Strategy generation method and device, computer equipment and computer readable storage medium

Abstract

The invention discloses a strategy generation method and device, computer equipment and a computer readable storage medium, relates to the field of computer networks, and can continuously optimize a user growth path to obtain a strategy with a higher adaptation degree and a better execution effect by fusing AB test and conversion funnel analysis and testing different strategies by using an artificial intelligence technology. The method comprises the steps of obtaining historical user behavior data, inputting the historical user behavior data into a conversion funnel model for business stage analysis, and obtaining a user conversion funnel output by the conversion funnel model; performing evaluation analysis on the user conversion funnel by adopting the historical user behavior data to obtain a conversion funnel analysis result; formulating a plurality of initial strategies according to the conversion funnel analysis result, and testing the plurality of initial strategies by adopting an AB test to obtain an AB test result; and adjusting the plurality of initial strategies by adopting the conversion funnel analysis result and the AB test result to obtain a target strategy.

Description

Technical Field

[0001] The present application relates to the field of computer network technology, and in particular to a strategy generation method, device, computer equipment and computer-readable storage medium. Background Art

[0002] With the continuous development of computer network technology, digital methods are constantly being implemented in the leasing industry. Small B retail business has the natural advantage of digital assistance due to its small-volume business characteristics. Among them, small B retail business mainly provides small-amount business services to small and micro enterprises and individual users. As the business market continues to mature, market competition has gradually shifted from attracting new customers to maintaining existing customers. Therefore, formulating effective strategies is the key to increasing market share and user satisfaction.

[0003] In the existing technology, through precise market segmentation and positioning of target user groups, C-end enterprises can more effectively convey brand information and product value, thereby improving brand awareness, consumer stickiness and sales conversion rate. This relies on the online transaction of C-end business, which can rely on APP (Application) to fully collect the entire process of the user life cycle. However, the applicant recognizes that the transaction of small B retail business also depends on offline business exhibitions to bring business leads, and the collection of users' offline behavior is a difficult point in the industry. Therefore, the strategies applicable to the C-end lack reusability in the small B field, which not only reduces the effectiveness of business activities, but also leads to poor user experience. Summary of the invention

[0004] In view of this, the present application provides a policy generation method, apparatus, computer device and computer-readable storage medium, the main purpose of which is to solve the problem that the policies applicable to the C-end lack reusability in the small B field, which not only reduces the effectiveness of business activities but also leads to poor user experience.

[0005] According to the first aspect of the present application, a strategy generation method is provided, the method comprising:

[0006] Acquire historical user behavior data, input the historical user behavior data into a conversion funnel model for business stage analysis, and obtain a user conversion funnel output by the conversion funnel model;

[0007] Using the historical user behavior data to evaluate and analyze the user conversion funnel to obtain a conversion funnel analysis result;

[0008] Formulate multiple initial strategies according to the conversion funnel analysis results, and test the multiple initial strategies using AB testing to obtain AB testing results;

[0009] The conversion funnel analysis results and the AB test results are used to adjust the multiple initial strategies to obtain a target strategy.

[0010] According to a second aspect of the present application, a strategy generation device is provided, the device comprising:

[0011] An analysis module is used to obtain historical user behavior data, input the historical user behavior data into a conversion funnel model for business stage analysis, and obtain a user conversion funnel output by the conversion funnel model;

[0012] An evaluation module, used to evaluate and analyze the user conversion funnel using the historical user behavior data to obtain a conversion funnel analysis result;

[0013] A testing module, used to formulate multiple initial strategies according to the conversion funnel analysis results, and to test the multiple initial strategies using AB testing to obtain AB testing results;

[0014] An adjustment module is used to adjust the multiple initial strategies using the conversion funnel analysis results and the AB test results to obtain a target strategy.

[0015] According to a third aspect of the present application, a computer device is provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the methods described in the first aspect when executing the computer program.

[0016] According to a fourth aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the methods described in the first aspect are implemented.

[0017] By means of the above technical scheme, the present application provides a strategy generation method, device, computer equipment and computer-readable storage medium. The present application obtains historical user behavior data, inputs the historical user behavior data into the conversion funnel model for business stage analysis, obtains the user conversion funnel output by the conversion funnel model, uses the historical user behavior data to evaluate and analyze the user conversion funnel, obtains the conversion funnel analysis results, formulates multiple initial strategies according to the conversion funnel analysis results, uses AB testing to test multiple initial strategies, obtains AB test results, uses the conversion funnel analysis results and AB test results to adjust multiple initial strategies, and obtains the target strategy. The machine learning algorithm is used to screen the historical user behavior data to obtain high-quality users, and the effect of AB testing is tracked through the conversion funnel, which provides data support for continuous adjustment of strategies, rapid understanding of market trends, and analysis of user growth paths, solving the problem that the entire process of the user life cycle cannot be fully collected by relying on APP. By integrating AB testing and conversion funnel analysis, and using artificial intelligence technology to quickly test different strategies, the user growth path can be continuously optimized to obtain strategies with higher adaptability and better execution effects.

[0018] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0020] Figure 1 A schematic diagram of an application environment for strategy generation provided by an embodiment of the present application is shown;

[0021] Figure 2 A schematic diagram of a method flow for generating a strategy provided by an embodiment of the present application is shown;

[0022] Figure 3A A schematic diagram of another method flow for generating a strategy provided by an embodiment of the present application is shown;

[0023] Figure 3B A schematic diagram of the architecture of an optimization strategy provided by an embodiment of the present application is shown;

[0024] Figure 4A A schematic diagram of a strategy generation structure provided by an embodiment of the present application is shown;

[0025] Figure 4B A schematic diagram showing another structure of strategy generation provided by an embodiment of the present application is shown;

[0026] Figure 5 A schematic diagram of the device structure of a computer device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0027] The exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0028] The strategy generation method provided by the embodiment of the present invention can be applied in Figure 1 In the application environment, the execution subject of this application is the strategy generation system, which relies on the computing power of the server to provide services to users and relies on the client to provide service programs for users corresponding to the server, and the client communicates with the server through the network. The server can obtain historical user behavior data, input the historical user behavior data into the conversion funnel model for business stage analysis, obtain the user conversion funnel output by the conversion funnel model, use the historical user behavior data to evaluate and analyze the user conversion funnel, obtain the conversion funnel analysis results, formulate multiple initial strategies according to the conversion funnel analysis results, use AB testing to test multiple initial strategies, obtain AB test results, use the conversion funnel analysis results and AB test results to adjust multiple initial strategies, obtain the target strategy, and feed the target strategy back to the client. The machine learning algorithm is used to screen the historical user behavior data to obtain high-quality users, and the effect of AB testing is tracked through the conversion funnel, which provides data support for continuous adjustment of strategies, rapid understanding of market trends, and analysis of user growth paths, solving the problem that the entire process of the user life cycle cannot be fully collected by relying on APP. By integrating AB testing and conversion funnel analysis, and using artificial intelligence technology to quickly test different strategies, the user growth path can be continuously optimized to obtain strategies with higher adaptability and better execution effects. Among them, the client can be but is not limited to various personal computers, laptops, smart phones, tablets and portable wearable devices. The server can be an independent server, or it can provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and servers for basic cloud computing such as big data and artificial intelligence platforms. The present invention is described in detail below through specific embodiments.

[0029] The present application embodiment provides a strategy generation method, such as Figure 2 As shown, the method includes:

[0030] 201. Obtain historical user behavior data, input the historical user behavior data into a conversion funnel model for business stage analysis, and obtain a user conversion funnel output by the conversion funnel model.

[0031] In the embodiment of the present application, the strategy generation system obtains historical user behavior data, inputs the historical user behavior data into the conversion funnel model for business stage analysis, and obtains the user conversion funnel output by the conversion funnel model. The strategy generation system defines the user conversion funnel according to the key stages of business development and customer acquisition of the financial leasing business. Each stage of the user conversion funnel represents the different progress of the user in the marketing and sales process, including the visitor stage, the lead stage, the business opportunity stage, the pre-lease stage, the mid-lease stage, and the post-lease stage, which can classify user samples and thus screen out high-qualified users.

[0032] 202. Use historical user behavior data to evaluate and analyze the user conversion funnel to obtain the conversion funnel analysis results.

[0033] In an embodiment of the present application, a strategy generation system uses historical user behavior data to evaluate and analyze the user conversion funnel to obtain a conversion funnel analysis result, wherein the conversion funnel analysis result includes the user conversion funnel, the benchmark conversion rate of each target stage in the user conversion funnel, the churn rate of each target stage in the user conversion funnel, the funnel bottleneck of the user conversion funnel, and the key indicators of the user conversion funnel, so that the strategy generation system can subsequently combine AB testing to verify the difference in conversion rate effects under different marketing strategies, and provide directional data guidance for formulating precise marketing strategies.

[0034] 203. Formulate multiple initial strategies based on the conversion funnel analysis results, use AB testing to test the multiple initial strategies, and obtain AB testing results.

[0035] In the embodiment of the present application, the strategy generation system formulates multiple initial strategies according to the conversion funnel analysis results, and uses AB testing to test the multiple initial strategies to obtain AB testing results. Through AB testing, the impact of different marketing strategies on the conversion rate of each stage can be tested, so as to screen out better marketing strategies.

[0036] 204. Use the conversion funnel analysis results and AB test results to adjust multiple initial strategies and obtain the target strategy.

[0037] In the embodiment of the present application, the strategy generation system uses the conversion funnel analysis results and AB test results to adjust multiple initial strategies to obtain the target strategy. By integrating AB testing and conversion funnel analysis, and using artificial intelligence technology to quickly test different marketing strategies, the user growth path can be continuously optimized to obtain a marketing strategy with higher adaptability and better marketing effect.

[0038] The method provided in the embodiment of the present application obtains historical user behavior data, inputs the historical user behavior data into the conversion funnel model for business stage analysis, obtains the user conversion funnel output by the conversion funnel model, uses the historical user behavior data to evaluate and analyze the user conversion funnel, obtains the conversion funnel analysis results, formulates multiple initial strategies according to the conversion funnel analysis results, uses AB testing to test the multiple initial strategies, obtains the AB test results, uses the conversion funnel analysis results and the AB test results to adjust the multiple initial strategies, and obtains the target strategy. The historical user behavior data is screened by a machine learning algorithm to obtain high-quality users, and the effect of the AB test is tracked through the conversion funnel, providing data support for continuous adjustment of strategies, rapid understanding of market trends, and analysis of user growth paths, solving the problem that the entire process of the user life cycle cannot be fully collected by relying on APP. By integrating AB testing and conversion funnel analysis, and using artificial intelligence technology to quickly test different strategies, the user growth path can be continuously optimized to obtain a strategy with higher adaptability and better execution effect.

[0039] Further, as a refinement and extension of the specific implementation of the above embodiment, in order to fully illustrate the specific implementation process of this embodiment, the embodiment of the present application provides another strategy generation method, such as Figure 3A As shown, the method includes:

[0040] 301. Build a conversion funnel model based on the multiple stages of the user's entire life cycle.

[0041] In an embodiment of the present application, the strategy generation system obtains multiple stages included in the user's entire life cycle, and sends the multiple stages to the terminal held by the user, wherein the multiple stages include the visitor stage, the lead stage, the business opportunity stage, the pre-lease stage, the mid-lease stage, and the post-lease stage. Specifically, the visitor stage represents the stage when a potential user first contacts the company's products or services, the lead stage represents the stage when telemarketing or online marketing reaches users with leasing intentions, the business opportunity stage represents the stage when the account manager reaches out offline and enters the sales negotiation process, the pre-lease stage represents the stage when the user enters the leasing object review process, the mid-lease stage represents the stage when the user enters the fund use supervision, lease contract signing, leased goods delivery, and payment payment stages, and the post-lease stage represents the stage when the user enters the repayment period after the lease starts. Then, the strategy generation system receives multiple target stages selected by the user based on the terminal held in the multiple stages, and uses multiple target stages to build a conversion funnel model. The strategy generation system defines the conversion funnel model based on the key stages of business development and customer acquisition in the financial leasing business. Each stage in the conversion funnel represents the different progress of users in the marketing and sales process. This enables the construction of a model that is close to the characteristics of the business, thereby obtaining a better marketing strategy.

[0042] 302. Obtain historical user behavior data, input the historical user behavior data into a conversion funnel model for business stage analysis, and obtain a user conversion funnel output by the conversion funnel model.

[0043] In an embodiment of the present application, the strategy generation system determines multiple target users and user data of each target user based on historical user behavior data, wherein the user data is the behavior information of the target user for the target business, and the target business is the business to be implemented. Then, the strategy generation system inputs the user data of multiple target users into the conversion funnel model, and the conversion funnel model determines the target stage to which each target user belongs according to the user data of each target user, and the conversion funnel model divides the target users belonging to the same target stage among the multiple target users into the same user group, and obtains multiple user groups, wherein the target stage is any one of the visitor stage, the lead stage, the business opportunity stage, the pre-rental stage, the mid-rental stage, and the post-rental stage. Finally, the strategy generation system obtains the user conversion funnel generated based on multiple user groups output by the conversion funnel model. Through the conversion funnel model, all users can be divided into different conversion stages according to user behavior data, and the boundaries of each stage are clarified, so that more valuable and high-qualified users that meet the requirements of the risk policy can be screened out, and the interference of unqualified users on the marketing team can be reduced, thereby generating a better marketing strategy.

[0044] 303. Use historical user behavior data to perform feature engineering calculations on the user conversion funnel to obtain a benchmark conversion rate for each target stage in the user conversion funnel.

[0045] In an embodiment of the present application, the strategy generation system uses historical user behavior data to perform feature engineering calculations on the user conversion funnel to obtain the benchmark conversion rate for each target stage in the user conversion funnel. It should be noted that for new users without historical user behavior data, the strategy generation system obtains the benchmark conversion rate by building a model prediction; for old users with historical user behavior data, the strategy generation system uses historical user behavior data to perform feature engineering calculations to obtain the benchmark conversion rate, that is, the strategy generation system performs feature engineering calculations on user behavior data based on classification labels such as the user's industry, region, lease stage, financing scale, cooperation status, and asset risk level. Through the benchmark conversion rate strategy generation system, the marketing strategy can be continuously optimized to obtain a better marketing strategy.

[0046] 304. Calculate the churn rate of each target stage in the user conversion funnel, and select at least one target stage whose churn rate is greater than or equal to a churn rate threshold from the multiple target stages as the funnel bottleneck of the user conversion funnel.

[0047] In the embodiment of the present application, the strategy generation system calculates the churn rate of each target stage in the user conversion funnel, and selects at least one target stage with a churn rate greater than or equal to a churn rate threshold from multiple target stages as the funnel bottleneck of the user conversion funnel. The target stage with a higher churn rate corresponds to a stage with an abnormally low conversion rate, which is the funnel bottleneck. Determining the funnel bottleneck facilitates the leasing operator to improve conversion in a more refined staged manner, and ultimately improves the overall conversion. For example, if the churn rate in the business opportunity stage is high, a corresponding marketing channel optimization strategy is formulated; if the churn rate in the pre-rental stage is high, a corresponding sales follow-up mechanism is optimized.

[0048] The calculation formula for the churn rate at each target stage is as follows:

[0049] Lead stage: 1-number of users entering the lead stage / number of users entering the visitor stage;

[0050] Business opportunity stage: 1-number of users entering the business opportunity stage / number of users entering the lead stage;

[0051] Pre-lease stage: 1-number of users entering the pre-lease stage / number of users entering the business opportunity stage;

[0052] Mid-rental stage: 1-number of users entering the mid-rental stage / number of users in the pre-rental stage;

[0053] Post-rental stage: 1-number of users entering the post-rental stage / number of users entering the mid-rental stage.

[0054] It should be noted that the criteria for judging the bottleneck of the funnel can also be defined according to the usage scenario, and this application does not impose specific restrictions. For example, by combining user segmentation and horizontally comparing the conversion rates of different customer groups at the same target stage, the conversion rate differences of each group can be further analyzed for different customer groups, and the low-conversion customer groups can be accurately located. That is, for the same target stage, the conversion rate differences of each group can be further analyzed for different customer groups, and the low-conversion customer groups can be accurately located. Then, combined with the AB test of the marketing strategy, the conversion rate effect differences under different marketing strategies can be verified, which can provide directional data guidance for the formulation of precision marketing strategies.

[0055] 305. Use multiple preset classification tags to refine and classify multiple user groups included in the user conversion funnel to obtain user classification results.

[0056] In the embodiment of the present application, the strategy generation system obtains multiple preset classification tags, and uses the multiple preset classification tags to refine and classify the multiple user groups included in the user conversion funnel to obtain user classification results. Among them, the preset classification tags include the user's industry, region, leasing stage, financing scale, asset risk level, etc., so that the strategy generation system can build accurate key indicators of the user conversion funnel.

[0057] 306. Calculate key indicators of the user conversion funnel based on user classification results and historical user behavior data.

[0058] In the embodiment of this application, financial services need to combine marketing and risk, both to improve the efficiency of renting and to balance user risks. Therefore, when analyzing the funnel, the strategy generation system calculates the key indicators of the user conversion funnel based on the user classification results and historical user behavior data, so as to evaluate the conversion efficiency and user quality at each stage. Among them, the key indicators are as follows:

[0059] Visits: the total number of visitors, indicating the traffic diversion effect of different channels;

[0060] Lead acquisition rate: the ratio of lead users to the total number of visitors;

[0061] Opportunity conversion rate: the ratio of lead users converted into business opportunity users;

[0062] User conversion rate: the ratio of business opportunity users to pre-rental users;

[0063] Starting rent rate: the ratio of pre-rental users to starting renters;

[0064] Normal asset ratio: the ratio of the non-focused assets of the renting user;

[0065] Marketing cycle length: the average time from lead generation to user entering the review process;

[0066] Length of operating cycle: the average time from when a user enters the review process to when the user starts renting.

[0067] It should be noted that the key indicators corresponding to different functional positions and business development stages will be different. Therefore, this application can select or customize key indicators according to actual scenarios, and no specific restrictions are made here.

[0068] Optionally, the strategy generation system can use statistical methods such as t-test and chi-square test to verify whether there are significant differences in key indicators of different user groups, thereby identifying which user groups have better conversion performance than other groups.

[0069] 307. The user conversion funnel, the benchmark conversion rate of each target stage in the user conversion funnel, the churn rate of each target stage in the user conversion funnel, the funnel bottleneck of the user conversion funnel, and the key indicators of the user conversion funnel are taken as the conversion funnel analysis results.

[0070] In the embodiment of the present application, the strategy generation system uses the user conversion funnel, the benchmark conversion rate of each target stage in the user conversion funnel, the churn rate of each target stage in the user conversion funnel, the funnel bottleneck of the user conversion funnel, and the key indicators of the user conversion funnel as the conversion funnel analysis results, and uses the conversion funnel analysis results to continuously optimize the marketing strategy, which can improve the adaptability and execution effect of the marketing strategy. It should be noted that the strategy generation system incorporates the feedback of the marketing team and the asset performance of the post-rental user into the analysis link, and optimizes the user screening model, which can improve the screening criteria and follow-up strategies for lead users and business opportunity users.

[0071] In an optional implementation scheme, the strategy generation system obtains the user conversion funnel, the benchmark conversion rate of each target stage in the user conversion funnel, the churn rate of each target stage in the user conversion funnel, and the key indicators of the user conversion funnel in the conversion funnel analysis results. Next, the strategy generation system obtains multiple user groups included in the user conversion funnel, and counts the number of users corresponding to each target stage according to the multiple user groups. Then, the strategy generation system displays the number of users corresponding to each target stage, the benchmark conversion rate of each target stage in the user conversion funnel, the churn rate of each target stage in the user conversion funnel, the key indicators of the user conversion funnel, and the target strategy. Using data visualization tools to create a visual report of the conversion funnel, the strategy generation system can display the number of users, conversion rate, churn rate and core indicators of each stage, helping management to quickly understand the performance of current marketing and sales activities. It should be noted that the core module of the conversion funnel report is the indicator decomposition and dimension drilling of key indicators, which facilitates the marketing team and company management to quickly locate the conversion rate of each funnel stage and the main reasons for the differences in the performance of key indicators of different customer groups.

[0072] 308. Formulate multiple initial strategies based on the conversion funnel analysis results, use AB testing to test the multiple initial strategies, and obtain AB testing results.

[0073] In the embodiment of the present application, the strategy generation system determines the test target, test variable, test group and test hypothesis according to the multiple initial strategies. Then, the strategy generation system uses AB test to test the multiple initial strategies based on the test target, test variable, test group and test hypothesis to obtain the AB test results.

[0074] The following example uses the recommendation of new leaseback products with different financing scales to illustrate the implementation plan of AB testing:

[0075] Determine the test goal: to improve the rental rate in the key indicator of the conversion funnel.

[0076] Select the test variable: Select the strategy variable to be tested as the financing scale, and make it clear that the AB test is to compare the starting lease rate performance of the new leaseback product under the recommended strategies of different financing packages. Among them, Plan A: recommends a small financing package with multiple withdrawals; Plan B: recommends a one-time large financing package.

[0077] Divide the test group: Randomly divide the target user group into two (or more) experimental groups, ensuring that the number of people in each group is roughly the same and comparable to reduce bias. It should be noted that in B2B (Business-to-Business, the exchange of products, services and information between enterprises through the Internet) marketing, the sample size requirement is usually higher than B2C (Business to Consumer, enterprises sell products or provide services directly to consumers) because the conversion rate is lower.

[0078] Formulate a test hypothesis: Assume that strategy A will perform better than strategy B.

[0079] Implementation test: Apply the designed strategy to two groups of users and test them in the same time period to ensure that other factors (such as promotion time, region, channel, etc.) remain consistent. For example: two different product strategies are sent to two groups of users of the same account managers at the same time, and the lead acquisition rate and rental rate of each group are recorded. It should be noted that for B2B business, considering the long decision cycle, the testing time may need to last from several weeks to a month.

[0080] Result verification: Use statistical methods such as t-test and chi-square test to verify the significance of AB test results to ensure that the test results are not caused by random factors.

[0081] Continuous testing and iteration: AB testing is a continuous optimization process. After confirming that the strategy of the microfinancing package is effective, other variables should be tested in detail, such as the industry and region in which the user is located, the user's asset risk, etc.

[0082] 309. Use the conversion funnel analysis results and AB test results to adjust multiple initial strategies and obtain the target strategy.

[0083] In an embodiment of the present application, the strategy generation system selects a first strategy from multiple initial strategies according to the AB test results, wherein the first strategy is the initial strategy with the highest score among the multiple initial strategies indicated by the AB test results. Next, the strategy generation system obtains multiple information acquisition channels, and based on the AB test, uses multiple information acquisition channels to test multiple target stages included in the user, obtains channel optimization test results, and uses the channel optimization test results to adjust the first strategy. Subsequently, the strategy generation system obtains the user conversion funnel and the key indicators of the user conversion funnel in the conversion funnel analysis results, and uses the key indicators of the user conversion funnel to build a machine learning model. Then, the strategy generation system inputs the user conversion funnel into the machine learning model for evaluation, obtains the user quality score, and uses the user quality score to screen the multiple user groups included in the user conversion funnel to obtain multiple high-quality users. Finally, the strategy generation system obtains the user data of multiple high-quality users, and uses the user data of multiple high-quality users to adjust the adjusted first strategy to obtain the target strategy.

[0084] like Figure 3BOne optimization strategy shown in the figure is to optimize the marketing channels first: for different channels such as peer lists and supplier push orders, AB tests are used to test the impact of different marketing channels on the conversion rate at each stage. Then, user scoring and screening are carried out: feature engineering calculations are performed on user sample data based on classification labels such as the user's industry, region, leasing stage, financing scale, cooperation status, and asset risk level, and a machine learning model is built based on key indicators to score the quality of potential customers, automatically screening out more valuable and high-qualified users who meet the requirements of risk policies, and reducing the interference of unqualified users on the marketing team. Then, sales follow-up optimization: if the conversion rate of the business opportunity before the sublease is low, consider increasing the follow-up frequency of the marketing team, or optimizing communication strategies, such as personalized communication. Through sales follow-up optimization, the marketing team can be provided with the user's historical cooperation information in the existing customer marketing scenario; and the marketing team can be provided with financing plans for similar customer groups such as the same industry and the same factory area for reference in the new customer marketing scenario. In the sales follow-up optimization stage, AB tests can also be used to continuously track the impact of different sales follow-up strategies on user conversion rates. This application feeds back user data by generating marketing result data. In this way, as the business iterates, user feature data and marketing strategy library are accumulated to achieve a closed loop of the user conversion funnel and dynamically track which marketing strategy is more effective in the market.

[0085] In summary, this application combines the AB testing method and the conversion funnel analysis framework to precipitate a set of indicator model construction and analysis dimension decomposition frameworks that are close to the business characteristics, which can locate the key nodes for finding and maintaining users in the leasing business, and provide data support for continuously adjusting marketing strategies, quickly understanding market trends, and analyzing user growth paths. In addition, this application incorporates the feedback from the marketing team and the asset performance of users after leasing into the analysis process, which can optimize the user screening model. At the same time, combined with data information from multiple stages of marketing, pre-rental, during lease, and post-rental, it can improve the screening criteria and follow-up strategies for lead users and business opportunity users.

[0086] The method provided in the embodiment of the present application obtains historical user behavior data, inputs the historical user behavior data into the conversion funnel model for business stage analysis, obtains the user conversion funnel output by the conversion funnel model, uses the historical user behavior data to evaluate and analyze the user conversion funnel, obtains the conversion funnel analysis results, formulates multiple initial strategies according to the conversion funnel analysis results, uses AB testing to test the multiple initial strategies, obtains the AB test results, uses the conversion funnel analysis results and the AB test results to adjust the multiple initial strategies, and obtains the target strategy. The historical user behavior data is screened by a machine learning algorithm to obtain high-quality users, and the effect of the AB test is tracked through the conversion funnel, providing data support for continuous adjustment of strategies, rapid understanding of market trends, and analysis of user growth paths, solving the problem that the entire process of the user life cycle cannot be fully collected by relying on APP. By integrating AB testing and conversion funnel analysis, and using artificial intelligence technology to quickly test different strategies, the user growth path can be continuously optimized to obtain a strategy with higher adaptability and better execution effect.

[0087] Further, as Figure 1 In the specific implementation of the method, the present application embodiment provides a strategy generation device, such as Figure 4A As shown, the device includes: an analysis module 401 , an evaluation module 402 , a test module 403 and an adjustment module 404 .

[0088] An analysis module 401 is used to obtain historical user behavior data, input the historical user behavior data into a conversion funnel model for business stage analysis, and obtain a user conversion funnel output by the conversion funnel model;

[0089] An evaluation module 402 is used to evaluate and analyze the user conversion funnel using the historical user behavior data to obtain a conversion funnel analysis result;

[0090] A testing module 403 is used to formulate multiple initial strategies according to the conversion funnel analysis result, and to test the multiple initial strategies using AB testing to obtain AB testing results;

[0091] The adjustment module 404 is used to adjust the multiple initial strategies using the conversion funnel analysis results and the AB test results to obtain a target strategy.

[0092] In specific application scenarios, such as Figure 4B As shown, the device also includes: a construction module 405.

[0093] Construction module 405 is used to obtain multiple stages included in the user's entire life cycle, and send the multiple stages to the terminal held by the user. The multiple stages include visitor stage, lead stage, business opportunity stage, pre-lease stage, mid-lease stage, and post-lease stage; receive multiple target stages selected by the user from the multiple stages based on the terminal held, and use the multiple target stages to construct the conversion funnel model.

[0094] In a specific application scenario, the analysis module 401 is used to determine multiple target users and user data of each target user based on the historical user behavior data, wherein the user data is the behavior information of the target user for the target business, and the target business is the business for which the strategy is to be implemented; the user data of the multiple target users are input into the conversion funnel model, and the conversion funnel model determines the target stage to which each target user belongs according to the user data of each target user, and the conversion funnel model divides the target users belonging to the same target stage among the multiple target users into the same user group to obtain multiple user groups, wherein the target stage is any one of the visitor stage, lead stage, business opportunity stage, pre-lease stage, mid-lease stage, and post-lease stage; and the user conversion funnel generated based on the multiple user groups output by the conversion funnel model is obtained.

[0095] In a specific application scenario, the evaluation module 402 is used to use the historical user behavior data to perform feature engineering calculations on the user conversion funnel to obtain a benchmark conversion rate for each target stage in the user conversion funnel; calculate the churn rate for each target stage in the user conversion funnel, obtain a churn rate threshold, and select at least one target stage with a churn rate greater than or equal to the churn rate threshold from multiple target stages as the funnel bottleneck of the user conversion funnel; obtain multiple preset classification labels, and use the multiple preset classification labels to perform detailed classification on the multiple user groups included in the user conversion funnel to obtain user classification results; calculate the key indicators of the user conversion funnel based on the user classification results and the historical user behavior data; and use the user conversion funnel, the benchmark conversion rate for each target stage in the user conversion funnel, the churn rate for each target stage in the user conversion funnel, the funnel bottleneck of the user conversion funnel, and the key indicators of the user conversion funnel as the conversion funnel analysis results.

[0096] In a specific application scenario, the test module 403 is used to determine the test objectives, test variables, test groups and test hypotheses according to the multiple initial strategies;

[0097] Based on the test objectives, the test variables, the test population and the test hypothesis, the multiple initial strategies are tested using AB testing to obtain the AB testing results.

[0098] In a specific application scenario, the adjustment module 304 is used to select a first strategy from the multiple initial strategies according to the AB test result, where the first strategy is the initial strategy with the highest score among the multiple initial strategies indicated by the AB test result; obtain multiple information acquisition channels, and based on the AB test, use the multiple information acquisition channels to test the multiple target stages included in the user, obtain a channel optimization test result, and use the channel optimization test result to adjust the first strategy;

[0099] The user conversion funnel and key indicators of the user conversion funnel are obtained from the conversion funnel analysis result, and a machine learning model is constructed using the key indicators of the user conversion funnel; the user conversion funnel is input into the machine learning model for evaluation to obtain a user quality score, and the user quality score is used to screen multiple user groups included in the user conversion funnel to obtain multiple high-qualified users; user data of the multiple high-qualified users are obtained, and the adjusted first strategy is adjusted using the user data of the multiple high-qualified users to obtain the target strategy.

[0100] In specific application scenarios, such as Figure 4B As shown, the device further includes: a display module 406.

[0101] The display module 406 is used to obtain the user conversion funnel, the benchmark conversion rate of each target stage in the user conversion funnel, the churn rate of each target stage in the user conversion funnel, and the key indicators of the user conversion funnel in the conversion funnel analysis results; obtain multiple user groups included in the user conversion funnel, and count the number of users corresponding to each target stage according to the multiple user groups; and display the number of users corresponding to each target stage, the benchmark conversion rate of each target stage in the user conversion funnel, the churn rate of each target stage in the user conversion funnel, the key indicators of the user conversion funnel, and the target strategy.

[0102] The device provided in the embodiment of the present application obtains historical user behavior data, inputs the historical user behavior data into the conversion funnel model for business stage analysis, obtains the user conversion funnel output by the conversion funnel model, uses the historical user behavior data to evaluate and analyze the user conversion funnel, obtains the conversion funnel analysis results, formulates multiple initial strategies according to the conversion funnel analysis results, uses AB testing to test the multiple initial strategies, obtains the AB test results, uses the conversion funnel analysis results and the AB test results to adjust the multiple initial strategies, and obtains the target strategy. The historical user behavior data is screened by a machine learning algorithm to obtain high-quality users, and the effect of the AB test is tracked through the conversion funnel, providing data support for continuous adjustment of strategies, rapid understanding of market trends, and analysis of user growth paths, solving the problem that the entire process of the user life cycle cannot be fully collected by relying on APP. By integrating AB testing and conversion funnel analysis, and using artificial intelligence technology to quickly test different strategies, the user growth path can be continuously optimized to obtain a strategy with higher adaptability and better execution effect.

[0103] It should be noted that for other corresponding descriptions of the functional units involved in the strategy generation device provided in the embodiment of the present application, reference can be made to Figure 1 , Figure 2 and FIG. 3A to FIG. 3B The corresponding description in will not be repeated here.

[0104] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0105] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0106] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

[0107] In an exemplary embodiment, see Figure 5, and also provides a computer device, which includes a bus, a processor, a memory and a communication interface, and may also include an input and output interface and a display device, wherein each functional unit can communicate with each other through the bus. The memory stores a computer program, and the processor is used to execute the program stored in the memory and execute the strategy generation method in the above embodiment.

[0108] A computer-readable storage medium stores a computer program, which implements the steps of the strategy generation method when executed by a processor.

[0109] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present application can be implemented by hardware, or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present application.

[0110] Those skilled in the art will appreciate that the accompanying drawings are merely schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily required for implementing the present application.

[0111] Those skilled in the art will appreciate that the modules in the device in the implementation scenario can be distributed in the device in the implementation scenario according to the implementation scenario description, or can be changed accordingly and located in one or more devices different from the implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple sub-modules.

[0112] The above application serial numbers are for description only and do not represent the advantages or disadvantages of the implementation scenarios.

[0113] The above disclosure only discloses several specific implementation scenarios of the present application. However, the present application is not limited thereto, and any changes that can be conceived by technicians in this field should fall within the scope of protection of the present application.

Claims

1. A strategy generation method, It is characterized in that include: Acquire historical user behavior data, input the historical user behavior data into a conversion funnel model for business stage analysis, and obtain a user conversion funnel output by the conversion funnel model; Using the historical user behavior data to evaluate and analyze the user conversion funnel to obtain a conversion funnel analysis result; Formulate multiple initial strategies according to the conversion funnel analysis results, and test the multiple initial strategies using AB testing to obtain AB testing results; The conversion funnel analysis results and the AB test results are used to adjust the multiple initial strategies to obtain a target strategy.

2. The strategy generation method according to claim 1, It is characterized in that Before inputting the historical user behavior data into the conversion funnel model to obtain the user conversion funnel output by the conversion funnel model, the method further includes: Acquire multiple stages of the user's entire life cycle, and send the multiple stages to the user's terminal, wherein the multiple stages include visitor stage, lead stage, business opportunity stage, pre-lease stage, mid-lease stage, and post-lease stage; A plurality of target stages selected by the user from the plurality of stages based on the terminal held by the user is received, and the conversion funnel model is constructed by using the plurality of target stages.

3. The strategy generation method according to claim 1, It is characterized in that The step of inputting the historical user behavior data into a conversion funnel model for business stage analysis to obtain a user conversion funnel output by the conversion funnel model includes: Determine multiple target users and user data of each target user according to the historical user behavior data, wherein the user data is behavior information of the target user for a target service, and the target service is a service for which a policy is to be implemented; Inputting user data of the multiple target users into the conversion funnel model, determining the target stage to which each target user belongs according to the user data of each target user by the conversion funnel model, and dividing the target users belonging to the same target stage among the multiple target users into the same user group by the conversion funnel model to obtain multiple user groups, wherein the target stage is any one of the visitor stage, the lead stage, the business opportunity stage, the pre-lease stage, the mid-lease stage, and the post-lease stage; The user conversion funnel generated based on the multiple user groups and output by the conversion funnel model is obtained.

4. The strategy generation method according to claim 3, It is characterized in that The use of the historical user behavior data to evaluate and analyze the user conversion funnel to obtain a conversion funnel analysis result includes: Using the historical user behavior data to perform feature engineering calculations on the user conversion funnel to obtain a benchmark conversion rate for each target stage in the user conversion funnel; Calculating the churn rate of each target stage in the user conversion funnel, obtaining a churn rate threshold, and selecting at least one target stage with a churn rate greater than or equal to the churn rate threshold from multiple target stages as the funnel bottleneck of the user conversion funnel; Acquire multiple preset classification tags, and use the multiple preset classification tags to refine and classify multiple user groups included in the user conversion funnel to obtain user classification results; Calculate the key indicators of the user conversion funnel based on the user classification results and the historical user behavior data; The user conversion funnel, the benchmark conversion rate of each target stage in the user conversion funnel, the churn rate of each target stage in the user conversion funnel, the funnel bottleneck of the user conversion funnel, and the key indicators of the user conversion funnel are used as the conversion funnel analysis results.

5. The strategy generation method according to claim 1, It is characterized in that The AB test is used to test the multiple initial strategies to obtain AB test results, including: Determining test objectives, test variables, test populations, and test hypotheses according to the plurality of initial strategies; Based on the test objectives, the test variables, the test population and the test hypothesis, the multiple initial strategies are tested using AB testing to obtain the AB testing results.

6. According to the strategy generation method of claim 1, It is characterized in that The step of adjusting the multiple initial strategies using the conversion funnel analysis results and the AB test results to obtain a target strategy includes: Selecting a first strategy from among the multiple initial strategies according to the AB test result, the first strategy being the initial strategy with the highest score among the multiple initial strategies indicated by the AB test result; Acquire multiple information acquisition channels, and based on the AB test, use the multiple information acquisition channels to test multiple target stages of the user to obtain channel optimization test results, and use the channel optimization test results to adjust the first strategy; Obtaining the user conversion funnel and key indicators of the user conversion funnel from the conversion funnel analysis result, and building a machine learning model using the key indicators of the user conversion funnel; Inputting the user conversion funnel into the machine learning model for evaluation to obtain a user quality score, and using the user quality score to screen multiple user groups included in the user conversion funnel to obtain multiple high-qualified users; The user data of the multiple high-qualified users is obtained, and the adjusted first policy is adjusted using the user data of the multiple high-qualified users to obtain the target policy.

7. The strategy generation method according to claim 1, It is characterized in that The method further comprises: Obtaining the user conversion funnel, the benchmark conversion rate of each target stage in the user conversion funnel, the churn rate of each target stage in the user conversion funnel, and the key indicators of the user conversion funnel from the conversion funnel analysis result; Acquire multiple user groups included in the user conversion funnel, and count the number of users corresponding to each target stage according to the multiple user groups; The number of users corresponding to each target stage, the benchmark conversion rate of each target stage in the user conversion funnel, the churn rate of each target stage in the user conversion funnel, the key indicators of the user conversion funnel, and the target strategy are displayed.

8. A strategy generating device, It is characterized in that include: An analysis module is used to obtain historical user behavior data, input the historical user behavior data into a conversion funnel model for business stage analysis, and obtain a user conversion funnel output by the conversion funnel model; An evaluation module, used to evaluate and analyze the user conversion funnel using the historical user behavior data to obtain a conversion funnel analysis result; A testing module, used to formulate multiple initial strategies according to the conversion funnel analysis results, and to test the multiple initial strategies using AB testing to obtain AB testing results; An adjustment module is used to adjust the multiple initial strategies using the conversion funnel analysis results and the AB test results to obtain a target strategy.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program. It is characterized in that When the processor executes the computer program, the steps of the strategy generation method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the strategy generation method according to any one of claims 1 to 7 are implemented.

Images