A method to improve the business efficiency of CRM system through automated processes

Abstract

A method for improving the efficiency of a CRM business system through an automated process comprises the following steps: S1: firstly, a CRM business system is established, comprising an information collection module, a business judgment module, an analysis module, an automatic circulation module, a monitoring module, a display module and a data protection module. After the customer information is collected, the business judgment module and the analysis module are carried out simultaneously; wherein: the analysis module comprises a data screening module, a preprocessing module, a model selection and fine-tuning module, an information extraction and analysis module and a classification and clustering module, and the analysis module is used to classify and output unstructured data in a structured form; S2: the information collection module collects customer-related information, and after being processed and integrated by the CRM business system, the information is automatically assigned to the owner to ensure that customer needs are responded to in a timely manner. The information is intuitively displayed through the display module, and the unstructured data is understood and used more accurately, thereby obtaining a deeper customer insight and improving the decision-making level of the CRM system.

Description

Technical Field

[0001] The present invention relates to the field of business management methods, and in particular to a method for improving the business efficiency of a CRM system through an automated process. Background Art

[0002] Customer relationship management (CRM) is a concept in economic management. Its essence is to select and manage customers to achieve the maximum long-term value. With the continuous development of the market, customer relationship management (CRM) systems came into being. Modern software systems are mainly used to manage the interaction between companies and customers. It integrates functions such as sales, marketing, customer service and support, aiming to improve customer satisfaction and loyalty, thereby promoting the sustainable growth of business. With the continuous development of information technology, CRM systems are gradually developing in the direction of automation, but currently they are mainly modularized. With the increase in customer needs and the types of services provided, data silos and system compatibility issues between different departments and systems of CRM systems are becoming more and more prominent. Secondly, CRM systems contain a large amount of unstructured data, such as customer interaction records, emails, social media, etc. The difficulty of processing and analyzing these data has also increased, resulting in the complexity of communication between management companies and customers and reduced accuracy, and the inability to achieve seamless connection, and it is easy to cause data leakage and other problems. Summary of the invention

[0003] The purpose of the present invention is to address the shortcomings of the prior art and solve the problem of unstructured data processing and analysis through end-to-end automated integration and advanced natural language processing and machine learning technologies.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] A method for improving the efficiency of a CRM business system through an automated process, applied to a client, comprises the following steps:

[0006] S1: First, a CRM business system is established, including: an information collection module, a business judgment module, an analysis module, an automatic circulation module, a monitoring module, a display module and a data protection module. After the customer information is collected, the business judgment module and the analysis module are performed synchronously;

[0007] Among them: the parsing module includes a data screening module, a preprocessing module, a model selection and fine-tuning module, an information extraction and analysis module, and a classification and clustering module. The parsing module is used to classify and output unstructured data in a structured form;

[0008] S2: The information collection module collects customer-related information, and after being processed and integrated by the CRM business system, it automatically assigns the owner and monitors whether the customer's owner follows up with the customer in a timely manner to ensure that customer needs are responded to in a timely manner. It is intuitively displayed through the display module, making it easy for decision makers to quickly understand and analyze customer data.

[0009] Preferably, the information collection module stores customer information in the database, passes the data to the business judgment module and the analysis module after the collection is completed, and then transfers the customer to different owners; the specific steps of the business judgment module are as follows:

[0010] a. Determine the region to which the customer belongs based on the customer's corporate registration location and IP address, and assign the customer to the corresponding branch or subsidiary based on the customer's region;

[0011] b. Determine whether a regional branch is configured in the region. If the customer's region does not have a regional branch, it will be transferred to the group headquarters customer public pool and assigned by the group headquarters staff. If it is configured, it will be automatically transferred to the corresponding branch and subsidiary, and determine whether the enterprise already exists in the database;

[0012] c. Determine whether the contact person exists and whether the company has the registered contact person. If the contact person does not exist, the system will automatically add the contact person to the company's customer name and transfer the contact person to the public pool of the branch or subsidiary in the city where the company is located. If the contact person exists, determine whether the company is in the public pool or the private pool.

[0013] d. Determine whether it is in the public pool or the private pool. If the corporate customer is in the private pool, notify the corresponding customer owner to follow up the corporate customer in a timely manner; if the corporate customer is in the public pool, notify the regional manager / manager to allocate / follow up;

[0014] e. Determine the customer type and the business personnel's expertise. Use text preprocessing and natural language analysis tools to analyze the company's business scope, company type, industry, and core product / service information, comprehensively determine the customer type, and make transfer recommendations based on the business personnel's preset expertise;

[0015] f. Provide future business directions. After the customer type and business personnel are determined, match them with the subsequent business that can be reported, and provide business personnel with references for business directions that can be explored.

[0016] Preferably, the automatic transfer module will automatically transfer the enterprise according to the result of judging the business conditions. The business personnel can see the transferred enterprises in their private customer management, and the system will notify the business personnel. If there is no follow-up within the deadline, it will be automatically transferred to the public pool and other business personnel will be notified to apply for it. The monitoring module will monitor whether the customer's owner follows up the customer in time by whether there is new information about the customer, such as follow-up information from phone records and email records, to ensure that customer needs are responded to in a timely manner.

[0017] Preferably, the data screening module uses the customer-specific ID to conduct primary targeted labeling and data screening on the unstructured data by constructing a large keyword semantic library for the emails, questions, feedback and opinions sent by the customer, as well as a large amount of invalid information in the custom text. Then, the preprocessing module performs secondary unstructured data processing on the initially screened data.

[0018] Preferably, the model selection and fine-tuning module performs Transformer model training on the secondary processed data, and then performs model fine-tuning to improve the performance and accuracy of the model by adjusting the learning rate, optimizing the network structure or increasing the training data; the information extraction and analysis module automatically identifies entities in the text, such as names of people, places, organizations, dates, and product names, through named entity recognition (NER), to help identify the basic information and related details of customers, build an association network between customers and products and services, provide a basis for optimizing sales strategies and providing personalized services, and output the analysis results in a structured form.

[0019] Preferably, the classification and clustering module classifies and clusters the structured text according to themes or categories, and manages the customer information in groups, so as to facilitate decision makers to quickly understand and analyze customer data.

[0020] Preferably, the data protection module includes a data desensitization module and a security protection module. The data desensitization module mainly shields sensitive data, and is divided into static data desensitization and dynamic data desensitization. Static data desensitization refers to the "movement and simulation replacement" of data, which is to extract data for desensitization processing, and then send it to downstream links for arbitrary access and reading and writing. The desensitized data is isolated from the production environment; the dynamic data desensitization refers to real-time desensitization processing while accessing sensitive data, and different desensitization schemes can be executed for different roles, different permissions, and different data types, thereby ensuring that the returned data is available and safe.

[0021] Preferably, the security protection module mainly adopts an anti-crawler mechanism to protect data security, and the anti-crawler mechanism mainly includes the following categories:

[0022] Header Validation

[0023] When a browser sends a request to a server, it sends a header. In the header, there are several values, and the values ​​are different for each browser. If the header is not equivalent to a regular browser, a robot can be easily identified, or if the header pattern used is inconsistent with the pattern of a known browser, it will be restricted or even blocked;

[0024] TCP / IP fingerprinting

[0025] When the crawler uses the Internet, TCP leaves many parameters that need to be set by the device / operating system used (such as TTL or initial window state). If these parameter values ​​are inconsistent, it means that the data is likely to be crawled;

[0026] Abnormal IP access restriction

[0027] Use IP blacklist to restrict access and block all requests from that specific data center so that crawlers will not be able to access the site;

[0028] Verification Code

[0029] The verification code is used to determine whether it is a robot access, but the machine can easily solve this type of verification code.

[0030] Preferably, the security protection module also includes a real-time monitoring module and a risk control module, which mainly monitor and control the risks of database data in real time to prevent the database system from being attacked, stolen or tampered with by hackers, and monitor, audit and replay the database in real time to truly display the complete operation process at that time, facilitate the analysis and tracing of system security issues, thereby protecting the security of data center data.

[0031] The beneficial effects of the present invention are:

[0032] 1. Through workflow automation technology, all business processes in the CRM system are automated to achieve end-to-end automation and integrate various links to ensure data consistency and accuracy. Customer business data and operations can be automatically pushed to the sales staff in the area for lead follow-up based on the customer's location, making it easier for local sales to follow up and communicate with customer leads efficiently, thereby breaking the data silos between different departments and systems and achieving seamless integration.

[0033] 2. Natural language processing (NLP) technology and pre-trained Transformer neural network large model algorithm are used to analyze and process unstructured data in the CRM system. Compared with existing technologies, this technical solution can more accurately understand and utilize unstructured data, thereby obtaining deeper customer insights and improving the decision-making level of the CRM system.

[0034] 3. For the large amount of sensitive data involved in the CRM system, such as customer information, transaction records, etc., this technical solution strengthens data security and privacy protection. Through data desensitization and data security modules, it prevents the database system from being attacked, stolen or tampered with by hackers, and monitors, audits and replays the database in real time to truly show the complete operation process at the time, which is convenient for analyzing and tracing system security issues, thereby protecting the security of data center data. Compared with existing technologies, this technical solution can better deal with emerging network attack methods and data leakage risks, and ensure the security of customer information. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a system workflow diagram of the present invention;

[0036] Figure 2 is a flow chart of the method of the present invention;

[0037] Figure 3 It is the flowchart of the parsing module;

[0038] Figure 4 This is a flowchart of the business judgment module; DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0040] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" etc., indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention. In the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0041] Embodiment 1

[0042] See also Figure 1-4 , a method for improving the efficiency of a CRM business system through an automated process, applied to a client, comprising the following steps:

[0043] A method for improving the efficiency of a CRM business system through an automated process, applied to a client, comprises the following steps:

[0044] S1: First, a CRM business system is established, including: an information collection module, a business judgment module, an analysis module, an automatic circulation module, a monitoring module, a display module and a data protection module. After the customer information is collected, the business judgment module and the analysis module are performed synchronously;

[0045] Among them: the parsing module includes a data screening module, a preprocessing module, a model selection and fine-tuning module, an information extraction and analysis module, and a classification and clustering module. The parsing module is used to classify and output unstructured data in a structured form;

[0046] S2: The information collection module collects customer-related information, and after being processed and integrated by the CRM business system, it automatically assigns the owner and monitors whether the customer's owner follows up with the customer in a timely manner to ensure that customer needs are responded to in a timely manner. It is intuitively displayed through the display module, making it easy for decision makers to quickly understand and analyze customer data.

[0047] The information collection module implements the storage of customer information, and the system front end supports all users to register themselves to use all enterprise service tools in the platform. When a user enters the platform, he needs to bind the enterprise information after registration. When the user binds the enterprise, the system will transmit the enterprise name data to the background to realize the storage of customer information. After the collection is completed, the data will be passed to the business judgment module and the analysis module, and then the customer will be transferred to different owners; the specific steps of the business judgment module are as follows:

[0048] a. Determine the region to which the customer belongs based on the customer's corporate registration location and IP address, and assign the customer to the corresponding branch or subsidiary based on the customer's region;

[0049] b. Determine whether a regional branch is configured in the region. If the customer's region does not have a regional branch, it will be transferred to the group headquarters customer public pool and assigned by the group headquarters staff. If it is configured, it will be automatically transferred to the corresponding branch and subsidiary, and determine whether the enterprise already exists in the database;

[0050] c. Determine whether the contact person exists and whether the company has the registered contact person. If the contact person does not exist, the system will automatically add the contact person to the company's customer name and transfer the contact person to the public pool of the branch or subsidiary in the city where the company is located. If the contact person exists, determine whether the company is in the public pool or the private pool.

[0051] d. Determine whether it is in the public pool or the private pool. If the corporate customer is in the private pool, notify the corresponding customer owner to follow up the corporate customer in a timely manner; if the corporate customer is in the public pool, notify the regional manager / manager to allocate / follow up;

[0052] e. Determine the customer type and the business personnel's expertise, use text preprocessing and natural language analysis tools to analyze the company's business scope, company type, industry, and core product / service information, comprehensively determine the customer type, and make transfer recommendations based on the business personnel's preset expertise; comprehensively determine the customer type and the company's most likely business direction, and transfer based on the business personnel's preset expertise. For example, company A belongs to the software service and development industry, and mainly develops ERP software. Business personnel B has been communicating with companies in the software development industry for a long time, and the preset business direction is the B-end software development industry, then company A will automatically transfer to business personnel B for follow-up.

[0053] f. Provide future business directions. After the customer type and business personnel are determined, match them with the subsequent business that can be reported, and provide business personnel with references for business directions that can be explored.

[0054] The automatic transfer module will automatically transfer the enterprise according to the results of the business conditions. The business personnel can see the transferred enterprises in their private customer management, and the system will notify the business personnel. If there is no follow-up within the deadline, it will be automatically transferred to the public pool and other business personnel will be notified to apply for it. The monitoring module will monitor whether the customer's owner follows up on the customer in a timely manner by checking whether there is new information about the customer, such as follow-up information from phone records and email records, to ensure that customer needs are responded to in a timely manner.

[0055] Reference Figure 3 , the data screening module uses the customer's exclusive ID as an important basis for customer identification, and analyzes all customer inquiries and product feedback. The emails sent by customers, the questions they ask, and the feedback they give are all long custom texts, which contain a lot of invalid information. By building a large keyword semantic library, the unstructured data is initially targeted and labeled to screen out corporate information with a high proportion of valid information. Then the preprocessing module performs secondary unstructured data processing on the initially screened data; the preliminary screening step only helps the model screen out a batch of completely invalid information data. Text preprocessing is still required to process the customer's unstructured data, including text cleaning (removing irrelevant information in the text, such as advertisements, spam content, etc.), correcting spelling errors (such as unifying text formats and terms), word segmentation and part-of-speech tagging (splitting the text into independent words or phrases, identifying the part of speech of each word, such as nouns, verbs, adjectives, etc., to provide a basis for subsequent analysis.

[0056] The model selection and fine-tuning module performs Transformer model training on the secondary processed data, and then fine-tunes the model to improve the performance and accuracy of the model by adjusting the learning rate, optimizing the network structure or increasing the training data. The information extraction and analysis module automatically identifies entities in the text through named entity recognition (NER), and analyzes the relationship between entities through relationship extraction technology, such as the relationship between customer satisfaction and product defects, customer feedback on different service personnel, etc., combined with keyword extraction, such as "purchase", "use", "recommend", etc., to automatically identify important keywords and phrases in the text, help the machine quickly understand the text body, and help understand customer behavior and preferences. Build an association network between customers and products and services to provide a basis for optimizing sales strategies and providing personalized services. Use the Transformer model to perform sentiment analysis on customer feedback and identify customers' emotional tendencies (such as positive, negative, and neutral. Identify customer intentions and needs, such as purchase intentions, consulting questions, complaints and suggestions, etc.

[0057] The classification and clustering module classifies and clusters structured text according to themes or categories, and manages customer information in groups, making it easier for decision makers to quickly understand and analyze customer data.

[0058] The data protection module includes a data desensitization module and a security protection module. The data desensitization module mainly shields sensitive data, such as ID number, mobile phone number, card number, customer name, customer address, email address, salary, etc., and deforms the data through desensitization rules (replacement, rearrangement, encryption, truncation, masking) to achieve reliable protection of privacy data. It is divided into static data desensitization and dynamic data desensitization. Static data desensitization refers to the "movement and simulation replacement" of data. It extracts data for desensitization and sends it to downstream links for arbitrary access and reading and writing. The desensitized data is isolated from the production environment; the dynamic data desensitization refers to real-time desensitization while accessing sensitive data. Different desensitization schemes can be executed for different roles, different permissions, and different data types to ensure that the returned data is available and safe.

[0059] The security protection module mainly uses anti-crawler mechanisms to protect data security. The anti-crawler mechanisms mainly include the following categories:

[0060] Header Validation

[0061] When a browser sends a request to a server, it also sends a header. Several values ​​in the header are different for each browser. The system is set up to identify robots by determining whether the header pattern is not equivalent to that of a regular browser, and by determining whether the pattern used is consistent with that of a known browser. If not, restrictions are enabled and the request is blocked.

[0062] TCP / IP fingerprinting

[0063] When the crawler uses the Internet, TCP leaves many parameters that need to be set by the device / operating system used (such as TTL or initial window state). If these parameter values ​​are inconsistent, it means that the data is likely to be crawled;

[0064] Abnormal IP access restriction

[0065] Use IP blacklist to restrict access and block all requests from that specific data center so that crawlers will not be able to access the site;

[0066] Verification Code

[0067] The verification code is used to determine whether it is a robot access, but the machine can easily solve this type of verification code.

[0068] The security protection module also includes a real-time monitoring module and a risk control module, which mainly monitor and control the database data in real time and prevent the database system from being attacked, stolen or tampered with by hackers. The real-time monitoring, auditing and playback of the database truly show the complete operation process at that time, which is convenient for analyzing and tracing system security issues, thereby protecting the security of data center data. The following control measures are mainly used: real-time request monitoring, user tracing, real-time operation monitoring, formulating database strategies, establishing a data security rule base, customizing tags, feature detection and audit rule detection, and establishing a data isolation mechanism.

[0069] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for improving the efficiency of a CRM business system by automating a process, characterized in that: The steps include: S1: First, a CRM business system is established, including: an information collection module, a business judgment module, an analysis module, an automatic circulation module, a monitoring module, a display module and a data protection module. After the customer information is collected, the business judgment module and the analysis module are performed synchronously; Among them: the parsing module includes a data screening module, a preprocessing module, a model selection and fine-tuning module, an information extraction and analysis module, and a classification and clustering module. The parsing module is used to classify and output unstructured data in a structured form; S2: The information collection module collects customer-related information, and after being processed and integrated by the CRM business system, it automatically assigns the owner and monitors whether the customer's owner follows up with the customer in a timely manner to ensure that customer needs are responded to in a timely manner. It is intuitively displayed through the display module, making it easy for decision makers to quickly understand and analyze customer data.

2. The method for improving the efficiency of a CRM business system through an automated process according to claim 1, characterized in that: The information collection module stores customer information in the database, passes the data to the business judgment module and the analysis module after the collection is completed, and then transfers the customer to different owners; the specific steps of the business judgment module are as follows: a. Determine the region to which the customer belongs based on the customer's corporate registration location and IP address, and assign the customer to the corresponding branch or subsidiary based on the customer's region; b. Determine whether a regional branch is configured in the region. If the customer's region does not have a regional branch, it will be transferred to the group headquarters customer public pool and assigned by the group headquarters staff. If it is configured, it will be automatically transferred to the corresponding branch and subsidiary, and determine whether the enterprise already exists in the database; c. Determine whether the contact person exists, and whether the company has the registered contact person. If the contact person does not exist, the system will automatically add the contact person under the name of the company customer, and transfer the contact person to the public pool of the branch and subsidiary companies in the city where the company is located; If there is a contact person, determine whether the enterprise is in the public pool or the private pool; d. Determine whether it is in the public pool or the private pool. If the corporate customer is in the private pool, notify the corresponding customer owner to follow up the corporate customer in a timely manner; if the corporate customer is in the public pool, notify the regional manager / manager to allocate / follow up; e. Determine the customer type and the business personnel's expertise. Use text preprocessing and natural language analysis tools to analyze the company's business scope, company type, industry, and core product / service information, comprehensively determine the customer type, and make transfer recommendations based on the business personnel's preset expertise; f. Provide future business directions. After the customer type and business personnel are determined, match them with the subsequent business that can be reported, and provide business personnel with references for business directions that can be explored.

3. The method for improving the efficiency of a CRM business system through an automated process according to claim 1, characterized in that: The automatic transfer module will automatically transfer the enterprise according to the results of the business conditions. The business personnel can see the transferred enterprises in their private customer management, and the system will notify the business personnel. If there is no follow-up within the deadline, it will be automatically transferred to the public pool and other business personnel will be notified to apply for it. The monitoring module will monitor whether the customer's owner follows up on the customer in a timely manner by checking whether there is new information about the customer, such as follow-up information from phone records and email records, to ensure that customer needs are responded to in a timely manner.

4. The method for improving the efficiency of a CRM business system through an automated process according to claim 1, characterized in that: The data screening module uses the customer's exclusive ID to conduct initial targeted labeling and screening of unstructured data in emails, consultation questions, feedback opinions, and custom texts sent by customers, based on a large amount of invalid information. The preprocessing module then performs secondary unstructured data processing on the initially screened data by building a large keyword semantic library.

5. The method for improving the efficiency of a CRM business system through an automated process according to claim 1, characterized in that: The model selection and fine-tuning module trains the Transformer model on the secondary processed data, and then fine-tunes the model to improve the performance and accuracy of the model by adjusting the learning rate, optimizing the network structure or increasing the training data. The information extraction and analysis module automatically identifies entities in the text, such as names of people, places, organizations, dates, and product names, through named entity recognition (NER), to help identify the basic information and related details of customers, build an association network between customers and products and services, provide a basis for optimizing sales strategies and providing personalized services, and output the analysis results in a structured form.

6. The method for improving the efficiency of a CRM business system through an automated process according to claim 1, characterized in that: The classification and clustering module classifies and clusters structured text according to themes or categories, and manages customer information in groups, making it easier for decision makers to quickly understand and analyze customer data.

7. The method for improving the efficiency of a CRM business system through an automated process according to claim 1, characterized in that: The data protection module includes a data desensitization module and a security protection module. The data desensitization module mainly shields sensitive data and is divided into static data desensitization and dynamic data desensitization. Static data desensitization refers to the "movement and simulation replacement" of data, which is to extract and desensitize the data and then send it to the downstream link for arbitrary access and reading and writing. The desensitized data is isolated from the production environment; The dynamic data desensitization refers to the real-time desensitization processing while accessing sensitive data. Different desensitization schemes can be implemented for different roles, different permissions, and different data types, thereby ensuring that the returned data is available and secure.

8. The method for improving the efficiency of a CRM business system through an automated process according to claim 7, characterized in that: The security protection module mainly uses anti-crawler mechanisms to protect data security. The anti-crawler mechanisms mainly include the following categories: Header Validation When a browser sends a request to a server, it sends a header. In the header, there are several values, and the values ​​are different for each browser. If the header is not equivalent to a regular browser, a robot can be easily identified, or if the header pattern used is inconsistent with the pattern of a known browser, it will be restricted or even blocked; TCP / IP fingerprinting When the crawler uses the Internet, TCP leaves many parameters that need to be set by the device / operating system used (such as TTL or initial window state). If these parameter values ​​are inconsistent, it means that the data is likely to be crawled; Abnormal IP access restriction Use IP blacklist to restrict access and block all requests from that specific data center so that crawlers will not be able to access the site; Verification Code Use the verification code to determine whether the access is from a robot.

9. The method for improving the efficiency of a CRM business system through an automated process according to claim 1, characterized in that: The security protection module also includes a real-time monitoring module and a risk control module, which mainly monitor and control the risks of database data in real time to prevent the database system from being attacked, stolen or tampered with by hackers, and monitor, audit and replay the database in real time to truly display the complete operation process at that time, which is convenient for analyzing and tracing system security issues, thereby protecting the security of data center data.

Images