Large language model-based sales promotion analysis device and method, and electronic equipment

Through a sales improvement analysis device based on a large language model, combining data analysis and external factors, a personalized sales strategy is generated, which solves the problem that traditional automobile sales rely on personal experience, realizes accurate demand analysis and scientific decision-making support, and improves transaction rate and sales efficiency.

CN118886952APending Publication Date: 2024-11-01SHANGHAI SHUHENG INFORMATION TECH CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411062303.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-05
Publication Date
2024-11-01

AI Technical Summary

Technical Problem

Traditional automobile sales rely on personal experience and cannot systematically summarize effective sales strategies. The existing technology cannot accurately analyze decision-making factors in the conversation content, and cannot reduce manual intervention and provide scientific decision-making support.

Method used

Adopt sales improvement analysis device based on large language models, including data collection and preprocessing modules, large language model modules, data analysis modules, external factor archiving modules and suggestions generation modules. Through large language models training and optimization, combined with external factors , generate personalized sales strategy suggestions, reduce personal experience dependence, and improve transaction rate and sales efficiency.

Benefits of technology

By deeply exploring customer decision-making factors, generating accurate personalized sales strategies, improving transaction rates, reducing dependence on personal experience, providing scientific decision-making support, and improving sales efficiency and system intelligence.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118886952A_ABST
    Figure CN118886952A_ABST
Patent Text Reader

Abstract

The invention relates to a sales promotion analysis device and method based on a large language model and electronic equipment, and the device comprises a data collection and preprocessing module which is used for collecting sales information and dialogue data, and carrying out the cleaning and content extraction of the data; the big language model module is used for training and optimizing a big language model by using the data preprocessed by the data collecting and preprocessing module; the data analysis module is used for performing feature extraction and transaction analysis on the processed data by using a large language model module, and deeply mining key factors influencing customer decisions in combination with current external influence factors; the external factor archiving module is used for introducing and archiving external factors at all time points; the suggestion generation module is used for generating personalized sales strategy suggestions for each customer according to the comprehensive analysis result and the current external influence factors; the method has remarkable effects in the aspects of improving the transaction rate, reducing personal experience dependence, providing accurate demand analysis and scientific decision support and the like.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a sales improvement analysis device, method and electronic equipment based on a large language model. [Background Technology]

[0002] In the traditional automobile sales field, salespeople mainly rely on personal experience and skills to communicate with customers and sell vehicles. However, this method relies heavily on the personal ability of salespeople and cannot systematically summarize effective sales strategies.

[0003] Although some technologies on the market have attempted to introduce artificial intelligence and data analysis to improve sales efficiency, they are unable to reduce manual intervention for consumers, nor can they provide accurate demand analysis and scientific decision-making support. At the same time, existing data analysis tools are often limited to simple analysis of labeled data and are unable to deeply explore potential decision-making factors in the conversation content. [Summary of the invention]

[0004] The purpose of the present invention is to solve the above-mentioned deficiencies and to provide a sales improvement analysis device, method and electronic device based on a large language model. By combining a modern large language model and scientific data analysis, it has significant effects in improving transaction rates, reducing reliance on personal experience, providing accurate demand analysis and scientific decision-making support, etc., and can provide sales personnel with targeted suggestions, thereby improving sales efficiency and transaction rates.

[0005] In one aspect, the present invention provides a sales improvement analysis device based on a large language model, comprising:

[0006] Data collection and preprocessing module: used to collect sales information and conversation data from sales staff and customers, and clean and extract content from these data;

[0007] Large language model module: Use the data preprocessed by the data collection and preprocessing module to perform LLM training and model optimization of the large language model;

[0008] Data analysis module: Use the large language model module to extract features and conduct transaction analysis on the processed data, and combine it with the external influencing factors at the time to deeply explore the key factors that affect customer decisions;

[0009] External factor archiving module: used to introduce and archive external factors at various time points for reference when generating personalized recommendations;

[0010] Recommendation generation module: used to generate personalized sales strategy recommendations for each customer based on comprehensive analysis results and current external influencing factors.

[0011] In a possible implementation, the data collection and preprocessing module also collects transaction data and external influencing factors at various time points, and the external influencing factors include but are not limited to promotional activity information and the latest loan policies.

[0012] In another possible implementation, the external factors in the external factor archiving module include, but are not limited to, historical promotional activity information, historical loan policies, current promotional activity information, and current loan policies.

[0013] In another possible implementation, a system feedback module is further included. The system feedback module is connected to the large language model module and the suggestion generation module respectively. The system feedback module is used to collect feedback data after the implementation of the sales strategy, as well as model retraining and optimization.

[0014] In another aspect, the present invention provides a sales improvement analysis method based on a large language model, comprising the following steps:

[0015] 1) Data collection and preprocessing: Collect sales information and conversation data from sales staff and customers, and clean and extract content from these data;

[0016] 2) Large language model training and optimization: Use the data preprocessed in step 1) to perform LLM training and model optimization of the large language model;

[0017] 3) Data analysis: Use a large language model to extract features and conduct transaction analysis on the processed data, and combine the external influencing factors at the time to deeply explore the key factors that affect customer decisions;

[0018] 4) External factor archiving: Introduce and archive external factors at various time points for reference when generating personalized recommendations;

[0019] 5) Recommendation generation: Generate personalized sales strategy recommendations for each customer based on comprehensive analysis results and current external influencing factors.

[0020] In a possible implementation, in step 1), transaction data and external influencing factors at various time points are also collected, and the external influencing factors include but are not limited to promotional activity information and the latest loan policies.

[0021] In another possible embodiment, step 2) comprises the following steps:

[0022] 201) Model training: The large language model is trained using a large amount of historical data, including but not limited to previous sales conversations, transaction status, and external factors at the time;

[0023] 202) Model optimization: The model is optimized through continuous feedback, enabling it to understand and analyze new conversation content more accurately.

[0024] In another possible implementation, after step 5), a system feedback step is also included: collecting feedback data after the implementation of the sales strategy for retraining and optimization of the model to continuously improve system performance and analysis accuracy.

[0025] According to a third aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a program, and when the program is executed by a processor, the above method is implemented.

[0026] In a fourth aspect of the present invention, an electronic device is proposed, comprising a memory and a processor, wherein the memory stores a program executable on the processor, and when the program is executed by the processor, the electronic device implements the above method.

[0027] Compared with the prior art, the present invention has the following advantages:

[0028] (1) Improving transaction rates: The present invention can generate highly personalized sales strategy suggestions for each customer by comprehensively considering the customer conversation content, transaction data, and external influencing factors (such as promotional activity information, the latest loan policies, etc.), thereby effectively improving the transaction rate;

[0029] (2) Reduce reliance on personal experience: The present invention uses a large language model to perform systematic data analysis and extract a set of universal and efficient sales strategies, which can reduce reliance on the personal experience of sales staff, shorten the learning time of new employees, and improve the level of the overall sales team;

[0030] (3) Accurate demand mining: The large language model of the present invention can deeply understand the implicit needs and concerns in customer conversations, and provide more accurate sales suggestions based on external influencing factors to meet customers' personalized needs;

[0031] (4) Scientific decision support: The present invention can provide scientific decision support through comprehensive analysis of historical and current external influencing factors (such as promotional activity information and loan policies), so that sales strategies can respond to market changes and policy adjustments in a timely manner, and improve response speed and accuracy;

[0032] (5) Continuous optimization: The present invention collects data and results from the actual sales process through a system feedback mechanism, and continuously retrains and optimizes the large language model, thereby forming a closed-loop improvement process, making the system more intelligent and accurate over time;

[0033] (6) Overall efficiency improvement: The present invention combines historical data and current information to automatically generate sales suggestions, which can reduce manual intervention and improve the efficiency and effectiveness of overall sales work;

[0034] (7) The present invention applies a large language model to the analysis of 4S store sales conversation data, and can automatically extract important information from the conversation and analyze its impact on the transaction rate;

[0035] (8) The present invention uses data analysis results to provide sales personnel with scientific sales strategy suggestions to improve the deal closing rate; through the feedback mechanism, the large language model and analysis process are continuously optimized, making the system more intelligent and accurate;

[0036] (9) The present invention not only analyzes the content of the conversation, but also takes into account multi-dimensional factors such as customer emotions and concerns, laying the foundation for providing more comprehensive sales suggestions;

[0037] In summary, the present invention not only focuses on sales conversation data, but also takes external factors into consideration in the analysis, so as to ensure the comprehensiveness and timeliness of the suggestions, which has important guiding significance for the work of sales personnel; in addition, the present invention combines modern large language models and scientific data analysis methods, which has significant effects in improving transaction rates, reducing reliance on personal experience, providing accurate demand analysis and scientific decision-making support, and is worthy of promotion and application. [Drawings]

[0038] Figure 1 It is a schematic diagram of the principle of the present invention;

[0039] Figure 2 It is a schematic diagram of the structure of the device of the present invention;

[0040] Figure 3 It is a schematic flow diagram of the method of the present invention;

[0041] Figure 4 It is a structural schematic diagram of the electronic device of the present invention. [Specific implementation method]

[0042] References to "one embodiment" or "some embodiments" etc. described in this specification mean that a particular feature, structure or characteristic described in conjunction with the embodiment is included in one or more embodiments of the present invention. Thus, the phrases "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. appearing in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0043] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described below in conjunction with the accompanying drawings and specific embodiments:

[0044] As attached Figure 1 To Attachment Figure 3 As shown, in some embodiments of the present invention, the present invention provides a sales improvement analysis device based on a large language model, which extracts important factors affecting the transaction rate by deeply analyzing the conversation data between 4S store sales staff and customers and combining the final transaction situation; the analysis device can provide targeted suggestions for sales staff, thereby improving sales efficiency and transaction rate. The device mainly includes the following modules:

[0045] Data collection and preprocessing module: used to collect sales information and conversation data from sales staff and customers, and clean and extract content from these data;

[0046] Large language model module: Use the data preprocessed by the data collection and preprocessing module to perform LLM training and model optimization of the large language model;

[0047] Data analysis module: Use the large language model module to extract features and conduct transaction analysis on the processed data, and combine it with the external influencing factors at the time to deeply explore the key factors that affect customer decisions;

[0048] External factor archiving module: used to introduce and archive external factors at various time points for reference when generating personalized recommendations;

[0049] Suggestion generation module: used to generate personalized sales strategy suggestions for each customer based on comprehensive analysis results and current external influencing factors;

[0050] System feedback module: connected to the large language model module and the suggestion generation module respectively, the system feedback module is used to collect feedback data after the implementation of the sales strategy, as well as model retraining and optimization.

[0051] As a further embodiment, in the data collection and preprocessing module, transaction data and external factors at various time points are also collected, and the external factors include but are not limited to promotional activity information and the latest loan policy. In the external factor archiving module, the external factors include but are not limited to historical promotional activity information, historical loan policy, current promotional activity information and current loan policy.

[0052] In some other embodiments, the present invention proposes a sales improvement analysis method based on a large language model, comprising the following steps:

[0053] (1) Data collection and preprocessing: Collect sales information and conversation data from sales staff and customers, and clean and extract content from these data; at the same time, collect transaction data and external influencing factors at various time points (such as promotional information, the latest loan policies, etc.).

[0054] (2) Large language model: Use the preprocessed data to train and optimize the large language model so that the model can understand and analyze the conversation content and related influencing factors.

[0055] (3) Data analysis: Use a large language model to extract features and perform transaction analysis on the processed data, and combine it with the external influencing factors at the time to deeply explore the key factors that influence customer decisions.

[0056] (4) External factor archiving: Archiving external factors at each time point for reference when generating personalized recommendations.

[0057] (5) Recommendation generation: Generate personalized sales strategy recommendations for each customer based on comprehensive analysis results and current external influencing factors.

[0058] (6) System feedback: Collect feedback data after the implementation of sales strategies for retraining and optimization of models to continuously improve system performance and analysis accuracy.

[0059] By applying the large language model to the analysis of sales conversation data in 4S stores, the present invention can automatically extract important information in the conversation and analyze its impact on the transaction rate; use the data analysis results to provide sales personnel with scientific sales strategy suggestions to improve the transaction rate; continuously optimize the large language model and analysis process through the feedback mechanism, making the system more and more intelligent and accurate; the present invention not only analyzes the content of the conversation, but also takes into account multi-dimensional factors such as customer emotions and concerns, laying the foundation for providing more comprehensive sales suggestions.

[0060] The present invention is further described below with reference to real scene examples:

[0061] Scenario description:

[0062] Liu is a car salesperson working in a 4S store in Shanghai. One day, he received a potential customer, Ms. Zhang, and the two sides talked for about 30 minutes. Ms. Zhang was very interested in new energy vehicles, but she didn't know much about the relevant policies and the installation process of charging piles, so she didn't make a purchase decision on the spot.

[0063] The actual application process is as follows:

[0064] 1. Data Collection and Preprocessing

[0065] Conversation records: Liu's sales conversation was recorded through the store's customer management system, including Ms. Zhang's description of her needs and concerns, such as new energy vehicle policies and questions about charging station installation.

[0066] Transaction data: Ms. Zhang did not complete any transaction on that day, and this was also recorded.

[0067] External factors information archiving: The system also records information such as the new energy vehicle subsidy policy and charging pile installation policy of the day, which has a potential impact on Ms. Zhang's decision.

[0068] 2. Large Language Model

[0069] Model training: The large language model is trained using a large amount of historical data, including previous sales conversations, transaction status, and external factors at the time.

[0070] Model optimization: The model is optimized through continuous feedback, enabling it to understand and analyze new conversation content more accurately.

[0071] 3. Data Analysis

[0072] Data analysis: Feature extraction: The model extracts some key features from the conversation between Liu and Ms. Zhang, such as Ms. Zhang’s lack of understanding of new energy vehicle policies and her concerns about the installation of charging piles.

[0073] Transaction analysis: Combined with data analysis of previous similar cases, the model found that Ms. Zhang paid great attention to the installation convenience of charging piles and the subsidy policy for new energy vehicles.

[0074] 4. External Factors Archive

[0075] Historical information: The new energy vehicle subsidy policy and charging pile installation policy at that time were recorded.

[0076] Latest Policies: The latest current policies are also recorded for easy reference and analysis.

[0077] 5. Suggestion Generation

[0078] Personalized recommendations: The model generated a detailed sales strategy recommendation that focused on addressing Ms. Zhang’s concerns about new energy vehicle policies and charging station installation.

[0079] Recommended models: It is recommended that Liu recommend several new energy vehicles that have received good market response and meet Ms. Zhang’s needs.

[0080] Policy interpretation: A detailed introduction to national and local subsidy policies for new energy vehicles, including vehicle purchase subsidies and tax incentives, including regional policies: such as the Shanghai region's policy of giving away new energy license plate quotas.

[0081] Charging pile installation: focus on explaining the service policy of free installation of charging piles, and explain the specific installation process and subsequent maintenance services, including the materials needed to install the charging piles, and the certificates required by the community property.

[0082] 6. System Feedback

[0083] Feedback data collection: Liu contacted Ms. Zhang again according to the suggestion and introduced the subsidy policy and charging pile installation policy of new energy vehicles in detail, which dispelled her concerns. Finally, Ms. Zhang decided to buy a new energy vehicle.

[0084] Model retraining and optimization: New transaction data is fed back into the system to help further optimize the model, making it more accurate when handling similar situations.

[0085] result:

[0086] Through this process, Liu successfully sold a new energy vehicle to Ms. Zhang, who was also very satisfied with the detailed policy interpretation and charging pile installation plan, which ultimately led to the transaction. Through continuous feedback and optimization, the system gradually improves its intelligence and accuracy, helping more sales staff to increase their transaction rates.

[0087] In some other embodiments of the present invention, an electronic device 400 is disclosed, as shown in the attached figure. Figure 4 As shown, the electronic device includes a processor and a memory, the processor and the memory are connected via a bus, and the memory stores a program that can be run on the processor. When the program is executed by the processor, the electronic device implements the above method. Specifically, the hardware composition of the electronic device includes: one or more processors 401; memory 402; display 403; one or more application programs (not shown); and one or more computer programs 404, and the above devices can be connected via one or more communication buses 405. The one or more computer programs 404 are stored in the above memory 402 and are configured to be executed by the one or more processors 401, and the one or more computer programs 404 include instructions.

[0088] The present invention also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a computer, the method described in the above method embodiment is implemented. Specific beneficial effects can be found in the above method embodiment.

[0089] The present invention also provides a computer program product, which implements the method described in the above method embodiment when executed by a computer. Specific beneficial effects can be found in the above method embodiment.

[0090] Through the description of the above implementation methods, technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0091] Each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0092] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: flash memory, mobile hard disk, read-only memory, random access memory, disk or optical disk and other media that can store program codes.

[0093] The above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit it. Under the idea of ​​the present invention, the technical features in the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other changes in different aspects of the present invention as described above. Although the present invention is described in detail with reference to the above embodiments, it should be understood by those skilled in the art that the technical solutions described in the above embodiments can still be modified, or some of the technical features can be replaced by equivalents. However, these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

[0094] The present invention is not limited to the above-mentioned implementation modes, and any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present invention shall be equivalent replacement modes and shall be included in the protection scope of the present invention.

Claims

1. A sales improvement analysis device based on a large language model, characterized in that: include: Data collection and preprocessing module: used to collect sales information and conversation data from sales staff and customers, and clean and extract content from these data; Large language model module: Use the data preprocessed by the data collection and preprocessing module to perform LLM training and model optimization of the large language model; Data analysis module: Use the large language model module to extract features and conduct transaction analysis on the processed data, and combine it with the external influencing factors at the time to deeply explore the key factors that affect customer decisions; External factor archiving module: used to introduce and archive external factors at various time points for reference when generating personalized recommendations; Recommendation generation module: used to generate personalized sales strategy recommendations for each customer based on comprehensive analysis results and current external influencing factors.

2. The device according to claim 1, characterized in that: The data collection and preprocessing module also collects transaction data and external influencing factors at various time points, which include but are not limited to promotional activity information and the latest loan policies.

3. The device according to claim 1, characterized in that: In the external factor archiving module, the external factors include but are not limited to historical promotional activity information, historical loan policies, current promotional activity information and current loan policies.

4. The device according to claim 1, characterized in that: It also includes a system feedback module, which is respectively connected to the large language model module and the suggestion generation module. The system feedback module is used to collect feedback data after the implementation of the sales strategy, as well as model retraining and optimization.

5. A sales improvement analysis method based on a large language model, characterized in that: The following steps are involved: 1) Data collection and preprocessing: Collect sales information and conversation data from sales staff and customers, and clean and extract content from these data; 2) Large language model training and optimization: Use the data preprocessed in step 1) to perform LLM training and model optimization of the large language model; 3) Data analysis: Use a large language model to extract features and conduct transaction analysis on the processed data, and combine the external influencing factors at the time to deeply explore the key factors that affect customer decisions; 4) External factor archiving: Introduce and archive external factors at various time points for reference when generating personalized recommendations; 5) Recommendation generation: Generate personalized sales strategy recommendations for each customer based on comprehensive analysis results and current external influencing factors.

6. The method according to claim 5, characterized in that: In step 1), transaction data and external influencing factors at various time points are also collected, and the external influencing factors include but are not limited to promotional activity information and the latest loan policies.

7. The method according to claim 5, characterized in that Step 2) comprises the following steps: 201) Model training: The large language model is trained using a large amount of historical data, including but not limited to previous sales conversations, transaction status, and external factors at the time; 202) Model optimization: The model is optimized through continuous feedback, enabling it to understand and analyze new conversation content more accurately.

8. The method according to claim 5, characterized in that After step 5), a system feedback step is also included: collecting feedback data after the implementation of the sales strategy for retraining and optimization of the model to continuously improve system performance and analysis accuracy.

9. A computer-readable storage medium, wherein a program is stored in the computer-readable storage medium, characterized in that: When the program is executed by a processor, the method according to any one of claims 5 to 8 is implemented.

10. An electronic device, characterized in that: The electronic device comprises a memory and a processor, wherein the memory stores a program executable on the processor, and when the program is executed by the processor, the electronic device implements the method as claimed in any one of claims 5 to 8.

Citation Information

Cited By

  • Marketing strategy optimization method based on big language model analysis and evaluation driving

    CN119991168A