Marketing data analysis method and device, electronic equipment and storage medium

Abstract

The invention discloses a marketing data analysis method and apparatus, an electronic device and a storage medium. The method comprises the steps of obtaining real-time circulation data of each marketing right and interest and right and interest behavior data of multiple clients in a target marketing activity process; constructing a graph model of the target marketing activity by using the real-time flow data of various marketing rights and interests; performing path tracking and node identification on the graph model of the target marketing activity by using a graph calculation algorithm to obtain a data consanguinity analysis result; wherein the data blood relationship analysis result comprises a tracking path and a key node; constructing a structured causal model based on the rights and interests behavior data of the plurality of customers; based on a structured causal model, carrying out anti-fact analysis on various marketing rights and interests and customer consumption behaviors through a causal inference algorithm to obtain causal relationship information of various marketing rights and interests and customer consumption behaviors; and adjusting the marketing strategy of the target marketing activity based on the data blood relationship analysis result and the causal relationship information of the target marketing activity.

Description

Technical Field

[0001] The present application relates to the field of data analysis technology, and in particular to a marketing data analysis method and device, electronic equipment, and storage medium. Background Art

[0002] Currently, most banks sell financial products and other products through their client applications, or promote asset-related businesses. In order to stimulate customers to purchase or increase the amount of bank assets, they often invest a certain amount of money in marketing activities on client applications, such as issuing various coupons or points and other benefits.

[0003] When conducting online activities, we currently rely on historical marketing experience and intuitive judgment of rights and products to formulate corresponding marketing strategies and distribute various rights in accordance with the marketing strategies. When the overall marketing effect does not meet the needs, we will rely on historical operations and intuitive judgment to adjust the strategy until the overall marketing effect meets the needs, that is, the sales of the product meet the expected requirements.

[0004] Therefore, the current method lacks accurate data support and in-depth data analysis, and cannot obtain the impact of various factors on consumption. Therefore, it may take frequent random strategy adjustments to meet expectations, and there may be situations where expectations are never met. In addition, the marketing effect meets the needs in a timely manner, but it is often not the best effect. It is just a compromise choice that cannot effectively wait for a better solution. Therefore, there is an urgent need for an accurate scientific analysis method for marketing-related data to provide accurate scientific data support for marketing formulation. Summary of the invention

[0005] Based on the above-mentioned deficiencies of the prior art, the present application provides a marketing data analysis method and device, an electronic device, and a storage medium to solve the problem that the prior art lacks the support of accurate scientific data.

[0006] In order to achieve the above objectives, this application provides the following technical solutions:

[0007] The first aspect of the present application provides a marketing data analysis method, comprising:

[0008] Acquire real-time flow data of each marketing benefit and benefit behavior data of multiple customers during the target marketing activity; wherein the customer benefit behavior data includes data of the customer's consumption behavior related to each marketing benefit;

[0009] Using the real-time flow data of various marketing rights, a graph model of the target marketing activity is constructed;

[0010] Using a graph computing algorithm to perform path tracing and node identification on the graph model of the target marketing activity, to obtain a data lineage analysis result of the target marketing activity; wherein the data lineage analysis result includes the traced path and the identified key nodes;

[0011] Building a structured causal model based on the equity behavior data of multiple customers;

[0012] Based on the structured causal model, a counterfactual analysis is performed on the causal relationship between the various marketing benefits and the consumption behavior of the customer through a causal inference algorithm to obtain causal relationship information between the various marketing benefits and the consumption behavior of the customer;

[0013] Based on the data lineage analysis results of the target marketing activity and the causal relationship information, the marketing strategy of the target marketing activity is adjusted.

[0014] Optionally, in the above-mentioned marketing data analysis method, the use of the flow data of various marketing rights to construct a graph model of the target marketing activity includes:

[0015] Each of the marketing rights, each marketing expense of the marketing rights, and each customer and each product transferred in the transfer data of each of the marketing rights are respectively taken as nodes;

[0016] Based on the circulation data of the various marketing rights, the nodes are connected, and attributes are configured for the nodes and edges.

[0017] Optionally, in the above-mentioned marketing data analysis method, constructing a structured causal model based on the equity behavior data of the plurality of customers includes:

[0018] Clean and standardize the equity behavior data of each of the above customers;

[0019] Based on the equity behavior data of each of the customers, define equity types, customer behavior types, and potential influencing factors;

[0020] The defined equity type, customer behavior type, and potential influencing factors are used as variables of the model, and combined with the determined covariates to construct the structured causal model.

[0021] Optionally, in the above-mentioned marketing data analysis method, based on the structured causal model, a counterfactual analysis is performed on the causal relationship between various marketing rights and the consumption behavior of the customer through a causal inference algorithm to obtain causal relationship information between various marketing rights and the consumption behavior of the customer, including:

[0022] Based on the structured causal model, calculating a propensity score for each of the customers;

[0023] Matching each of the customers based on their propensity scores to obtain a matched treatment group and a control group;

[0024] respectively calculating potential outcomes for each of the customers in the treatment group;

[0025] Calculating a causal effect result based on the deviation of the potential result of each of the clients in the treatment group from the result of each of the matched clients in the control group;

[0026] The causal effect results are analyzed to obtain causal relationship information between the various marketing benefits and the customer's consumption behavior.

[0027] A second aspect of the present application provides a marketing data analysis device, comprising:

[0028] An acquisition unit, used to acquire real-time flow data of each marketing benefit and benefit behavior data of multiple customers during a target marketing activity; wherein the customer benefit behavior data includes data of the customer's consumption behavior related to each marketing benefit;

[0029] A graph model building unit, configured to build a graph model of the target marketing activity by using the real-time flow data of various marketing rights;

[0030] A lineage analysis unit, configured to use a graph computing algorithm to perform path tracing and node identification on the graph model of the target marketing activity, and obtain a data lineage analysis result of the target marketing activity; wherein the data lineage analysis result includes the traced path and the identified key nodes;

[0031] A causal model building unit, used to build a structured causal model based on the equity behavior data of a plurality of said customers;

[0032] A causal analysis unit, configured to perform a counterfactual analysis on the causal relationship between the various marketing benefits and the consumption behavior of the customer based on the structured causal model by using a causal inference algorithm to obtain causal relationship information between the various marketing benefits and the consumption behavior of the customer;

[0033] An adjustment unit is used to adjust the marketing strategy of the target marketing activity based on the data lineage analysis result of the target marketing activity and the causal relationship information.

[0034] Optionally, in the above-mentioned marketing data analysis device, the graph model building unit includes:

[0035] A node construction unit, used to respectively use each of the marketing rights, each marketing expense of each of the marketing rights, and each customer and each product to which the circulation data of each of the marketing rights flows as nodes;

[0036] A connection unit, used to connect the nodes based on the flow data of the various marketing rights;

[0037] The configuration unit is used to configure attributes for each of the nodes and edges.

[0038] Optionally, in the above-mentioned marketing data analysis device, the causal model building unit includes:

[0039] A data preprocessing unit, used for cleaning and standardizing the equity behavior data of each of the customers;

[0040] A factor definition unit, used to define equity type, customer behavior type, and potential influencing factors based on the equity behavior data of each of the customers;

[0041] The model building unit is used to use the defined equity type, the customer behavior type, and the potential influencing factors as variables of the model, and combine them with the determined covariates to build the structured causal model.

[0042] Optionally, in the above-mentioned marketing data analysis device, the cause-effect analysis unit includes:

[0043] A score calculation unit, used to calculate a propensity score for each of the customers based on the structured causal model;

[0044] A matching unit, used for matching each of the customers based on the propensity score of each of the customers to obtain a matched treatment group and a control group;

[0045] A potential result calculation unit, used for respectively calculating the potential result of each of the customers in the processing group;

[0046] A causal effect calculation unit, configured to calculate a causal effect result based on the deviation between the potential result of each of the clients in the treatment group and the result of each of the matched clients in the control group;

[0047] The interpretation unit is used to analyze the causal effect results to obtain the causal relationship information between the various marketing rights and the customer's consumption behavior.

[0048] A third aspect of the present application provides an electronic device, including:

[0049] Memory and processor;

[0050] Wherein, the memory is used to store programs;

[0051] The processor is used to execute the program, and when the program is executed, it is specifically used to implement the marketing data analysis method as described in any one of the above.

[0052] A fourth aspect of the present application provides a computer storage medium for storing a computer program, which, when executed by a processor, is used to implement a marketing data analysis method as described in any one of the above.

[0053] The present application provides a marketing data analysis method to obtain the real-time flow data of each marketing right and the rights behavior data of multiple customers during the target marketing activities. Among them, the rights behavior data of customers include the data of the consumption behavior of customers related to each marketing right. Then, the real-time flow data of various marketing rights are used to build a graph model of the target marketing activities, and the graph computing algorithm is used to track the path and identify the nodes of the graph model of the target marketing activities to obtain the data lineage analysis results of the target marketing activities. Among them, the data lineage analysis results include the tracked path and the identified key nodes, so that by performing lineage analysis on the marketing data, it is known how the funds are converted into rights and the specific impact of various rights on consumption, thereby providing accurate data for the adjustment of each node in the marketing strategy. In addition, based on the rights and interests behavior data of multiple customers, a structured causal model is constructed, and based on the structured causal model, a counterfactual analysis of the causal relationship between various marketing rights and customers' consumption behavior is performed through a causal inference algorithm to obtain the causal relationship information between various marketing rights and customers' consumption behavior, that is, the causal relationship between various marketing rights and customers' consumption behavior is obtained, thereby providing accurate data for adjusting the marketing rights. Finally, based on the data lineage analysis results and causal relationship information of the target marketing activities, the marketing strategy of the target marketing activities is adjusted, thereby providing accurate scientific data for the formulation of marketing strategies by combining lineage analysis and causal relationship analysis, and then the marketing strategy can be accurately adjusted without frequent adjustments, and the best effect can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0055] Figure 1 A flowchart of a marketing data analysis method provided in an embodiment of the present application;

[0056] Figure 2 A flowchart of a method for constructing a graphical model of a target marketing activity provided in an embodiment of the present application;

[0057] Figure 3 A flowchart of a method for constructing a structured causal model provided in an embodiment of the present application;

[0058] Figure 4 A flow chart of a method for analyzing the causal relationship between marketing rights and consumer behavior provided in an embodiment of the present application;

[0059] Figure 5 A schematic diagram of the architecture of a marketing data analysis device provided in an embodiment of the present application;

[0060] Figure 6 A schematic diagram of the architecture of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0062] In this application, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0063] The present application embodiment provides a marketing data analysis method, such as Figure 1 As shown, the following steps are included:

[0064] S101. Obtain real-time flow data of each marketing benefit during a target marketing activity and benefit behavior data of multiple customers.

[0065] Optionally, the categories of marketing benefits can be divided according to needs, such as coupons, points, etc., and coupons of different amounts, etc., can also belong to different marketing benefits.

[0066] The real-time circulation data of marketing rights refers to the information that is transferred to each node in the whole process from the allocation of marketing rights to the use of rights in real time. Specifically, it may include but is not limited to the allocation of marketing rights, the generation of marketing rights, that is, the generation of a corresponding number of marketing rights according to the funds, the distribution, collection and use of marketing rights, etc.

[0067] The customer's rights and interests behavior data includes the customer's consumption behavior data related to each marketing right, that is, the user's consumption and use of the marketing rights received, such as the type and quantity of the marketing rights received, the type and quantity of the products purchased, etc.

[0068] S102. Use the real-time flow data of various marketing rights to build a graph model of target marketing activities.

[0069] It should be noted that step S102 and step S103 are processes for analyzing data blood relationship, while step S104 and step S105 are processes for analyzing causal relationship. The two processes are related and do not affect each other. Therefore, in the embodiment of the present application, the execution order of the two analysis processes is only an optional method, and other orders may also be used.

[0070] Since the graph model can directly represent and process the relationships between the various flow nodes of marketing rights and interests, it can facilitate the lineage analysis of marketing data. Therefore, it is necessary to first use the real-time flow data of marketing rights and interests to construct the various flow nodes involved in the real-time flow data of marketing rights and interests, and connect the flow nodes according to the flow order of each flow node in the real-time flow data of marketing rights and interests, so as to obtain the graph model of the target marketing activity.

[0071] It should also be noted that constructing the graph model of the target marketing activity may be constructing a completely new graph model, or may be further updating a previously constructed model to thereby construct an updated graph model.

[0072] Optionally, in another embodiment of the present application, a specific implementation of step S102 is as follows: Figure 2 As shown, the following steps are included:

[0073] S201. Take each marketing right, each marketing expense of each marketing right, and each customer and each product that flows through the circulation data of each marketing right as nodes.

[0074] It should be noted that in the marketing process, the marketing funds are mainly allocated first, that is, the marketing funds are divided into multiple funds and then transferred to various marketing rights, thereby generating various marketing rights. The marketing rights are transferred to users through distribution and collection, and finally the users use the rights to purchase products, thereby transferring to the final products. Therefore, in the embodiment of the present application, corresponding nodes are mainly constructed for each marketing right, each marketing fund, each customer and each product.

[0075] S202. Based on the circulation data of various marketing rights, connect the nodes and configure attributes for each node and each edge.

[0076] Specifically, based on the circulation data of various marketing rights, the circulation process of rights can be determined, so that each node can be connected in the order of circulation. In order to clarify and distinguish each node and each edge connecting the nodes, it is necessary to configure corresponding attributes for each node and each edge.

[0077] Among them, node attributes can include corresponding attribute information such as equity type, customer purchase history, etc. Edge attributes can include the specific time, quantity, and amount of equity transfer.

[0078] S103. Use a graph computing algorithm to perform path tracing and node identification on a graph model of the target marketing activity to obtain a data lineage analysis result of the target marketing activity.

[0079] Among them, the data lineage analysis results include the tracked path and the identified key nodes.

[0080] It should be noted that the path tracked based on real-time circulation data can reflect the real-time circulation of equity, so as to determine the blood relationship between various nodes in the process of equity circulation, as well as the circulation efficiency of equity and whether it has been circulated to products, etc., so it can be used as reference data for adjusting marketing strategies. The key nodes reflect the key nodes in the process of equity circulation, so as to determine which nodes to focus on for adjustment, so as to obtain a better marketing strategy.

[0081] Optionally, data lineage analysis can be performed using network centrality analysis algorithms and cluster analysis algorithms. Among them, the network centrality analysis algorithm can determine the most important nodes in the graph model, that is, the key nodes, such as which benefits are the most popular and which customers are the most active. The cluster analysis algorithm can identify node clusters with similar attributes or behaviors and analyze which customer groups are more interested in a certain type of benefit. Community detection algorithms, such as modularity maximization algorithms, are often used for analysis.

[0082] S104. Construct a structured causal model based on the equity behavior data of multiple customers.

[0083] It should be noted that causality is different from statistical correlation, that is, it is different from relevance. Statistical correlation refers to the degree of association between two or more variables, while causality involves how one variable (cause) affects another variable (result). Usually, correlation analysis cannot reflect the direction of causality, or cannot distinguish between cause and effect. Therefore, in order to accurately analyze the impact of various types of marketing rights on the consumption of various products, in the embodiments of the present application, the causal relationship between various types of marketing rights and the consumption of various products is analyzed, so as to more accurately reflect the specific impact of various types of marketing rights on the consumption of various products through causal relationships.

[0084] In an embodiment of the present application, causal analysis is performed by constructing a structured causal model. A structured causal model is a mathematical model that uses a graph structure to represent the causal relationship between variables. Therefore, in an embodiment of the present application, the causal relationship between model rights and consumer behavior is represented by this model. Specifically, the rights and interests behavior data of multiple customers include situations in which customers use various marketing rights to purchase products, so based on the rights and interests behavior data of multiple customers, a causal relationship between model rights and consumer behavior reflected in the initial rights and interests behavior data is constructed.

[0085] Optionally, in another embodiment of the present application, a specific implementation of step S104 is as follows: Figure 3 As shown, the following steps are included:

[0086] S301. Clean and standardize the equity behavior data of each customer.

[0087] Specifically, in order to facilitate the processing of equity behavior data, it is necessary to first clean and standardize it so as to unify it into a processable format. For example, in the purchase behavior, "not purchased" is coded as "0" and the behavior of "purchase" is coded as "1".

[0088] S302. Based on the equity behavior data of each customer, define the equity type, customer behavior type, and potential influencing factors.

[0089] Based on the equity behavior data of each customer, define the types of marketing equity involved in the data and the types of customer behavior involved, and also need to consider other influencing factors, such as customer asset level and past purchase history, so that the corresponding results can be accurately calculated for the input variables through a structured causal model.

[0090] S303. Defined equity types, customer behavior types, and potential influencing factors are used as variables of the model, and combined with determined covariates to construct a structured causal model.

[0091] S105. Based on the structured causal model, a counterfactual analysis is performed on the causal relationship between various marketing rights and the customer's consumption behavior through a causal inference algorithm to obtain causal relationship information between various marketing rights and the customer's consumption behavior.

[0092] Counterfactual analysis refers to setting conditions that are contrary to the facts, analyzing the differences between the obtained results and the actual results, so as to determine the causal relationship between the corresponding conditions and the results. Therefore, in order to determine whether a certain type of marketing rights and interests and a certain customer's consumption behavior really have a causal relationship, in an embodiment of the present application, a counterfactual analysis is performed on the causal relationship between various marketing rights and interests and the customer's consumption behavior based on a structured causal model. For example, a user uses a certain marketing right to purchase a certain amount of products, and in turn analyzes the situation of purchasing the product when the user does not receive the right under other conditions, such as whether the product is purchased and the amount of the purchase.

[0093] Optionally, in another embodiment of the present application, a specific implementation of step S105 is as follows: Figure 4 As shown, the following steps are included:

[0094] S401. Calculate the propensity score of each customer based on the structured causal model.

[0095] It should be noted that in order to conduct comparative analysis of the results, it is necessary to divide the treatment group and the comparison group, that is, the customer group that received the benefits and the customer group that did not receive the benefits. Since the propensity score equals the probability of receiving the treatment under a given covariate, that is, the probability of receiving the benefits in the embodiment of the present application, each customer can be grouped by calculating the propensity score of each client.

[0096] The calculation method of the propensity score can be expressed as: (e(X)=P(T=1|X)), where (X) is a covariate.

[0097] S402: Match each customer based on their propensity score to obtain a matched treatment group and control group.

[0098] Specifically, each customer can be grouped based on his or her propensity score, and then the treatment group and the control group can be matched through nearest neighbor matching or propensity score weighting method, that is, matching the treatment group and the control group that are contrasted with each other, that is, matching users who are as similar as possible in other aspects except for the receipt of benefits.

[0099] S403, respectively calculating the potential results of each customer in the processing group.

[0100] Specifically, the potential results of each customer in the treatment group can be calculated based on the potential result model in the structured causal model, that is, the specific purchase situation of the product by each customer in the treatment group assuming that they have not received the marketing rights.

[0101] S404. Calculate the causal effect results based on the deviation between the potential results of each customer in the treatment group and the results of each matched customer in the control group.

[0102] Matched customers refer to users who are similar in other aspects in terms of rights and interests collection. Therefore, we can compare the consumption of customers when they actually have not collected their rights and interests with the consumption of customers when they are assumed to have not collected their rights and interests, and thus obtain the impact of corresponding rights and interests on consumption behavior, that is, the causal effect result.

[0103] Optionally, multiple deviations may be calculated based on the deviations of the potential outcomes for each customer in the treatment group from the outcomes for each matched customer in the control group, and then averaged for the same type of marketing equity and product to obtain the causal effect results for each type of marketing equity on each type of product.

[0104] S405. Analyze the causal effect results to obtain causal relationship information between various marketing rights and customers' consumption behaviors.

[0105] It should be noted that the larger the causal effect result, the greater the effect of whether the marketing rights are received or not. It means that if the rights are not received, it will have little impact on the user's consumption behavior and cannot make their consumption behavior closer to the real consumption behavior of not receiving the rights. Therefore, it means that the marketing rights have little impact on the product, and therefore the causal relationship between the marketing rights and the consumption behavior of the product is weak. Therefore, by analyzing the causal effect results, we can get whether there is a causal relationship between various marketing rights and the customer's consumption behavior, and the size of the relationship, and explain the causal relationship between various marketing rights and the customer's consumption behavior.

[0106] S106. Adjust the marketing strategy of the target marketing activity based on the data lineage analysis results and causal relationship information of the target marketing activity.

[0107] Specifically, through the results of data lineage analysis, we can fully understand how each fund is converted into different marketing rights and how these rights affect customers' consumption behavior, so as to evaluate the corresponding effects of various marketing rights on consumption, and can identify key users and rights, so we can make targeted adjustments to the allocation of funds for rights and the allocation of rights for various users. Similarly, the causal relationship information includes the causal relationship between various marketing rights and customers' consumption behavior, so when making adjustments, we focus on improving the allocation of funds for marketing rights with strong causal relationships.

[0108] The embodiment of the present application provides a marketing data analysis method to obtain the real-time flow data of each marketing right and the rights behavior data of multiple customers during the target marketing activity. Among them, the rights behavior data of customers include the data of the consumption behavior of customers related to each marketing right. Then, the real-time flow data of various marketing rights are used to build a graph model of the target marketing activity, and the graph model of the target marketing activity is tracked and nodes are identified using a graph computing algorithm to obtain the data lineage analysis results of the target marketing activity. Among them, the data lineage analysis results include the tracked path and the identified key nodes, so that by performing lineage analysis on the marketing data, it is known how the funds are converted into rights and the specific impact of various rights on consumption, thereby providing accurate data for the adjustment of each node in the marketing strategy. In addition, based on the rights and interests behavior data of multiple customers, a structured causal model is constructed, and based on the structured causal model, a counterfactual analysis of the causal relationship between various marketing rights and customers' consumption behavior is performed through a causal inference algorithm to obtain the causal relationship information between various marketing rights and customers' consumption behavior, that is, the causal relationship between various marketing rights and customers' consumption behavior is obtained, thereby providing accurate data for adjusting the marketing rights. Finally, based on the data lineage analysis results and causal relationship information of the target marketing activities, the marketing strategy of the target marketing activities is adjusted, thereby providing accurate scientific data for the formulation of marketing strategies by combining lineage analysis and causal relationship analysis, and then the marketing strategy can be accurately adjusted without frequent adjustments, and the best effect can be achieved.

[0109] Another embodiment of the present application provides a marketing data analysis device, such as Figure 5 As shown, including:

[0110] The acquisition unit 501 is used to acquire the real-time flow data of each marketing right and the right behavior data of multiple customers during the target marketing activity, wherein the right behavior data of customers includes the consumption behavior data of customers related to each marketing right.

[0111] The graph model building unit 502 is used to build a graph model of a target marketing activity by utilizing the real-time flow data of various marketing rights.

[0112] The lineage analysis unit 503 is used to use a graph computing algorithm to perform path tracing and node identification on the graph model of the target marketing activity to obtain a data lineage analysis result of the target marketing activity, wherein the data lineage analysis result includes the traced path and the identified key nodes.

[0113] The causal model building unit 504 is used to build a structured causal model based on the equity behavior data of multiple customers.

[0114] The causal analysis unit 505 is used to perform counterfactual analysis on the causal relationship between various marketing rights and the customer's consumption behavior based on the structured causal model and through a causal inference algorithm to obtain causal relationship information between various marketing rights and the customer's consumption behavior.

[0115] The adjustment unit 506 is used to adjust the marketing strategy of the target marketing activity based on the data lineage analysis results and causal relationship information of the target marketing activity.

[0116] Optionally, in the marketing data analysis device provided in another embodiment of the present application, the graph model building unit includes:

[0117] The node construction unit is used to respectively use each marketing right, each marketing expense of each marketing right, and each customer and each product flowing through the flow data of each marketing right as nodes.

[0118] The connection unit is used to connect various nodes based on the flow data of various marketing rights.

[0119] Configuration unit, used to configure properties for each node and each edge.

[0120] Optionally, in the marketing data analysis device provided in another embodiment of the present application, the causal model building unit includes:

[0121] The data preprocessing unit is used to clean and standardize the equity behavior data of each customer.

[0122] The factor definition unit is used to define the equity type, customer behavior type, and potential influencing factors based on the equity behavior data of each customer.

[0123] The model building unit is used to construct a structured causal model by taking the defined equity type, customer behavior type, and potential influencing factors as variables of the model and combining them with the determined covariates.

[0124] Optionally, in the marketing data analysis device provided in another embodiment of the present application, the causal analysis unit includes:

[0125] The score calculation unit is used to calculate the propensity score of each customer based on the structured causal model.

[0126] The matching unit is used to match each customer based on the propensity score of each customer to obtain a matched treatment group and a matched control group.

[0127] The potential result calculation unit is used to calculate the potential result of each customer in the processing group respectively.

[0128] The causal effect calculation unit is used to calculate the causal effect results based on the deviation of the potential results of each customer in the treatment group and the results of each matched customer in the control group.

[0129] The explanation unit is used to analyze the causal effect results and obtain the causal relationship information between various marketing rights and customers' consumption behaviors.

[0130] It should be noted that the specific working process of each unit provided in the above embodiments of the present application can refer to the implementation process of the corresponding steps in the above method embodiments, and will not be repeated here.

[0131] Another embodiment of the present application provides an electronic device, such as Figure 6 As shown, including:

[0132] Memory 601 and processor 602 .

[0133] The memory 601 is used to store programs.

[0134] The processor 602 is used to execute the program stored in the memory 601. When the program is executed, it is specifically used to implement the marketing data analysis method provided in any one of the above embodiments.

[0135] Another embodiment of the present application provides a computer storage medium for storing a computer program, which, when executed by a processor, is used to implement a marketing data analysis method as provided in any of the above embodiments.

[0136] Computer storage media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0137] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0138] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A marketing data analysis method, characterized in that: include: Acquire real-time flow data of each marketing benefit and benefit behavior data of multiple customers during the target marketing activity; wherein the customer benefit behavior data includes data of the customer's consumption behavior related to each marketing benefit; Using the real-time flow data of various marketing rights, a graph model of the target marketing activity is constructed; Using a graph computing algorithm to perform path tracing and node identification on the graph model of the target marketing activity, to obtain a data lineage analysis result of the target marketing activity; wherein the data lineage analysis result includes the traced path and the identified key nodes; Building a structured causal model based on the equity behavior data of multiple customers; Based on the structured causal model, a counterfactual analysis is performed on the causal relationship between the various marketing benefits and the consumption behavior of the customer through a causal inference algorithm to obtain causal relationship information between the various marketing benefits and the consumption behavior of the customer; Based on the data lineage analysis results of the target marketing activity and the causal relationship information, the marketing strategy of the target marketing activity is adjusted.

2. The method according to claim 1, characterized in that The method of utilizing the circulation data of various marketing rights to construct a graph model of the target marketing activity includes: Each of the marketing rights, each marketing expense of the marketing rights, and each customer and each product transferred in the transfer data of each of the marketing rights are respectively taken as nodes; Based on the circulation data of the various marketing rights, the nodes are connected, and attributes are configured for the nodes and edges.

3. The method according to claim 1, characterized in that The constructing of a structured causal model based on the equity behavior data of the plurality of customers includes: Clean and standardize the equity behavior data of each of the above customers; Based on the equity behavior data of each of the customers, define equity types, customer behavior types, and potential influencing factors; The defined equity type, customer behavior type, and potential influencing factors are used as variables of the model, and combined with the determined covariates to construct the structured causal model.

4. The method according to claim 1, characterized in that Based on the structured causal model, a counterfactual analysis is performed on the causal relationship between the various marketing rights and the customer's consumption behavior through a causal inference algorithm to obtain causal relationship information between the various marketing rights and the customer's consumption behavior, including: Based on the structured causal model, calculating a propensity score for each of the customers; Matching each of the customers based on the propensity score of each of the customers to obtain a matched treatment group and a control group; respectively calculating potential outcomes for each of the customers in the treatment group; Calculating a causal effect result based on the deviation of the potential result of each of the clients in the treatment group from the result of each of the matched clients in the control group; The causal effect results are analyzed to obtain causal relationship information between the various marketing benefits and the customer's consumption behavior.

5. A marketing data analysis device, characterized in that: include: An acquisition unit, used to acquire real-time flow data of each marketing benefit and benefit behavior data of multiple customers during a target marketing activity; wherein the customer benefit behavior data includes data of the customer's consumption behavior related to each marketing benefit; A graph model building unit, configured to build a graph model of the target marketing activity by using the real-time flow data of various marketing rights; A lineage analysis unit, configured to use a graph computing algorithm to perform path tracing and node identification on the graph model of the target marketing activity, and obtain a data lineage analysis result of the target marketing activity; wherein the data lineage analysis result includes the traced path and the identified key nodes; A causal model building unit, used to build a structured causal model based on the equity behavior data of a plurality of said customers; A causal analysis unit, configured to perform a counterfactual analysis on the causal relationship between the various marketing benefits and the consumption behavior of the customer based on the structured causal model by using a causal inference algorithm to obtain causal relationship information between the various marketing benefits and the consumption behavior of the customer; An adjustment unit is used to adjust the marketing strategy of the target marketing activity based on the data lineage analysis result of the target marketing activity and the causal relationship information.

6. The device according to claim 5, characterized in that The graph model building unit includes: A node construction unit, used to respectively use each of the marketing rights, each marketing expense of each of the marketing rights, and each customer and each product to which the circulation data of each of the marketing rights flows as nodes; A connection unit, used to connect the nodes based on the flow data of the various marketing rights; The configuration unit is used to configure attributes for each of the nodes and edges.

7. The device according to claim 5, characterized in that The causal model building unit comprises: A data preprocessing unit, used for cleaning and standardizing the equity behavior data of each of the customers; A factor definition unit, used to define equity type, customer behavior type, and potential influencing factors based on the equity behavior data of each of the customers; The model building unit is used to use the defined equity type, the customer behavior type, and the potential influencing factors as variables of the model, and combine them with the determined covariates to build the structured causal model.

8. The device according to claim 5, characterized in that The causal analysis unit comprises: A score calculation unit, used to calculate a propensity score for each of the customers based on the structured causal model; A matching unit, used for matching each of the customers based on the propensity score of each of the customers to obtain a matched treatment group and a control group; A potential result calculation unit, used for respectively calculating the potential result of each of the customers in the processing group; A causal effect calculation unit, configured to calculate a causal effect result based on the deviation between the potential result of each of the clients in the treatment group and the result of each of the matched clients in the control group; The interpretation unit is used to analyze the causal effect results to obtain the causal relationship information between the various marketing rights and the customer's consumption behavior.

9. An electronic device, characterized in that: include: Memory and processor; Wherein, the memory is used to store programs; The processor is used to execute the program, and when the program is executed, it is specifically used to implement the marketing data analysis method as described in any one of claims 1 to 4.

10. A computer storage medium, characterized in that: Used to store a computer program, which, when executed by a processor, is used to implement the marketing data analysis method as described in any one of claims 1 to 4.

Images