Marketing strategy optimization processing method and system based on user grouping

Abstract

The invention provides a marketing strategy optimization processing method and system based on user grouping. The marketing strategy optimization processing method based on user grouping comprises the following steps: accessing business data of users and user vehicles of a plurality of business systems to a customer service operation platform, and constructing ID maps of the users; using the related business data of the user and the user vehicle to construct a label of the user; performing grouping processing on the users according to the labels of the users, and obtaining target grouping packets corresponding to the users; obtaining a multi-dimensional portrait of a target group corresponding to the target grouping packet through a deep analysis and insight mode based on a label corresponding to the user; and according to the target grouping packet corresponding to the user, making a differentiated operation strategy for groups with different value layers, accessing the user through multiple channels, recovering effect data, and optimizing the strategy by taking the effect data as guidance. The system comprises modules corresponding to the steps of the method.

Description

Technical Field

[0001] The present invention provides a method and a system for optimizing a marketing strategy based on user clustering, belonging to the technical field of strategy optimization processing. Background Art

[0002] In the economic downturn cycle, the external environment has become more complex, competition in the automotive industry has intensified, and there is a risk of elimination; digitization has improved the efficiency of physical industries, and traditional industries are facing accelerated transformation. The high-quality demand has promoted the expansion and structural optimization of the service industry, and the integration of digital intelligence and the service industry has brought new growth opportunities. With the demand for digital transformation of automobile enterprises, the role of software system platforms in the operation of automobile industry enterprises has changed from a tool to improve enterprise operation efficiency to a driving force to promote enterprise innovation. Software in the fields of automotive marketing and after-sales service will further penetrate into the entire life cycle of car owners' car purchase, use, and maintenance, and integrate with numerous software and application scenarios. Through software, automobile enterprises can better understand consumer needs, enhance the consumer experience, and improve consumer satisfaction. According to statistics, the software market in the automotive after-sales market will continue to expand in the future, with an average annual growth rate exceeding 25%, and each automobile enterprise is increasing its investment in the digital transformation of the automotive after-sales market.

[0003] The customer service operation system has undergone many changes, evolving from the past single and passive service to the current refined and proactive service model. In the future, enterprises will face more challenges in the field of customer service. In the past: there was a single product or service, limited customer touchpoints and sales channels, more passive responses, and a lack of proactive service awareness. Now: the service journey is three-dimensional and diverse, consumption channels are more abundant, the customer reach link is shorter, and it is necessary to work meticulously to actively occupy the user's mind. In the future: deepen the service depth, broaden the service breadth, apply and popularize VR&AR technologies, and empower customer service with data analysis.

[0004] The digital economy and the real economy are accelerating their integration, the digital investment in the automotive after-sales market is continuously increasing, and the customer service operation system is facing profound changes. As the marketing environment is changing, traditional marketing models face the following challenges: (1) The number of online touchpoints has increased. Affected by the mobile Internet, users are accelerating their migration to the online world. While users' attention is becoming more fragmented, it has become more difficult to integrate resources and achieve effective communication. (2) The marketing cost has increased. With the disappearance of the traffic dividend and the increase in market saturation, market competition has become increasingly fierce, and the difficulty and cost of acquiring new customers at the traffic end have gradually increased. (3) The conversion effect has deteriorated. With the continuous influx of diverse content, the rough customer acquisition and one-size-fits-all user operation are difficult to continue, and undifferentiated marketing can no longer easily impress target customers. Summary of the Invention

[0005] The present invention provides a method and system for optimizing marketing strategies based on user grouping to solve the technical problems in the above-mentioned prior art. The technical solutions adopted are as follows:

[0006] A method for optimizing marketing strategies based on user grouping, the method for optimizing marketing strategies based on user grouping includes:

[0007] Access the relevant business data of users and user vehicles in multiple business systems to a customer service operation platform, and construct an ID map corresponding to the users;

[0008] Construct labels corresponding to users by using the relevant business data of users and user vehicles;

[0009] Group the users according to the labels corresponding to the users, obtain the target grouping package corresponding to the users, and display the users and the data information with upstream and downstream dependency relationships according to the target grouping package;

[0010] Based on the labels corresponding to the users, obtain multi-dimensional portraits of the target groups corresponding to the target grouping package through in-depth analysis and insight methods;

[0011] According to the target grouping package corresponding to the users, formulate differentiated operation strategies for groups with different value hierarchies, reach the users through multiple channels and collect effect data, and optimize the strategies guided by the effect data.

[0012] Further, accessing the relevant business data of users and user vehicles in multiple business systems to a customer service operation platform and constructing an ID map corresponding to the users includes:

[0013] Access the original system business data corresponding to the users and user vehicles in multiple business systems to a customer service operation platform; wherein, the multiple business systems include but are not limited to a marketing system, an ERP system, and a vehicle system;

[0014] Perform data cleaning on the original system business data, and transmit the original system business data after data cleaning to the target source; wherein, the data cleaning includes but is not limited to screening, de-duplication, and replacement;

[0015] Configure IDmapping logic, set the priority in the matching process of multi-source data according to the source and data integrity of the original system business data in combination with actual business rules, and at the same time, establish a unique identifier for the users and their corresponding user vehicles, wherein the unique identifier is OneID;

[0016] Based on the user ID information, the fragmented data of multiple accounts of the same user are associated, and the fragmented data are integrated to generate integrated data corresponding to the user ID information; wherein the user ID information includes but is not limited to the user ID, mobile phone number, device number, etc.

[0017] Furthermore, data quality monitoring is performed on the integrated data corresponding to the user ID information, including:

[0018] Scan the integrated data corresponding to the user ID information to obtain the initial abnormal characteristic elements contained in the integrated data; wherein the initial abnormal characteristic elements include the number of missing key fields and the weight value corresponding to the missing data and the data garbled rate contained in the integrated data and the weight value corresponding to the garbled data;

[0019] According to the number of missing key fields combined with the corresponding weight values, a row matrix X corresponding to the key fields is formed. c ;

[0020] According to the number of garbled fields and the corresponding weight values, a column matrix Y corresponding to the garbled fields is formed. c ;

[0021] According to the row matrix X c and column matrix Y c The number of elements in which the number of elements is larger is used as a reference, and the number of dimensions of the matrix with fewer elements is padded with the number 1 to generate the padded row matrix X and column matrix Y; wherein the structures of the padded row matrix X and column matrix Y are as follows:

[0022] X=[x 01 ,x 02 ,…,x α ]

[0023] Among them, X represents the row matrix after the padding operation corresponding to the key field; x 01 、x 02 , ..., x α They respectively represent the weight values ​​corresponding to the elements contained in the row matrix after the padding operation;

[0024]

[0025] Among them, Y represents the column matrix after the padding operation corresponding to the garbled field; 01 ,y 02 , ..., y β They respectively represent the weight values ​​corresponding to the elements contained in the column matrix after the padding operation;

[0026] The initial abnormality factor is obtained by using the padded row matrix X and column matrix Y combined with the key field missing rate and garbled code rate;

[0027] Among them, the initial abnormal factor is obtained through the following formula:

[0028]

[0029] Among them, K represents the initial abnormal factor; σ represents the number of elements included in the padded row matrix X; μ represents the number of elements included in the padded column matrix Y; x i represents the weight value corresponding to the i-th element included in the padded row matrix X; y i represents the weight value corresponding to the i-th element included in the padded column matrix Y; P 01 and P 02 respectively represent the data missing rate and data garbled rate corresponding to the integrated data; || || represents the norm operation symbol corresponding to the matrix;

[0030] Compare the initial abnormal factor with a preset factor threshold;

[0031] When the initial abnormal factor exceeds the preset factor threshold, data anomaly alarm is carried out;

[0032] When the initial abnormal factor does not exceed the preset factor threshold, the time difference between the corresponding integrated data update time and the fragmented data update time is monitored in real time each time the fragmented data is updated, and the number of missing key fields after data update, the weight value corresponding to the missing data, the data garbled rate included in the integrated data, and the weight value corresponding to the garbled data;

[0033] Use the initial abnormal factor combined with the time difference between the integrated data update time and the fragmented data update time, and the number of missing key fields after data update, the weight value corresponding to the missing data, the data garbled rate included in the integrated data, and the weight value corresponding to the garbled data to obtain the abnormal factor corresponding to each data update;

[0034] Among them, the abnormal factor is obtained through the following formula:

[0035]

[0036] Among them, Q represents the abnormal factor; w represents the number of updated fragmented data corresponding to the current data update; T ci represents the time difference between the completion time of the i-th updated fragmented data and the completion time of the data corresponding to the fragmented data in the integrated data; K represents the initial abnormal factor; K g represents the initial abnormal factor corresponding to the current integrated data after completion of update;

[0037] Compare the abnormal factor with a preset abnormal factor threshold;

[0038] When the abnormal factor exceeds the preset abnormal factor threshold, data anomaly alarm is carried out.

[0039] Furthermore, relevant business data of users and their vehicles are used to construct corresponding tags for users, including:

[0040] Retrieve the behavior data and attribute data corresponding to the user, and create tags corresponding to the user and the user's vehicle according to the preset business logic in combination with different model types;

[0041] Present and manage the content of the tags corresponding to each user and the user's vehicle in a tag-structured manner, where the tag structure includes the number, classification, hierarchical relationship, etc. of the tags.

[0042] Furthermore, the different model types include the AIPL model, the 5A model, and the RFM model; among them, the structures of the AIPL model, the 5A model, and the RFM model are as follows:

[0043] In the AIPL model, A represents the brand awareness population; I represents the brand interest population; P represents the brand purchase population; L represents the brand loyalty population;

[0044] The 5A model includes parameters A1, A2, A3, A4, and A5. Among them, A1 represents customers who passively receive information; A2 represents customers with an increased brand impression; A3 represents customers who are driven by curiosity to actively search for information; A4 represents customers who take action; A5 represents customers who are loyal to the brand and promote it.

[0045] The RFM model evaluates the high or low values of R, F, and M for each user and maps them to different intervals, thereby dividing users into 8 types of user value types; among them, the R value represents the most recent consumption, reflecting the activity of a customer; the F value represents the consumption frequency, reflecting the loyalty of a customer; the M value represents the consumption amount, reflecting the contribution of a customer; and the 8 types of user value types include important value customers, important return customers, important deep cultivation customers, important retention customers, potential customers, new customers, general maintenance customers, and lost customers.

[0046] Furthermore, the users are grouped according to the tags corresponding to the users, and the target group package corresponding to the users is obtained, and the users and the data information with upstream and downstream dependency relationships are displayed according to the target group package, including:

[0047] Create the target group package corresponding to the user through the preset grouping rules or the uploaded list; among them, the target group package is used to gain insights or push to downstream marketing channels, so as to achieve precise reach of target customers.

[0048] Automatically generate a group movement record according to the target group package, wherein the group movement record is used to display the operation record, historical performance and trend of the group, so as to timely understand the latest situation of the group task;

[0049] According to the target grouping package, the blood relationship details corresponding to the target grouping package are viewed, and the data sources, population packages and label resources with upstream and downstream dependencies are displayed with the target grouping package as the center.

[0050] Furthermore, the method of creating a target grouping package corresponding to the user is as follows:

[0051] Rule creation: Create groups by combining labels, behaviors, crowd packages, detailed data, and user attribute data through visual components;

[0052] Upload creation: quickly create a target grouping package by uploading a local seed population file.

[0053] Furthermore, based on the tags corresponding to the users, a multi-dimensional portrait of the target group corresponding to the target grouping package is obtained through in-depth analysis and insight, including:

[0054] Retrieve the tags corresponding to the user, and display the main information of the user and the user's vehicle for the tags corresponding to the user, wherein the main information includes the basic information corresponding to the user, the covering tags and the behavior timeline, and the group to which the user belongs; quickly understand the key information of a user through the individual portrait, so as to carry out further marketing actions for the user;

[0055] Retrieve the target grouping packages corresponding to users and their vehicles, generate insight reports for the target groups corresponding to the main bodies of users and their vehicles, and mine group characteristics. At the same time, use cross-listening and drill-down analysis to gain in-depth insights into the user's characteristics in various dimensions, where the group characteristics are used to guide business decisions;

[0056] Extract the explicit and implicit features of the target grouping packages corresponding to users and their vehicles, and mine the label combinations matching the target grouping packages based on the explicit and implicit features. At the same time, generate grouping packages that meet the preset requirements according to indicators such as prominence, coverage, and number of groups for further analysis and decision-making;

[0057] Create lifecycle tags and gain insights into the user lifecycle through lifecycle tags. At the same time, export target segmentation packages according to user stages to conduct refined operations, promote user purchasing decisions, and continuously improve user loyalty to the brand.

[0058] Further, according to the target clustering package corresponding to the user, formulate differentiated operation strategies for groups with different value hierarchies, reach users through multiple channels and recover effect data, and optimize the strategies guided by the effect data, including:

[0059] Formulate operation strategies according to the clustering attributes and behavior prediction results of the target clustering package corresponding to the user;

[0060] Connect to systems such as CRM to obtain a panoramic customer portrait, and execute marketing strategies according to the panoramic customer portrait;

[0061] Execute personalized operation strategies by constructing a churn prediction model and understanding the churn population to maximize the extension of the user life cycle.

[0062] Further, the marketing strategy optimization processing method based on user clustering further includes:

[0063] Real-time monitor the data transmission operation parameters between each business system and the customer service operation platform, where the data transmission operation parameters include the standard deviation of the data retrieval response duration per unit time, the ratio between the process and the available CPU per unit time, the API request frequency per unit time, and the security incident occurrence rate per unit time; and the value range of the unit time is 3min - 8min;

[0064] Use the data transmission operation parameters to form a feature vector for each unit time corresponding to each business system; where the feature vector structure is as follows:

[0065] A = [T b , P c , F, P a

[0066] Where, A represents the feature vector corresponding to each unit time; T b represents the standard deviation of the data retrieval response duration per unit time; P c represents the ratio between the process and the available CPU per unit time; F represents the API request frequency per unit time; P a represents the security incident occurrence rate per unit time;

[0067] Perform standard deviation processing on the feature vector for each unit time corresponding to each business system to generate a standardized feature vector, and the standardized feature vector for each unit time corresponding to each business system is obtained through the following formula:

[0068]

[0069] Where, A b ​represents the feature vector after standardization processing; δ(T b ), δ(P c ), δ(F) and δ(P a ) respectively represent the standardized parameters corresponding to the standard deviation of data retrieval response duration per unit time, the ratio between the process and the available CPU per unit time, the API request frequency per unit time, and the incidence rate of security events per unit time;

[0070] The feature coefficients are obtained by using the standardized feature vectors corresponding to all the elapsed unit times of each business system. Among them, the feature coefficients are obtained through the following formula:

[0071]

[0072] Among them, ξ represents the feature coefficient corresponding to each business system; n represents the number of all elapsed unit times corresponding to each business system; ||A bi || 2 represents the Manhattan norm corresponding to the standardized feature vector corresponding to the i-th unit time, the cumulative effect of the overall load; ||A bi || 1 represents the Euclidean norm corresponding to the standardized feature vector corresponding to the i-th unit time, reflecting the overall strength or importance of the feature vector; ε represents a preset minimum constant, used to prevent ||A bi || 1 from being 0; δ(T b ) i , δ(P c ) i , δ(F) i and δ(P a ) i respectively represent the standardized parameters corresponding to the data retrieval response duration standard deviation, the ratio between the process and the available CPU, the API request frequency, and the incidence rate of security events corresponding to the i-th unit time; α and β respectively represent the first adjustment coefficient and the second adjustment coefficient, and the value ranges of the first adjustment coefficient and the second adjustment coefficient are 0.53 - 0.72, 0.47 - 0.79;

[0073] The comprehensive load coefficient is obtained by using the feature coefficient corresponding to each business system;

[0074] Among them, the comprehensive load coefficient is obtained through the following formula:

[0075]

[0076] Among them, S represents the comprehensive load coefficient; m represents the number of business systems; ξ iDenote the characteristic coefficient corresponding to the i-th service coefficient; ξ b Denote the standard deviation of the characteristic coefficients corresponding to m service coefficients; ξ bi Denote the standard deviation of the characteristic coefficients corresponding to the i-th service coefficient for n unit time periods;

[0077] Compare the comprehensive load coefficient with a preset comprehensive coefficient threshold;

[0078] When the comprehensive load coefficient exceeds the preset comprehensive coefficient threshold, it is determined that the communication operation load of the customer service operation platform is overloaded, and a load overload warning is issued.

[0079] A marketing strategy optimization processing system based on user grouping, the marketing strategy optimization processing system based on user grouping includes:

[0080] An ID graph construction module, configured to access relevant service data of users and user vehicles of multiple service systems to the customer service operation platform, and construct an ID graph corresponding to the users;

[0081] A label construction module, configured to construct labels corresponding to users by using relevant service data of users and user vehicles;

[0082] A target grouping package acquisition module, configured to group users according to the labels corresponding to the users, acquire a target grouping package corresponding to the users, and display the users and the data information with upstream and downstream dependency relationships thereof according to the target grouping package;

[0083] A multi-dimensional portrait acquisition module, configured to acquire a multi-dimensional portrait of a target group corresponding to a target grouping package based on the labels corresponding to the users through in-depth analysis and insight methods;

[0084] An operation strategy acquisition and optimization module, configured to formulate differentiated operation strategies for groups with different value hierarchies according to the target grouping packages corresponding to the users, reach users through multiple channels and recover effect data, and optimize the strategies guided by the effect data.

[0085] Advantages of the present invention:

[0086] The marketing strategy optimization processing method and system based on user grouping proposed in the present invention help enterprises break data silos, establish unified human and vehicle files, enable enterprises to drive full-link marketing and deep operations with data, and realize digital transformation and growth of enterprises. Enterprises can build user tags and portraits through the customer service operation platform, stratify and group users, and use group insight capabilities to explore group significant characteristics, deeply understand the target user group, and formulate targeted operation strategies for users with different characteristics. Through the refined operation of customers, they can change from traffic thinking to user thinking, so as to master the initiative of marketing. At the same time, by establishing a set of customer service operation platforms, through the integration and connection of various marketing-related systems, all human and vehicle data can be fully entered into the lake; by connecting the human and vehicle IDs of different systems, users, vehicles and other individuals can be uniquely identified; tags and groupings can be established, and portrait analysis can be performed on user groups and individuals to realize rapid selection of specific groups; through machine learning and model algorithms, modeling of segmented groups can be provided for analysis and decision-making; according to the attributes and consumption predictions of different groups of people, accurate matching marketing strategies can be established to realize accurate marketing of target groups.

[0087] On the other hand, the marketing strategy optimization processing method and system based on user grouping also have the following beneficial technical effects:

[0088] (1) The marketing strategy optimization processing method based on user grouping and the system data fusion function provided by the present invention identify the identity identifiers of different business systems and different sources as the same subject through data technology, thereby breaking down data silos and achieving data integration.

[0089] (2) The marketing strategy optimization processing method based on user grouping, system label setting and label system function provided by the present invention builds a closed loop of label optimization through the full life cycle of labels + label application effects. Through a one-stop label construction method and a visual interface interaction method, self-service creation and management of labels can be achieved, realizing human-machine collaboration and efficient completion of the 360° label system construction.

[0090] (3) The marketing strategy optimization processing method based on user grouping and the system user grouping function provided by the present invention can self-service the target group based on tags and data, accurately and quickly circle the crowd package, and meet the diverse analysis and operation needs.

[0091] (4) The customer operation platform algorithm model function provided by the present invention can quickly obtain prediction results based on the configured data content, thereby supporting business goals such as decision-making, optimization, and prediction, and exerting greater business value.

[0092] (5) The marketing strategy optimization processing method and system insight analysis function provided by the present invention enable salespersons to view detailed customer profiles through this function for key customers and provide more targeted services according to customer situations. In view of the portrait differences among different user groups, significant difference features are compared and analyzed, and the marketing strategy can be continuously optimized. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Figure 1 is a flowchart of the marketing strategy optimization processing method of the present invention;

[0094] Figure 2 is a system schematic diagram of the marketing strategy optimization processing system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0095] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only for the purpose of illustrating and explaining the present invention, and are not intended to limit the present invention.

[0096] An embodiment of the present invention provides a marketing strategy optimization processing method based on user clustering. As Figure 1 shown, the marketing strategy optimization processing method based on user clustering includes:

[0097] S1. Access the relevant business data of users and user vehicles in multiple business systems to a customer service operation platform, and construct an ID map corresponding to the users;

[0098] S2. Construct labels corresponding to the users by using the relevant business data of the users and user vehicles;

[0099] S3. Perform clustering processing on the users according to the labels corresponding to the users, obtain a target clustering package corresponding to the users, and display the users and the data information with upstream and downstream dependency relationships thereof according to the target clustering package;

[0100] S4. Obtain multi-dimensional portraits of the target groups corresponding to the target clustering package through in-depth analysis and insight based on the labels corresponding to the users;

[0101] S5. According to the target clustering package corresponding to the users, formulate differentiated operation strategies for groups with different value hierarchies, reach the users through multiple channels and recover the effect data, and optimize the strategies guided by the effect data.

[0102] The working principle of the above technical solution is as follows: The above technical solution of this embodiment establishes a unified personnel and vehicle file, empowers enterprises to drive the whole-link marketing and in-depth operation with data, and realizes the digital transformation and growth of enterprises. Enterprises can build user tags and portraits through the customer service operation platform, stratify and group users, and through the group insight ability, explore the significant features of the group, deeply understand the target user group, formulate targeted operation strategies for users with different characteristics, and transform from the traffic thinking to the user thinking through the refined operation of customers, so as to master the initiative of marketing. At the same time, by establishing a set of customer service operation platforms, through the integration and connection with various marketing-related systems, all personnel and vehicle data are fully imported into the lake; through the connection of personnel and vehicle IDs in different systems, the uniqueness identification of individuals such as users and vehicles is realized; tags and groups are established, and portrait analysis is carried out on user groups and individuals to achieve the rapid selection of specific populations; through machine learning and model algorithms, the modeling of segmented populations is provided for analysis and decision-making; according to the attributes and consumption predictions of different populations, a precisely matched marketing strategy is established to achieve the precise marketing of the target population.

[0103] First, for data integration and fusion, comprehensively access the data of business systems related to personnel and vehicles, and through data cleaning and processing, achieve the connection of personnel and vehicle OneID and establish a unique identifier. Then, build tags and groups. Tags are the basis of the customer service operation platform. Through the establishment of a tag system, a basis is established for user grouping, group insight, and marketing applications. Subsequently, select user groups. According to the obtained tags or tag system, select user groups. Then, conduct user insight and analysis. Through in-depth analysis and insight into the multi-dimensional portraits of the target group, achieve global insight and analysis. Finally, for marketing applications, formulate differentiated operation strategies for groups with different value levels, reach users through multiple channels and recover the effect data, and optimize the strategies guided by data to continuously improve the conversion effect.

[0104] The effect of the above technical solution is: the marketing strategy optimization processing method based on user grouping proposed in this embodiment helps enterprises break data silos, establish unified human and vehicle files, enable enterprises to drive full-link marketing and deep operations with data, and realize digital transformation and growth of enterprises. Enterprises can build user tags and portraits through the customer service operation platform, stratify and group users, and use group insight capabilities to explore group significant characteristics, deeply understand the target user group, and formulate targeted operation strategies for users with different characteristics. Through the refined operation of customers, they can change from traffic thinking to user thinking, so as to master the initiative of marketing. At the same time, through the integration and connection of various marketing-related systems, all human and vehicle data can be fully entered into the lake; by connecting the human and vehicle IDs of different systems, users, vehicles and other individuals can be uniquely identified; tags and groupings can be established, and portrait analysis can be performed on user groups and individuals to realize the rapid selection of specific groups; through machine learning and model algorithms, modeling of segmented groups can be provided for analysis and decision-making; according to the attributes and consumption predictions of different groups, accurate matching marketing strategies can be established to realize accurate marketing of target groups.

[0105] In one embodiment of the present invention, relevant business data of users and user vehicles of multiple business systems are connected to a customer service operation platform, and an ID map corresponding to the user is constructed, including:

[0106] S101, connecting the system business original data corresponding to the user and the user's vehicle in multiple business systems to the customer service operation platform; wherein the multiple business systems include but are not limited to a marketing system, an ERP system, and a vehicle system;

[0107] S102, performing data cleaning processing on the original data of the system business, and transmitting the original data of the system business after data cleaning processing to the target source; wherein the data cleaning processing includes but is not limited to screening, deduplication and replacement;

[0108] S103, configure IDmapping logic, and set the priority in the multi-source data matching process according to the source and data integrity of the system business original data combined with the actual business rules, and at the same time, establish a unique identifier of the user and its corresponding user vehicle, wherein the unique identifier is OneID;

[0109] S104. Associating fragmented data of multiple accounts of the same user based on user ID information, and integrating the fragmented data to generate integrated data corresponding to the user ID information; wherein the user ID information includes but is not limited to user ID, mobile phone number, device number, etc.

[0110] The working principle of the above technical solution is as follows: Original data access: The original data of all systems related to people and vehicles, such as the marketing system, ERP system, and vehicle system, is accessed to the customer service operation platform.

[0111] Data cleaning and processing: Perform cleaning operations such as screening, duplicate removal, and replacement on the data. After data cleaning is completed, the processed data can be output to the target source.

[0112] Data connection to build the OneID system: Configure the IDmapping logic, set the priority in the matching process of multi-source data according to the data source, data integrity, and actual business rules, and establish the unique identifiers (i.e., OneID) of people and vehicles.

[0113] Build an ID map: Based on ID information such as user ID, mobile phone number, and device number, associate the fragmented data of multiple accounts of the same user, and integrate the multi-party data of the enterprise.

[0114] The effects of the above technical solution are as follows: By accessing the original data related to users and their vehicles in multiple business systems to the customer service operation platform, cross-system data integration is achieved. This provides users with a comprehensive and unified data view, enabling the enterprise to more comprehensively understand the situation of users and their vehicles. Data cleaning and processing steps (such as screening, duplicate removal, and replacement) effectively improve the quality of the data. This ensures the accuracy of subsequent analysis and applications, and avoids decision-making mistakes caused by data errors or duplicates. Configuring the ID mapping logic and setting priorities according to the data source and data integrity enables the enterprise to more effectively manage multi-source data. This helps to ensure the accuracy and consistency of the data, while improving the efficiency of data processing. By establishing the unique identifiers (OneID) of users and their corresponding vehicles, precise identification of users and their vehicles is achieved. This provides the enterprise with a more accurate user portrait, helping the enterprise to better carry out personalized services and precision marketing. Based on user ID information (such as user ID, mobile phone number, device number, etc.), associate the fragmented data of multiple accounts of the same user and integrate it. This helps the enterprise to more comprehensively understand the behavior and preferences of users, so as to provide more personalized services and products. The integrated data provides strong support for the enterprise's business decisions. Through the analysis and mining of the data, the enterprise can discover potential market opportunities, optimize business processes, improve the user experience, etc., thereby enhancing the competitiveness of the enterprise.

[0115] In summary, through measures such as integrating the data of multiple business systems, improving data quality, establishing unique identifiers, and integrating fragmented data, the technical solution provides the enterprise with a comprehensive, accurate, and efficient data management solution, which helps the enterprise to better understand users, optimize business decisions, and enhance competitiveness.

[0116] Specifically, data quality monitoring is performed on the integrated data corresponding to the user ID information, including:

[0117] Scan the integrated data corresponding to the user ID information to obtain the initial abnormal characteristic elements contained in the integrated data; wherein the initial abnormal characteristic elements include the number of missing key fields and the weight value corresponding to the missing data and the data garbled rate contained in the integrated data and the weight value corresponding to the garbled data;

[0118] According to the number of missing key fields combined with the corresponding weight values, a row matrix X corresponding to the key fields is formed. c ;

[0119] According to the number of garbled fields and the corresponding weight values, a column matrix Y corresponding to the garbled fields is formed. c ;

[0120] According to the row matrix X c and column matrix Y c The number of elements in which the number of elements is larger is used as a reference, and the number of dimensions of the matrix with fewer elements is padded with the number 1 to generate the padded row matrix X and column matrix Y; wherein the structures of the padded row matrix X and column matrix Y are as follows:

[0121] X=[x 01 ,x 02 ,…,x α ]

[0122] Among them, X represents the row matrix after the padding operation corresponding to the key field; x 01 、x 02 , ..., x α They respectively represent the weight values ​​corresponding to the elements contained in the row matrix after the padding operation;

[0123]

[0124] Among them, Y represents the column matrix after the padding operation corresponding to the garbled field; 01 ,y 02 , ..., y β They respectively represent the weight values ​​corresponding to the elements contained in the column matrix after the padding operation;

[0125] The initial abnormality factor is obtained by using the padded row matrix X and column matrix Y combined with the key field missing rate and garbled code rate;

[0126] The initial abnormality factor is obtained by the following formula:

[0127]

[0128] Wherein, K represents the initial anomaly factor; σ represents the number of elements contained in the row matrix X after padding; μ represents the number of elements contained in the column matrix Y after padding; x i represents the weight value corresponding to the i-th element contained in the row matrix X after padding; y i represents the weight value corresponding to the i-th element contained in the column matrix Y after padding; P 01 and P 02 respectively represent the data missing rate and data garble rate corresponding to the integrated data; || || represents the norm operation symbol corresponding to the matrix;

[0129] Compare the initial anomaly factor with a preset factor threshold;

[0130] When the initial anomaly factor exceeds the preset factor threshold, data anomaly alarm is carried out;

[0131] When the initial anomaly factor does not exceed the preset factor threshold, the time difference between the corresponding integrated data update time and the fragmented data update time is monitored in real time each time the fragmented data is updated, as well as the number of missing key fields after data update, the weight values corresponding to the missing data, the data garble rate contained in the integrated data, and the weight values corresponding to the garbled data;

[0132] Use the initial anomaly factor combined with the time difference between the integrated data update time and the fragmented data update time, as well as the number of missing key fields after data update, the weight values corresponding to the missing data, the data garble rate contained in the integrated data, and the weight values corresponding to the garbled data to obtain the anomaly factor corresponding to each data update;

[0133] Wherein, the anomaly factor is obtained through the following formula:

[0134]

[0135] Wherein, Q represents the anomaly factor; w represents the number of updated fragmented data corresponding to the current data update; T ci represents the time difference between the time when the i-th updated fragmented data is completed and the time when the data corresponding to the fragmented data in the integrated data is completed; K represents the initial anomaly factor; K g represents the initial anomaly factor corresponding to the current integrated data after completion of the update;

[0136] Compare the anomaly factor with a preset anomaly factor threshold;

[0137] When the anomaly factor exceeds the preset anomaly factor threshold, data anomaly alarm is carried out.

[0138] The working principle of the above technical solution is as follows: Scan the integrated data to identify and extract initial abnormal feature elements, including the number of missing key fields and their weight values, and the data garbled rate and its weight value. Construct a row matrix X based on the number of missing key fields and the corresponding weight values. c Construct a column matrix Y based on the number of garbled fields and the corresponding weight values. c Based on the row matrix X c and the column matrix Y c According to the number of elements in the matrices, use the number 1 to complete the dimension number of the matrix with fewer elements, generating the completed row matrix X and column matrix Y.

[0139] Using the completed row matrix X and column matrix Y, combined with the data missing rate and the garbled rate, calculate the initial abnormal factor K through a specific formula. Compare the initial abnormal factor K with the preset factor threshold. If K exceeds the threshold, an alarm for data abnormality is issued; otherwise, enter the real-time monitoring stage.

[0140] Real-time monitor the update situation of fragmented data, record the time difference between the update time of the integrated data and the update time of the fragmented data, as well as the number of missing key fields and the garbled rate after the update. Combine the initial abnormal factor K, the time difference, the number of missing key fields and the garbled rate after the update, and calculate the abnormal factor Q corresponding to each data update through a specific formula. Compare the abnormal factor Q with the preset abnormal factor threshold. If Q exceeds the threshold, an alarm for data abnormality is issued.

[0141] The technical effects of the above technical solution are as follows: By real-time monitoring the update situation of fragmented data, data quality problems can be discovered in a timely manner, improving the real-time performance of data monitoring. By comprehensively considering multiple factors such as the number of missing key fields, the garbled rate, and the time difference, calculate the abnormal factor, improving the accuracy of data quality monitoring. Using the number 1 to complete the dimension number of the matrix with fewer elements makes the row matrix X and the column matrix Y consistent in structure, facilitating subsequent calculations. This completion method is simple and flexible, and does not increase the additional computational complexity. By setting the factor threshold and the abnormal factor threshold, when the data quality exceeds the acceptable range, the alarm mechanism can be automatically triggered, improving the reliability of data processing. This technical solution can adjust the identification rules of key fields and garbled fields and the setting of weight values according to actual business needs, having good scalability. At the same time, this technical solution is applicable to various types of data integration scenarios, having high applicability.

[0142] By comprehensively considering multiple dimensions such as the number of missing key fields, the garbled rate, and the time difference, the technical solution can more accurately identify anomalies in the data. This multi-dimensional anomaly recognition method can more comprehensively reflect the data quality problems compared with the single-dimensional recognition method. When calculating the initial anomaly factor and the anomaly factor, the technical solution uses a simple and clear formula, avoiding complex iterative calculations or optimization algorithms, thus improving the calculation efficiency. This enables the technical solution to process and analyze a large amount of data in a short time and meet the real-time requirements. When constructing the matrix and calculating the anomaly factor, the technical solution makes full use of the existing data and computing resources, avoiding unnecessary resource waste. For example, by using the number 1 to complete the dimension of the matrix, it not only ensures the consistency of the matrix structure but also avoids introducing additional data or computational volume. Once data anomalies are detected, the technical solution can immediately trigger an alarm mechanism to notify relevant personnel for processing. This timely anomaly handling mechanism helps reduce the impact of data errors on business decisions and improve the reliability and efficiency of data processing.

[0143] By scanning and analyzing the initial abnormal feature elements such as the number of missing key fields, the weight value of missing data, the data garble rate, and the weight value of garbled data in the integrated data, various abnormal situations existing in the data can be accurately captured, key problems can be avoided from being missed, and the accuracy of data anomaly judgment is improved. At the same time, by calculating the initial anomaly factor and the anomaly factor through the above formula, the data anomaly situation is quantified, making the evaluation of data quality more accurate, being able to more accurately reflect the actual quality status of the data, and providing a more reliable basis for subsequent decision-making. At the same time, when the initial anomaly factor does not exceed the threshold, information such as the time difference between the integrated data update time and the fragmented data update time during each fragmented data update can be monitored in real time, potential problems in the data update process can be discovered in a timely manner, ensuring the real-time nature of data quality monitoring and being able to quickly respond to data changes. Whether the initial anomaly factor or the anomaly factor corresponding to each data update exceeds the preset threshold, data anomaly alarms can be issued in a timely manner, enabling relevant personnel to understand data quality problems in the first time, quickly take measures to handle them, and reducing the impact time of data anomalies on the business. In addition, the above technical solution not only considers the two aspects of missing key fields and data garbling, but also combines their corresponding weight values, as well as factors such as the time difference during the data update process, evaluates the data quality from multiple dimensions, avoids the limitations of single-dimensional evaluation, and makes data quality monitoring more comprehensive. By continuously obtaining the anomaly factor corresponding to each data update, the change situation of data quality can be dynamically tracked. Whether the data is in the initial integration stage or subsequent update process, continuous monitoring can be carried out to comprehensively ensure the stability of data quality. During the data update process, the anomaly factor can be dynamically calculated according to factors such as the number of updated fragmented data, the update time difference, and the abnormal situation after the update, and effective quality monitoring can be carried out for data updates of different scales and frequencies, with strong adaptability and flexibility. By setting the preset factor threshold and anomaly factor threshold, the sensitivity to data anomalies can be flexibly adjusted according to different business requirements and data quality standards, making data quality monitoring more in line with the requirements of the actual business scenario and improving the adaptability and configurability of the system.

[0144] In summary, the technical effects of the above technical solution in terms of performance indicators are not only reflected in the real-time nature and accuracy of data quality monitoring, the flexibility of matrix dimension filling, the reliability of the alarm mechanism, as well as scalability and applicability, but also further include potential technical effects such as the accuracy of anomaly recognition, the improvement of calculation efficiency, the optimization of resource utilization, the timeliness of anomaly handling, and visualization and interpretability. These technical effects together constitute the comprehensive advantages of the technical solution in terms of performance indicators. At the same time, through comprehensive consideration of multiple factors, the technical solution realizes comprehensive, accurate, and real-time data quality monitoring of the integrated data corresponding to the user ID information, improving the reliability and efficiency of data processing.

[0145] An embodiment of the invention constructs tags corresponding to a user by using relevant business data of the user and the user's vehicle, including:

[0146] S201. Retrieve the behavior data and attribute data corresponding to the user, and create tags corresponding to the user and the user's vehicle according to the preset business logic in combination with different model types;

[0147] S202. Present and manage the content of the tags corresponding to each user and the user's vehicle in a tag-structured manner, where the tag structure includes the number, classification, hierarchical relationship, etc. of the tags.

[0148] Among them, the different model types include the AIPL model, the 5A model, and the RFM model; among them, the structures of the AIPL model, the 5A model, and the RFM model are as follows:

[0149] In the AIPL model, A represents the brand awareness population; I represents the brand interest population; P represents the brand purchase population; L represents the brand loyalty population;

[0150] In the 5A model, it includes parameters A1, A2, A3, A4, and A5. Among them, A1 represents the customers who passively receive information; A2 represents the customers with an increased brand impression; A3 represents the customers who are driven by curiosity to actively search for information; A4 represents the customers who take actions; A5 represents the customers who are loyal to the brand and promote it;

[0151] The RFM model evaluates the high or low values of R, F, and M for each user and maps them to different intervals, thereby dividing users into 8 types of user value types; among them, the R value represents the last consumption, reflecting the activity degree of a customer; the F value represents the consumption frequency, reflecting the loyalty of a customer; the M value represents the consumption amount, reflecting the contribution degree of a customer; and the 8 types of user value types include important value customers, important return customers, important deep cultivation customers, important retention customers, potential customers, new customers, general maintenance customers, and lost customers.

[0152] The working principle of the above technical solution is: Tag setting: Based on data such as behavior and attributes, create tags corresponding to people and vehicles based on business logic or model capabilities.

[0153] Tag system: The tag system is composed of tags, and presents and manages the content of the tags, including the number, classification, hierarchical relationship, etc. of the tags, in a structural manner.

[0154] Based on business logic or model capabilities, establish a tag model, specifically as follows:

[0155] AIPL Model: A means of quantitatively and link-operating brand population assets. Specifically: A (Awareness) represents the brand awareness population; I (Interest) represents the brand interest population; P (Purchase) represents the brand purchase population; L (Loyalty) represents the brand loyalty population

[0156] 5A Model: A marketing model proposed by Philip Kotler in "Marketing Revolution 4.0". Specifically: A1 Awareness means that customers passively receive information; A2 Appeal means customers with increased brand impressions; A3 Ask means customers who are driven by curiosity to actively search for information; A4 Act means customers who take actions; A5 Advocate means customers who are loyal to the brand and promote it.

[0157] RFM Model: By evaluating the R value, F value, and M value of each user and corresponding them to different intervals, users are divided into 8 types of user value types. Specifically: important value customers, important reclaimed customers, important deep-cultivated customers, important retained customers, potential customers, new customers, general maintained customers, and lost customers. Among them, R represents Recency, reflecting a customer's activity level; F represents Frequency, reflecting a customer's loyalty; M represents Monetary, reflecting a customer's contribution degree.

[0158] The effects of the above technical solution are as follows: By retrieving the corresponding behavioral data and attribute data of users and creating tags for users and their vehicles according to the preset business logic in combination with different model types (such as AIPL model, 5A model, and RFM model), the refined construction of user portraits can be achieved. These tags can more accurately reflect the user's status, interests, behavior patterns, etc., and help enterprises understand users more deeply. Presenting and managing the tags corresponding to each user and their vehicle in a tag-structured manner (such as the number, classification, and hierarchical relationship of tags) makes the tag system clearer and more orderly. This helps enterprises use tags more efficiently for user analysis, precision marketing, etc., and improves work efficiency. With different model types such as AIPL model, 5A model, and RFM model, enterprises can divide users more meticulously, so as to formulate more precise marketing strategies. For example, for brand-loyal customers (L) or customers at the A5 level, enterprises can launch more loyalty reward programs or high-end services; for potential customers or new customers, enterprises can adopt more attractive promotional strategies to guide them to become loyal users. The RFM model divides user value types by evaluating the user's R value (recency), F value (frequency), and M value (monetary value), providing an enterprise with a scientific and objective method for evaluating user value. This helps enterprises identify high-value users, potential users, and churned users, etc., so as to formulate targeted user maintenance and recovery strategies.

[0159] Through the construction of a user tag system, this technical solution realizes the in-depth mining and analysis of user data. This provides rich data support for enterprises, enabling enterprises to make decisions based on data and improving the accuracy and scientific nature of decisions. By dividing users meticulously and conducting precision marketing, enterprises can better meet the needs and expectations of users and enhance the user experience. At the same time, by identifying and paying attention to high-risk groups such as churned users, enterprises can take timely measures to recover these users, thereby maintaining user satisfaction and loyalty.

[0160] In summary, through measures such as constructing a user tag system, realizing efficient tag management, precision marketing strategies, scientific user value evaluation, and data-driven decision-making, this technical solution provides comprehensive, scientific, and efficient user management and marketing strategy support for enterprises, helping enterprises enhance competitiveness and achieve sustainable development.

[0161] In an embodiment of the present invention, users are grouped according to the tags corresponding to the users, a target group package corresponding to the users is obtained, and data information of the users and their upstream and downstream dependency relationships is displayed according to the target group package, including:

[0162] S301. Create a target segmentation package corresponding to the user through a preset segmentation rule or an uploaded list. Among them, the target segmentation package is used for insight or pushing to downstream marketing channels, so as to achieve precise reach of target customers.

[0163] S302. Automatically generate a segmentation movement record according to the target segmentation package. Among them, the segmentation movement record is used to display the operation record, historical performance, trends, etc. of this segmentation, so as to timely understand the latest situation of the segmentation task.

[0164] S303. View the lineage details corresponding to the target segmentation package according to the target segmentation package, and display the data sources, population packages, and label resources with upstream and downstream dependency relationships centered on the target segmentation package.

[0165] Among them, the method for creating a target segmentation package corresponding to the user is as follows:

[0166] Rule creation: Through visual components, use tags, behaviors, population packages, detailed data, and user attribute data for combination to create a segmentation.

[0167] Upload creation: Quickly create a target segmentation package by uploading a local seed population file.

[0168] The working principle of the above technical solution is: First, create a segmentation package: Create a segmentation package through certain rules or an uploaded list. After the creation of the target segmentation package, it can be used for insight or pushing to downstream marketing channels, so as to achieve precise reach of target customers. The method for creating a segmentation package is as follows:

[0169] 1) Rule creation: Through visual components, use tags, behaviors, population packages, detailed data, and user attribute data for combination to create a segmentation.

[0170] 2) Upload creation: Quickly create a target segmentation package by uploading a local seed population file.

[0171] Seed population file: According to the needs of specific business scenarios, the population with the same needs and interests for goods and services is called the seed population, which generally includes the basic attributes of users such as gender, age, occupation, region, etc., as well as behavior preference information, etc.

[0172] Expanded population / similar population: The population with the same characteristics as the seed population is called the expanded population.

[0173] Obtaining the target segmentation: Find the expanded population by uploading the seed population, and then use the expanded population as the target user for operation reach. When there are multiple seed populations, the expanded populations of each seed population can be found first, and then the intersection of the expanded populations of each seed population is taken as the final target user for delivery.

[0174] Then, group operation records: Based on the above-mentioned population package corresponding to the user, group operation records can be automatically generated to display the operation records, historical performance, trends, etc. of the group, so as to timely understand the latest situation of the group task.

[0175] Finally, group data lineage: Based on the above-mentioned population package corresponding to the user, view the lineage details of the population package, and display the data sources, population packages, and label resources with upstream and downstream dependencies centered on the population package.

[0176] The effects of the above technical solutions are as follows: Through the preset grouping rules or uploaded lists, the target grouping package corresponding to the user can be flexibly created. This method not only supports grouping based on combinations of tags, behaviors, population packages, detailed data, and user attribute data, but also allows for the quick creation of groups by uploading local files, thus achieving the precise division of target customers. This helps enterprises more accurately identify target user groups and improve the pertinence and effectiveness of marketing activities. The automatically generated group movement records display the operation records, historical performance, trends, etc. of the group. This enables enterprises to timely understand the latest progress of the group task, monitor the group effect, and adjust and optimize the group strategy. This improves the transparency and efficiency of group management. View the corresponding lineage details according to the target grouping package, and display the data sources, population packages, and label resources with upstream and downstream dependencies centered on the target grouping package. This visualization of data lineage and dependencies helps enterprises understand the source, flow, and transformation process of data, ensuring data accuracy and consistency. At the same time, it also helps enterprises better manage and utilize data resources and improve the level of data governance. The target grouping package can be used to gain insights into user behavior, preferences, and needs, providing accurate user portraits and segmentation strategies for downstream marketing channels. This helps enterprises formulate more targeted marketing strategies, improve the conversion rate and ROI of marketing activities. At the same time, through automation and optimization of group management, enterprises can more efficiently manage and operate user groups and reduce marketing costs. The functions such as group management, data lineage visualization, and marketing effect monitoring provided by this technical solution provide rich data support and analysis tools for enterprises. This helps enterprises better understand user and market dynamics, and formulate more scientific decisions and strategic plans.

[0177] In summary, through measures such as precise user grouping, group task monitoring and management, data lineage and dependency visualization, and marketing efficiency improvement, this technical solution provides comprehensive and efficient user management and marketing strategy support for enterprises. This helps enterprises better understand and utilize user data, improve the pertinence and effectiveness of marketing activities, and achieve business growth and sustainable development.

[0178] In one embodiment of the present invention, based on the tags corresponding to the user, a multi-dimensional portrait of the target group corresponding to the target segmentation package is obtained through in-depth analysis and insight methods, including:

[0179] S401. Retrieve the tags corresponding to the user, and display the main information of the user and the user's vehicle for the tags corresponding to the user. Among them, the main information includes the basic information corresponding to the user, the covered tags, the behavior timeline, and the segmentation group where the user belongs; quickly understand the key information of a user through the individual portrait, so as to perform further marketing actions for the user.

[0180] S402. Retrieve the target segmentation package corresponding to the user and the user's vehicle, generate an insight report for the target group corresponding to the main body of the user and the user's vehicle, and mine the group characteristics. At the same time, use methods such as cross-listening and drill-down analysis to deeply understand the characteristics of each dimension of the user. Among them, the group characteristics are used to guide the enterprise operation decision-making.

[0181] S403. Extract the explicit characteristics and implicit characteristics of the target segmentation package corresponding to the user and the user's vehicle, and mine the tag combination that matches the target segmentation package according to the explicit characteristics and implicit characteristics. At the same time, generate a segmentation package that meets the preset requirements according to indicators such as significance, coverage rate, and the number of segmentation groups, so as to facilitate further analysis and decision-making.

[0182] S404. Create a life cycle tag, and gain insights into the user life cycle through the life cycle tag. At the same time, export the target segmentation package according to the user stage for refined operation, promote the user's purchase decision-making, and continuously improve the user's loyalty to the brand.

[0183] The working principle of the above technical solution is as follows: Individual portrait analysis: Based on the tag settings, generate a detailed display of the main information of a person, vehicle, etc., including the basic information of the main body, the covered tags, the behavior timeline, and the segmentation group where the main body belongs. Quickly understand the key information of a user through the individual portrait, so as to perform further marketing actions for the user.

[0184] Group portrait insight: Based on the target segmentation package corresponding to the user, segment the main body of the person, vehicle, etc., circle the target group to generate an insight report, and guide the enterprise marketing decision-making by mining the significant characteristics of the group. At the same time, cross-listening, drill-down analysis and other methods can be used to deeply understand the characteristics of each dimension of the user.

[0185] Multi-characteristic analysis: Based on models and algorithms, gain insights into the explicit and implicit characteristics of the segmentation group, and mine the tag combination that best matches the segmentation group. At the same time, generate a segmentation package that meets specific requirements according to indicators such as significance, coverage rate, and the number of segmentation groups, so as to facilitate further analysis and decision-making.

[0186] Lifecycle Analysis: By creating lifecycle tags and gaining insights into the user lifecycle based on model tags. At the same time, export the segmented packages according to the user stages for refined operation, promote the user's purchase decision-making, and continuously enhance the user's loyalty to the brand.

[0187] The effects of the above technical solution are as follows: By retrieving the tags corresponding to the user and displaying the main information of the user and the user's vehicle (such as basic information, covered tags, behavior timeline, and the segmented group where the user is located), the enterprise can quickly understand the key information of the user, and thus carry out more accurate and personalized marketing actions for this user. This helps to improve the user experience, increase user stickiness, and promote the user's purchase decision-making. Generating an insight report for the target segmented package corresponding to the user and the user's vehicle, mining group characteristics, and using methods such as cross-analysis and drill-down analysis for in-depth insights helps the enterprise better understand the behavior patterns, preferences, and needs of the target group. This information can provide strong support for the enterprise to make business decisions, optimize product design, and improve marketing strategies. Extracting the explicit and implicit characteristics of the target segmented package and mining the matching tag combinations based on these characteristics helps the enterprise more accurately identify the target user group and optimize the segmentation strategy. At the same time, generating a segmented package that meets the preset requirements through indicators such as significance, coverage rate, and the number of segmented groups can further improve the accuracy and effectiveness of segmentation, providing strong guarantee for subsequent marketing and operation. Creating lifecycle tags and gaining insights into the user lifecycle through lifecycle tags helps the enterprise understand the needs and behavior changes of users at different stages. Exporting the target segmented package according to the user stage and carrying out refined operation can provide more appropriate services and products for users at different stages, thus promoting the user's purchase decision-making and enhancing the user's loyalty to the brand. This technical solution realizes the refinement of user portraits, the mining of group characteristics, the optimization of tag combinations, and the management of the user lifecycle through in-depth analysis and insights of user data. These measures help the enterprise make more scientific decisions, improve operation efficiency, reduce marketing costs, and achieve sustainable development.

[0188] In summary, this technical solution provides comprehensive and efficient data analysis and marketing strategy support for the enterprise through measures such as refining user portraits, deeply understanding group characteristics, optimizing tag combinations and segmentation strategies, and implementing user lifecycle management. This helps the enterprise better understand and utilize user data, improve marketing effectiveness, enhance user loyalty, and achieve business growth and competitive advantages.

[0189] In one embodiment of the present invention, according to the target segmented package corresponding to the user, formulate differentiated operation strategies for groups with different value hierarchies, reach users through multiple channels and collect effect data, and optimize the strategies guided by the effect data, including:

[0190] S501. Develop an operation strategy based on the segmentation attributes and behavior prediction results of the target segmentation package corresponding to the user.

[0191] S502. Interface with systems such as CRM to obtain a panoramic customer portrait and execute a marketing strategy based on the panoramic customer portrait.

[0192] S503. Execute a personalized operation strategy by building a churn prediction model and gaining insights into the churned population to maximize the extension of the user lifecycle.

[0193] The working principle of the above technical solution is: refined operation: based on the target segmentation package corresponding to the user, formulate specific operation strategies according to the segmentation attributes and behavior prediction to achieve more accurate, intelligent, and personalized reach, promoting conversion and repeat purchase.

[0194] Empowerment in the sales scenario: Interface with systems such as CRM to assist sales personnel in obtaining a panoramic customer portrait, formulating a marketing strategy, and effectively improving the lead conversion rate.

[0195] Recovery of churned users: By building a churn prediction model, gaining insights into the churned population, and formulating a personalized operation strategy, maximize the extension of the user lifecycle.

[0196] The effects of the above technical solution are as follows: Based on the segmentation attributes and behavior prediction results of the target segmentation package corresponding to the user, it is possible to formulate differentiated operation strategies for groups at different value levels. This precise strategy formulation helps to improve the pertinence and effectiveness of marketing activities and avoid waste of resources. By interfacing with systems such as CRM to obtain a panoramic customer portrait, this provides enterprises with more comprehensive and in-depth user information. Based on this information, enterprises can execute more accurate and personalized marketing strategies to meet the personalized needs of users, enhancing the user experience and satisfaction. Building a churn prediction model and gaining insights into the churned population helps enterprises promptly identify potential churned users and adopt personalized operation strategies for retention. This can not only maximize the extension of the user lifecycle but also improve the user retention rate and loyalty, bringing continuous value to the enterprise. This technical solution supports multi-channel reach to users, including various methods such as social media, email, SMS, phone calls, etc. This helps to ensure that information can be accurately and timely conveyed to users, improving the information arrival rate and reading rate. At the same time, by recovering the effect data, enterprises can evaluate the execution effect of the marketing strategy and provide data support for subsequent strategy optimization. Guided by the recovered effect data, enterprises can continuously optimize the operation strategy. This data-driven decision-making method helps to improve the accuracy and scientific nature of decision-making, reduce the trial-and-error cost, and enhance the enterprise operation efficiency and market competitiveness. Through the implementation of the above technical solution, enterprises can more accurately target the target user group, formulate effective marketing strategies, and reach users through multiple channels. This helps to improve the overall operation efficiency, reduce the marketing cost, and increase the return on investment (ROI).

[0197] In summary, through measures such as precisely formulating operation strategies, obtaining a panoramic portrait of customers, building a churn prediction model, reaching users through multiple channels and recovering effect data, and optimizing strategies driven by data, this technical solution provides comprehensive and efficient operation support for enterprises. This helps enterprises better understand and utilize user data, improve operation efficiency and market competitiveness, and achieve sustainable development.

[0198] In one embodiment of the present invention, the method for optimizing marketing strategies based on user segmentation further includes:

[0199] Step 1: Real-time monitor the data transmission operation parameters between each business system and the customer service operation platform. Among them, the data transmission operation parameters include the standard deviation of the data retrieval response duration per unit time, the ratio between the process and the available CPU per unit time, the API request frequency per unit time, and the security incident occurrence rate per unit time; and, the value range of the unit time is 3 min - 8 min;

[0200] Step 2: Use the data transmission operation parameters to form a feature vector for each unit time corresponding to each business system; among them, the structure of the feature vector is as follows:

[0201] A = [T b , P c , F, P a

[0202] Among them, A represents the feature vector corresponding to each unit time; T b represents the standard deviation of the data retrieval response duration per unit time; P c represents the ratio between the process and the available CPU per unit time; F represents the API request frequency per unit time; P a represents the security incident occurrence rate per unit time;

[0203] Step 3: Perform standard deviation processing on the feature vector for each unit time corresponding to each business system to generate a standardized feature vector, and, the standardized feature vector for each unit time corresponding to each business system is obtained through the following formula:

[0204]

[0205] Among them, A b represents the standardized feature vector; δ(T b ), δ(P c ), δ(F) and δ(P a) respectively represent the standardized parameters corresponding to the standard deviation of data retrieval response duration per unit time, the ratio between the process and available CPU per unit time, the API request frequency per unit time, and the incidence rate of security events per unit time;

[0206] Step 4: Obtain the feature coefficients by using the standardized feature vectors corresponding to all elapsed unit times of each business system, where the feature coefficients are obtained through the following formula:

[0207]

[0208] where ξ represents the feature coefficient corresponding to each business system; n represents the number of all elapsed unit times corresponding to each business system; ||A bi || 2 represents the Manhattan norm corresponding to the standardized feature vector corresponding to the i-th unit time, the cumulative effect of the overall load; ||A bi || 1 represents the Euclidean norm corresponding to the standardized feature vector corresponding to the i-th unit time, reflecting the overall strength or importance of the feature vector; ε represents a preset minimum constant used to prevent ||A bi || 1 from being 0; δ(T b ) i 、δ(P c ) i 、δ(F) i and δ(P a ) i respectively represent the standardized parameters corresponding to the standard deviation of data retrieval response duration, the ratio between the process and available CPU, the API request frequency, and the incidence rate of security events corresponding to the i-th unit time; α and β respectively represent the first adjustment coefficient and the second adjustment coefficient, and the value ranges of the first adjustment coefficient and the second adjustment coefficient are 0.53 - 0.72, 0.47 - 0.79;

[0209] Step 5: Obtain the comprehensive load coefficient by using the feature coefficients corresponding to each business system;

[0210] where the comprehensive load coefficient is obtained through the following formula:

[0211]

[0212] where S represents the comprehensive load coefficient; m represents the number of business systems; ξ i represents the feature coefficient corresponding to the i-th business coefficient; ξ b represents the standard deviation of the feature coefficients corresponding to m business coefficients; ξ biIt represents the standard deviation of the characteristic coefficients corresponding to the n unit times of the i-th service coefficient;

[0213] Step 6: Compare the comprehensive load coefficient with a preset comprehensive coefficient threshold;

[0214] Step 7: When the comprehensive load coefficient exceeds the preset comprehensive coefficient threshold, it is determined that the communication operation load of the customer service operation platform is overloaded, and a load overload warning is given.

[0215] The working principle of the above technical solution is as follows: Real-time collect the data transmission operation parameters between each business system and the customer service operation platform, including the standard deviation of the data retrieval response duration, the ratio of the process to the available CPU, the API request frequency, and the security event occurrence rate. These parameters are collected in units of 3 minutes to 8 minutes to ensure the real-time and accuracy of the data. Use the collected data transmission operation parameters to construct a feature vector for each business system within each unit time. The feature vector contains information in four dimensions: the standard deviation of the data retrieval response duration, the ratio of the process to the available CPU, the API request frequency, and the security event occurrence rate. Perform standard deviation processing on the feature vectors of each business system to generate the standardized feature vectors. The standardization process helps to eliminate the dimensional differences between different parameters, making subsequent calculations more accurate. Use the standardized feature vectors corresponding to all unit times that each business system has experienced to calculate the characteristic coefficients. The characteristic coefficients comprehensively consider the Manhattan norm and Euclidean norm of the feature vectors, as well as a preset minimum constant, to reflect the overall load situation of the business system. Calculate the comprehensive load coefficient based on the characteristic coefficients of each business system, where the comprehensive load coefficient reflects the overall communication operation load situation of all business systems. Compare the comprehensive load coefficient with the preset comprehensive coefficient threshold. When the comprehensive load coefficient exceeds the preset comprehensive coefficient threshold, it is determined that the communication operation load of the customer service operation platform is overloaded, and a load overload warning is given.

[0216] The effects of the above technical solution are as follows: By real-time monitoring the data transmission operation parameters between each business system and the customer service operation platform, potential communication operation load problems can be discovered in a timely manner. When the load is overloaded, the system can automatically give a warning, which helps the enterprise to take measures in a timely manner to avoid service interruption or performance degradation. By constructing feature vectors, performing standardization processing, and calculating characteristic coefficients, the communication operation load situation of each business system can be accurately evaluated. The calculation of the comprehensive load coefficient further reflects the overall load situation of all business systems, providing strong support for the enterprise's decision-making. Through real-time monitoring and warning, the enterprise can discover and solve communication operation load problems in a timely manner, thereby improving the stability and reliability of the system. This helps to enhance the user experience and strengthen the enterprise's market competitiveness.

[0217] On the other hand, by monitoring multiple key data transmission operation parameters in real time (such as the standard deviation of data retrieval response duration, the ratio of process to available CPU, API request frequency, and security incident occurrence rate), this technical solution can comprehensively reflect the communication operation status between the business system and the customer service operation platform. Processing the eigenvector with standard deviation eliminates the dimensional differences between different parameters, making the calculation process more unified and accurate. By calculating the comprehensive load factor, this technical solution can comprehensively consider the overall load situation of all business systems, thus more accurately evaluating the communication operation load of the customer service operation platform. This technical solution can monitor the data transmission operation parameters in real time. Once an abnormal situation is detected, it can immediately give an early warning, thereby improving the timeliness of the early warning. When the comprehensive load factor exceeds the preset comprehensive factor threshold, the system will automatically give an early warning without manual intervention, further improving the efficiency and accuracy of the early warning.

[0218] Through real-time monitoring and early warning, enterprises can timely discover and solve potential communication operation load problems, thus avoiding service interruption or performance degradation. According to the monitoring and early warning results, enterprises can reasonably adjust resource allocation, such as adding servers, optimizing the network architecture, etc., to improve the stability and reliability of the system. By ensuring the stable operation of the customer service operation platform, this technical solution helps to improve the user experience and enhance the market competitiveness of the enterprise. The data and calculation results collected by this technical solution can provide a reference for the optimization of marketing strategies. For example, by analyzing the load situation of different business systems, enterprises can adjust marketing strategies to reduce the pressure on the customer service operation platform. Combining with user segmentation technology, enterprises can formulate more precise marketing strategies for different user groups, thereby improving the marketing effect and conversion rate.

[0219] In summary, the technical effects of this technical solution and calculation process in terms of performance indicators are mainly reflected in improving evaluation accuracy, enhancing early warning timeliness, strengthening system stability and reliability, and supporting marketing strategy optimization. These technical effects help enterprises better monitor and manage the communication operation load of the customer service operation platform, improve the user experience and market competitiveness. At the same time, this technical solution realizes the accurate evaluation and early warning of the communication operation load of the customer service operation platform by monitoring data transmission operation parameters in real time, constructing eigenvectors, calculating eigen coefficients and comprehensive load factors. This helps enterprises timely discover and solve potential problems, improve the stability and reliability of the system, and provide strong support for the optimization of marketing strategies.

[0220] An embodiment of the present invention proposes a marketing strategy optimization processing system based on user segmentation, as Figure 2 shown, the marketing strategy optimization processing system based on user segmentation includes:

[0221] ID graph construction module, used to connect the relevant business data of users and user vehicles of multiple business systems to the customer service operation platform, and build an ID graph corresponding to the user;

[0222] A label building module, used to build a label corresponding to the user by using relevant business data of the user and the user's vehicle;

[0223] A target grouping package acquisition module is used to group users according to the tags corresponding to the users, acquire the target grouping packages corresponding to the users, and display the data information of the users and their upstream and downstream dependencies according to the target grouping packages;

[0224] The multi-dimensional portrait acquisition module is used to obtain the multi-dimensional portrait of the target group corresponding to the target group package through in-depth analysis and insight based on the tags corresponding to the users;

[0225] The operation strategy acquisition optimization module is used to formulate differentiated operation strategies for groups of different value layers according to the target grouping packages corresponding to the users, reach users through multiple channels and collect effect data, and optimize strategies based on the effect data.

[0226] The working principle of the above technical solution is as follows: The above technical solution of this embodiment establishes a unified human and vehicle file, enabling enterprises to drive full-link marketing and deep operations with data, and realize digital transformation and growth of enterprises. Enterprises can build user tags and portraits through the customer service operation platform, stratify and group users, and use group insight capabilities to explore the significant characteristics of the group, deeply understand the target user group, and formulate targeted operation strategies for users with different characteristics. Through the refined operation of customers, it changes from traffic thinking to user thinking, so as to master the initiative of marketing. At the same time, by establishing a customer service operation platform, through the integration and connection of various marketing-related systems, all human and vehicle data are fully entered into the lake; by connecting the human and vehicle IDs of different systems, users, vehicles and other individuals can be uniquely identified; tags and groups are established, and portrait analysis is performed on user groups and individuals to realize the rapid selection of specific groups; through machine learning and model algorithms, modeling of segmented groups is provided for analysis and decision-making; according to the attributes and consumption predictions of different groups, a precise matching marketing strategy is established to realize precise marketing of target groups.

[0227] First, data integration and fusion, fully access the business system data related to people and cars, and through data cleaning and processing, realize the connection of people and cars OneID, and recommend unique identification. Then, build tags and grouping. Tags are the basis of the customer service operation platform. Through the establishment of a tag system, a foundation is established for user grouping, group insights and marketing applications. Subsequently, circle user groups and circle user groups based on the acquired tags or tag systems. Then, user insights and analysis, through in-depth analysis and insight into the multi-dimensional portraits of the target groups, achieve full-domain insights and analysis. Finally, marketing applications, formulate differentiated operation strategies for groups with different value layers, reach users through multiple channels and collect effect data, use data as a guide to optimize strategies, and continuously improve conversion effects.

[0228] The effect of the above technical solution is: the marketing strategy optimization processing system based on user grouping proposed in this embodiment helps enterprises break data silos, establish unified human and vehicle files, enable enterprises to drive full-link marketing and deep operations with data, and realize digital transformation and growth of enterprises. Enterprises can build user tags and portraits through the customer service operation platform, stratify and group users, and use group insight capabilities to explore the significant characteristics of the group, deeply understand the target user group, and formulate targeted operation strategies for users with different characteristics. Through the refined operation of customers, it changes from traffic thinking to user thinking, so as to master the initiative of marketing. At the same time, through the integration and connection of various marketing-related systems, all human and vehicle data are fully entered into the lake; by connecting the human and vehicle IDs of different systems, users, vehicles and other individuals can be uniquely identified; tags and groupings are established, and portrait analysis is performed on user groups and individuals to realize the rapid selection of specific groups; through machine learning and model algorithms, modeling of segmented groups is provided for analysis and decision-making; according to the attributes and consumption predictions of different groups, a marketing strategy with precise matching is established to realize precise marketing of target groups.

[0229] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A marketing strategy optimization processing method based on user grouping, characterized in that: The marketing strategy optimization processing method based on user grouping includes: Connect the relevant business data of users and user vehicles of multiple business systems to the customer service operation platform, and build an ID map corresponding to the user; Use the relevant business data of the user and the user's vehicle to build the user's corresponding label; The user is grouped according to the tag corresponding to the user, a target grouping package corresponding to the user is obtained, and data information of the user and its upstream and downstream dependencies is displayed according to the target grouping package; Based on the user's corresponding tags, obtain a multi-dimensional portrait of the target group corresponding to the target group package through in-depth analysis and insight; According to the target grouping packages corresponding to the users, differentiated operation strategies are formulated for groups with different value layers. Users are reached through multiple channels and effect data is collected, and the optimization strategy is guided by the effect data.

2. The marketing strategy optimization processing method based on user grouping according to claim 1 is characterized in that: Connect the relevant business data of users and user vehicles of multiple business systems to the customer service operation platform, and build an ID map corresponding to the user, including: Connecting system business raw data corresponding to the user and the user's vehicle in multiple business systems to the customer service operation platform; wherein the multiple business systems include a marketing system, an ERP system, and a vehicle system; Perform data cleaning on the original data of the system business, and transmit the original data of the system business after data cleaning to the target source; wherein the data cleaning process includes screening, deduplication and replacement; Configure the IDmapping logic, and set the priority in the multi-source data matching process according to the source and data integrity of the system business original data combined with the actual business rules. At the same time, establish the unique identification of the user and its corresponding user vehicle, where the unique identification is OneID; Based on the user ID information, the fragmented data of multiple accounts of the same user are associated, and the fragmented data are integrated to generate integrated data corresponding to the user ID information; wherein the user ID information includes the user ID, mobile phone number and device number.

3. The marketing strategy optimization processing method based on user grouping according to claim 1 is characterized in that: Use the relevant business data of the user and the user's vehicle to build the user's corresponding label, including: Retrieve the user's corresponding behavior data and attribute data, and create labels corresponding to the user and the user's vehicle based on the preset business logic and different model types; The tag content corresponding to each user and the user's vehicle is presented and managed according to the tag structure, wherein the tag structure includes the number, classification and hierarchical relationship of the tags.

4. The marketing strategy optimization processing method based on user grouping according to claim 3 is characterized in that: The different types of models include AIPL model, 5A model and RFM model; wherein the structures of the AIPL model, 5A model and RFM model are as follows: In the AIPL model, A represents the brand awareness population; I represents the brand interest population; P represents the brand purchase population; L represents the brand loyalty population; The 5A model includes parameters A1, A2, A3, A4 and A5, where A1 indicates that customers passively accept information; A2 indicates customers whose brand impression increases; A3 indicates customers who actively search for information driven by curiosity; A4 indicates customers who take action; and A5 indicates customers who are loyal to the brand and promote it. The RFM model evaluates the R value, F value, and M value of each user and matches them to different intervals, thereby dividing users into eight user value types; among them, the R value represents the most recent consumption, reflecting the activity level of a customer; the F value represents the consumption frequency, reflecting the loyalty of a customer; the M value represents the consumption amount, reflecting the contribution of a customer; and the eight user value types include important value customers, important returned customers, important deep-cultivation customers, important retained customers, potential customers, new customers, general maintained customers, and lost customers.

5. The marketing strategy optimization processing method based on user grouping according to claim 1 is characterized in that: The user is grouped according to the tag corresponding to the user, a target grouping package corresponding to the user is obtained, and data information of the user and its upstream and downstream dependencies is displayed according to the target grouping package, including: Create a target grouping package corresponding to the user through preset grouping rules or uploaded lists; wherein the target grouping package is used for insight or push to downstream marketing channels, so as to achieve accurate reach of target customers; Automatically generate a group movement record according to the target group package, wherein the group movement record is used to display the operation record, historical performance and trend of the group; According to the target grouping package, the blood relationship details corresponding to the target grouping package are viewed, and the data sources, population packages and label resources with upstream and downstream dependencies are displayed with the target grouping package as the center.

6. The marketing strategy optimization processing method based on user grouping according to claim 5 is characterized in that: The method for creating the target grouping package corresponding to the user is as follows: Rule creation: Create groups by combining labels, behaviors, crowd packages, detailed data, and user attribute data through visual components; Upload creation: quickly create a target grouping package by uploading a local seed population file.

7. The marketing strategy optimization processing method based on user grouping according to claim 1 is characterized in that: Based on the tags corresponding to the users, a multi-dimensional portrait of the target group corresponding to the target grouping package is obtained through in-depth analysis and insight, including: Retrieving the tags corresponding to the user, and displaying the main information of the user and the user's vehicle for the tags corresponding to the user, wherein the main information includes the basic information corresponding to the user, the covering tags and the behavior timeline, and the group to which the user belongs; Retrieve the target grouping packages corresponding to users and their vehicles, generate insight reports for the target groups corresponding to the main bodies of users and their vehicles, and mine group characteristics. At the same time, use cross-listening and drill-down analysis methods to gain in-depth insights into the user's characteristics in various dimensions, where the group characteristics are used to guide business decisions; Extract the explicit and implicit features of the target grouping packages corresponding to the users and their vehicles, and mine the label combinations matching the target grouping packages based on the explicit and implicit features. At the same time, generate grouping packages that meet the preset requirements according to the prominence, coverage and number of groups. Create lifecycle tags and use them to gain insights into the user lifecycle. At the same time, export target segmentation packages according to user stages for refined operations.

8. The marketing strategy optimization processing method based on user grouping according to claim 1 is characterized in that: According to the target grouping packages corresponding to the users, differentiated operation strategies are formulated for groups with different value layers, users are reached through multiple channels and effect data is collected, and the optimization strategies are guided by the effect data, including: Formulate operation strategies based on the grouping attributes and behavior prediction results of the target grouping package corresponding to the user; Connect to the CRM system to obtain a panoramic picture of customers and implement marketing strategies based on the panoramic picture of customers; By building a churn prediction model and gaining insights into churn populations, we can implement personalized operational strategies to maximize the user life cycle.

9. The marketing strategy optimization processing method based on user grouping according to claim 1 is characterized in that: The marketing strategy optimization processing method based on user grouping also includes: Real-time monitoring of the data transmission operation parameters between each business system and the customer service operation platform, wherein the data transmission operation parameters include the standard deviation of the data retrieval response time per unit time, the ratio between the process and the available CPU per unit time, the API request frequency per unit time, and the security incident occurrence rate per unit time; and the value range of the unit time is 3min-8min; Using the data transmission operation parameters to form a characteristic vector for each unit time corresponding to each business system; Performing standard deviation processing on the characteristic vector of each unit time corresponding to each business system to generate a standardized characteristic vector; Obtain characteristic coefficients using the standardized characteristic vectors corresponding to all the unit times experienced by each business system; Obtain the comprehensive load factor using the characteristic coefficient corresponding to each business system; comparing the comprehensive load factor with a preset comprehensive factor threshold; When the comprehensive load coefficient exceeds a preset comprehensive coefficient threshold, it is determined that the communication operation load of the customer service operation platform is overloaded, and a load overload warning is issued.

10. A marketing strategy optimization processing system based on user grouping, characterized in that: The marketing strategy optimization processing system based on user grouping includes: ID graph construction module, used to connect the relevant business data of users and user vehicles of multiple business systems to the customer service operation platform, and build an ID graph corresponding to the user; A label building module, used to build a label corresponding to the user by using relevant business data of the user and the user's vehicle; A target grouping package acquisition module is used to group users according to the tags corresponding to the users, acquire the target grouping packages corresponding to the users, and display the data information of the users and their upstream and downstream dependencies according to the target grouping packages; The multi-dimensional portrait acquisition module is used to obtain the multi-dimensional portrait of the target group corresponding to the target group package through in-depth analysis and insight based on the tags corresponding to the users; The operation strategy acquisition optimization module is used to formulate differentiated operation strategies for groups of different value layers according to the target grouping packages corresponding to the users, reach users through multiple channels and collect effect data, and optimize strategies based on the effect data.

Images