Intelligent enterprise sales system and method based on sales skill improvement

Abstract

The invention relates to the technical field of sales management systems, and discloses an enterprise sales intelligent system based on sales skill improvement and a method thereof, and the system is used for integrating a sales skill learning platform and an enterprise sales management platform. The first pushing module is used for taking at least one of item object dimension acquisition data, merchant individual dimension attribute acquisition data and merchant preference dimension acquisition data as input data, and inputting the input data into a first pushing model in advance, and the first pushing model outputs multiple groups of first feature data in one-to-one correspondence; and pushing multiple groups of first sales skill learning strategies in one-to-one correspondence to merchants according to the multiple groups of first feature data. According to the invention, merchants can be helped to quickly master and apply sales skills in a short time.

Description

Technical Field

[0001] The present invention relates to the technical field of sales management systems, and in particular to an enterprise sales intelligence system and method based on sales skill improvement. Background Art

[0002] The learning of sales skills is aimed at improving sales staff's ability to communicate effectively, understand customer needs, provide solutions and facilitate transactions. By participating in training courses specially designed by training schools, unified training and online learning platforms of enterprises, merchants can master core skills from customer development, demand analysis to negotiation skills and relationship management. By mastering core skills and focusing on customers, they can achieve a win-win situation of personal career growth and corporate performance.

[0003] Whether it is training organized by the enterprise or courses designed by training schools, most of them are currently taught to merchants at the same sales skill stage, and such teaching only teaches merchants some theoretical sales skills. Although there are some learning platforms that provide courses covering a variety of different sales skill stages, the courses on these platforms often require merchants to spend more energy to choose and learn. Merchants not only have to spend time to understand, organize and memorize a lot of information, but even if they remember these contents, it is difficult to effectively transform them into actual sales practice in a short period of time. In this way, the time, resources and funds invested by both merchants and enterprises have not been quickly transformed into actual sales results, resulting in a lower return on investment in training. Summary of the invention

[0004] The present invention is intended to provide an enterprise sales intelligence system and method based on sales skill improvement to help merchants quickly master and apply sales skills in a short period of time.

[0005] In order to achieve the above object, the present invention adopts the following technical scheme:

[0006] The first solution: an enterprise sales intelligence system based on sales skill improvement, including: an order recording module, used to record the real-time progress of orders and information on communication with customers as order process information, and use the order process information as project object dimension collection data; a merchant information recording module, used to obtain the merchant's own sales qualification information, and use the merchant's own sales qualification information as merchant individual dimension attribute collection data; a browsing information recording module, used to obtain the merchant's browsing and learning of sales skills information, and use the merchant's browsing and learning of sales skills information as merchant preference dimension collection data; a first output module, used to use at least one of the project object dimension collection data, the merchant individual dimension attribute collection data and the merchant preference dimension collection data as input data, and pre-input the input data into the first push model, the first push model outputs multiple groups of one-to-one corresponding first feature data; according to the multiple groups of first feature data, multiple groups of one-to-one corresponding first sales skill learning strategies are pushed to the merchants respectively.

[0007] The principles and advantages of this solution are: different from the existing technology, first of all, this system provides a new learning model for learning sales skills, integrating the sales skills learning platform and the enterprise sales management platform into an integrated system, which not only makes it convenient for merchants to learn sales skills while managing orders, but also allows them to directly apply the learning of sales skills to sales management, thereby increasing the speed at which merchants apply sales skills.

[0008] Secondly, by using the collected data of three different dimensions as input data, the collected data of the project object dimension, the collected data of the merchant individual dimension attribute and the collected data of the merchant preference dimension, the object data of the project, the subject data of the merchant individual attribute and the merchant preference data are associated, that is, the multi-dimensional data from the subject and the object are combined to improve the accuracy of the output of the first push model, so that each different merchant can get a sales skill strategy that is more suitable for the training required in the combination of personal learning stage and work stage.

[0009] At the same time, merchants learn sales skills at each stage in a dynamic and step-by-step manner. Based on the real-time progress of orders, the merchant's own sales qualifications and browsing information, all of which are changing trends, different sales skill strategies are output according to data from different periods to ensure that merchants can learn different sales skill strategies at different stages of their work, thereby ensuring that merchants can achieve the effect of learning and using them immediately. Not only based on the merchant's work needs, but also based on the merchant's own qualifications and preferences, personalized, precise, and dynamic multiple sets of first-class sales skill learning strategies are pushed to merchants.

[0010] Next, the sales skills learning platform and the enterprise sales management platform are integrated and connected through the first push model. The first push model can directly obtain the project object dimension collection data and merchant individual dimension attribute collection data used in the order record module in the enterprise sales management platform, and can also directly obtain the merchant preference dimension collection data of the sales skills learning platform; not only is it convenient, simple and direct to obtain data, which improves the calculation efficiency of the first push model, but it can also closely combine the two to output multiple sets of first sales skills learning strategies for merchants to learn in real time during the working stage. In this way, a new learning model is provided, whether it is the merchant or the enterprise, the time, resources and funds invested can be quickly converted into actual sales results, and the return on investment of training is high.

[0011] Finally, it is possible to use multi-dimensional data to judge what merchants need to learn from the merchant's own subject or objective project dynamic progress in order to push multiple groups of different types of first sales skill learning strategies. This can not only help sales staff improve their sales skills in a short period of time, but also adaptively provide dynamic and accurate learning materials for each sales staff's different learning stages.

[0012] Preferably, it also includes a first similarity calculation module; the first similarity calculation module is used to use the similarity calculation module to take at least one of the project object dimension collection data, the merchant individual dimension attribute collection data and the merchant preference dimension collection data as input data, and the input data has at most 7 different combination types, and the project object dimension collection data, the merchant individual dimension attribute collection data and the merchant preference dimension collection data are combined as the collection data of the first combination type; randomly select no less than 2 groups and no more than 4 groups of different combination types from the collection data of other combination types except the first combination type as the collection data of the second combination type; calculate the similarity between the collection data of the first combination type and the collection data of the second combination type; take the collection data of the first combination type and the collection data of the second combination type as different input data, and then execute the first output module.

[0013] Beneficial effect: The first similarity calculation module is not only used to randomly select multiple groups of different combination types as the collected data of the second combination type, but also to output the multiple groups of different combination types as the collected data of the second combination type.

[0014] Preferably, it also includes a feedback module and a second output module; the feedback module is executed after the first output module is executed; the feedback module is used to collect the number of clicks of merchants on multiple groups of different first sales skill learning strategies; and obtain the first sales skill learning strategy corresponding to the highest number of clicks; if the collected data corresponding to the first sales skill learning strategy is the collected data of the second combination type, the collected data of the second combination type is used as the collected data of the target combination type; if the collected data corresponding to the first sales skill learning strategy is the collected data of the first combination type, one of the collected data of the second combination type with the highest similarity to the collected data of the first combination type is used as the collected data of the target combination type; the second output module pre-inputs the collected data of the target combination type into the first push model, and the first push model outputs the target feature data; and pushes the target sales skill learning strategy to the merchant according to the target feature data.

[0015] Beneficial effect: By executing the feedback module and the second output module, the output rate can be increased while ensuring the accuracy of the output.

[0016] Preferably, the project object dimension collected data includes order content, order number, order flag, order status, payment type, creation time, payment time, order amount, customer communication frequency, number of customer communications, number of collaborations with customers, and order keyword information in communication with customers.

[0017] Beneficial effects: Collecting this data can not only understand the merchant's order status and order process, but also accurately analyze the sales skills that the merchant currently needs to learn based on the merchant's real-time order status; at the same time, the frequency of communication with customers, the number of communications, the number of cooperation times and the order keyword information in the communication can accurately reflect the communication between the merchant and the customer to accurately understand the status of the order, thereby objectively and accurately determining and pushing the first sales skill learning strategy to the merchant.

[0018] Preferably, the individual attribute collection data includes merchant account opening time, merchant sales data, merchant sales time, merchant completion time for each order and regular reports on merchant orders.

[0019] Beneficial effects: These data help to accurately assess merchants’ performance and the learning stage they are in. They can provide merchants with targeted first-level sales skill learning strategies based on merchants’ evaluation of their own work and the data reflected in their work.

[0020] Preferably, the merchant preference dimension collected data includes the merchant's browsing history of learning sales skills, the time period when the merchant learned sales skills, the browsing duration of the merchant learning sales skills, and the starting point and ending point of the merchant watching the sales skills video segment.

[0021] Beneficial effects: These data help personalize the first sales skill learning strategy and ensure that the learning resources provided are more in line with the actual needs and preferences of merchants, thereby improving learning efficiency and knowledge absorption.

[0022] Preferably, the first sales skills learning strategy includes sales skills learning articles, sales skills learning audios, and sales skills learning videos.

[0023] Preferably, the similarity between the collected data of the first combination type and the collected data of the second combination type is calculated. Specifically,

[0024]

[0025] Among them, C n and D n The same type of character string set between the collected data of the first combination type and the collected data of the second combination type is represented respectively, G match Indicates the number of characters that match between two strings, K t represents half of the number of misplaced characters, n represents the number of characters in the minimum string and is a positive integer, Ω represents the similarity between two strings and is less than or equal to n.

[0026] Beneficial effect: The similarity between the collected data of the first combination type and multiple groups of collected data of the second combination type is calculated through the formula, and other collected data with the highest similarity to the collected data of the first combination type are found. Since the amount of collected data of the first combination type is the largest, its collected data is used as input data. While ensuring accuracy, the calculation time of the first push model is reduced and the output speed of the first push model is improved.

[0027] Preferably, the feedback module is executed after a first time period after the first output module is executed.

[0028] Beneficial effect: The feedback module is executed after the first time period because the number of clicks by merchants on the first sales skill learning strategy needs to be collected first, so that the feedback module is not executed until the number of clicks reaches a certain number to ensure the accuracy of executing the feedback module.

[0029] The second solution: an enterprise sales intelligence method based on sales skill improvement, characterized in that the enterprise sales intelligence system based on sales skill improvement described in the first solution is executed, including: S1, collecting project object dimension collection data, merchant individual dimension attribute collection data and merchant preference dimension collection data; S2, using at least one of the project object dimension collection data, merchant individual dimension attribute collection data and merchant preference dimension collection data as input data, and pre-inputting the input data into a first push model, and the first push model outputs one-to-one corresponding first feature data; S3, pushing multiple groups of first sales skill learning strategies to merchants based on the first feature data.

[0030] Beneficial effects: It is able to judge what merchants need to learn from the merchant's own subject or objective project dynamic progress through multiple dimensions to push multiple groups of first sales skill learning strategies, and then accurately push them based on the customer's preference for a certain group. It can not only help sales staff improve their sales skills in a short period of time, but also adaptively provide dynamic and accurate learning materials for sales staff at different learning stages. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A schematic diagram of an enterprise sales intelligence system based on sales skill improvement according to at least one embodiment;

[0032] Figure 2 A structural block diagram of a device according to at least one embodiment. DETAILED DESCRIPTION

[0033] Further explanation of the nouns of the specific implementation method:

[0034] The enterprise sales management platform is an integrated enterprise management platform that combines customer relationship management (CRM) and sales record functions to optimize sales processes, improve sales efficiency and enhance team collaboration. It can centrally manage customer information, automate daily tasks and track sales progress.

[0035] The sales skills learning platform is a comprehensive training platform for sales personnel from basic to advanced. Through interactive courses, case studies and expert guidance, students can master key sales skills such as customer communication, demand analysis, negotiation strategies and relationship management.

[0036] The reference numerals in the drawings of the specification include:

[0037] Processor 101 , input device 102 , output device 103 , memory 104 , and bus 105 .

[0038] Embodiment 1

[0039] like Figure 1As shown, the enterprise sales intelligence system based on sales skill improvement provided in this embodiment includes an order recording module, a merchant information recording module, a browsing information recording module, a first output module, a first similarity calculation module, a feedback module, and a second output module. The enterprise sales intelligence system based on sales skill improvement provided in this embodiment integrates the sales skill learning platform and the enterprise sales management platform into an integrated design, which enables merchants to learn sales skills while managing orders, so as to achieve the purpose of applying what they have learned and integrating knowledge and practice.

[0040] The order record module is used to record the real-time progress of orders and the information communicated with customers as order process information, and collects order process information as the project object dimension data. In this system, it is possible to monitor the real-time progress of orders and communicate with customers online through this system. The object dimension collected data includes order content, order number, order mark, order status, payment type, creation time, payment time, order amount, customer communication frequency, number of customer communications, number of cooperation with customers, and order keyword information communicated with customers.

[0041] The merchant information recording module is used to obtain the merchant's own sales qualification information and use the merchant's own sales qualification information as the merchant's individual dimension attribute collection data. The individual attribute collection data includes the merchant's account opening time, merchant sales data, merchant sales time, merchant's completion time for each order, and regular reports on merchant orders.

[0042] The browsing information recording module is used to obtain the information of merchants browsing and learning sales skills, and use the information of merchants browsing and learning sales skills as merchant preference dimension collection data. The merchant preference dimension collection data includes the browsing records of merchants learning sales skills, the time period of merchants learning sales skills, the browsing time of merchants learning sales skills, and the starting point and ending point of the merchant watching the sales skills video segment.

[0043] The first similarity calculation module is used to use at least one of the project object dimension collection data, the merchant individual dimension attribute collection data and the merchant preference dimension collection data as input data, wherein the input data has at most 7 different combination types, and combine the project object dimension collection data, the merchant individual dimension attribute collection data and the merchant preference dimension collection data as the collection data of the first combination type. Randomly select no less than 2 groups and no more than 4 groups of different combination types from the collection data of other combination types except the first combination type as the collection data of the second combination type; calculate the similarity between the collection data of the first combination type and the collection data of the second combination type.

[0044] Specifically, the total amount of data collected from the project object dimension, the merchant individual dimension attribute collection data, and the merchant preference dimension collection data is judged with the first data judgment amount. If the total amount of data is greater than the first data judgment amount, 2 groups of different combination types are randomly selected as the collection data of the second combination type. Here, there are 2 groups of collection data of the second combination type. If the total amount of data is less than 1 / 2 of the first data judgment amount, 3 groups of different combination types are randomly selected as the collection data of the second combination type. So far, there are 3 groups of collection data of the second combination type. If the total amount of data is less than 1 / 4 of the first data judgment amount, 4 groups of different combination types are randomly selected as the collection data of the second combination type. So far, there are 4 groups of collection data of the second combination type. In this embodiment, the size of the first data judgment amount can be a conventional adjustment setting in the art. In this way, when the total amount of data is large, the accuracy of the collection data of the first combination type is high at this time, and the combination of random selection is reduced to reduce the amount of data calculation and improve the calculation and output speed. When the total amount of data is small, the accuracy of the collection data of the first combination type is not high at this time, and the combination of random selection is increased to improve the accuracy of the output while ensuring that the calculation and output speed are small.

[0045] The project object dimension collection data, merchant individual dimension attribute collection data and merchant preference dimension collection data are represented in the form of strings as features. Secondly, the pairwise string similarity is used as the similarity between the collection data of the first combination type and the collection data of the second combination type. The formula for calculating the pairwise string similarity using the model is as follows:

[0046]

[0047] In formula (1), C n and D n The same type of character string set between the collected data of the first combination type and the collected data of the second combination type is represented respectively, G match Indicates the number of characters that match between two strings, K t represents half of the number of misplaced characters, n represents the number of characters in the minimum string and is a positive integer, Ω represents the similarity between two strings and is less than or equal to n.

[0048] The formula for the maximum distance between two strings is as follows:

[0049]

[0050] In formula (2), only when C n and D n The two characters are matched when their characters are the same and the distance does not exceed λ. n With D nThe matched characters are compared, and λ / 2 is the number of character conversions Kt. In this embodiment, the similarity between the collected data of the first combination type and the collected data of the second combination type can also be calculated by using other existing similarity algorithms.

[0051] The first output module is used to take at least one of the project object dimension collection data, the merchant individual dimension attribute collection data and the merchant preference dimension collection data as input data, and pre-input the input data into the first push model, and the first push model outputs multiple groups of first feature data; according to the multiple groups of first feature data, multiple groups of first sales skill learning strategies that correspond one to one are pushed to the merchants respectively.

[0052] Specifically, the collected data of the first combination type and the collected data of the second combination type are respectively used as different input data. In the present embodiment, there are 3 types of collected data of the second combination type. That is, 4 different input data are pre-input into the first push model, and the first push model outputs 4 groups of first feature data; according to the 4 groups of first feature data, 4 groups of first sales skill learning strategies corresponding to each other are pushed to the merchants respectively. In the present embodiment, the first push model can be an existing content recommendation algorithm model, or an existing filtering algorithm model based on collaborative recommendation, or a hybrid recommendation algorithm, or a deep learning model such as a convolutional neural network (CNN), a recurrent neural network (RNN) or a deep neural network (DNN). Regardless of whether an existing algorithm model is used, the algorithm model has been pre-created and debugged by conventional technical means.

[0053] A feedback module is used to collect the number of clicks by merchants on multiple groups of different first sales skill learning strategies; and obtain the first sales skill learning strategy corresponding to the highest number of clicks; if the collected data corresponding to the first sales skill learning strategy is the collected data of the second combination type, the collected data of the second combination type is used as the collected data of the target combination type; if the collected data corresponding to the first sales skill learning strategy is the collected data of the first combination type, one of the collected data of the second combination type that has the highest similarity to the collected data of the first combination type is used as the collected data of the target combination type.

[0054] Specifically, after the first time period after the execution of the first output module, the feedback module is executed. The feedback module collects the data volume of the merchant clicking on the first sales skill learning strategy recommended by the first push model, and finds one of the first sales skill learning strategies with the highest click volume. The first sales skill learning strategy corresponds to a set of first feature data, and the first feature data is the target feature data. It is determined whether the target feature data corresponds to the collected data of the first combination type or one of the collected data of the second combination type. If the highest click volume is one of the collected data of the second combination type, the collected data of the second combination type is used as the collected data of the target combination type. If the highest click volume is the collected data of the first combination type, the collected data of the second combination type with the highest similarity between the collected data of the first combination type and the collected data of the second combination type is used as the collected data of the target combination type.

[0055] The second output module pre-inputs the collected data of the target combination type into the first push model, and the first push model outputs the target feature data; and pushes the target sales skill learning strategy to the merchant according to the target feature data. In this embodiment, the first time period is 72 hours. The first feature data and the target feature data are both web links, and the web links also correspond to different sales skill learning strategies. The output first sales skill learning strategy and target sales skill learning strategy both include sales skill learning articles, sales skill learning audios, and sales skill learning videos.

[0056] The second output module pre-inputs the collected data of the target combination type into the first push model, and the first push model outputs the target feature data; and pushes the target sales skill learning strategy to the merchant according to the target feature data.

[0057] For example, the project object dimension collection data, merchant individual dimension attribute collection data and merchant preference dimension collection data are set as three different data A, B and C respectively. Among them, ABC data is the collection data of the first combination type, and 3 groups of different combination types are randomly selected as the collection data of the second combination type. It is planned to extract 3 groups of different combination types such as A, B and AC as the collection data of the second combination type. Then ABC data, A data, B data and AC data are used as 4 types of input data respectively, and the input data are pre-input into the first push model. The first push model outputs 4 groups of different first feature data, and pushes 4 groups of first sales skill learning strategies corresponding to each other to the merchants according to the 4 groups of different first feature data. After the first time period, the group of first sales skill learning strategies with the highest click volume is obtained, and the group of first sales skill learning strategies with the highest click volume is proposed as the input data represented by AC data, then the collection data of the target combination type is the AC data in the collection data of the second combination type. The second output module takes the AC data as input data and then inputs it into the first push model to push the target sales skill learning strategy to the merchant. If the first sales skill learning strategy with the highest number of clicks is the first combination type of collected data represented by ABC data, then the second combination type of collected data with the highest similarity to ABC data will be used as input data, and it is assumed that ABC data has the highest similarity to AC data, that is, AC data will be used as the target combination type of collected data.

[0058] In one formulation and implementation of sales skill learning strategies, the actions of sales champions in this industry at each stage of each order are obtained through the existing big data algorithm model, and the actions are characterized by the successful experience of each sales champion's sales strategy at different stages of each order. The sales experience of each sales champion at each different stage is summarized and key information is extracted using an AI model, and the key information is used as different sales skill learning strategies, and one-to-one correspondence is made with each order and each different stage of each order. The experience of the sales champion not only comes from the judgment of different types of orders and different stages of orders, but also includes the psychological characteristics of customers at different stages and the experience summarized before. At the same time, labels can be attached according to the behavior of different users to represent the implementation of accurate sales skill learning strategies for users of this behavior. Finally, the summarized successful experience of the sales champion is covered in the sales skill learning strategies that need to be learned in this industry, so as to transform practical experience into sales skill learning strategies and achieve the goal of unity of knowledge and action.

[0059] The beneficial effects of this embodiment

[0060] Compared with the prior art, this embodiment provides a new enterprise sales intelligent management system, which integrates the sales skills learning platform and the enterprise sales management platform into a new comprehensive management and learning model. Not only can it dynamically output multiple groups of different first sales skills learning strategies through multiple dimensions, including subjective or objective data, but the sales skills learning strategy can also help merchants improve their sales skills in a short period of time in combination with the current work progress, so as to achieve the purpose of learning and applying, and integrating knowledge and practice. At the same time, based on the feedback from merchants on the first sales skills learning strategy, multiple dimensional data is optimized and the amount of data is reduced, which not only reduces the calculation time of the first push model and improves the output speed of the first push model, but also ensures the accuracy of the output target sales skills learning strategy.

[0061] First, generally, the enterprise sales management platform stores the project object dimension collection data of the order record module and the merchant individual dimension attribute collection data of the merchant information record module; the sales skill learning platform stores the merchant preference dimension collection data of the browsing information record module. The enterprise sales intelligence system of this embodiment can directly obtain data of different dimensions on the two platforms, which is not only simple and convenient, but also has a cleaner data volume and improves the data capture efficiency of the entire enterprise sales intelligence system.

[0062] Secondly, the data collected in the project object dimension is a real-time dynamic objective data, including the real-time progress of orders and customer communication information. The data collected in the merchant individual dimension attribute is about the merchant's own sales qualification information, which is the merchant's own data of the main attribute, and is also a stage data. The data collected in the merchant preference dimension is a subjective data of the merchant, targeting the merchant's subjective preferences and needs. By using the characteristic data of three different dimensions as input data, it is possible to accurately and dynamically push to the merchant the multiple sets of first sales skill learning strategies it needs. The first sales skill learning strategy is not only closely related to the merchant's order process, its own qualifications and preferences, but also achieves the effect of combining the three dimensions of subjectivity, objectivity and self-subjectivity.

[0063] Next, through the first similarity calculation module, firstly, according to the size of the number, multiple groups of second combination type of collected data of different combination types are randomly captured, and then different first sales skill learning strategies with the second combination type of collected data as multiple groups of one-to-one corresponding input data are pushed to the merchants one by one. The accuracy of the data is determined according to the number of clicks on the first sales skill learning strategy by customers after the first time period. Generally, when the data is relatively clean and the noise is less, the number of clicks on the first sales skill learning strategy corresponding to the first combination type of collected data should be the highest. When it is confirmed that the number of clicks on the first sales skill learning strategy corresponding to the first combination type of collected data is the highest, it reflects that the data is relatively clean and the noise is less. Random selection is also to check whether the data is clean and whether the output is true and accurate. When it is verified that the three groups of data are relatively clean at this time, the collected data of one of the second combination types with the highest similarity to the collected data of the first combination type is used as the collected data of the target combination type. This is because the amount of collected data of the first combination type is too large, and the amount of data is relatively clean at this time. Therefore, reducing the amount of input data can not only ensure the accuracy of the output, but also reduce the amount of calculation and improve the output speed of the first push model. Therefore, it is a better choice to use the collected data of the second combination type that has the highest similarity with the collected data of the first combination type as the collected data of the target combination type.

[0064] On the other hand, when the merchant's highest click volume is the first sales skill strategy corresponding to the collected data of one of the second combination types, it is considered that the data noise of the collected data of the first combination type is large and the data is not clean. At this time, the collected data of one of the second combination types with the highest click volume is used as the target data, which not only reduces the impact of data noise on the output accuracy, which is equivalent to improving the output accuracy, but also reduces the amount of data calculation and improves the output speed of the first push model.

[0065] Finally, by executing the first output module, then executing the feedback module, and then executing the second output module, a feedback mechanism based on logical judgment is formed, which can achieve the best of the best, reduce data volume calculation, and improve the output efficiency of the first push model. That is, according to the combination of three dimensions, multiple groups of different first sales skill learning strategies are pushed, and then the merchants’ click volume on multiple groups of sales skill learning strategies is fed back, the collected data of the first combination type or the collected data of the second combination type is optimized, and then a second push is performed to achieve the purpose of accurate, dynamic feedback, and rapidity.

[0066] Embodiment 2

[0067] Different from the first embodiment, the enterprise sales intelligence system based on sales skill improvement also includes a sales strategy adjustment module. The sales strategy adjustment module is used to add sales strategies, search for sales strategies, and delete sales strategies. The sales strategy details cover the sales strategy itself and the specific information about the products sold. During the operation, the merchants, the sales strategies formulated, and the product conditions can be effectively screened, accurate searches can be carried out, and reset operations can be performed.

[0068] The order record module is also used to completely record all the product information of the merchant, and the product specifications and attributes of the products selected by the merchant can be viewed.

[0069] The beneficial effects of this embodiment

[0070] Different from the prior art, this embodiment and embodiment 1 combine the entire enterprise sales management system and the learning platform for improving sales skills, which can not only meet the needs of processing the sales order module on the enterprise sales management system, but also adjust the sales strategy. The sales strategy can be continuously updated by adjusting the sales strategy to adapt to the latest learning environment, and the sales strategy can be used to improve the sales ability of merchants.

[0071] Embodiment 3

[0072] Different from the first embodiment, the enterprise sales intelligence system based on sales skills improvement also includes a merchant preference dimension collection data optimization module; the merchant preference dimension collection data optimization module is used to judge the merchant's learning progress based on the starting point and ending point of the merchant watching the sales skills video segment, and use the merchant's learning progress represented by the starting point and ending point of the merchant watching the sales skills video segment as a separate type of merchant preference dimension collection data.

[0073] Specifically, determine the number of times and duration of the merchant watching the sales skills video in the second time period; if the number of views and duration are not less than the first target number of views and the first target viewing duration, respectively, parse the video content between the starting point and the ending point of the last sales skills video segment watched, and analyze the merchant's learning progress based on the video content; use the learning progress as the only data in the data collected from the merchant preference dimension.

[0074] First, the video content is extracted by using the existing video text extraction technology, and the video content is input into the existing text processing algorithm model to obtain the output result, which is the refined content of the video content, and the refined content is used as the only data in the merchant preference dimension collection data. In this embodiment, the first target viewing number is 2 times, the first target viewing time is 10 minutes, and the second time period is 48 hours.

[0075] The beneficial effects of this embodiment

[0076] Unlike the prior art, the data collected in the merchant preference dimension has many types of data, a large amount of data, and is easily mixed with noise, and is not all clean data. Reading sales skill learning articles and audio in the merchant preference dimension collected data cannot fully indicate the merchant's preferences in the second time period. When a merchant browses a sales skill video, if the number of views is large and the browsing time is long, it indicates the merchant's preferences in the second time period. The refined content of the sales skill video is processed to form the only data in the merchant preference dimension collected data. That is, in this embodiment, the merchant preference dimension collected data is refined content. In this way, not only the data samples of the merchant preference dimension collected data are cleaner, the sample noise is greatly reduced, and the data samples of the merchant preference dimension collected data are optimized. At the same time, the amount of calculation of the merchant preference dimension collected data as input data in the first push model calculation is reduced to improve the calculation speed of the first push model.

[0077] Embodiment 4

[0078] The present embodiment provides an enterprise sales intelligence method based on sales skill improvement, which is used to execute the enterprise sales intelligence system based on sales skill improvement described in Example 1, including: S1, collecting project object dimension collection data, merchant individual dimension attribute collection data and merchant preference dimension collection data; S2, using at least one of the project object dimension collection data, merchant individual dimension attribute collection data and merchant preference dimension collection data as training data, and pre-inputting the training data into a first push model, and the first push model outputs multiple groups of first feature data; S3, pushing a first sales skill learning strategy group to the merchant according to the first feature data.

[0079] Embodiment 5

[0080] like Figure 2 The device may include: one or more processors 101, one or more input devices 102, one or more output devices 103 and a memory 104, wherein the processors 101, the input devices 102, the output devices 103 and the memory 104 are interconnected via a bus 105. The memory 104 is used to store a computer program, wherein the computer program includes program instructions, and the processor 101 is configured to call the program instructions to execute the method steps described in the above embodiments.

[0081] It should be understood that in the embodiment of the present invention, the processor 101 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0082] The input device 102 may include a keyboard, etc., and the output device 103 may include a display (LCD, etc.), a speaker, etc.

[0083] The memory 104 may include a read-only memory and a random access memory, and provides instructions and data to the processor 101. A portion of the memory 104 may also include a non-volatile random access memory. For example, the memory 104 may also store information on the device type.

[0084] In a specific implementation, the processor 101, input device 102, and output device 103 described in the embodiment of the present invention can execute the implementation method described in the embodiment of the product recommendation method provided by the embodiment of the present invention, which will not be repeated here.

[0085] It should be noted that for a more specific description of the working process of the device, please refer to the aforementioned method embodiment part, which will not be repeated here.

[0086] Embodiment 6

[0087] This embodiment provides a computer-readable storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, an enterprise sales intelligence method based on sales skill improvement is implemented.

[0088] The computer-readable storage medium may be the internal storage unit of the device described in the foregoing embodiments, such as a hard disk or memory of the system. The computer-readable storage medium may also be an external storage device of the system, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the system. The computer-readable storage medium is used to store other programs and data required by the computer program. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0089] The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0090] In this specification, a lot of specific details are described. However, it is understood that embodiments of the present invention can be put into practice without these specific details. In some examples, known methods, systems and techniques are not shown in detail, so as not to obscure the understanding of this specification. In the description of this specification, the description of reference terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" etc. means that the specific features, methods, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of this specification.

[0091] In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, systems, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples without contradiction.

[0092] Finally, 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. 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 invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. Enterprise sales intelligence system based on sales skills improvement, characterized by: A system that integrates sales skills learning platform and enterprise sales management platform, including: An order recording module is used to record the real-time progress of orders and information communicated with customers as order process information, and collect data using the order process information as the project object dimension; The merchant information recording module is used to obtain the merchant's own sales qualification information and collect the merchant's own sales qualification information as the merchant's individual dimension attribute data; A browsing information recording module is used to obtain information about merchants browsing and learning sales skills, and collect data using the information about merchants browsing and learning sales skills as merchant preference dimensions; The first output module is used to take at least one of the project object dimension collection data, the merchant individual dimension attribute collection data and the merchant preference dimension collection data as input data, and pre-input the input data into the first push model, and the first push model outputs multiple groups of one-to-one corresponding first feature data; according to the multiple groups of first feature data, multiple groups of one-to-one corresponding first sales skill learning strategies are pushed to the merchants respectively.

2. The enterprise sales intelligence system based on sales skill improvement according to claim 1 is characterized in that: Also includes a first similarity calculation module; A first similarity calculation module is used to use the similarity calculation module to take at least one of the project object dimension collection data, the merchant individual dimension attribute collection data and the merchant preference dimension collection data as input data, wherein the input data has at most 7 different combination types, and combine the project object dimension collection data, the merchant individual dimension attribute collection data and the merchant preference dimension collection data as collection data of a first combination type; Randomly select no less than 2 groups and no more than 4 groups of different combination types from the collected data of other combination types excluding the first combination type as the collected data of the second combination type; calculate the similarity between the collected data of the first combination type and the collected data of the second combination type; The collected data of the first combination type and the collected data of the second combination type are respectively used as different input data, and then the first output module is executed.

3. The enterprise sales intelligence system based on sales skill improvement according to claim 2 is characterized in that: Also includes a feedback module and a second output module; Execute the feedback module after executing the first output module; A feedback module, used to collect merchants' clicks on multiple groups of different first sales skill learning strategies; And obtain the first sales skill learning strategy corresponding to the highest number of clicks; If the collected data corresponding to the first sales skill learning strategy is collected data of the second combination type, using the collected data of the second combination type as collected data of the target combination type; If the collected data corresponding to the first sales skill learning strategy is collected data of the first combination type, one of the collected data of the second combination type having the highest similarity to the collected data of the first combination type is used as collected data of the target combination type; A second output module pre-inputs the collected data of the target combination type into the first push model, and the first push model outputs the target feature data; A target sales skill learning strategy is pushed to the merchant based on the target feature data.

4. The enterprise sales intelligence system based on sales skill improvement according to claim 1 is characterized in that: The data collected in the project object dimension include order content, order number, order mark, order status, payment type, creation time, payment time, order amount, customer communication frequency, number of customer communications, number of collaborations with customers, and order keyword information in communication with customers.

5. The enterprise sales intelligence system based on sales skill improvement according to claim 1 is characterized in that: The individual attribute collection data includes merchant account opening time, merchant sales data, merchant sales time, merchant completion time for each order and regular reports on merchant orders.

6. The enterprise sales intelligence system based on sales skill improvement according to claim 1 is characterized in that: The data collected in the merchant preference dimension include the merchant's browsing history of learning sales skills, the time period when the merchant learned sales skills, the browsing time of the merchant learning sales skills, and the starting point and ending point of the merchant watching the sales skills video segment.

7. The enterprise sales intelligence system based on sales skill improvement according to claim 1 is characterized in that: The first sales skills learning strategy includes sales skills learning articles, sales skills learning audios, and sales skills learning videos.

8. The enterprise sales intelligence system based on sales skill improvement according to claim 2 is characterized in that: Calculate the similarity between the collected data of the first combination type and the collected data of the second combination type. Specifically, Among them, C n and D n The same type of character string set between the collected data of the first combination type and the collected data of the second combination type is represented respectively, G match Indicates the number of characters that match between two strings, K t represents half of the number of misplaced characters, n represents the number of characters in the minimum string and is a positive integer, Ω represents the similarity between any two strings and is less than or equal to n.

9. The enterprise sales intelligence system based on sales skill improvement according to claim 3 is characterized in that: The feedback module is executed after a first time period after the first output module is executed.

10. An enterprise sales intelligence method based on sales skill improvement is characterized by: The enterprise sales intelligence system based on sales skill improvement implemented in any one of claims 1 to 9 comprises: S1, collects data from the project object dimension, merchant individual dimension attribute dimension, and merchant preference dimension; S2, using at least one of the project object dimension collected data, the merchant individual dimension attribute collected data and the merchant preference dimension collected data as training data, and pre-inputting the training data into the first push model, so that the first push model outputs multiple groups of first feature data; S3: Pushing a first sales skill learning strategy group to the merchant based on the first feature data.

Images