Digital marketing customer obtaining method and system oriented to traditional enterprise multi-platform data

Abstract

The invention discloses a digital marketing customer obtaining method and system oriented to traditional enterprise multi-platform data, and belongs to the technical field of computers. The method comprises the following steps: collecting heterogeneous data by calling APIs of a plurality of external platforms; converting the heterogeneous data into standard features by using a preset unified data specification, and storing the standard features into a central database; integrating a third-party statistical tool to collect user behavior data, and calculating real-time and delayed return on investment according to different conversion attribution time windows; taking the fused features and the ROI data as input, and adopting a logistic regression model to train and output a guest obtaining probability prediction value; and automatically generating a multi-dimensional marketing strategy suggestion according to the predicted value. According to the method, the problems that traditional enterprise marketing data are dispersed and are difficult to integrate, analyze and accurately predict are solved. The invention also relates to a system for realizing the method.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a digital marketing technology that integrates data from multiple platforms, specifically a digital marketing customer acquisition method for traditional enterprises and a system for implementing this method. Background Technology

[0002] Currently, when traditional enterprises conduct digital marketing, their data is typically scattered across multiple independent platforms, such as website building platforms, cloud server platforms, corporate email platforms, and various marketing channel back-ends. These platforms use different data formats, creating data silos. This makes it difficult for enterprises to analyze marketing effectiveness from a holistic perspective, accurately calculate return on investment (ROI), and formulate customer acquisition strategies that rely heavily on experience, lacking data-driven predictive models.

[0003] Existing technologies lack a solution that can effectively integrate the aforementioned multi-source heterogeneous data and combine real-time and long-term ROI analysis to accurately predict customer acquisition probability. Summary of the Invention

[0004] The present invention aims to solve the problems in the prior art mentioned above, and to provide a digital marketing method and system that can break down data silos and achieve accurate customer acquisition prediction.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A digital marketing customer acquisition method for traditional enterprises using multi-platform data, characterized by the following steps: Multi-platform data collection steps: obtain product information by calling the APIs of Guangzhou Fanke's five major product systems, obtain cloud server and email information by calling the APIs of Shanghai Meicheng Internet Cloud Service Platform, obtain domain name and virtual host information by calling the APIs of West Digital Platform, and obtain online media management backend information by calling the APIs of Ruanmeng Platform. Cross-platform data fusion steps: Heterogeneous data from the above multiple platforms are transformed through a preset unified data specification. The unified data specification includes at least mapping "product information" to "core resource characteristics", mapping "cloud server configuration" to "service capability characteristics", and mapping "domain name host information" to "brand asset characteristics", and outputting it to the central database. User behavior integration and ROI calculation steps: Taking the company's official website as the core entry point, by integrating Baidu Statistics and CNZZ's third-party statistics code, collect user traffic data, registration data and order data on multiple independent sites (Youjijian, Youcaiyun, Youqiyou, Youruanwen, Youhudong, Youtuoke), and based on the data, calculate real-time ROI and delayed ROI according to different conversion attribution time windows. Customer acquisition probability prediction steps: Using the aforementioned "core resource characteristics", "service capability characteristics", "brand asset characteristics" and comprehensive ROI data that integrates short-term and long-term conversion value as input features, a logistic regression model is used for training. The training objective of the logistic regression model is the user's repurchase behavior, and the predicted customer acquisition probability value is output. Marketing strategy generation steps: Based on the customer acquisition probability prediction value, automatically generate marketing strategy suggestions including at least six dimensions: exposure, click, inquiry, registration, payment, and repeat purchase, and push them to the operations staff.

[0006] Furthermore, the present invention also provides a system for implementing the above method, the system comprising: A multi-platform data acquisition gateway for establishing communication connections with the platform APIs of Guangzhou Fanke, Shanghai Meicheng Internet, West Digital, and Ruanmeng; A cross-platform data fusion engine, connected to the acquisition gateway, is embedded with the unified data specification for performing standardized data mapping; A user behavior and ROI analysis module for integrating data from third-party statistical tools and performing ROI calculations; A customer acquisition probability prediction server, connected to the fusion engine and ROI analysis module, is used to run the logistic regression model; A marketing strategy generator, connected to the prediction server, is used to output dimensional marketing strategy recommendations.

[0007] Compared with existing technologies, the beneficial effects of this invention are as follows: by integrating multi-source heterogeneous data through unified standards, it breaks down data silos in traditional enterprise marketing; by combining real-time and delayed ROI analysis, it makes customer acquisition effect evaluation more accurate; and by using a logistic regression model to train and predict the features of the fused data, it improves the accuracy and efficiency of marketing strategies. Attached Figure Description Figure 1 is a flowchart of a digital marketing customer acquisition method for traditional enterprises using multi-platform data, provided by an embodiment of the present invention. The method begins with step S101, the multi-platform data acquisition step, which specifically involves calling the API interfaces of platforms such as Guangzhou Fanke, Shanghai Meicheng Internet, West Digital, and Ruanmeng to obtain heterogeneous data such as products, servers, and domain names. Subsequently, in step S102, the cross-platform data fusion step, the heterogeneous data is parsed and mapped into standard features such as "core resource features" using a pre-defined unified data specification (such as an extensible XML configuration file). Next, in step S103, the user behavior integration and ROI calculation step, tools such as Baidu Statistics are integrated to calculate real-time and delayed ROI based on different conversion attribution time windows (e.g., 7 days, 30 days, 90 days). Then, in step S104, the customer acquisition probability prediction step, the fused features and ROI data are used as input, a logistic regression model is trained, and a predicted customer acquisition probability value is output. Finally, in step S105, the marketing strategy generation step, multi-dimensional marketing strategy suggestions are automatically generated and pushed based on the predicted values.

[0008] Figure 2 is a system structure block diagram of an embodiment of the present invention for implementing the above method. The system 200 includes: a multi-platform data acquisition gateway 201, used to establish communication connections with APIs of platforms such as Guangzhou Fanke, Shanghai Meicheng Internet, West Digital, and Ruanmeng; a cross-platform data fusion engine 202, connected to the acquisition gateway 201, embedding the unified data specification, used to perform standardized data mapping; a user behavior and ROI analysis module 203, used to integrate data from third-party tools such as Baidu Statistics and perform ROI calculation; a customer acquisition probability prediction server 204, connected to the fusion engine 202 and the ROI analysis module 203, used to run the logistic regression model; and a marketing strategy generator 205, connected to the prediction server 204, used to output dimensional marketing strategy suggestions. The above modules communicate and interact with data through a system bus or network.

Claims

1. A digital marketing customer acquisition method for traditional enterprises using multi-platform data, characterized in that, The method integrates data from enterprise website building platforms, cloud server platforms, enterprise email platforms, and various marketing channel platforms, and constructs a unified analysis model, including the following steps: Multi-platform data collection steps: obtain product information by calling the APIs of Guangzhou Fanke's five major product systems, obtain cloud server and email information by calling the APIs of Shanghai Meicheng Internet Cloud Service Platform, obtain domain name and virtual host information by calling the APIs of West Digital Platform, and obtain online media management backend information by calling the APIs of Ruanmeng Platform. Cross-platform data fusion steps: Heterogeneous data from the above multiple platforms are transformed through a preset unified data specification. The unified data specification includes at least mapping "product information" to "core resource characteristics", mapping "cloud server configuration" to "service capability characteristics", and mapping "domain name host information" to "brand asset characteristics", and outputting it to the central database. User behavior integration and ROI calculation steps: Using the company's official website as the core entry point, by integrating Baidu Statistics third-party statistical code, collect user traffic data, registration data and order data from multiple independent sites, and based on the data, calculate real-time ROI and delayed ROI according to different conversion attribution time windows (e.g., 1 to 90 days). Customer acquisition probability prediction steps: Using the aforementioned "core resource characteristics", "service capability characteristics", "brand asset characteristics" and comprehensive ROI data that integrates short-term and long-term conversion value as input features, a logistic regression model is used for training. The training objective of the logistic regression model is the user's repurchase behavior, and the predicted customer acquisition probability value is output. Marketing strategy generation steps: Based on the customer acquisition probability prediction value, automatically generate marketing strategy suggestions including at least six dimensions: exposure, click, inquiry, registration, payment, and repeat purchase, and push them to the operations staff.

2. The method according to claim 1, characterized in that, In the cross-platform data fusion step, the unified data specification is an extensible XML format configuration file used to define the mapping relationship between JSON or XML data fields returned by APIs from different sources and standard fields in the central database.

3. The method according to claim 1, characterized in that, In the customer acquisition probability prediction step, the input features of the logistic regression model also include: historical conversion rate data from multiple dimensions such as exposure, clicks, inquiries, registrations, and conversions, obtained from third-party statistical tools and analyzed based on different retrospective time windows.

4. A system for implementing the method of any one of claims 1 to 3, characterized in that, include: A multi-platform data acquisition gateway for establishing communication connections with platform APIs of companies such as Guangzhou Fanke, Shanghai Meicheng Internet, West Digital, and Ruanmeng. A cross-platform data fusion engine, connected to the acquisition gateway, is embedded with the unified data specification for performing standardized data mapping; A user behavior and ROI analysis module for integrating data from third-party statistical tools and performing ROI calculations; A customer acquisition probability prediction server, connected to the fusion engine and ROI analysis module, is used to run the logistic regression model; A marketing strategy generator, connected to the prediction server, is used to output dimensional marketing strategy recommendations.