Business recommendation method and system based on CRM customer portrait

Abstract

The invention relates to the field of customer relationship management (CRM), in particular to a business recommendation method and system based on a CRM customer portrait, and the method comprises the steps: inputting the key information and behavior data of a customer into a pre-established CRM customer portrait for matching when a server side receives an access request from a client, and obtaining the grouping information of the customer; and performing upper-layer application management on the customer grouping information, formulating a targeted strategy according to characteristics of different customer grouping information, pushing corresponding target customers to related sales personnel, determining demands and value degrees of the target customers, and formulating a personalized business recommendation scheme. Therefore, the problem that content cannot be screened and recommended according to user requirements in the prior art is solved, accurate grouping of users is achieved by deeply mining user portraits in a CRM system, deep association with the system is established on the basis, the response rate and satisfaction of clients are improved through an accurate pushing and follow-up mechanism of a foreground, and the user experience is improved. And the continuous growth of businesses is promoted.

Description

Technical Field

[0001] The present invention relates to the field of customer relationship management (CRM). Specifically, it relates to a business recommendation method and system based on a CRM customer profile. Background Art

[0002] With the increasingly fierce market competition, enterprises pay more and more attention to the maintenance and management of customer relationships. As an important tool for customer relationship management, the CRM system has been widely used in various industries. However, traditional CRM systems often simply record basic customer information and transaction history, lacking in-depth analysis of customer behavior and accurate grouping, resulting in a lack of pertinence and effectiveness in enterprise marketing and services.

[0003] In the existing field of customer relationship management (CRM), although the CRM system can collect a large amount of customer information, the utilization of this information is often not deep and accurate enough. There are many problems in the current CRM system in terms of customer grouping. For example, the grouping method is single, mostly based on simple purchase amount or purchase frequency for grouping, and it cannot comprehensively integrate various characteristics of customers. This leads to poor pertinence of marketing activities, inability to effectively meet the personalized needs of different customer groups, waste of a large amount of marketing resources, low customer response rate and satisfaction, and seriously affects enterprise business growth. In addition, the data flow between the existing CRM system and other business systems is not smooth, and it is impossible to achieve efficient collaborative work, making it difficult to make timely and accurate decisions based on real-time customer situations.

[0004] To solve this problem, the industry has begun to pay attention to the application of customer profile technology. By integrating multi-dimensional information of customers, such as basic information, behavior data, consumption habits, etc., a user profile constructs a comprehensive view of customers, providing the possibility for enterprises' precise marketing and services. However, how to effectively use the user profile for user grouping and apply the grouping results to the actual marketing and service processes is still an urgent problem to be solved. Summary of the Invention

[0005] Embodiments of the present invention provide a business recommendation method and system based on a CRM customer profile, which can at least solve the problems of inaccurate customer grouping, unsmooth data flow between systems, and inability to achieve precise marketing and services in the existing CRM system. By deeply mining the value of customer data, precise customer grouping, deep association between systems, and precise push and follow-up are realized, thereby improving enterprise marketing effects, enhancing customer satisfaction, and promoting business growth.

[0006] The object of the present invention is achieved by the following technical solutions:

[0007] According to an embodiment of the present invention, there is provided a business recommendation method based on a CRM customer profile, including:

[0008] When the server side receives an access request from the client side, it inputs the customer's key information and behavior data into the pre-established CRM customer portrait for matching to obtain the customer's grouping information;

[0009] Conduct upper-layer application management on the customer grouping information, formulate targeted strategies according to the characteristics of different customer grouping information, and push the corresponding target customers to relevant salespersons for the salespersons to determine the needs and value levels of the target customers and formulate personalized business recommendation plans.

[0010] In an exemplary embodiment, before receiving the access request from the client side, it further includes: generating a CRM customer portrait based on the customer's historical key information and behavior data, establishing an association with the server side through an API interface, obtaining the customer's grouping information from the CRM customer portrait, and real-time synchronizing the customer grouping information to the server side to achieve data interaction.

[0011] In an exemplary embodiment, the above-mentioned customer basic information includes name, contact information, industry, and enterprise scale; the transaction information includes: purchase amount, purchase frequency, types of purchased products and preferences; the access request sending records include browsing records, consultation records, frequency of participating in marketing activities, and interaction situations.

[0012] In an exemplary embodiment, the above-mentioned customer key information includes customer basic information and transaction information; the customer behavior data includes the customer's concerned product information and access request sending records.

[0013] In an exemplary embodiment, after obtaining the customer's key information and behavior data, it further includes: strictly cleaning the collected data to remove duplicate, incorrect, and invalid data.

[0014] In an exemplary embodiment, the construction of generating a CRM customer portrait based on the customer's historical key information and behavior data includes:

[0015] Obtain historical customer key information and behavior data, use data analysis algorithms, combine the enterprise's business requirements and market strategies, divide customers into multiple groups with similar characteristics, enable each customer group to customize conditional screening, arbitrary combination of all customer attributes, and set characteristic indicators for different customer groups;

[0016] Define high-value customers as those whose cumulative purchase amount exceeds a predetermined amount, and whose purchase frequency within a preset time range is not less than a predetermined quantity, and who have a high degree of attention to high-profit products;

[0017] Potential growth customers are those who browse high-potential product pages more than a predetermined number within a set period and have had consultation behaviors but have not yet made purchases;

[0018] Inactive customers are those who have not conducted any transactions and have a very low interaction frequency for a period exceeding a preset time range.

[0019] In one exemplary embodiment, the above-mentioned pushing the corresponding target customers to relevant salespersons for the salespersons to determine the needs and value levels of the target customers and formulate personalized business recommendation plans includes: analyzing the current customer grouping information and value levels using a trained recommendation model to generate a list of business recommendation plans; optimizing the recommendation model based on the user's response information to the list; wherein, the response information includes feedback information on whether there is an opinion on the business recommendation plan.

[0020] In one exemplary embodiment, the above-mentioned obtaining a trained recommendation model includes:

[0021] Obtain the set feature indicators of different customer groups, use the customer grouping information and feature indicators as the nodes of the graph respectively, and establish an edge between the customer and its key information and historical behavior data to generate an original interaction graph representing the user grouping information.

[0022] Obtain the customer needs and value levels corresponding to the customer grouping information in the original interaction graph, and establish a grouping feature graph; based on the encoder, learn the correlation between the user grouping information and the customer needs and value levels in the original interaction graph to obtain the user and grouping feature representations.

[0023] By comparing the association degree between the user representation in the original interaction graph and the representation of the corresponding node in the grouping feature graph;

[0024] Adopt a sample selector to select a preset number of associated users from the descending order result based on the association degree to form a positive sample set, and the remaining users form a negative sample set, and define a loss function for contrastive learning between the positive sample set and the negative sample set.

[0025] By calculating the prediction scores of the user grouping information and value levels, formulate business recommendation plans for different customer needs, and define a loss function for supervised learning based on the differences of different business recommendation plans for users; by optimizing the loss function of supervised learning and the loss function of contrastive learning, determine the recommendation loss.

[0026] With the goal of minimizing the recommendation loss, conduct model training until a trained recommendation model that meets the above goal is obtained.

[0027] In one exemplary embodiment, the above-mentioned generating a list of business plan recommendations includes:

[0028] Perform a dot product on the vector representation of the business recommendation plan related to the user grouping information and value level output by the model to obtain the recommendation degree score of the business recommendation plan.

[0029] Determine the recommendation priority of the business recommendation plan according to the high or low prediction score, and generate a business plan recommendation list based on the Top-N sorting of the priority;

[0030] Generating a business plan recommendation includes: calculating the similarity scores of customers in different groups according to the customer grouping information and value degree representation learned by the model; using the user with the highest score as the first item in the recommendation list, and arranging the others in descending order of scores to form a business recommendation plan list and feedback it to the user for the user to select.

[0031] According to an embodiment of the present invention, there is provided a business recommendation system based on a CRM customer portrait, including:

[0032] A grouping module, configured to, when the server side receives an access request from the client side, input the key information and behavior data of the customer into a pre-established CRM customer portrait for matching to obtain the customer grouping information;

[0033] A business recommendation module, configured to perform upper-layer application management on the customer grouping information, formulate targeted strategies according to the characteristics of different customer grouping information, and push the corresponding target customers to relevant salespersons for the salespersons to determine the needs and value degrees of the target customers and formulate personalized business recommendation plans.

[0034] According to an embodiment of the present invention, there is provided a computer-readable storage medium, in which a computer program is stored, and wherein when the computer program is executed by a processor, the steps of the method described in any one of the above first aspects are implemented.

[0035] According to an embodiment of the present invention, there is provided a computer device, the computer device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the method described in any one of the first aspects are implemented.

[0036] Through accurate customer grouping and accurate push, this application enables marketing activities to accurately reach the target customer group, greatly improves the pertinence of marketing activities, is expected to greatly improve the conversion rate of marketing activities, effectively saves marketing resources, and improves the return on marketing investment.

[0037] The personalized follow-up service of this application can better meet the personalized needs of different customer groups, make customers feel the attention and importance of the enterprise, contribute to the establishment of long-term and stable customer relationships, and enhance customer loyalty.

[0038] Through the in-depth excavation of high-value customers and the effective conversion of potential growth customers, it can promote the significant growth of the enterprise's business income, improve the business income growth rate, enhance the enterprise's market competitiveness, and lay a solid foundation for the sustainable development of the enterprise.

[0039] In addition, the proposed invention can achieve a deep association between the server side and the CRM system, ensure two-way data flow, enable each department of the enterprise to work collaboratively based on unified customer data, improve work efficiency, reduce information asymmetry, and enhance the overall operational efficiency of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. In all the drawings, similar elements or parts are generally denoted by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0041] Figure 1 Flowchart of a business recommendation method based on a CRM customer profile provided by the present invention;

[0042] Figure 2 Block diagram of the hardware structure of a mobile terminal for a business recommendation method based on a CRM customer profile;

[0043] Figure 3 Schematic diagram of the structure of a business recommendation system based on a CRM customer profile provided by the present invention;

[0044] Figure 4 Schematic diagram of the structure of a computer device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] The embodiments of the present invention will be described in detail below with reference to the drawings and in conjunction with the embodiments.

[0046] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence.

[0047] In this embodiment, a business recommendation method based on a CRM customer profile is provided. Figure 1 It is a flowchart of a business recommendation method based on a CRM customer profile according to an embodiment of the present invention, and the process includes the following steps:

[0048] Step S11, when the server side receives an access request from the client side, input the key information and behavior data of the customer into a pre-established CRM customer profile for matching to obtain the grouping information of the customer;

[0049] Step S12: Conduct upper-level application management on customer grouping information. Based on the characteristics of different customer grouping information, formulate targeted strategies, and push the corresponding target customers to relevant salespersons for the salespersons to determine the needs and value levels of the target customers and formulate personalized business recommendation plans.

[0050] Optionally, before executing the process step S11 of the above embodiments, before receiving an access request from the client, it further includes: generating a CRM customer portrait based on the customer's historical key information and behavior data, establishing an association with the server through an API interface, obtaining the customer's grouping information from the CRM customer portrait, and synchronizing the customer grouping information to the server in real time to achieve data interaction.

[0051] The customer basic information in the process step S11 in the above embodiments includes name, contact information, industry, and enterprise scale; the transaction information includes: purchase amount, purchase frequency, types of purchased products and preferences; the access request sending records include browsing records, consultation records, frequency of participating in marketing activities, and interaction situations.

[0052] In the above embodiments, the customer key information includes customer basic information and transaction information; the customer behavior data includes the customer's concerned product information and access request sending records.

[0053] After obtaining the customer's key information and behavior data in the process step S11 in the above embodiments, it further includes: strictly cleaning the collected data to remove duplicate, incorrect, and invalid data.

[0054] The construction of generating a CRM customer portrait based on the customer's historical key information and behavior data in the process step S11 in the above embodiments includes:

[0055] Obtain historical customer key information and behavior data, use data analysis algorithms, combine the enterprise's business needs and market strategies, divide customers into multiple groups with similar characteristics, enable each customer group to customize conditional screening of all customer attributes, make arbitrary combinations, and set characteristic indicators for different customer groups;

[0056] Define high-value customers as those whose cumulative purchase amount exceeds a predetermined amount, whose purchase frequency within a preset time range is not less than a predetermined quantity, and who have a high degree of attention to high-profit products;

[0057] Potential growth customers are those who have browsed high-potential product pages more than a predetermined quantity within a set period and have had consultation behaviors but have not yet made purchases;

[0058] Dormant customers are those who have not conducted any transactions and have extremely low interaction frequencies beyond a preset time range.

[0059] In the process step S12 in the above embodiments, the step of pushing the corresponding target customers to relevant salespersons for the salespersons to determine the needs and value degrees of the target customers and formulate personalized business recommendation plans includes: analyzing the current customer grouping information and value degrees by using the obtained recommendation model to generate a list of business recommendation plans; optimizing the recommendation model based on the response information of the user to the list; wherein the response information includes the opinion feedback information on the business recommendation plan.

[0060] In the above embodiments, the obtaining the recommendation model by training includes:

[0061] Obtain the set characteristic indicators of different customer groups, use the customer grouping information and characteristic indicators as the nodes of the graph respectively, and establish an edge between the customer and his key information and historical behavior data to generate an original interaction graph for representing the user grouping information;

[0062] Obtain the customer needs and value degrees corresponding to the customer grouping information in the original interaction graph, and establish a grouping feature graph; based on the encoder, learn the correlation between the user grouping information and the customer needs and value degrees in the original interaction graph to obtain the user and grouping feature representations;

[0063] By comparing the association degree between the user representation in the original interaction graph and the representation of the corresponding node in the grouping feature graph;

[0064] Adopt a sample selector, select a preset number of associated users from the descending order result based on the association degree to form a positive sample set, and the remaining users form a negative sample set, and define a loss function for contrastive learning between the positive sample set and the negative sample set;

[0065] By calculating the prediction scores of the user grouping information and value degrees, formulate business recommendation plans for different customer needs, and define a loss function for supervised learning based on the differences of different business recommendation plans for the user; by optimizing the loss function of supervised learning and the loss function of contrastive learning, determine the recommendation loss;

[0066] With the goal of minimizing the recommendation loss, perform model training until the obtained trained recommendation model meets the said goal.

[0067] In the above embodiments, the generating the list of business plan recommendations includes:

[0068] Perform a dot product on the vector representation of the business recommendation plan related to the user grouping information and value degrees output by the model to obtain the recommendation degree score of the business recommendation plan;

[0069] Determine the recommendation priority of the business recommendation plan according to the high and low of the prediction score, and generate a list of business plan recommendations based on the Top-N sorting of the priority;

[0070] Generating business solution recommendations includes: calculating the similarity scores of customers in different groups based on the customer grouping information and value degree representations learned by the model; taking the user with the highest score as the first item in the recommendation list, and arranging the others in descending order of scores to form a business recommendation solution list and feedback it to the user for selection.

[0071] The method embodiments provided in the above embodiments of the present application can be executed on a mobile terminal, a computer terminal, or a similar computing device. Taking running on a mobile terminal as an example, Figure 2 is the mobile terminal hardware structure block diagram of the business recommendation method based on the CRM customer portrait in the embodiments of the present invention. As Figure 2 shown, the mobile terminal may include one or more ( Figure 2 only one is shown in the figure) processors 101 (the processor 101 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 102 for storing data. Among them, the above mobile terminal may further include a transmission device 103 for communication functions and an input / output device 104. Those of ordinary skill in the art can understand that Figure 2 the structure shown is only schematic and does not limit the structure of the above mobile terminal. For example, the mobile terminal may further include more or fewer components than Figure 2 shown in the figure, or have a different configuration from Figure 2 shown in the figure.

[0072] The memory 102 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for determining the abnormal prompt information of nodes in the embodiments of the present invention. The processor 101 executes various functional applications and data processing by running the computer program stored in the memory 102, that is, implements the above method. The memory 102 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some instances, the memory 102 may further include a memory remotely set relative to the processor 101, and these remote memories can be connected to the mobile terminal through a network. Examples of the above network include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0073] The transmission device 103 is used to receive or send data via a network. Specific examples of the above-mentioned network may include a wireless network provided by a communication provider of a mobile terminal. In one example, the transmission device 103 includes a network adapter (Network Interface Controller, abbreviated as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 103 may be a Radio Frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0074] To better illustrate the implementation manners of the present invention, a specific embodiment is given below.

[0075] An e-commerce enterprise adopts the precise customer grouping and marketing service system based on the CRM customer portrait module of the present invention. First, the enterprise obtains multi-dimensional information of customers by mining the customer portrait module in the CRM system. Then, according to indicators such as the consumption ability, purchase frequency, and interaction activity of customers, the customers are divided into multiple groups such as high-value customers, potential growth customers, and dormant customers.

[0076] Next, the enterprise establishes a seamless connection with the CRM customer portrait module in the background to ensure the real-time synchronization of customer grouping information. On this basis, the enterprise formulates targeted marketing strategies and service plans. For example, for the potential growth customer group, the enterprise designs exclusive marketing activities, such as issuing coupons and providing limited-time discounts.

[0077] Finally, through the precise push mechanism, the target customers are pushed to relevant salespersons. The consultant formulates a personalized follow-up plan according to the portrait and grouping characteristics of the customers. Through this precise push and follow-up mechanism, the enterprise significantly improves the customer response rate and satisfaction, and promotes the continuous growth of the business.

[0078] Based on the above specific implementation manners, the present application provides the following embodiments.

[0079] This embodiment proposes a business recommendation method based on a CRM customer portrait, and the specific steps are as follows:

[0080] I. Customer grouping:

[0081] 1. Data collection: In the CRM system of an enterprise, all relevant information of customers in the past year is regularly collected through data collection tools, including obtaining transaction information from the sales module, obtaining consultation records from the customer service module, extracting behavior information from website browsing logs, etc. A total of 100 customer data records are collected.

[0082] 2. Data Cleaning: By combining data cleaning algorithms with manual review, 5 duplicate data entries were removed, 10 incorrect data entries were corrected, and 2 invalid data entries were filtered. Finally, 83 high-quality data entries were obtained.

[0083] 3. Model Construction and Grouping: Using the Python language, clustering analysis and decision tree algorithm programs were written. Combining the enterprise's business goals and market strategies, a customer grouping model was constructed. The high-value customer indicator was set as the cumulative purchase amount exceeding 500,000 yuan and the purchase frequency not less than 3 times in the past 6 months. The potential growth customer indicator was set as browsing specific high-potential product pages more than 10 times in the past 4 weeks and having consultation behavior, etc. The cleaned data was input into the pre-generated CRM customer portrait, and after calculation, the customer grouping results were obtained. Among them, there were 50 high-value customers, 20 potential growth customers, 30 dormant customers, etc. After manual review, the grouping of individual abnormal customers was fine-tuned to ensure the accuracy and reasonableness of the grouping.

[0084] 4. Establishment of Tag System: Tags were established for customers in each group. For example, the high-value customer tag is "HV - [customer number]", the potential growth customer tag is "PG - [customer number]", and the dormant customer tag is "SC - [customer number]", which is convenient for quick identification and management in subsequent system operations.

[0085] II. Association between Server Side and CRM Customer Portrait

[0086] Interface Development: The technical team used the Java language to develop the docking interface between Huiqitong and the CRM system, adopting the API architecture to ensure the generality and scalability of the interface. After 3 weeks of development, the basic functions of the interface were built and preliminary joint debugging was carried out.

[0087] Data Synchronization Test: Simulate various scenarios such as adding new customers, changing customer information, and adjusting customer groups for data synchronization testing. For example, add 100 customer data entries, modify the contact information and purchase records of 50 customers, and adjust the groups of 30 customers. After testing, data synchronization delay problems were found. By optimizing the database query statements and network transmission settings, the delay problems were solved, ensuring that the data was synchronized within 1 minute, and the accuracy and integrity both reached 100%.

[0088] Upper-level Application Management Setup: In the Huiqitong system, set management permissions for different customer groups. For example, only marketing managers have the permission to create and edit exclusive marketing activities for potential growth customers. Create a time-limited discount marketing activity for potential growth customers, set activity rules, participating products, activity time, etc. through the configuration interface, and accurately push the activity to the corresponding customer group.

[0089] III. Precise Push and Follow-up Implementation

[0090] Front-end function optimization: Design the front-end customer grouping display interface of Huiqitong, presenting customer information of each group in the form of a table, including customer name, affiliated group, key portrait information, etc. Integrate follow-up plan templates, such as "Deep Product Consultation Template" and "Customized Service Plan Template" for high-value customers. When the consultant logs in to the system, the system automatically pushes the target customer list according to the customer grouping.

[0091] Training and promotion: Organize a 3-day training for consultants. Through classroom lectures, practical operation demonstrations, and case analyses, let the consultants be familiar with the accurate push and follow-up functions. Conduct an assessment after the training to ensure that the consultants can proficiently master the system operation and follow-up skills.

[0092] Implementation and feedback: Within one month after going live, the consultants follow up on the customers pushed by the system. Collect customer evaluations of the service through an online questionnaire. A total of 89 valid questionnaires are received, and the customer satisfaction rate reaches 85%. Analyze the business conversion data and find that the conversion rate of potential growth customers has increased by 20%. According to the feedback results, optimize the follow-up plan template and add more personalized recommendation content to further improve the service effect.

[0093] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases, the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that makes a contribution to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc), including several instructions for causing a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0094] In this embodiment, a business recommendation system based on CRM customer portraits is also provided. This system is used to implement the above embodiments and preferred implementation methods, and those that have been described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation by hardware, or a combination of software and hardware, is also possible and contemplated.

[0095] Figure 3 is the structural block diagram of the business recommendation system based on CRM customer portraits according to the embodiment of the present invention. As Figure 3 shown, the system includes:

[0096] The grouping module 210 is configured to, when the server side receives an access request from the client side, input the key information and behavior data of the customer into a pre-established CRM customer portrait for matching, so as to obtain the grouping information of the customer;

[0097] The business recommendation module 220 is configured to perform upper-layer application management on the customer grouping information, formulate targeted strategies according to the characteristics of different customer grouping information, and push the corresponding target customers to relevant salespersons, so that the salespersons can determine the needs and value degrees of the target customers and formulate personalized business recommendation plans.

[0098] It should be noted that the above-mentioned various modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above-mentioned modules are all located in the same processor; or, the above-mentioned various modules are respectively located in different processors in any combination form.

[0099] An embodiment of the present invention also provides a computer-readable storage medium, in which a computer program is stored. Wherein, the computer program is configured to execute the steps in any one of the above method embodiments when running.

[0100] In an exemplary embodiment, the above computer-readable storage medium may include, but is not limited to: USB flash drives, read-only memories (ROM), random access memories (RAM), mobile hard disks, magnetic disks, or optical disks and other various media that can store computer programs.

[0101] An embodiment of the present invention also provides that in one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as Figure 4 shown. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Wherein, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through technologies such as WIFI, mobile cellular networks, NFC (near field communication), or others. The computer program, when executed by the processor, implements a business recommendation method based on a CRM customer portrait.

[0102] Those skilled in the art can understand, Figure 4The structure shown is only a block diagram of some of the structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have a different component arrangement.

[0103] Specific examples in this embodiment may refer to the examples described in the above embodiments and exemplary embodiments, and will not be repeated here.

[0104] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. They can be implemented by program codes executable by the computing device. Thus, they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a different order than here, or they can be separately made into individual integrated circuit modules, or multiple modules or steps of them can be made into a single integrated circuit module to implement. In this way, the present invention is not limited to any specific combination of hardware and software.

[0105] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A business recommendation method based on CRM customer portrait, characterized in that: include: When the server receives an access request from the client, it inputs the customer's key information and behavior data into the pre-established CRM customer profile for matching and obtains the customer's grouping information; Carry out upper-level application management of customer group information, formulate targeted strategies according to the characteristics of different customer group information, and push the corresponding target customers to relevant sales personnel, so that the sales personnel can determine the needs and value of the target customers and formulate personalized business recommendation plans.

2. The method according to claim 1, characterized in that Before receiving the access request from the client, the process also includes: generating a CRM customer portrait based on the customer's historical key information and behavior data, establishing an association with the server through an API interface, obtaining the customer's grouping information from the CRM customer portrait, and synchronizing the customer grouping information to the server in real time to achieve data interaction.

3. The method according to claim 1, characterized in that The basic customer information includes name, contact information, industry, and company size; the transaction information includes: purchase amount, purchase frequency, types of products purchased and preferences; the access request sending records include browsing records, consultation records, frequency of participation in marketing activities and interaction status.

4. The method according to claim 1, characterized in that The customer key information includes basic customer information and transaction information; the customer behavior data includes the user's concerned product information and access request sending records.

5. The method according to claim 1, characterized in that The acquisition of the key information and behavior data of the customer also includes: strictly cleaning the collected data to remove duplicate, erroneous and invalid data.

6. The method according to claim 2, characterized in that The construction of the CRM customer profile generated based on the customer's historical key information and behavior data includes: Obtain historical customer key information and behavior data, use data analysis algorithms, and combine the company's business needs and market strategies to divide customers into multiple groups with similar characteristics, so that each customer group can customize all customer attributes to perform conditional screening, any combination, and set characteristic indicators for different customer groups; High-value customers are defined as customers whose cumulative purchase amount exceeds the predetermined amount, whose purchase frequency is not less than the predetermined amount within the preset time frame, and who have a high degree of interest in high-profit products; Potential growth customers are customers who have browsed high-potential product pages more than a predetermined number of times within a set period and have consulted but have not yet made a purchase; Dormant customers are customers who have not made any transactions for more than a preset time frame and whose interaction frequency is very low.

7. The method according to claim 1, characterized in that The corresponding target customers are pushed to relevant sales personnel for the sales personnel to determine the needs and value levels of the target customers and formulate personalized business recommendation plans, including: using the recommendation model obtained through training to analyze the current customer grouping information and value levels, and generate a list of business recommendation plans; optimizing the recommendation model based on the user's response information to the list; wherein the response information includes feedback information on whether there is an opinion on the business recommendation plan.

8. The method according to claim 7, characterized in that The training to obtain the recommendation model includes: Obtain characteristic indicators for different customer groups, use customer group information and characteristic indicators as nodes of the graph, establish an edge between the customer and its key information and historical behavior data, and generate an original interaction graph for representing user group information; Obtain the customer needs and value levels corresponding to the customer grouping information in the original interaction graph, and establish a grouping feature graph; Based on the encoder, learn the correlation between the user grouping information in the original interaction graph and the customer needs and value levels, and obtain user and group feature representations; By comparing the correlation between the user representation in the original interaction graph and the representation of its corresponding node in the grouped feature graph; A sample selector is used to select a preset number of related users from the descending sorting results based on the relevance to form a positive sample set, and the remaining users form a negative sample set, and a loss function for comparative learning between the positive sample set and the negative sample set is defined; By calculating the predicted scores of user grouping information and value levels, business recommendation plans for different customer needs are formulated, and the loss function of supervised learning is defined based on the differences in user responses to different business recommendation plans; the recommendation loss is determined by optimizing the loss function of supervised learning and the loss function of contrastive learning; and the model is trained with the goal of minimizing the recommendation loss until a recommendation model that meets the goal is obtained after training.

9. The method according to claim 7, characterized in that Generating the service plan recommendation list includes: performing dot multiplication on the user grouping information output by the model and the service recommendation plan vector representation related to the value degree to obtain the recommendation degree score of the service recommendation plan; Determine the recommendation priority of the business recommendation plan based on the prediction score, and generate a business plan recommendation list based on the Top-N sorting of the priority; Generating business plan recommendations includes: calculating the similarity scores of customers in different groups based on the customer grouping information and value representation learned by the model; placing the user with the highest score as the first item in the recommendation list, and arranging the others in order of score to form a business recommendation plan list to be fed back to the user for selection.

10. A business recommendation system based on CRM customer portrait, characterized in that: include: The grouping module is used to input the key information and behavior data of the customer into the pre-established CRM customer profile for matching when the server receives the access request from the client, so as to obtain the customer's grouping information; The business recommendation module is used to perform upper-level application management on customer group information, formulate targeted strategies according to the characteristics of different customer group information, and push the corresponding target customers to relevant sales personnel so that the sales personnel can determine the needs and value of the target customers and formulate personalized business recommendation plans.

Images