Tag cube-based customer group retention analysis method and apparatus, and electronic device

Abstract

The invention provides a customer group retention analysis method and device based on a tag cube and electronic equipment, and the method comprises the steps: carrying out the analysis of customer group tag data through employing the tag cube, generating and operating a customer group query statement corresponding to a target customer group in a preset database, and determining a target customer list corresponding to the target customer group; and performing SQL (Structured Query Language) assembly conversion on the target customer list and the retention data query statements by utilizing the tag cube, generating and running retention summary query statements so as to determine target user retention data tables corresponding to retention dates under each calculation date, and according to the target user retention data tables between each calculation date and the corresponding retention date, obtaining the target customer list according to the target user retention data tables between each calculation date and the corresponding retention date. And determining a retention analysis result of the target customer group for the retention analysis event. According to the method and the device, retention analysis of the target customer group is realized through the label cube and the retention analysis function, so that complex data calling and calculation are avoided, and the analysis accuracy and the analysis efficiency of the retention event are improved.

Description

Technical Field

[0001] The present application relates to the technical field of retention analysis models, and in particular to a customer retention analysis method, device and electronic device based on a label cube. Background Art

[0002] In the process of refined customer operations, it is impossible to monitor the entire marketing process, such as starting the application service to "payment success", and detecting the customer conversion rate in each process. Especially for relatively standardized business processes, long cycles, and more links, a technology is needed to intuitively discover and explain the problems in the business process, and at the same time scientifically reflect the important analysis model of the reasons for user loss from the initial behavior state of the customer group to the retention behavior stage.

[0003] Based on this, the existing technology provides a customer retention analysis model for measuring and analyzing user loyalty and stickiness to products or services. However, the existing customer retention analysis model may involve complex data processing and calculation in actual operation, especially when processing large-scale data sets, which may require advanced technology and computing resources. Summary of the invention

[0004] In view of this, the purpose of the present application is to at least provide a customer retention analysis method, device and electronic device based on the label cube, which realizes the retention analysis of the target customer group by combining the label cube with the retention analysis function, avoids complex data calls and calculations, and improves the accuracy and efficiency of retention event analysis.

[0005] This application mainly includes the following aspects:

[0006] In a first aspect, an embodiment of the present application provides a customer retention analysis method based on a label cube, comprising: determining rule configuration data corresponding to a retention analysis event through a customer retention analysis interface, the rule configuration data comprising customer label data, initial behavior event label data, retained behavior event label data, a calculation period and retention days, the calculation period comprising at least one continuous calculation date, the calculation date being the date on which the initial behavior event occurs, the retention days being the effective analysis period corresponding to each calculation date, the effective analysis period comprising multiple retention dates; performing SQL parsing processing on the initial behavior event label data, the retained behavior event label data, the retention analysis period and the retention days using a pre-created label cube model and a preset retention analysis function, generating a retention data query statement, the retention data query statement being used to determine a first user retention data table corresponding to each retention date under each calculation date, the first user retention data table corresponding to each retention date under each calculation date. The user retention data table is the intersection between users who have a retention behavior event on the retention date and users who have an initial behavior event on the corresponding calculation date; the customer group label data is parsed using the label cube to generate and run the customer group query statement corresponding to the target customer group in the preset database to determine the target customer list corresponding to the target customer group, and the customer group label data indicates the customer group query rule; the target customer list and the retention data query statement are SQL assembled and converted using the label cube to generate and run the retention summary query statement to determine the target user retention data table corresponding to each retention date under each calculation date, and the target user retention data table is the user intersection between the first user retention data table corresponding to the retention date and its corresponding calculation date and the target customer list; according to the target user retention data table between each calculation date and its corresponding retention date, the retention analysis result of the target customer group for the retention analysis event is determined.

[0007] In a possible implementation, the initial behavior event label data indicates an initial behavior event that satisfies first constraint information, and the retained behavior event label data indicates a retained behavior event that carries second constraint information, the first constraint information is a constraint condition corresponding to the initial behavior event, and the second constraint information is a constraint condition for the retained behavior event, wherein a retained data query statement is generated in the following manner: using a label cube to perform SQL parsing on the initial behavior event label data to generate a first user query statement, the first user query statement is used to locate a first user data set that triggers the initial behavior event on a calculation date; combining the retention date, using the label cube to perform SQL parsing on the retained behavior event label data to generate a second user query statement, the second user query statement is used to locate a second user data set that triggers a retained behavior event within each retention date; using a preset retained data analysis function, performing SQL assembly processing on the first user query statement and the second user query statement to generate a retained data query statement, the first user retained data table includes a first user intersection between the second user data set corresponding to the retention date and the first user data set corresponding to the calculation date to which it belongs.

[0008] In a possible implementation, the first user query statement and the second user query statement are generated in the following manner: in combination with the calculation cycle, the initial behavior event is subjected to SQL parsing conversion using the label cube to generate a first physical table access statement, the first physical table access statement is used to locate the first physical table corresponding to each calculation date, and the first physical table includes the user who has the initial behavior event on the calculation date; the first constraint information is subjected to SQL parsing conversion using the label cube to generate the initial behavior event constraint statement; the first physical table access statement and the initial behavior event constraint statement are assembled to generate the first user query statement; in combination with the retention days and the calculation cycle, the retained behavior event is subjected to SQL parsing conversion using the label cube to generate the second physical table access statement, the second physical table access statement is used to locate the second physical table corresponding to each retention date, and the second physical table includes the user who has the retained behavior event within the retention date; the second constraint information is subjected to SQL parsing conversion using the label cube to generate the retained behavior event constraint statement; the second physical table access statement and the retained behavior event constraint statement are assembled to generate the second user query statement.

[0009] In a possible implementation, before using the signature cube to perform SQL assembly conversion on the target customer list and the retention data query statement, the method also includes: determining whether there is comparison dimension label data in the rule configuration data, the comparison dimension label data indicating the analysis dimension for retention analysis on the target customer group; if there is comparison dimension label data in the rule configuration data, using the label cube model to perform SQL assembly conversion on the retention data query statement with different comparison dimensions respectively, and determine the target retention data query statement under different comparison dimensions, the target retention data query statement is used to locate the second user retention data table, and the second user retention data table includes users in the first user retention data table who meet the characteristics indicated by the comparison dimension.

[0010] In a possible implementation, a retention summary query statement is generated in the following manner: using a signature cube, SQL assembly conversion is performed on the target customer list and the target retention data query statement corresponding to each comparison dimension to generate a retention summary query statement for the target customer group under each comparison dimension.

[0011] In a possible implementation, the retention analysis result of the target customer group under each comparison dimension is determined in the following manner: counting the number of initial behavior users who have an initial behavior event within the calculation date; for each retention date corresponding to the calculation date, executing: counting the number of retained users corresponding to the target user retention data table corresponding to the retention date; calculating the percentage between the number of retained users corresponding to the retention date and the number of initial behavior users, and determining the percentage as the retention rate of the target customer group corresponding to the retention date.

[0012] In a possible implementation, the method further includes: displaying the retention analysis results in a retention chart, the retention chart at least showing the calculation date on which the initial behavior event occurred, the number of users on which the initial behavior event occurred on the calculation date, the number of retained users corresponding to each retention date under each calculation date, and the retention rate; determining whether the retention days need to be adjusted based on the retention chart, and if so, returning to redefine the retention days.

[0013] In a second aspect, an embodiment of the present application further provides a customer retention analysis device based on a label cube, the device comprising: a first determination module, for determining rule configuration data corresponding to a retention analysis event through a customer retention analysis interface, the rule configuration data comprising customer label data, initial behavior event label data, retained behavior event label data, calculation cycle and retention days, the calculation cycle comprising at least one continuous calculation date, the calculation date being the date of occurrence of the initial behavior event, the retention days being the effective analysis cycle corresponding to each calculation date, the effective analysis cycle comprising multiple retention dates; a first assembly module, for performing SQL parsing processing on the initial behavior event label data, the retained behavior event label data, the retention analysis cycle and the retention days using a pre-created label cube model and a preset retention analysis function, and generating a retention data query statement, the retention data query statement being used to determine the first user retention data table corresponding to each retention date under each calculation date, the first user retention The data table is the intersection between users who have a retention behavior event on the retention date and users who have an initial behavior event on the corresponding calculation date; the second assembly module is used to use the label cube to parse the customer group label data, generate and run the customer group query statement corresponding to the target customer group in the preset database, and determine the target customer list corresponding to the target customer group. The customer group label data indicates the customer group query rule; the third assembly module is used to use the label cube to perform SQL assembly conversion on the target customer list and the retention data query statement, generate and run the retention summary query statement to determine the target user retention data table corresponding to each retention date under each calculation date. The target user retention data table is the user intersection between the first user retention data table corresponding to the retention date and its corresponding calculation date and the target customer list; the retention analysis module is used to determine the retention analysis results of the target customer group for the retention analysis event based on the target user retention data table between each calculation date and its corresponding retention date.

[0014] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and the machine-readable instructions are executed by the processor to execute the steps of the customer retention analysis method based on the label cube in the above-mentioned first aspect or any possible implementation manner of the first aspect.

[0015] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of customer retention analysis based on the label cube in the above-mentioned first aspect or any possible implementation of the first aspect are executed.

[0016] The embodiment of the present application provides a method, device and electronic device for analyzing customer group retention based on a label cube, including: parsing customer group label data using a label cube, generating and running a customer group query statement corresponding to the target customer group in a preset database, and determining a target customer list corresponding to the target customer group; using a label cube to perform SQL assembly conversion on the target customer list and the retention data query statement, generating and running a retention summary query statement to determine the target user retention data table corresponding to each retention date under each calculation date, and determining the retention analysis results of the target customer group for the retention analysis event based on the target user retention data table between each calculation date and its corresponding retention date. The present application realizes retention analysis of the target customer group by combining the label cube with the retention analysis function, thereby avoiding complex data calls and calculations, and improving the accuracy and efficiency of retention event analysis.

[0017] The benefits of this application are:

[0018] Combining the label cube model with the retention analysis function, the SQL statement assembly is quickly completed with the help of the label cube model in the form of data tags, and further introduced into the retention analysis function. It can not only quickly complete the retention analysis of the target customer group based on the retention analysis needs, avoiding complex data calls and calculations, improving the accuracy of retention event analysis and optimizing data storage and query efficiency, but also locate the real-time customer group by combining the target customer group rules, ensuring the real-time and effectiveness of retention analysis. In addition, by providing different dimensional labels based on user feature classification, it can not only achieve a fine division of users, but also make it easier for users to conduct a comprehensive analysis of the retention status of the target customer group in different dimensions, so as to obtain more comprehensive insights into user retention behavior within the optimal retention days.

[0019] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 A flow chart of a customer retention analysis method based on a tag cube provided in an embodiment of the present application is shown;

[0022] Figure 2 A flowchart of a method for generating a query statement for retained data provided in an embodiment of the present application is shown;

[0023] Figure 3 A flow chart of a method for generating a customer group query statement provided in an embodiment of the present application is shown;

[0024] Figure 4 A flowchart of a method for generating a target retention data query statement provided in an embodiment of the present application is shown;

[0025] Figure 5 A functional module diagram of a customer retention analysis device based on a label cube provided in an embodiment of the present application is shown;

[0026] Figure 6 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0027] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can be implemented out of sequence, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art, under the guidance of the content of the present application, can add one or more other operations to the flowchart, or remove one or more operations from the flowchart.

[0028] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0029] In the process of refined customer operations, it is impossible to monitor the entire marketing process, such as starting the application service to "payment success", detecting the customer conversion rate in each process. Especially for processes with relatively standardized business processes, long cycles, and more links, a technology is needed to intuitively discover and explain the problems in the business process, and at the same time scientifically reflect the important analysis model of the reasons for user loss from the initial behavior state of the customer group to the retention behavior stage.

[0030] Based on this, the existing technology provides a customer retention analysis model for measuring and analyzing user loyalty and stickiness to products or services. It mainly has the following values: identifying loyal users. The retention analysis model can help identify loyal users who continue to use products or services over a period of time; evaluating the effect of product iteration. By comparing the retention rates of different versions of products or services in the same time period, it can be evaluated whether the product iteration has brought positive results; optimizing user base: retention analysis can reveal the reasons for user churn, help product teams understand and optimize user experience, thereby improving user retention rate; customized analysis: according to different business needs and user characteristics, the retention analysis model can be customized to adapt to specific analysis goals; measuring product market performance: retention analysis can measure the long-term value and market performance of the product to users, and help companies formulate user retention strategies and product improvement measures; identifying churned users: retention analysis can also help identify user groups that may or have left, so that corresponding measures can be taken to reduce user churn.

[0031] However, existing customer retention analysis models may involve complex data processing and calculations in actual operation, especially when dealing with large-scale data sets, which may require advanced technology and computing resources.

[0032] Based on this, the embodiments of the present application provide a customer retention analysis method, device and electronic device based on a label cube, which realizes the retention analysis of the target customer group by combining the label cube with the retention analysis function, thereby avoiding complex data calls and calculations, and improving the accuracy and efficiency of retention event analysis, as follows:

[0033] See also Figure 1 , Figure 1 FIG. 1 is a flowchart of a method for analyzing customer retention based on a tag cube provided in an embodiment of the present application. Figure 1 As shown, the method provided in the embodiment of the present application comprises the following steps:

[0034] S100. Determine the rule configuration data corresponding to the retention analysis event through the customer group retention analysis interface.

[0035] The rule configuration data includes customer group label data, initial behavior event label data, retained behavior event label data, calculation cycle and retention days. The calculation cycle includes at least one continuous calculation date, and the calculation date is the date on which the initial behavior event occurs. The retention days are the effective analysis period corresponding to each calculation date, and the effective analysis period includes multiple retention dates.

[0036] S200, using a pre-created label cube model and a preset retention analysis function to perform SQL parsing on the initial behavior event label data, the retained behavior event label data, the retention analysis cycle and the retention days, and generate a retention data query statement.

[0037] Among them, the retention data query statement is used to determine the first user retention data table corresponding to each retention date under each calculation date, and the first user retention data table is the intersection between users who have a retention behavior event on the retention date and users who have an initial behavior event on the corresponding calculation date.

[0038] S300: parse the customer group label data using the label cube, generate and run a customer group query statement corresponding to the target customer group in a preset database, and determine a target customer list corresponding to the target customer group.

[0039] Among them, the customer group label data indicates the customer group query rules.

[0040] S400: Use the label cube to perform SQL assembly conversion on the target customer list and the retention data query statement, generate and run the retention summary query statement, so as to determine the target user retention data table corresponding to each retention date under each calculation date.

[0041] The target user retention data table is the user intersection between the first user retention data table corresponding to the retention date and the calculation date to which it belongs and the target customer list.

[0042] S500: Determine the retention analysis result of the target customer group for the retention analysis event according to the target user retention data table between each calculation date and its corresponding retention date.

[0043] The label cube model is a fact- and dimension-based database model that meets the needs of users to query and analyze data from multiple angles and levels. Its basic application is to implement OLAP. It is a technology commonly used in data analysis and indexing, which can establish multi-dimensional indexes for original data.

[0044] Analyzing data through the label cube model can speed up data query efficiency. In the intelligent label portrait platform system, the label cube model metadata content management includes table name, dimension column, metric column, threshold column, associated label category, label information, and finally forms a mapping relationship between SQL and metadata and data dictionary, and saves it to the relational database to realize the logical mapping from the application layer to the database layer and as a business rule parsing template.

[0045] In steps S100 to S500, the present application provides rule configuration data corresponding to retention analysis events through a retention analysis interface, and combines the label cube with the retention analysis model or retention analysis function to create a label-based retention data analysis method. This can avoid complex data calls and calculations on the basis of the original retention analysis model, thereby improving the accuracy and efficiency of retention event analysis.

[0046] In a specific embodiment, the retention analysis interface includes customer group configuration items, initial behavior event configuration items, retention behavior event configuration items, calculation cycle configuration items, and retention days configuration items, wherein customer group label data is generated in the following manner:

[0047] In response to a selection operation performed on a customer group configuration item, multiple customer group labels are displayed, each preset customer group label indicates a customer group (e.g., a lost customer group, a young customer group with a diamond card for a credit card), a selection operation on a target customer group label is received, and customer group label data is generated, the customer group label data indicating the target customer group.

[0048] The initial behavior event label data indicates an initial behavior event that satisfies the first constraint information. The first constraint information is a constraint condition corresponding to the initial behavior event. In a specific embodiment, the initial behavior event label data is generated in the following manner:

[0049] In response to a selection operation performed on an initial behavior event configuration item, an initial behavior event configuration interface is displayed, the initial behavior event configuration interface includes multiple event configuration controls, different event configuration controls are used to configure different types of events, specifically, the multiple event configuration controls include but are not limited to at least one of the following items: a contact event configuration control, a business event configuration control, a contract event configuration control, a product event configuration control, and a transaction event configuration control; in response to a selection operation performed on a target event configuration control, multiple event selection items corresponding to the target event configuration control are displayed; in response to a selection operation performed on a target event selection item, an event list corresponding to the target event selection item is displayed, the event list includes multiple behavior events; in response to a selection operation performed on a target behavior event, the target behavior event is determined as an initial behavior event; in response to an edit operation performed on the initial behavior event, first constraint information corresponding to the initial behavior event is generated, the first constraint information may constrain the time and place of occurrence of the initial behavior event, etc., and no specific restrictions are made here.

[0050] For example, the multiple event selection items corresponding to the contact event configuration control include but are not limited to at least one of the following items: an application event selection item, a participation event selection item, a browsing event selection item, and a login event selection item, receiving a selection operation performed on the application event selection item, and expanding multiple behavioral events under the application event category, such as a retail customer applying for a credit card or a retail customer applying for a loan.

[0051] The retained behavior event label data indicates a retained behavior event that satisfies the second constraint information. The second constraint information is the constraint condition of the retained behavior event. The second constraint information can constrain the location where the retained behavior event occurs, etc. No specific restrictions are made here. The configuration method of the retained behavior event label data is the same as the configuration method of the initial behavior event label data, which will not be repeated here.

[0052] In another preferred embodiment, in response to the configuration operation performed on the retention analysis period configuration item, the calculation period and retention days corresponding to the retention analysis event are determined. For example, the retention days provide at least 7 days, 14 days, one month, one year from the current date, and a customized retention day setting method.

[0053] In addition, in response to the configuration operation performed on the retention days configuration item, the retention days corresponding to each retention date are determined.

[0054] In a preferred embodiment, see Figure 2 , Figure 2 A flowchart of a method for generating a query statement for retained data provided by an embodiment of the present application is shown. Figure 2 As shown, in step S200, a retained data query statement is generated in the following manner:

[0055] S2001. Use the label cube to perform SQL parsing on the initial behavior event label data to generate a first user query statement.

[0056] The first user query statement is used to locate a first user data set that triggers an initial behavior event on a calculation date.

[0057] S2002. In combination with the retention date, the tag cube is used to perform SQL parsing on the retention behavior event tag data to generate a second user query statement.

[0058] The second user query statement is used to locate a second user data set that triggers a retention behavior event within each retention date.

[0059] S2003. Using a preset retained data analysis function, perform SQL assembly processing on the first user query statement and the second user query statement to generate a retained data query statement.

[0060] The first user retention data table includes a first user intersection between a second user data set corresponding to a retention date and a first user data set corresponding to a calculation date to which the second user data set belongs.

[0061] In a specific embodiment of the present application, assuming that the retention days are 3 days and the calculation period is January 6, 2023 to January 7, 2023, it can be determined that the calculation dates within the calculation period include January 6, 2023 and January 7, 2023, a total of two days. Combined with the retention days, taking the calculation date January 6, 2023 as an example, the retention dates corresponding to January 6, 2023 are determined to be January 6, 2023 to January 9, 2023, which are respectively defined as the retention day corresponding to January 6, 2023, the first retention day, the second retention day and the third retention day.

[0062] Therefore, in this application, through the first user query statement corresponding to January 6, 2023, the first user data set corresponding to January 6, 2023 can be generated, and the first user data set includes the user ID of the user who had an initial behavior event on January 6, 2023. Through the second user query statement, the second user data set corresponding to each retention date can be generated, including the second user data sets corresponding to the retention day, the first retention day, the second retention day and the third retention day corresponding to January 6, 2023, respectively. Taking the first retention day corresponding to January 6, 2023 as an example, the second user data set corresponding to the first retention day includes the user ID of the user who had a retention behavior event on January 7, 2023.

[0063] In the present application, a preset retention data analysis function is used to perform SQL assembly on the first user query statement corresponding to each calculation date and the second user query statement corresponding to each retention date under the calculation date for each calculation date to generate a retention data query statement. If the retention data query statement is run, taking the above-mentioned calculation date January 6, 2023 and the corresponding retention date January 7, 2023 as an example, the retention data query statement takes the first user intersection between the first user data set corresponding to January 6, 2023 and the second user data set corresponding to January 7, 2023 during the running process, that is, the intersection between the user who had an initial behavior event on January 6, 2023 and the user who had a retention behavior event on January 7, 2023 is taken as the first user retention data table corresponding to the first day of retention with January 6, 2023 as the calculation date.

[0064] In a preferred embodiment, step S2001 includes:

[0065] In combination with the calculation cycle, the label cube is used to perform SQL parsing conversion on the initial behavior event to generate a first physical table access statement, the first physical table access statement is used to locate the first physical table corresponding to each calculation date, the first physical table includes users who have initial behavior events on the calculation date, the label cube is used to perform SQL parsing conversion on the first constraint information to generate an initial behavior event constraint statement, the first physical table access statement and the initial behavior event constraint statement are assembled and processed to generate a first user query statement.

[0066] Specifically, for different business needs or analysis needs, when creating the initial behavior event label data, the first constraint information corresponding to the initial behavior event can be added with the help of the customer retention analysis interface to implement the constraint restrictions on the initial behavior event, such as adding a constraint on the location of the initial behavior event. On this basis, the label cube processes the initial behavior event label data in two processes, so that the first user query statement finally assembled can gradually obtain the first user data set. Specifically, the initial behavior event needs to be converted into a first physical table access statement first to locate users (not constrained) who only have initial behavior events on the calculation date, and then the first constraint information is converted into an initial behavior event constraint statement, and then the initial behavior event constraint statement is combined with the first physical table access statement, that is, a layer of constraint conditions is added to the users located by the first physical table access statement to form a first user query statement.

[0067] In combination with the retention days and the calculation cycle, the label cube is used to perform SQL parsing conversion on the retention behavior events to generate a second physical table access statement. The second physical table access statement is used to locate the second physical table corresponding to each retention date. The second physical table includes users who have retention behavior events within the retention date. The label cube is used to perform SQL parsing conversion on the second constraint information to generate a retention behavior event constraint statement. The second physical table access statement and the retention behavior event constraint statement are assembled to generate a second user query statement.

[0068] The second user query statement is generated in the same manner as the first user query statement, and will not be described in detail herein.

[0069] See also Figure 3 , Figure 3 A flow chart of a method for generating a customer group query statement provided in an embodiment of the present application is shown. Figure 3 As shown, step S300 includes:

[0070] S3001. Determine the target customer group rule corresponding to the target customer group label according to the mapping relationship between the customer group label and the customer group rule.

[0071] In the present application, different customer group classification rules are pre-defined, and each customer group classification rule is used to locate a corresponding customer group. The database is constantly updated, which means that even if the customer group resource package is uploaded, the customer group resource package is not real-time. If you only select the corresponding customer group resource package as the target customer group for analysis, the analysis results will not meet the current actual needs. Therefore, in the present application, a mapping relationship between customer group labels and customer group rules is pre-established, so that after determining the customer group labels, the customer group rules used to locate the target customer group can be directly called.

[0072] S3002. With the help of the label cube and the arrayJoin function, the selected target customer group rules are further subjected to SQL parsing conversion to generate corresponding customer group query statements.

[0073] Among them, the customer group query statement makes it easy to locate the accurate target customer list in real time when processing retention analysis events, ensuring the real-time and effectiveness of subsequent retention analysis results.

[0074] In a preferred embodiment, see Figure 4 , Figure 4 A flow chart of a method for generating a target retention data query statement provided by an embodiment of the present application is shown. Figure 4 As shown, before executing step S400, the method further includes:

[0075] SA1. Determine whether there is comparison dimension label data in the rule configuration data.

[0076] The comparison dimension label data indicates the analysis dimension for performing retention analysis on the target customer group.

[0077] SA2. If there is comparison dimension label data in the rule configuration data, the label cube model is used to perform SQL assembly conversion on the retained data query statements respectively with different comparison dimensions to determine the target retained data query statements under different comparison dimensions.

[0078] SA3. If there is no comparison dimension label data in the rule configuration data, execute step S400.

[0079] The target retention data query statement is used to locate a second user retention data table, where the second user retention data table includes users in the first user retention data table who meet the characteristics indicated by the comparison dimension.

[0080] In one example, the customer retention analysis interface of this application also provides a comparison dimension configuration item. The user can decide whether to configure the generation of comparison dimension labels based on the actual needs of the retention analysis event. Specifically, the comparison dimension label data is generated in the following way:

[0081] In response to a selection operation performed on a comparison dimension configuration item, a plurality of dimension tags are displayed, and in response to a selection operation performed on a plurality of target dimension tags, comparison dimension tag data are generated.

[0082] In a specific embodiment, multiple dimension tags are system built-in tags, and users can gradually select target dimension tags according to the types described in the tags. Specifically, users select dimension tags according to the actual comparison requirements of the retention analysis event. For example, the comparison dimensions required by the retention analysis event are gender = male and gender = female. There is no specific restriction on the comparison dimensions here, and users are allowed to select multiple dimension tags at the same time to achieve the final dimension comparison.

[0083] Specifically, users are segmented in advance according to their characteristics (such as behavior, demographic information, etc.) to form multi-dimensional labels, which is more suitable for meeting the retention data analysis needs of different customers and achieving the expansion of the analysis direction.

[0084] In summary, before executing step S400, the present application needs to first determine whether there is comparison dimension label configuration data in the rule configuration data. If there is no comparison dimension label data, it means that the retention analysis event does not need to be displayed for dimension comparison. If there is comparison dimension label configuration data, it is also necessary to use the label cube to group, assemble and transform the retention data query statements according to the dimension labels indicated in the comparison dimension label data, so as to split them into target retention data query statements under different comparison dimensions.

[0085] For example, if the comparison dimension label data includes two dimensions, male and female, then it is necessary to use the label cube to assemble the retention data query statement with the male and female dimensions respectively, and finally generate the target retention data query statements corresponding to the male and female dimensions respectively. For the target retention data query statement corresponding to the male dimension, it is used to locate the male users in the first user retention data table, and the third user retention data table is formed by the male users in the first user retention data table.

[0086] In this application, by providing comparison dimension label data and combining it with label cubes for further analysis and assembly, multi-dimensional retention analysis of the target customer group can be achieved. While solving the complexity of multi-dimensional comparison processing, more comprehensive user retention analysis results can be obtained, which is more convenient for users to perform a series of subsequent operations.

[0087] In a preferred embodiment, step S400 includes:

[0088] Using the label cube, SQL assembly and conversion are performed on the target customer list and the target retention data query statements corresponding to each comparison dimension to generate a retention summary query statement for the target customer group under each comparison dimension.

[0089] In this application, in this application, after determining the target retention data query statement corresponding to each comparison dimension, it is necessary to use the label cube to associate and assemble the target customer list with the target retention data query statement corresponding to each comparison dimension to obtain the retention summary query statement corresponding to each comparison dimension, thereby realizing retention analysis under different comparison dimensions.

[0090] In a preferred embodiment, step S500 includes:

[0091] Count the number of initial behavior users who have initial behavior events within the calculation date. For each retention date corresponding to the calculation date, execute: Count the number of retained users corresponding to the target user retention data table corresponding to the retention date, calculate the percentage between the number of retained users corresponding to the retention date and the number of initial behavior users, and determine the percentage as the retention rate of the target customer group corresponding to the retention date.

[0092] In a specific embodiment, the number of initial behavior users = the number of users recorded in the first user data set. Specifically, assuming that the number of retention days = 3, and the calculation date is June 1, 2023, then through the above method, it is determined that for users who trigger the corresponding initial behavior event on June 1, 2023, the statistical results are as follows:

[0093] The retention rate on the day of retention (June 1, 2023) = the number of retained users on that day / the number of initial behavior users corresponding to June 1, 2023 × 100%;

[0094] The first-day retention rate corresponding to the first day of retention (June 2, 2023) = the number of retained users on the first day / the number of initial behavior users corresponding to June 1, 2023 × 100%;

[0095] The retention rate on the second day of retention (June 3, 2023) = the number of retained users on the second day / the number of initial behavior users corresponding to June 1, 2023 × 100%;

[0096] The third-day retention rate corresponding to the third day of retention (June 4, 2023) = the number of retained users on the third day / the number of initial behavior users corresponding to June 1, 2023 × 100%.

[0097] The retention analysis method corresponding to each calculation date is the same as above and will not be repeated here.

[0098] Specifically, choosing the right retention days is crucial for retention analysis. Setting retention days that are too short may not accurately reflect user behavior, while setting retention days that are too long may introduce noise.

[0099] In a possible implementation, the method further includes:

[0100] The retention analysis results are displayed in a retention chart. The retention chart at least shows the calculation date when the initial behavior event occurred, the number of users who had the initial behavior event on the calculation date, the number of retained users corresponding to each retention date under each calculation date, and the retention rate. Based on the retention chart, determine whether the retention days need to be adjusted. If the retention days need to be adjusted, return to redetermine the retention days.

[0101] In this application, according to different comparison dimensions, the retention analysis results corresponding to each comparison dimension are displayed in the form of retention charts in the display area corresponding to the comparison dimension, or a switching control can be set to switch the retention analysis results under different comparison dimensions.

[0102] For each retention analysis result, according to the number of retained users and the retention rate corresponding to each retention date under any calculation date, the preset effective analysis rules are used to determine whether the effective analysis period or the retention days setting is reasonable (for example, the effective analysis rule may be to recognize the retention days before the retention rate reaches 0 as the effective retention period. Under this rule, assuming that the retention date is set to 7 days, but the retention analysis result indicates that the retention rate is 0 on the third day of retention, it is determined that the retention days are set too long, indicating that the retention days setting is unreasonable). If the retention days setting is unreasonable, a retention days adjustment prompt is generated to remind the user to reset the retention days.

[0103] It can be seen that this application helps users readjust the retention days by introducing preset effective analysis rules, thereby determining the optimal retention days corresponding to the retention analysis events, and eliminating the influence of noise, so that the final retention analysis results can more accurately reflect user behavior.

[0104] Based on the same application concept, the embodiment of the present application also provides a customer retention analysis method device based on a label cube corresponding to the customer retention analysis method based on a label cube provided in the above embodiment. Since the principle of solving the problem by the device in the embodiment of the present application is similar to the customer retention analysis method based on a label cube in the above embodiment of the present application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0105] See also Figure 5 , Figure 5 FIG. 1 shows a functional module diagram of a customer retention analysis device based on a label cube provided in an embodiment of the present application. Figure 5 As shown, the device comprises:

[0106] The first determination module 600 is used to determine the rule configuration data corresponding to the retention analysis event through the customer retention analysis interface. The rule configuration data includes customer label data, initial behavior event label data, retention behavior event label data, calculation period and retention days. The calculation period includes at least one continuous calculation date, and the calculation date is the date of occurrence of the initial behavior event. The retention days are the effective analysis period corresponding to each calculation date, and the effective analysis period includes multiple retention dates.

[0107] The first assembly module 610 is used to use a pre-created label cube model and a preset retention analysis function to perform SQL parsing processing on the initial behavior event label data, the retained behavior event label data, the retention analysis period and the retention days to generate a retention data query statement. The retention data query statement is used to determine the first user retention data table corresponding to each retention date under each calculation date. The first user retention data table is the intersection between users who have a retention behavior event on the retention date and users who have an initial behavior event on the corresponding calculation date.

[0108] The second assembly module 620 is used to parse the customer group label data using the label cube, generate and run the customer group query statement corresponding to the target customer group in the preset database, determine the target customer list corresponding to the target customer group, and the customer group label data indicates the customer group query rule.

[0109] The third assembly module 630 is used to use the label cube to perform SQL assembly conversion on the target customer list and the retention data query statement, generate and run the retention summary query statement to determine the target user retention data table corresponding to each retention date under each calculation date. The target user retention data table is the user intersection between the first user retention data table corresponding to the retention date and the calculation date to which it belongs and the target customer list.

[0110] The retention analysis module 640 is used to determine the retention analysis results of the target customer group for the retention analysis event based on the target user retention data table between each calculation date and its corresponding retention date.

[0111] Based on the same application idea, please refer to Figure 6 , Figure 6 FIG. 1 is a schematic diagram showing the structure of an electronic device provided by an embodiment of the present application. Figure 6 As shown, the electronic device 700 includes: a processor 710, a memory 720 and a bus 730. The memory 720 stores machine-readable instructions executable by the processor 710. When the electronic device 700 is running, the processor 710 and the memory 720 communicate through the bus 730. The machine-readable instructions are executed by the processor 710 when it is running, such as the steps of the customer retention analysis method based on the label cube provided in any of the above embodiments.

[0112] Based on the same application concept, an embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the customer retention analysis method based on the label cube provided in the above embodiment are executed.

[0113] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0114] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0115] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0116] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0117] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A customer retention analysis method based on label cube, characterized in that: The method comprises: Determine the rule configuration data corresponding to the retention analysis event through the customer group retention analysis interface, wherein the rule configuration data includes customer group label data, initial behavior event label data, retention behavior event label data, calculation period and retention days, wherein the calculation period includes at least one continuous calculation date, the calculation date is the date of occurrence of the initial behavior event, and the retention days are the effective analysis period corresponding to each calculation date, and the effective analysis period includes multiple retention dates; Using a pre-created label cube model and a preset retention analysis function, SQL parsing is performed on the initial behavior event label data, the retained behavior event label data, the retention analysis cycle, and the retention days to generate a retention data query statement, wherein the retention data query statement is used to determine the first user retention data table corresponding to each retention date under each calculation date, wherein the first user retention data table is the intersection between users who have a retention behavior event on the retention date and users who have an initial behavior event on the corresponding calculation date; Utilizing the label cube to parse the customer group label data, generating and running a customer group query statement corresponding to the target customer group in a preset database, determining a target customer list corresponding to the target customer group, wherein the customer group label data indicates a customer group query rule; Using the label cube, the target customer list and the retention data query statement are SQL assembled and converted to generate and run a retention summary query statement to determine a target user retention data table corresponding to each retention date under each calculation date, wherein the target user retention data table is a user intersection between a first user retention data table corresponding to a retention date and its corresponding calculation date and the target customer list; According to the target user retention data table between each calculation date and its corresponding retention date, the retention analysis results of the target audience for the retention analysis event are determined.

2. The method according to claim 1, characterized in that The initial behavior event label data indicates an initial behavior event that satisfies first constraint information, and the retained behavior event label data indicates a retained behavior event that carries second constraint information, wherein the first constraint information is a constraint condition corresponding to the initial behavior event, and the second constraint information is a constraint condition for the retained behavior event. The retained data query statement is generated in the following manner: Using the label cube to perform SQL parsing on the initial behavior event label data to generate a first user query statement, where the first user query statement is used to locate a first user data set that triggers the initial behavior event on a calculation date; In combination with the retention date, the retention behavior event label data is subjected to SQL parsing using the label cube to generate a second user query statement, where the second user query statement is used to locate a second user data set that triggers a retention behavior event within each retention date; Using the preset retention data analysis function, the first user query statement and the second user query statement are SQL assembled to generate a retention data query statement, wherein the first user retention data table includes the first user intersection between the second user data set corresponding to the retention date and the first user data set corresponding to the calculation date to which it belongs.

3. The method according to claim 2, characterized in that The first user query statement and the second user query statement are generated in the following manner: In combination with the calculation cycle, the initial behavior event is subjected to SQL parsing conversion using the label cube to generate a first physical table access statement, wherein the first physical table access statement is used to locate a first physical table corresponding to each calculation date, wherein the first physical table includes users who have the initial behavior event on the calculation date; Using the label cube to perform SQL parsing conversion on the first constraint information to generate an initial behavior event constraint statement; Assembling the first physical table access statement and the initial behavior event constraint statement to generate a first user query statement; In combination with the retention days and the calculation cycle, the retention behavior event is subjected to SQL parsing conversion using the label cube to generate a second physical table access statement, where the second physical table access statement is used to locate a second physical table corresponding to each retention date, where the second physical table includes users who have the retention behavior event within the retention date; Using the label cube to perform SQL parsing conversion on the second constraint information to generate a retention behavior event constraint statement; The second physical table access statement and the retained behavior event constraint statement are assembled to generate a second user query statement.

4. The method according to claim 1, characterized in that Before using the signing cube to perform SQL assembly conversion on the target customer list and the retained data query statement, the method further includes: Determining whether there is comparison dimension label data in the rule configuration data, the comparison dimension label data indicating an analysis dimension for performing retention analysis on a target customer group; If there is comparison dimension label data in the rule configuration data, the label cube model is used to perform SQL assembly conversion on the retention data query statement and different comparison dimensions respectively, and the target retention data query statement under different comparison dimensions is determined. The target retention data query statement is used to locate the second user retention data table, and the second user retention data table includes users in the first user retention data table who meet the characteristics indicated by the comparison dimension.

5. The method according to claim 4, characterized in that Generate a retention summary query statement in the following way: By using the signature cube, the target customer list and the target retention data query statements corresponding to each comparison dimension are respectively subjected to SQL assembly conversion to generate a retention summary query statement of the target customer group under each comparison dimension.

6. The method according to claim 4, characterized in that Determine the retention analysis results of the target audience in each comparison dimension by: Count the number of initial behavior users who have initial behavior events within the calculation date; For each retention date corresponding to the calculation date, execute: count the number of retained users corresponding to the target user retention data table corresponding to the retention date; The percentage between the number of retained users corresponding to the retention date and the number of users with the initial behavior is calculated, and the percentage is determined as the retention rate of the target customer group corresponding to the retention date.

7. The method according to claim 1, characterized in that The method further comprises: The retention analysis result is presented in a retention chart, wherein the retention chart at least displays the calculation date on which the initial behavior event occurs, the number of users on which the initial behavior event occurs on the calculation date, the number of retained users corresponding to each retention date under each calculation date, and the retention rate; According to the retention chart, determine whether the retention days need to be adjusted. If the retention days need to be adjusted, return to re-determine the retention days.

8. A customer retention analysis device based on a label cube, characterized in that: The device comprises: A first determination module is used to determine the rule configuration data corresponding to the retention analysis event through the customer group retention analysis interface, wherein the rule configuration data includes customer group label data, initial behavior event label data, retention behavior event label data, calculation period and retention days, wherein the calculation period includes at least one continuous calculation date, the calculation date is the date of occurrence of the initial behavior event, and the retention days are the effective analysis period corresponding to each calculation date, and the effective analysis period includes multiple retention dates; A first assembly module, for performing SQL parsing processing on the initial behavior event label data, the retained behavior event label data, the retention analysis period and the retention days using a pre-created label cube model and a preset retention analysis function, and generating a retention data query statement, wherein the retention data query statement is used to determine the first user retention data table corresponding to each retention date under each calculation date, wherein the first user retention data table is the intersection between users who have a retention behavior event on the retention date and users who have an initial behavior event on the corresponding calculation date; A second assembly module is used to parse the customer group label data using the label cube, generate and run a customer group query statement corresponding to the target customer group in a preset database, determine a target customer list corresponding to the target customer group, and the customer group label data indicates a customer group query rule; A third assembly module is used to use the label cube to perform SQL assembly conversion on the target customer list and the retention data query statement, generate and run a retention summary query statement, so as to determine the target user retention data table corresponding to each retention date under each calculation date, wherein the target user retention data table is a user intersection between the first user retention data table corresponding to the retention date and the calculation date to which it belongs and the target customer list; The retention analysis module is used to determine the retention analysis results of the target customer group for the retention analysis event based on the target user retention data table between each calculation date and its corresponding retention date.

9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the memory communicate through the bus. When the machine-readable instructions are run by the processor, the steps of the customer retention analysis method based on the label cube as described in any one of claims 1 to 7 are executed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the customer retention analysis method based on the label cube as described in any one of claims 1 to 7 are executed.

Images