Brand monitoring method and system

Abstract

The invention discloses a brand monitoring method and system, and relates to the technical field of brand monitoring, and the method comprises the steps: obtaining customer interaction data and customer behavior data, dividing customer question keyword data, calculating the occurrence probability data, calculating the purchase obstacle probability, and calculating the total residence time of a customer on a payment interface; the method comprises the steps of calculating a corresponding payment page jump rate, establishing a sales volume prediction model based on customer behavior data, obtaining a predicted sales volume, sending an adjustment signal according to the predicted sales volume, and adjusting the monitoring frequency. Through real-time analysis of customer interaction data and behaviors, the purchase demand is accurately captured, sales obstacles are timely found, and the payment experience is optimized; the sales prediction model established based on customer behaviors can predict the sales trend, intelligently adjust the monitoring frequency and strategy, adjust the product stock and optimize the page loading speed, so that the operation efficiency and customer satisfaction are improved, the increase of sales performance is effectively promoted, and brand merchants have more advantages in competition.

Description

Technical Field

[0001] The present invention relates to the technical field of brand monitoring, and in particular to a brand monitoring method and system. Background Art

[0002] With the rapid development of e-commerce, brand merchants are facing increasingly fierce market competition and diversified consumer demands. In order to effectively improve customer satisfaction and sales performance, brand merchants need to deeply understand customers' purchasing behavior and potential needs. However, traditional sales monitoring methods often focus on post-analysis of sales data, lacking real-time monitoring and in-depth analysis of customers' pre-purchase behaviors and problems. This makes it difficult for merchants to capture market changes and subtle differences in customer needs in a timely manner, thus missing the best time to adjust sales strategies and product layout.

[0003] At present, a Chinese invention patent with application number CN202211642327.X discloses a method for monitoring online sales data. The invention includes establishing a data table in a relational database, setting up a company ID and a product URL respectively, and including the following steps: determining whether the product URL of the page is obtained, and ending if the product URL is not obtained, otherwise generating a company ID; traversing all product display pages to the database; after the system obtains all product information, it returns structured information according to an algorithm template preset by the system; cyclically capturing the corresponding fields according to the information returned by the server, and updating the corresponding fields in the data table; calculating the sales data and generating a visual report.

[0004] The above technologies make it difficult to establish a connection between the types of questions customers ask before purchasing and the behavior and sales of customers who did not complete the purchase in order to monitor and adjust. Summary of the invention

[0005] The technical problem solved by the present invention is that it is difficult in the prior art to establish a connection between the types of questions asked by customers before purchasing and the behavior of customers who have not completed the purchase and sales volume for monitoring and adjustment.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: A brand monitoring method comprises the following steps: Step S1, obtaining customer interaction data and customer behavior data, extracting keywords from the customer interaction data, and obtaining customer question keyword data; Step S2, dividing the customer question keyword data and calculating the occurrence probability data, obtaining the total number of customers who asked basic questions or advanced questions and the number of customers who did not complete the purchase after asking basic questions or advanced questions, and calculating the purchase obstacle probability; Step S3, calculating the total stay time of the customer on the payment interface, recording the user's bounce behavior caused by page loading speed or complex payment process, and calculating the corresponding payment page bounce rate; Step S4, establishing a sales forecasting model based on customer behavior data to obtain forecasted sales; Step S5, sending an adjustment signal and adjusting the monitoring frequency according to the predicted sales volume.

[0007] Preferably, the step S1 includes the following sub-steps: Step S101, obtaining customer behavior data, wherein the customer behavior data includes the total number of access records, the duration of each access, and the specific area clicked; Step S102, obtaining customer interaction data, wherein the customer interaction data is question data input by the customer, extracting keywords from the question data, and outputting the data as customer question keyword data, wherein the customer question keyword data includes size data, color data, and material data, and calculating the percentage of the customer question keyword data.

[0008] Preferably, the mathematical expression of the customer keyword proportion data in step S102 is: ; in, For the percentage data, The number of times the customer question keyword data appears. is the total number of access records, Keyword data for customer questions.

[0009] Preferably, step S2 includes the following sub-steps: Step S201, dividing customer question keyword data into basic questions and advanced questions according to a preset basic question keyword database and advanced question keyword database; Step S202, calculating the occurrence probability data of basic questions and advanced questions, the mathematical expression of the occurrence probability data is: ; is the probability data, For basic or advanced questions, Keyword data in basic or advanced questions, For keyword data Total number of times raised; Step S203, obtaining the total number of customers who raised basic questions or advanced questions and the number of customers who did not complete the purchase after raising the basic questions or advanced questions, and calculating the purchase barrier probability, the mathematical expression of the purchase barrier probability is: ; in, is the purchase barrier probability, is the total number of customers who asked basic or advanced questions. The number of customers who asked basic or advanced questions but did not complete a purchase.

[0010] Preferably, step S3 includes the following sub-steps: Step S301, obtaining the number of times the customer visits the payment interface and the corresponding residence time, and calculating the total residence time of the customer on the payment interface. The mathematical expression of the total residence time on the payment interface is: ; in, is the total dwell time on the payment interface, is the number of visits to the payment interface. For the The duration of the payment page during each visit, is a natural number greater than 0.

[0011] Preferably, the step S3 further includes step S302, recording the user jump behavior caused by the page loading speed or the complex payment process, obtaining the number of customers who jump out of the payment page and the total number of customers who visit the payment page, and calculating the corresponding payment page jump rate. The mathematical expression of the payment page jump rate is: ; in, To pay for page bounce rate, The number of customers who bounced from the payment page. The total number of customers who visited the payment page.

[0012] Preferably, the step S4 establishes a sales forecasting model based on customer behavior data, and the mathematical expression of the sales forecasting model is: ; in, To predict sales, The ratio of basic questions to advanced questions, , and is the preset regression coefficient, The preset basic sales volume.

[0013] Preferably, step S5 includes the following sub-steps: Step S501, obtaining predicted sales, if the predicted sales is greater than the preset ideal sales, no adjustment signal is issued, if the predicted sales is less than the preset ideal sales, an adjustment signal is issued, the adjustment signal includes a product adjustment signal and a page loading speed adjustment signal; Step S502: If the predicted sales volume is less than the preset ideal sales volume, the monitoring frequency is adjusted. The mathematical expression of the monitoring frequency is: ; in, To monitor the frequency, As the basic monitoring frequency, is the preset adjustment factor.

[0014] Preferably, the product adjustment signal is: Respectively obtain customer question keyword data corresponding to the page with the highest payment page bounce rate and the page with the lowest payment page bounce rate, output as low-probability purchase product keywords and high-probability purchase product keywords, reduce the stock quantity of products corresponding to the low-probability purchase product keywords by a first percentage, and increase the stock quantity of products corresponding to the high-probability purchase product keywords by a second percentage; The mathematical expression of the page loading speed adjustment signal is: ; in, is the page loading time, is the page data size, Given the network bandwidth, the page load time is lower than the preset industry average.

[0015] A brand monitoring system includes a data collection module, a data processing module, a behavior analysis module, a model building module and a monitoring adjustment module; The data collection module is used to obtain customer interaction data and customer behavior data, extract keywords from the customer interaction data, and obtain customer question keyword data; The data processing module is used to divide the customer question keyword data and calculate the occurrence probability data, obtain the total number of customers who asked basic questions or advanced questions and the number of customers who did not complete the purchase after asking basic questions or advanced questions, and calculate the purchase obstacle probability; The behavior analysis module is used to calculate the total time customers stay on the payment interface, record user bounce behaviors caused by page loading speed or complex payment processes, and calculate the corresponding payment page bounce rate; The model building module is used to build a sales forecasting model based on customer behavior data to obtain forecasted sales; The monitoring adjustment module is used to send an adjustment signal and adjust the monitoring frequency according to the predicted sales volume.

[0016] Beneficial effects of the present invention: The present invention analyzes customer interaction data and behavior in real time, accurately captures purchasing needs, promptly discovers sales obstacles, and optimizes payment experience. At the same time, the sales forecasting model established based on customer behavior can predict sales trends, intelligently adjust monitoring frequency and strategy, adjust product inventory and optimize page loading speed, which not only improves operational efficiency and customer satisfaction, but also effectively promotes the growth of sales performance, giving brand merchants more advantages in competition. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a brand monitoring method provided by an embodiment of the present invention; Figure 2 A basic flow chart of a brand monitoring system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0019] Example 1, reference Figure 1 , provides a brand monitoring method, comprising the following steps: Step S1, obtaining customer interaction data and customer behavior data, extracting keywords from the customer interaction data, and obtaining customer question keyword data.

[0020] Step S2, divide the customer question keyword data and calculate the occurrence probability data, obtain the total number of customers who asked basic questions or advanced questions and the number of customers who did not complete the purchase after asking basic questions or advanced questions, and calculate the purchase obstacle probability.

[0021] Step S3, calculate the total time the customer stays on the payment interface, record the user's bounce behavior caused by page loading speed or complex payment process, and calculate the corresponding payment page bounce rate.

[0022] Step S4, establishing a sales forecasting model based on customer behavior data to obtain forecasted sales.

[0023] Step S5, sending an adjustment signal and adjusting the monitoring frequency according to the predicted sales volume.

[0024] Step S1 includes the following sub-steps: Step S101, obtaining customer behavior data, which includes the total number of access records, the duration of each access, and the specific area clicked.

[0025] Step S101 can obtain the total number of customer visit records, the duration of each visit, and the specific area clicked, so as to gain an in-depth understanding of the customer's browsing habits and interests, and provide data support for subsequent precision marketing and personalized recommendations.

[0026] Step S102, obtaining customer interaction data, which is the question data input by the customer, extracting keywords from the question data and outputting them as customer question keyword data, which includes size data, color data and material data, and calculating the percentage of the customer question keyword data.

[0027] The mathematical expression of the customer keyword proportion data in step S102 is: ; in, For the percentage data, The number of times the customer question keyword data appears. is the total number of access records, Keyword data for customer questions.

[0028] Step S102 can accurately identify the customer's purchasing needs and concerns by extracting keywords from the customer's input questions. Calculating the percentage of these keywords helps companies understand the overall preferences and trends of customer groups and provides a basis for product development and sales strategy adjustments.

[0029] Step S1 can comprehensively understand customers’ purchasing intentions and concerns by carefully collecting and analyzing customers’ behavior and interaction data. This helps companies grasp market trends more accurately, optimize products and services, and improve customer satisfaction and conversion rates.

[0030] Step S2 includes the following sub-steps: Step S201 , classifying customer question keyword data into basic questions and advanced questions according to a preset basic question keyword database and an advanced question keyword database.

[0031] Step S201 can classify customer questions into basic questions and advanced questions by matching customer question keyword data with a preset basic question keyword database and an advanced question keyword database. This step helps enterprises to more clearly understand the types of customer questions and demand levels, and provides a basis for subsequent problem analysis and resolution.

[0032] Step S202, calculating the occurrence probability data of basic questions and advanced questions, the mathematical expression of the occurrence probability data is: ; is the probability data, For basic or advanced questions, Keyword data in basic or advanced questions, For keyword data The total number of times it was raised.

[0033] Step S202 calculates the probability of occurrence of basic and advanced questions, which can reveal the main problems and concerns encountered by customers during the purchase process. These data help companies identify the prevalence and importance of problems, so as to give priority to solving the problems that have the greatest impact on customers and improve customer experience.

[0034] Step S203, obtaining the total number of customers who raised basic questions or advanced questions and the number of customers who did not complete the purchase after raising the basic questions or advanced questions, and calculating the purchase barrier probability. The mathematical expression of the purchase barrier probability is: ; in, is the purchase barrier probability, is the total number of customers who asked basic or advanced questions. The number of customers who asked basic or advanced questions but did not complete a purchase.

[0035] Step S203 calculates the probability of purchase barriers by obtaining the total number of customers who asked basic or advanced questions and the number of customers who did not complete the purchase, and can quantify the degree of influence of different question types on customer purchase decisions. This step helps companies identify key obstacles in the purchase process and develop targeted marketing strategies and service improvement plans to reduce purchase barriers and increase conversion rates.

[0036] Step S2 aims to identify and solve obstacles that customers may encounter during the purchase process by further segmenting and analyzing the keyword data of customer questions. By dividing questions into basic questions and advanced questions and calculating their probability of occurrence, companies can more accurately locate customers' concerns and pain points. At the same time, by calculating the probability of purchase obstacles, companies can evaluate the impact of different question types on customer purchase decisions, thereby developing targeted solutions to improve customer satisfaction and conversion rates.

[0037] Step S3 includes the following sub-steps: Step S301, obtaining the number of times the customer visits the payment interface and the corresponding residence time, and calculating the total residence time of the customer on the payment interface. The mathematical expression of the total residence time on the payment interface is: ; in, is the total dwell time on the payment interface, is the number of visits to the payment interface. For the The duration of the payment page during each visit, is a natural number greater than 0.

[0038] Step S301 obtains the number of times a customer visits the payment interface and the corresponding dwell time, and calculates the total dwell time of the customer on the payment interface, so as to understand the customer's patience and attention allocation during the payment process. This data helps enterprises evaluate the complexity of the payment process and the user's acceptance, and provides a basis for subsequent optimization.

[0039] Step S3 also includes step S302, recording the user jump behavior caused by page loading speed or complex payment process, obtaining the number of customers who jump out of the payment page and the total number of customers who visit the payment page, and calculating the corresponding payment page jump rate. The mathematical expression of the payment page jump rate is: ; in, To pay for page bounce rate, The number of customers who bounced from the payment page, The total number of customers who visited the payment page.

[0040] Step S302 records the user's bounce behavior caused by page loading speed or complex payment process, and calculates the corresponding payment page bounce rate, which can intuitively reflect the problems in the payment process. A high bounce rate may mean that the page loads slowly, the payment steps are cumbersome, or the user encounters other obstacles. This data helps companies quickly locate problems and take targeted measures to improve them, such as optimizing server performance, simplifying payment steps, etc., to reduce the bounce rate and increase the payment success rate.

[0041] Step S3 aims to optimize the user experience and improve the payment success rate by deeply analyzing the customer's behavior data on the payment interface and identifying potential problems in the payment process. By calculating the total time customers stay on the payment interface and the payment page bounce rate, companies can evaluate the efficiency of the payment process and user experience, and then take improvement measures, such as optimizing page loading speed and simplifying the payment process, to reduce the bounce rate and improve customer satisfaction and payment conversion rate.

[0042] Step S4 establishes a sales forecasting model based on customer behavior data. The mathematical expression of the sales forecasting model is: ; in, To predict sales, The ratio of basic questions to advanced questions, , and is the preset regression coefficient, The preset basic sales volume.

[0043] Step S4: By deeply analyzing customer behavior data, the sales forecasting model can more accurately capture market trends and changes in consumer demand, thereby improving the accuracy of sales forecasts. Accurate sales forecasts help companies arrange inventory reasonably and avoid inventory backlogs or out-of-stock situations. This can not only reduce inventory costs, but also ensure the timely supply of products and improve customer satisfaction. The sales forecasting model based on customer behavior data can provide companies with insights into market demand, consumer preferences, etc., thereby supporting companies to formulate more accurate and effective sales strategies and improve sales performance. By using customer behavior data for sales forecasting, companies can respond to market changes more quickly, adjust product structure and marketing strategies, and thus gain an advantage in the fierce market competition. Establishing a sales forecasting model is an important step for companies to transform to data-driven decision-making. Through the model prediction results, companies can make decisions more scientifically and reduce errors and uncertainties in human judgment.

[0044] Step S5 includes the following sub-steps: Step S501, obtain the predicted sales volume. If the predicted sales volume is greater than the preset ideal sales volume, no adjustment signal is issued. If the predicted sales volume is less than the preset ideal sales volume, an adjustment signal is issued. The adjustment signal includes a product adjustment signal and a page loading speed adjustment signal.

[0045] Step S501: By comparing the predicted sales volume with the preset ideal sales volume, the enterprise can judge the effectiveness of the current sales strategy. When the predicted sales volume is lower than the ideal sales volume, timely sending adjustment signals, including product adjustment signals and page loading speed adjustment signals, helps the enterprise to quickly respond to market changes, optimize product portfolio and payment experience, thereby improving sales volume and customer satisfaction. The specific implementation of product adjustment signals, that is, adjusting product inventory according to customer question keyword data corresponding to the high and low bounce rate of the payment page, helps the enterprise to accurately grasp market demand, reduce inventory backlogs, and improve capital turnover.

[0046] Step S502: If the predicted sales volume is less than the preset ideal sales volume, the monitoring frequency is adjusted. The mathematical expression of the monitoring frequency is: ; in, To monitor the frequency, As the basic monitoring frequency, is the preset adjustment factor.

[0047] The product adjustment signal is: The customer question keyword data corresponding to the page with the highest payment page bounce rate and the page with the lowest payment page bounce rate are respectively obtained, and the data are output as low-probability product purchase keywords and high-probability product purchase keywords. The inventory quantity of the products corresponding to the low-probability product purchase keywords is reduced by a first percentage, and the inventory quantity of the products corresponding to the high-probability product purchase keywords is increased by a second percentage.

[0048] The mathematical expression of the page loading speed adjustment signal is: ; in, is the page loading time, is the page data size, Given the network bandwidth, the page load time is lower than the preset industry average.

[0049] Step S502: When the predicted sales volume is lower than the ideal sales volume, adjusting the monitoring frequency can ensure that the enterprise can more closely track market dynamics and changes in customer demand. By increasing the frequency and depth of monitoring, the enterprise can timely discover potential sales obstacles and opportunities, and provide data support for formulating more effective sales strategies. Adjusting the monitoring frequency can also help the enterprise improve its sensitivity to market changes and ensure that it maintains a leading position in a highly competitive market environment.

[0050] Step S5 aims to optimize product structure and improve user experience based on sales forecast results by issuing adjustment signals and adjusting monitoring frequency, thereby increasing sales. This step ensures that enterprises can flexibly adjust strategies according to market demand and maintain competitiveness and profitability.

[0051] By extracting and analyzing keywords in customer interaction data, this method can accurately capture the customer's concerns about the product before purchase. This information helps merchants to deeply understand the customer's purchasing preferences and needs, and provide strong support for product design and optimization. By dividing the customer's question keyword data and calculating the probability of occurrence, this method can identify the proportion of customers who ask basic questions or advanced questions, as well as the proportion of these customers who have not completed the purchase. This helps merchants to promptly discover obstacles in the sales process, such as product information asymmetry, customer doubts that have not been eliminated, etc., so as to take corresponding measures to improve the conversion rate. By calculating the total time customers stay on the payment interface and the payment page bounce rate, this method can reveal potential problems in the payment process. Merchants can optimize the payment experience based on this information, reduce customer bounce rate, and increase payment success rate. The sales forecast model established based on customer behavior data can predict future sales trends and potential demand. This helps merchants adjust inventory and marketing strategies in advance, avoid inventory backlogs and out-of-stock risks, and improve sales performance at the same time. This method can automatically adjust the monitoring frequency and send adjustment signals according to the predicted sales, such as product adjustment signals and page loading speed adjustment signals. This helps merchants to achieve intelligent monitoring and dynamic adjustment, respond quickly to market changes, and improve operational efficiency and sales performance.

[0052] Example 2, reference Figure 2 , provides a brand monitoring system, including a data acquisition module, a data processing module, a behavior analysis module, a model building module and a monitoring adjustment module.

[0053] The data collection module is used to obtain customer interaction data and customer behavior data, extract keywords from customer interaction data, and obtain customer question keyword data.

[0054] The data collection module is the cornerstone of the brand monitoring system. It is responsible for comprehensively and accurately obtaining customer interaction data and customer behavior data, including customer visit records, click behaviors, input questions, etc. By extracting keywords from customer interaction data, this module can generate customer question keyword data, providing a basis for subsequent data processing and analysis. The effect of this module is to ensure the integrity, accuracy and timeliness of the data, and provide support for subsequent analysis and decision-making.

[0055] The data processing module is used to divide the customer question keyword data and calculate the occurrence probability data, obtain the total number of customers who asked basic or advanced questions and the number of customers who did not complete the purchase after asking basic or advanced questions, and calculate the purchase obstacle probability.

[0056] The data processing module performs in-depth processing on the collected keyword data of customer questions. It can not only classify questions into basic questions and advanced questions, but also calculate the probability of occurrence of these questions, revealing the general needs and concerns of customers. At the same time, the module also counts the total number of customers who have raised questions and the number of customers who have not completed their purchases, calculates the probability of purchase obstacles, and helps companies identify key obstacles in the sales process. The effect of this module is to provide valuable data insights, providing a basis for companies to formulate targeted marketing strategies and service improvement plans.

[0057] The behavior analysis module is used to calculate the total time customers stay on the payment interface, record user bounce behaviors caused by page loading speed or complex payment processes, and calculate the corresponding payment page bounce rate.

[0058] The behavior analysis module focuses on analyzing customer behavior on the payment interface. It calculates the total time customers stay on the payment interface, records user bounce behaviors caused by page loading speed or complex payment processes, and calculates the payment page bounce rate. The effect of this module is to reveal potential problems in the payment process, thereby helping companies optimize the payment experience, reduce the bounce rate, and increase the payment success rate.

[0059] The model building module is used to build a sales forecasting model based on customer behavior data to obtain predicted sales.

[0060] The model building module builds a sales forecast model based on customer behavior data. This model can predict future sales trends and provide companies with important market insights. By comparing predicted sales with actual sales, companies can evaluate the effectiveness of sales strategies and adjust strategies in a timely manner to respond to market changes. The effect of this module is to improve the accuracy of sales forecasts and provide support for companies to formulate more accurate sales plans and inventory management strategies.

[0061] The monitoring adjustment module is used to send adjustment signals and adjust the monitoring frequency according to the predicted sales volume.

[0062] The monitoring and adjustment module is the core of the brand monitoring system. It sends adjustment signals based on predicted sales to optimize sales strategies and improve user experience. At the same time, the module also adjusts the monitoring frequency according to market changes to ensure that companies can promptly discover potential sales opportunities and obstacles. The effect of this module is to ensure that companies can quickly respond to market changes, flexibly adjust strategies, and maintain competitiveness and profitability.

[0063] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program codes. Among them, the storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0064] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A brand monitoring method, characterized in that: The steps include: Step S1, obtaining customer interaction data and customer behavior data, extracting keywords from the customer interaction data, and obtaining customer question keyword data; Step S2, dividing the customer question keyword data and calculating the occurrence probability data, obtaining the total number of customers who asked basic questions or advanced questions and the number of customers who did not complete the purchase after asking basic questions or advanced questions, and calculating the purchase obstacle probability; Step S3, calculating the total stay time of the customer on the payment interface, recording the user's bounce behavior caused by page loading speed or complex payment process, and calculating the corresponding payment page bounce rate; Step S4, establishing a sales forecasting model based on customer behavior data to obtain forecasted sales; Step S5, sending an adjustment signal and adjusting the monitoring frequency according to the predicted sales volume.

2. A brand monitoring method as claimed in claim 1, characterized in that: The step S1 includes the following sub-steps: Step S101, obtaining customer behavior data, wherein the customer behavior data includes the total number of access records, the duration of each access, and the specific area clicked; Step S102, obtaining customer interaction data, wherein the customer interaction data is question data input by the customer, extracting keywords from the question data, and outputting the data as customer question keyword data, wherein the customer question keyword data includes size data, color data, and material data, and calculating the percentage of the customer question keyword data.

3. A brand monitoring method as claimed in claim 2, characterized in that: The mathematical expression of the customer keyword proportion data in step S102 is: ； in, For the percentage data, The number of times the customer question keyword data appears. is the total number of access records, Keyword data for customer questions.

4. A brand monitoring method as claimed in claim 3, characterized in that: The step S2 includes the following sub-steps: Step S201, dividing customer question keyword data into basic questions and advanced questions according to a preset basic question keyword database and advanced question keyword database; Step S202, calculating the occurrence probability data of basic questions and advanced questions, the mathematical expression of the occurrence probability data is: ； For the probability data, For basic or advanced questions, Keyword data in basic or advanced questions, For keyword data Total number of times raised; Step S203, obtaining the total number of customers who raised basic questions or advanced questions and the number of customers who did not complete the purchase after raising the basic questions or advanced questions, and calculating the purchase barrier probability, the mathematical expression of the purchase barrier probability is: ； in, is the purchase barrier probability, is the total number of customers who asked basic or advanced questions. The number of customers who asked basic or advanced questions but did not complete a purchase.

5. A brand monitoring method as claimed in claim 4, characterized in that: The step S3 includes the following sub-steps: Step S301, obtaining the number of times the customer visits the payment interface and the corresponding residence time, and calculating the total residence time of the customer on the payment interface. The mathematical expression of the total residence time on the payment interface is: ； in, is the total dwell time on the payment interface, is the number of visits to the payment interface. For the The duration of the payment page during each visit, is a natural number greater than 0.

6. A brand monitoring method as claimed in claim 5, characterized in that: The step S3 also includes step S302, recording the user's bounce behavior caused by the page loading speed or the complex payment process, obtaining the number of customers who bounced out of the payment page and the total number of customers who visited the payment page, and calculating the corresponding payment page bounce rate. The mathematical expression of the payment page bounce rate is: ； in, To pay for page bounce rate, The number of customers who bounced from the payment page, The total number of customers who visited the payment page.

7. A brand monitoring method as claimed in claim 6, characterized in that: The step S4 establishes a sales forecasting model based on the customer behavior data. The mathematical expression of the sales forecasting model is: ； in, To predict sales, The ratio of basic questions to advanced questions, , and is the preset regression coefficient, The preset basic sales volume.

8. A brand monitoring method as claimed in claim 7, characterized in that: The step S5 includes the following sub-steps: Step S501, obtaining predicted sales, if the predicted sales is greater than the preset ideal sales, no adjustment signal is issued, if the predicted sales is less than the preset ideal sales, an adjustment signal is issued, the adjustment signal includes a product adjustment signal and a page loading speed adjustment signal; Step S502: If the predicted sales volume is less than the preset ideal sales volume, the monitoring frequency is adjusted. The mathematical expression of the monitoring frequency is: ； in, To monitor the frequency, As the basic monitoring frequency, is the preset adjustment factor.

9. A brand monitoring method as claimed in claim 8, characterized in that: The product adjustment signal is: Respectively obtain customer question keyword data corresponding to the page with the highest payment page bounce rate and the page with the lowest payment page bounce rate, output as low-probability purchase product keywords and high-probability purchase product keywords, reduce the stock quantity of products corresponding to the low-probability purchase product keywords by a first percentage, and increase the stock quantity of products corresponding to the high-probability purchase product keywords by a second percentage; The mathematical expression of the page loading speed adjustment signal is: ； in, is the page loading time, is the page data size, Given the network bandwidth, the page load time is lower than the preset industry average.

10. A brand monitoring system, applied to a brand monitoring method as claimed in any one of claims 1 to 9, characterized in that: It includes data collection module, data processing module, behavior analysis module, model building module and monitoring and adjustment module; The data collection module is used to obtain customer interaction data and customer behavior data, extract keywords from the customer interaction data, and obtain customer question keyword data; The data processing module is used to divide the customer question keyword data and calculate the occurrence probability data, obtain the total number of customers who asked basic questions or advanced questions and the number of customers who did not complete the purchase after asking basic questions or advanced questions, and calculate the purchase obstacle probability; The behavior analysis module is used to calculate the total time customers stay on the payment interface, record user bounce behaviors caused by page loading speed or complex payment processes, and calculate the corresponding payment page bounce rate; The model building module is used to build a sales forecasting model based on customer behavior data to obtain forecasted sales; The monitoring adjustment module is used to send an adjustment signal and adjust the monitoring frequency according to the predicted sales volume.

Images