule is obtained to replace the target rule in the advertising delivery strategy, which can dynamically optimize the advertising delivery strategy.

[0020] Further, the storing of the conditions and the execution strategies as rules in association with each other includes: Arrange the rules according to preset priorities; The adjusting the target rule according to the expected proportion of the execution effect comprises: If the expected proportion of the execution effect is less than a fourth preset value, lowering the priority of the target rule; If the proportion of the execution effects that meet expectations is greater than a fifth preset value, the priority of the target rule is increased.

[0021] From the above description, it can be seen that for situations where different execution strategies exist for the same conditions or situations that may satisfy execution strategies under different conditions, the rules can be sorted and matched according to the pre-set priority, which can avoid the situation where multiple rules are hit and manual selection is required. Furthermore, it can be combined with the execution effect. When the execution effect does not meet the expected proportion too high, the priority of the target rule is lowered, or when the execution effect meets the expected proportion at a high rate, the priority of the target rule is increased. For the rules that continue to be at a lower priority, a second confirmation is performed, thereby realizing dynamic adjustment of the rule execution priority and increasing the probability that the target rule execution result meets the expected result.

[0022] Furthermore, the execution strategy includes: Increase the bid by a fixed percentage, change keywords, and change the delivery channel.

[0023] From the above description, we can see that the execution strategy includes increasing bids, changing keywords and changing delivery channels, which includes the main operational methods of advertising delivery and provides adjustment methods covering different directions for different conditions.

[0024] Further, adjusting the target rule according to the expected proportion of the execution effect includes: If the expected proportion of the execution effect exceeds the preset proportion, the target rule is retained; otherwise, the fixed proportion, keywords or delivery channels of the execution strategy in the target rule are adjusted.

[0025] From the above description, it can be seen that if the proportion of the execution effects corresponding to the rule that achieve the expectations is lower than the preset proportion, it means that there may be problems with the setting of the execution strategy in the rule. The execution strategy should be adjusted accordingly, so as to dynamically optimize the rules themselves and adjust the rules with poor execution effects in time.

[0026] Further, executing the target execution strategy includes: Obtaining an advertisement identifier of the target advertisement, and searching for a request corresponding to the target advertisement in an advertisement platform according to the advertisement identifier; The request is modified according to the target execution policy.

[0027] From the above description, it can be seen that when executing the target strategy, the corresponding request is retrieved in the corresponding advertising platform according to the advertising identifier of the target advertisement, and the request is directly modified, and the specific implementation process of the execution strategy is automated. After the rules are confirmed, the target execution strategy can be automatically executed.

[0028] A rule-based advertising delivery terminal is also provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements each step of the above-mentioned rule-based advertising delivery method when executing the computer program.

[0029] The above-mentioned rule-based advertising delivery method of the present invention can be applied to existing advertising delivery methods, such as advertising delivery on various e-commerce platforms or advertising delivery in search bars, pop-up windows and other locations of various applications, which is explained below through specific implementation methods.

[0030] The process of advertising delivery can be started manually.

[0031] Please refer to Figure 1-4 , Embodiment 1 of the present invention is: A rule-based advertising method comprises the following steps: S1. Taking the exposure, click volume and order volume as conditions, set execution strategies corresponding to different conditions, and associate the conditions and execution strategies and store them as rules.

[0032] In an optional implementation, the execution strategy includes: increasing the bid by a fixed ratio, changing keywords, and changing the delivery channel. Among them, the execution strategy corresponds to the adjustable parameters in the advertisement. The parameters corresponding to an advertisement usually include product content, bid, keywords, and delivery channels; product content is the content that the advertisement needs to display, and the bid is the bid under the corresponding keyword when the advertisement is delivered. The bid will affect the display position of the advertisement under the corresponding keyword; the delivery channel is the way the advertisement is displayed, such as associating in the search box, displaying on the search results page, etc. Further, it can also be applied to other methods. For example, if it is delivered on the homepage, it is usually not through bidding but purchasing a fixed position. In this case, whether to withdraw from the homepage can be used as an adjustable parameter, and the execution strategy corresponds to withdrawing from the homepage or purchasing a homepage position. That is, the execution strategy and the advertisement are in a corresponding relationship. Different types of advertisements can correspond to different execution strategies under the same conditions.

[0033] For product advertising, the main goal is to sell products. The main thing to pay attention to for an advertisement is how many products are sold, that is, the number of orders. However, the most important factor affecting the sale of products is exposure. Only when the product is exposed can it be clicked, and then orders can be placed. Therefore, the setting of rules is mainly based on the number of orders, and combined with exposure and clicks, so that the reasons for the change in the number of orders can be located when changing the advertising strategy. Therefore, it is divided into no exposure and no orders, no clicks and no orders; then for products with orders, ads with few orders but high exposure or click-through rates, and high cost prices, that is, low conversion rates, this type of need to adjust the bidding position, that is, adjust the bidding to adjust the ad position to increase the conversion rate. We set this type of rule as "clicks and orders"; by analogy, there are also "clicks but no orders", which may be that the price competitiveness of the product under the set keywords is insufficient, and the keywords need to be adjusted; or there is exposure but no orders, which means that the product corresponding to the advertisement may not match the keywords of the advertisement, for example, consumers search for baby beds, but the product of double beds has the keyword of baby beds. Classify the rules according to different conditions; after setting all the rules, enter the rules into the system.

[0034] In an optional implementation, rules can also be extracted from recorded manual operations. For example, if there is an advertisement report with a very good exposure rate, but no clicks, the manual operation process will consider whether the keywords of the advertisements are not good. For example, a user wants to buy a baby bed, but the keyword baby bed is placed in the advertisement corresponding to the double bed product on sale. This is obviously inconsistent with the actual user experience. In the manual operation process, the word baby bed is negated, that is, when searching for baby beds, double beds will not appear, which improves the conversion rate; the corresponding rule is clicks <10, exposure >100, replace the keywords, and record the manual operation as a rule. Or clicks <10, exposure <100, indicating that the effect of this placement is not very good, and we need to increase the bid to improve the competitiveness of the advertising ranking, so that the advertisement can be placed in the front to increase exposure.

[0035] In an optional implementation, if it is a newly opened store, there may not be any corresponding data yet. The condition of the rule can be that when there is no exposure, click volume, or order volume, the keywords of the same product in the historical advertisements are obtained for recommendation, and the highest bid value set is output to gain rapid exposure.

[0036] S2. According to the advertisement update cycle of the target advertisement, the current exposure, the current click volume, and the current order volume are obtained as matching parameters; In an optional implementation, please refer to Figure 2, and also includes time synchronization of the acquired data. Because there is a time difference problem for overseas e-commerce platforms, the time is converted to a unified time zone based on the different local times of the stores corresponding to the actual advertisements and the different data return times, which is convenient for analysis. For example, the time difference between European stores and the Eastern Time Zone is 6 hours, and the time difference between American stores and the Eastern Time Zone is as high as 12 hours; if the report time on the e-commerce platform report data is local time, then the local time is converted to the Eastern Time Zone 8, and the data of the same day under the Eastern Time Zone 8 is obtained. Specifically, during the execution process, you can customize the time difference corresponding to stores in different regions, and automatically perform batch conversion of time after obtaining the report.

[0037] S3. Obtain a target condition and a target execution strategy in a target rule corresponding to the target condition according to the matching parameter.

[0038] In an optional implementation manner, S31 is also included after S3.

[0039] S31. Associate and save the target rule, target advertisement, current exposure, current click volume, and current order volume as historical execution data.

[0040] There is no restriction on the execution order of steps S2 and S3, and they can be executed together or one after another.

[0041] S4. Execute the target execution strategy.

[0042] In an optional implementation, S4 includes S401-S402.

[0043] S401: Acquire the advertisement identifier of the target advertisement, and retrieve a request corresponding to the target advertisement in an advertisement platform according to the advertisement identifier.

[0044] S402: Modify the request according to the target execution policy.

[0045] Specifically, each e-commerce platform will provide an open ad modification interface. According to the suggestions in the execution strategy returned by the rules, the ad modification interface can be automatically called to edit the ad, so as to achieve the effect of automatically adjusting the ad. That is, after matching the corresponding rules for the ad, the current bid, keyword, channel and other information corresponding to the ad can be obtained in the system according to the ad's identifier (activity ID / keyword ID / delivery ID). For example, if the target execution strategy in the current target rule is to increase the bid adjustment by 10%, then we can get the filtered ID that meets the conditions, query the original bid in the system, perform calculations, and obtain the calculation results. Then, we can write a request that conforms to the format of the e-commerce platform and modify the ad through the provided ad modification interface.

[0046] In an optional implementation, S41-S42 are also included after S4.

[0047] S41. After a preset number of advertisement update cycles, the current new exposure, new click volume, and new order volume of the target advertisement are obtained again, associated with the target rule corresponding to the target advertisement, and saved as update execution data.

[0048] In an optional implementation, the preset number is 1, that is, the new exposure, new click volume and new order volume are obtained as update execution data every time an advertisement update cycle is compared with the current order volume, current exposure and current click volume before execution to refine the analysis data. That is, the execution result can be obtained in each advertisement update cycle to avoid the judgment of the execution effect of the previously matched rules being affected after the matching rules are updated due to changes in exposure, order volume and click volume during the advertisement update cycle. That is, if rule A is matched in the first advertisement update cycle, if the comparison is performed only after one advertisement update cycle, it can be confirmed that the impact is mainly caused by rule A, and if the advertisement parameters are obtained after two or three advertisement update cycles, in the second or third advertisement update cycle, because the change in the advertisement parameters may trigger other rules B or C, it cannot be uniquely determined that the change in the advertisement parameters is caused by rule A.

[0049] The advertising cycle can be one week, three days, etc., depending on the attributes of the product. If the product is highly seasonal or has large price fluctuations, a shorter advertising cycle can be set; otherwise, a longer advertising cycle can be set. Alternatively, a shorter advertising cycle can be set for newly launched products, and it can be flexibly adjusted according to the specific situation.

[0050] S42. Obtaining the execution effect of the target advertisement under the target rule according to the comparison result between the historical execution data and the updated execution data, including steps S421-S426.

[0051] In the next cycle after the advertisement is modified according to the hit rule, the system can automatically analyze the historical data through step S42.

[0052] S421, obtaining a first growth rate of the new exposure compared to the current exposure, a second growth rate of the new click volume equivalent to the current click volume, and a third growth rate of the new order volume equivalent to the current order volume; S422: If the third growth rate is greater than the third preset value, the output execution effect reaches the expectation.

[0053] Furthermore, if the third growth rate is greater than the third preset value, further obtain the first conversion rate of the new order volume / new click volume, the second conversion rate of the new order volume / new exposure volume, the third conversion rate of the current order volume / current click volume, and the fourth conversion rate of the current order volume / current exposure volume; determine whether the third conversion rate is greater than the first conversion rate and whether the fourth conversion rate is greater than the second conversion rate. If they are both, the output execution effect meets expectations; otherwise, determine whether the first conversion rate and the second conversion rate are respectively greater than their corresponding preset thresholds; if so, the output execution effect meets expectations; otherwise, the output execution effect meets expectations but attention needs to be paid to the conversion rate. That is, when the order volume meets expectations, the conversion rate of the orders is also monitored. If the conversion rate is too low, the conversion rate is also marked as too low when the order volume meets expectations, and the cause is further sought.

[0054] S423: If the third growth rate is less than the third preset value, the first growth rate is greater than the first preset value, and the second growth rate is greater than the second preset value, then the output conversion rate does not meet expectations; S424: if the third growth rate is less than the third preset value, the first growth rate is greater than the first preset value, and the second growth rate is less than the second preset value, outputting that the keyword does not match; S425: If the third growth rate is less than the third preset value and the first growth rate is less than the first preset value, the output advertisement bidding needs to be adjusted; S426. Otherwise, mark the execution effect as requiring verification. That is, if any of the above steps S422-S425 are not met, the execution result requiring verification is output, indicating that manual judgment is required to determine whether the execution effect meets expectations. The judgment can also be made through a pre-trained machine model, which is not limited here. For example, if a single indicator is too low, it needs to be manually marked, and the corresponding mark can be moved to match the same situation.

[0055] It can be seen that since the advertising placement of each product is different, it is difficult to analyze manually what kind of adjustments the advertisements need to make; and there are also multiple types of products in a store, each product may correspond to multiple advertisements with different keywords or delivery channels, and the effect of each advertisement after delivery is also different, so in order to quantify and analyze the execution effect of the execution strategy in the rule. In the process of setting the rules, quantifying the execution effect and establishing a screening mechanism can optimize the rules themselves, as shown in the following embodiments.

[0056] Embodiment 2 of the present invention is: A rule-based advertising delivery method is different from the first embodiment in that: S1 also includes S11-S12 after S1.

[0057] S11, obtaining a preset number of combination rules with different conditions from the rules.

[0058] S12: Combining the combination rules into an advertisement delivery strategy to obtain at least one advertisement delivery strategy.

[0059] S3 includes: obtaining a target condition according to the matching parameter in an advertisement delivery strategy and a target execution strategy in a target combination rule corresponding to the target condition.

[0060] According to the description of S1 of an embodiment, it can be known that in the end a lot of rules will be obtained, and the rules may be applicable to different types of advertisements. Then, multiple rules applicable to the same type of advertisements are set into an advertisement delivery strategy. Then, when matching the rules, this advertisement delivery strategy can be used for advertisement delivery, thereby improving the compatibility of the matched rules with the advertisements themselves, thereby increasing the probability that the effect after the rule execution can achieve the expected effect.

[0061] After S42, steps S43-S44 are also included.

[0062] S43. Obtain all execution effects corresponding to the target rule at preset intervals.

[0063] For example, if the execution cycle of an advertisement is once a week, the execution effects between the two weeks before and after can be obtained. If the preset cycle is set to three months, the execution effect comparison between the two weeks within the three months corresponding to the rule can be collected, that is, the data of the past three months that hit the rule can be analyzed, and the exposure, click volume, order volume and other information of the advertisement before and after the modification according to the execution strategy can be compared. The historical data expired before the preset cycle is saved in the system. When the rule is hit, the identification information of the rule can be obtained to trace the historical execution effect of the rule. Since the advertisement update cycle is executed once a week, that is, we execute an advertisement once a week, the system will compare the data of this week and last week. At the same time, the data of this week and last week for all the cases that hit the rule within three months are compared. That is, the setting of the preset cycle should include an even number of advertisement update cycles to ensure that the execution effect data can be obtained normally.

[0064] The system queries the database to obtain the hit rules of last week. The ad IDs that can hit the rules are grouped according to the hit rules and analyzed in the order of the rules. Since the same ad may not hit this rule again next week after the data changes, only the impact of the execution strategy in this rule is analyzed one week after it is executed. Get all the ads under this rule, and find the data performance of last week from the database according to the ad ID, that is, the current order volume, current exposure rate, and current click volume, and compare the data of this week (after the rule is executed), that is, the new order volume, new click volume, and new exposure volume.

[0065] S44: Adjust the target rule according to the expected proportion of the execution effect.

[0066] It can be seen that the execution effect obtained by the method in step S42 can determine whether the target rule and the target advertisement with an associated relationship are compatible. If the execution effect is not good, it may be that the target rule is not applicable to the target advertisement, and it cannot directly indicate that there is a problem with the target rule itself. Therefore, all execution effects corresponding to the target rule are further obtained at preset intervals, not limited to the execution effects corresponding to a specific target advertisement. If the expected ratio is low, it means that the applicable scenario of the target rule itself is small or the conditions or execution strategy configuration therein is not appropriate and needs to be adjusted.

[0067] In an optional implementation, fields are distinguished to determine the execution effect of the target rule, that is, the historical execution data obtained in S42 and the updated execution data are grouped and compared according to fields. For example, the increase in exposure, increase in clicks, and increase in orders are divided into three fields and counted separately; for each rule, the proportion of advertisements that meet the corresponding fields is calculated. If it exceeds a certain proportion, it is considered that the execution effect of the rule has met expectations.

[0068] For example, if the ads with increased exposure account for 60% of all ads, it is considered that this rule can achieve a relatively good optimization effect on the ads and the execution effect meets expectations. The analysis results are stored in the database. Find all the performance of the rule for three months, and output the conclusion of the rule performance for the past three months each month. Furthermore, manual review conclusions or machine models trained by manual labeling can be introduced for review.

[0069] Identify that there is a significant correlation between the bids of different ad slots and the bids of the ads placed by the product itself, that is, for ad slots in the same channel, the higher the bid price, the better the position of the ad slot and the higher the exposure; at the same time, the set rules are evaluated manually or through execution effects on a regular basis. By comparing the advertising performance before and after the application of the rules, the actual effect of the rules is evaluated. For rules with poor performance, their parameters are automatically adjusted. For example, for rules that adjust bids, the bid is mainly to adjust the percentage. If the rule has performed poorly in the past month, the bid percentage can be appropriately increased. For example, the original current bid * 10%, and then the rule is adjusted to the current bid * 12%, until the preset bid cap, that is, the maximum bid size that can be given. The bid cap can be a percentage or a specific value. If it is a specific value, it needs to be calculated based on the adjusted bid ratio to determine whether it exceeds it. If the performance is still poor after the upper limit is increased or it is considered to be replaced with a new rule, the rule will be deleted.

[0070] In an optional implementation, S44 includes S441-S442.

[0071] S441: If the expected proportion of the execution effect is less than a fourth preset value, obtaining the target advertisement delivery strategy where the target rule is located; S442: Acquire a replacement rule with the same condition as the target rule, and replace the target rule with the replacement rule in the target advertisement delivery strategy.

[0072] In an optional implementation, S44 may further include: if the expected proportion of the execution effect exceeds a preset proportion, retaining the target rule; otherwise, adjusting the fixed proportion, keywords or delivery channels of the execution strategy in the target rule.

[0073] Embodiment 3 of the present invention is: A rule-based advertising delivery method is different from the first embodiment in that: S13 is also included after S1.

[0074] S13: Arrange the rules according to preset priorities.

[0075] S3 includes: matching the target condition according to the matching parameters in the order of the preset priority in the rule to obtain the target condition and the target execution strategy in the target rule corresponding to the target condition.

[0076] S44 includes S443-S444.

[0077] S443: If the proportion of the execution effects that meet expectations is less than a fourth preset value, lower the priority of the target rule.

[0078] S444. If the expected proportion of the execution effect is greater than the fifth preset value, the priority of the target rule is increased. That is, the implementation method in Example 2 and the implementation method in Example 3 can be selected according to the characteristics of different advertising systems. For example, there are many product types corresponding to advertisements and many adjustable parameters in the advertising platform, which can be used as adjustment items for the advertising execution strategy. In this scenario, there are many rules. The method in Example 2 can be used to manage the advertising delivery strategy to increase the matching speed and improve the matching effect. If the adjustable parameters provided by the platform are relatively simple or the store has few types of goods in the advertisement, the solution in Example 3 can be used. It only needs to set the priority and dynamically adjust the priority. There is no need to distinguish the advertising execution strategy corresponding to the rule, which simplifies the data processing process.

[0079] Please refer to Figure 2 , Embodiment 4 of the present invention is: A rule-based advertising delivery terminal 1 includes a processor 2, a memory 3, and a computer program stored in the memory 3 and executable on the processor 2. When the processor 2 executes the computer program, each step in the first embodiment is implemented.

[0080] In summary, the present invention provides a rule-based advertising delivery method and terminal, which takes exposure, clicks and orders as conditions, sets execution strategy composition rules corresponding to different conditions, obtains the corresponding exposure, clicks and orders after the advertisement is released, and judges whether it can match the conditions. If it matches, it means that the advertisement meets the execution conditions of the corresponding execution strategy, and adjusts the advertisement itself according to the execution conditions in the rules corresponding to the conditions, so as to realize automatic adjustment of the advertising plan. And the advertisement and the matched target rule are associated and saved. After the advertisement is adjusted according to the rule, the exposure, clicks and orders of the advertisement corresponding to the rule after adjustment can be obtained according to the saved association relationship. According to the changes in exposure, clicks and orders, the execution effect of the rule in the advertising scenario can be analyzed. For a rule, if multiple advertisements are matched, the execution effects of the rule under different advertisements can be uniformly analyzed. If the expected execution effect of the rule is low, it means that the rule itself is not suitable. If only the expected execution effect of the rule is high, it means that the rule and some advertisements may not match. Then, a blocking mark can be added to the corresponding advertisement to block the corresponding rule, and the inappropriate rule will not be matched with the advertisement, thereby increasing the probability that the execution effect of the rule reaches the expected result. In addition, the rules can be packaged into advertising delivery strategies for different types of advertisements. The corresponding types of advertisements only match the rules in their corresponding advertising delivery strategies, further increasing the probability that the execution effect reaches the expected result. The matching priority method can also be used to set the priority of the rules. Moreover, after obtaining the execution effect, the rules that continue to fail to meet the standards in the advertising delivery strategy can be replaced or the priority can be lowered according to the execution effect until the rules that have not been matched for a long time and have poor execution effect are modified, thereby increasing the flexibility of rule setting and realizing dynamic adjustment of the rules.

[0081] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A rule-based advertising method, characterized in that: Includes steps: Using exposure, clicks, and orders as conditions, set execution strategies corresponding to different conditions, and associate the conditions and execution strategies and store them as rules; According to the target advertisement's advertisement update cycle, the current exposure, current click volume, and current order volume are obtained as matching parameters; and obtaining a target condition and a target execution strategy in a target rule corresponding to the target condition according to the matching parameter; The target execution policy is executed.

2. A rule-based advertising delivery method according to claim 1, characterized in that: The step of obtaining the target condition and the target execution strategy in the target rule corresponding to the target condition according to the matching parameter includes: The target rule, the target advertisement, the current exposure, the current click volume, and the current order volume are associated and saved as historical execution data; The executing of the target execution strategy comprises: After a preset number of advertisement update cycles, the current new exposure, new click volume, and new order volume of the target advertisement are obtained again, associated with the target rule corresponding to the target advertisement, and saved as update execution data; The execution effect of the target advertisement under the target rule is obtained according to the comparison result between the historical execution data and the updated execution data.

3. A rule-based advertising delivery method according to claim 2, characterized in that: The obtaining of the execution effect of the target advertisement under the target rule according to the comparison result between the historical execution data and the updated execution data includes: Obtain a first growth rate of the new exposure compared to the current exposure, a second growth rate of the new click volume equivalent to the current click volume, and a third growth rate of the new order volume equivalent to the current order volume; If the third growth rate is greater than the third preset value, the output execution effect reaches the expected result; If the third growth rate is less than the third preset value, the first growth rate is greater than the first preset value, and the second growth rate is greater than the second preset value, then the output conversion rate does not meet expectations; If the third growth rate is less than the third preset value, the first growth rate is greater than the first preset value and the second growth rate is less than the second preset value, outputting that the keyword does not match; If the third growth rate is less than the third preset value and the first growth rate is less than the first preset value, the output advertisement bidding needs to be adjusted; Otherwise, mark the execution effect as requiring verification.

4. A rule-based advertising delivery method according to claim 2, characterized in that: After obtaining the execution effect of the target advertisement under the target rule according to the comparison result between the historical execution data and the updated execution data, the method further includes: Obtain all execution effects corresponding to the target rule at a preset interval; The target rule is adjusted according to the expected proportion of the execution effect.

5. A rule-based advertising delivery method according to claim 4, characterized in that: The step of associating the condition with the execution strategy and storing it as a rule comprises: Acquire a preset number of combination rules whose conditions are different from each other from the rules; Combining the combination rules into an advertisement delivery strategy to obtain at least one advertisement delivery strategy; The adjusting the target rule according to the expected proportion of the execution effect comprises: If the expected proportion of the execution effect is less than a fourth preset value, obtaining the target advertisement delivery strategy where the target rule is located; A replacement rule having the same condition as the target rule is obtained, and the target rule is replaced with the replacement rule in the target advertisement delivery strategy.

6. A rule-based advertising delivery method according to claim 4, characterized in that: The step of associating the condition with the execution strategy and storing it as a rule comprises: Arrange the rules according to preset priorities; The adjusting the target rule according to the expected proportion of the execution effect comprises: If the expected proportion of the execution effect is less than a fourth preset value, lowering the priority of the target rule; If the proportion of the execution effects that meet expectations is greater than a fifth preset value, the priority of the target rule is increased.

7. A rule-based advertising delivery method according to claim 4, characterized in that: The execution strategies include: Increase the bid by a fixed percentage, change keywords, and change the delivery channel.

8. A rule-based advertising delivery method according to claim 7, characterized in that: The adjusting the target rule according to the expected proportion of the execution effect comprises: If the expected proportion of the execution effect exceeds the preset proportion, the target rule is retained; otherwise, the fixed proportion, keywords or delivery channels of the execution strategy in the target rule are adjusted.

9. The rule-based advertising delivery method according to claim 1, characterized in that: The executing target execution strategy comprises: Obtaining an advertisement identifier of the target advertisement, and searching for a request corresponding to the target advertisement in an advertisement platform according to the advertisement identifier; The request is modified according to the target execution policy.

10. A rule-based advertising delivery terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the rule-based advertising delivery method described in any one of claims 1-9 is implemented.

Images