Personalized sales-based 4S store customer intelligent commodity recommendation system

Abstract

The invention relates to the technical field of personalized sales, in particular to a personalized sales-based 4S store customer intelligent commodity recommendation system, which comprises a data acquisition module, and is characterized in that the data acquisition module is used for acquiring information of a customer through an identity information acquisition device installed at a store entrance; and collecting gender and age appearance information of the customer entering the store and posture information of height and weight of the customer entering the store. According to the method, accurate recommendation of 4S store intelligent commodities is realized through search records during online browsing of clients and access records during offline store access, the working efficiency of salesmen is effectively improved, an applicable personalized sales scheme is formulated according to different purchase intentions of the clients, and client loss caused by blind recommendation is avoided; and meanwhile, by accurately judging the intention to purchase products, the consumption habits and the value types of the clients during online browsing and combining offline access records, targeted personalized marketing is formulated, so that the order forming rate of the 4S store can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of personalized sales, and particularly to an intelligent commodity recommendation system for 4S store customers based on personalized sales. Background Art

[0002] With the development of the Internet, the sales model of 4S stores has also changed greatly. The general sales model can no longer meet the development needs of offline 4S stores. The traditional sales method cannot achieve personalized recommendations according to the actual needs of customers, resulting in an increasing customer churn rate in 4S stores year by year.

[0003] Therefore, an intelligent commodity recommendation system for 4S store customers based on personalized sales is proposed, which takes the customer's intended purchase as the center and improves the recommendation accuracy of 4S store commodities through personalized customization, providing strong sales support for 4S store salespersons. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides an intelligent commodity recommendation system for 4S store customers based on personalized sales, which solves the technical problem of high customer churn rate caused by poor recommendation accuracy of traditional 4S store commodities.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: An intelligent commodity recommendation system for 4S store customers based on personalized sales, including a data acquisition module, which is used to collect the gender, age and appearance information of customers entering the store and the body information of the height and weight of customers entering the store through an identity information acquisition device installed at the store entrance;

[0006] A search module, which searches for customer's intended purchase clues according to the appearance information and body information of the user collected by the data acquisition module, including:

[0007] A lead acquisition unit, which is used to obtain the appearance information of the gender and age of the customer's face, and obtain the physical information of the customer's height and weight of the figure, and classify and summarize the appearance and physical information of the customer; a lead comparison module, which queries the browsing records of the PC side and APP side of the customer's online browsing and the offline 4S store visit records according to the customer information obtained in the lead acquisition unit, and compares the data of the online browsing records and the offline visit records; a lead mining module, which is used to mine the comparison results of the data obtained in the lead comparison module, including: a search times recognition unit, which is used to recognize the number of times of keywords for searching target products on the PC side and APP side of the customer's online browsing, so as to accurately predict the products that consumers intend to purchase; a click times recognition unit, which is used to recognize the click times of the browsing records of the PC side and APP side of the customer's online browsing through the search results of the target product keywords, so as to accurately judge the consumption habits of the customer; a residence duration recognition unit, which is used to recognize the browsing residence duration of the products that meet the intention to purchase on the PC side and APP side of the customer's online browsing, so as to mine the user value type of the customer; an access data summary unit, which is used to summarize the intention to purchase of the products in the offline 4S store visit records, so as to quickly obtain the purchase demand of the customer during the offline visit;

[0008] A generation module, which is used to formulate the prediction results of the lead acquisition unit and the mining results in the lead mining module into a personalized sales plan;

[0009] A sending module, according to the prediction results of the lead acquisition unit and the mining results of the lead mining module, formulates a personalized sales plan through the generation module and sends it to the sales terminal of the 4S store.

[0010] Preferably, the lead acquisition unit recognizes the customer's facial appearance information, analyzes the customer's gender and age, and makes a preliminary prediction of the products intended to be purchased according to the customer's gender and age; it is also used to recognize the customer's physical information, analyze the customer's height and weight, and make a secondary prediction of the products intended to be purchased according to the customer's height and weight; and sorts the customer's intention to purchase products, and finally sends it to the sales terminal of the reception staff in the 4S store for the sales staff in the 4S store to timely obtain the customer's purchase intention;

[0011] The search times recognition unit is used to calculate the keyword density W of the number of times of keywords for searching target products on the PC side and APP side of the customer's online browsing, according to the calculation formula:

[0012]

[0013] Among them, T is the total length of keyword characters, T = the number of keyword occurrences * keyword length; G is the total length of the search text content; combined with the keyword density W, accurate prediction of the products that consumers intend to purchase is achieved; and the prediction results of the products that consumers intend to purchase are sent to the sales terminals of the receptionists in the 4S store for the sales staff in the 4S store to make precise recommendations on the products that customers intend to purchase.

[0014] Preferably, the click - count recognition unit is used to analyze the click counts of the search results of the target products browsed by the customer online on the PC side and the APP side, and record them as Q. Combined with the click - count purchase ratio P of the customer, the consumption habits of the customer are judged.

[0015] When the click count Q is within the interval of the click - count purchase ratio P[0, 200], the search results on the PC side and the APP side meet the search requirements, and this customer is a decisive consumer; if the click count Q is within the interval of the click - count purchase ratio P[200, +∞), at this time, the search results on the PC side and the APP side cannot meet the search requirements and it is difficult to match and find the desired item, and this user is a cautious consumer; and the judgment results of the customer's consumption habits are sent to the sales terminals of the receptionists in the S store for the sales staff in the 4S store to timely obtain the consumption habits of the customers and formulate targeted sales plans according to the consumption habits of the customers.

[0016] Preferably, the residence - duration recognition unit is used to analyze the browsing duration and the sliding - interface frequency of the customer's online browsing on the PC side and the APP side when clicking on the target product, and mine the user value type according to the RF model; record the customer's most recent browsing duration as R; record the frequency of sliding the interface in the customer's most recent browsing record as F; if is a value customer; if is a developing customer; if is a maintaining customer; if is a retaining customer; where is the average duration of the customer's browsing interface; is the average frequency of the customer's sliding the interface; according to the customer's value type, the mining results are sent to the sales terminals of the receptionists in the 4S store for the sales staff in the S store to quickly locate the customer's value type and formulate an adaptable sales plan.

[0017] Preferably, the access - data summarizing unit is used to summarize the access - record information of the customer's recent offline visits to the 4S store, and send the summary results to the sales terminals of the receptionists in the 4S store for the sales staff in the 4S store to quickly obtain the purchase needs of the customer during offline visits and formulate targeted sales plans.

[0018] Preferably, when the customer value type is to retain customers, according to the actual needs of the customers, the actual products of the offline 4S store are combined with the customers' needs to determine whether the offline products of the 4S store meet the customers' needs. When they meet the customers' needs, precise product recommendations are made to reduce customer loss caused by search errors.

[0019] By means of the above technical solution, the present invention provides a 4S store customer intelligent commodity recommendation system based on personalized sales, which at least has the following beneficial effects:

[0020] The present invention realizes precise recommendation of 4S store intelligent commodities through the search records during customers' online browsing and the access records during their offline store visits, effectively improving the work efficiency of salespersons. According to different purchase intentions of customers, personalized sales plans are formulated to avoid customer loss caused by blind recommendations. At the same time, through precise judgment of the customers' intended purchase products, consumption habits and value types during online browsing, combined with offline access records, targeted personalized marketing is formulated, improving the accuracy of 4S store salespersons' acquisition of customer demand information, providing strong sales support for 4S store salespersons, and effectively improving the order success rate of 4S stores. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present application, and constitute a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation of the present application. In the drawings:

[0022] Figure 1 is a system flow block diagram of the present invention;

[0023] Figure 2 is a structural block diagram of the search module of the present invention;

[0024] Figure 3 is a structural block diagram of the clue mining module of the present invention;

[0025] Figure 4 is a sector statistical chart of the proportion of the number of single purchase clicks in the big data of the present invention.

[0026] In the figure: 1, data acquisition module; 2, search module; 21, clue acquisition unit; 22, clue comparison module; 23, clue mining module; 231, search times identification unit; 232, click times identification unit; 233, residence duration identification unit; 234, access data summary unit; 3, generation module; 4, sending module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0028] Embodiment 1

[0029] Please refer to Figure 1 - Figure 4 , a 4S store customer intelligent commodity recommendation system based on personalized sales, including a data collection module 1. The data collection module 1 is used to identify the customer identity information according to the face of the customer in the store through the identity information collection device installed at the store entrance. The identity information collection device is a face recognition system and a visual recognition device. The height and weight information of the customer in the store is collected through the recognition device, and the gender, age, appearance information of the customer entering the store and the body information of the height and weight of the customer entering the store are collected, and the gender, age, height and weight information of the customer are collected.

[0030] A search module 2, which searches for customer intention purchase clues according to the user appearance information and body information collected by the data collection module 1, including: a clue acquisition unit 21, which is used to acquire the appearance information of the gender and age of the customer's face, and acquire the body information of the height and weight of the customer's body, and classify and summarize the appearance and body information of the customer, and summarize the age, gender and height and weight of the customer. Specifically, male consumers who come to the 4S store pay more attention to intelligent commodities such as power and handling in the car, and female consumers pay more attention to the appearance and safety performance of intelligent commodities; middle-aged consumers prefer mature and steady intelligent commodities, and young consumers prefer fashionable and novel intelligent commodities; height and weight are initially judged accordingly according to the comfort of intelligent commodities that consumers focus on in big data and the identity information of different customers in the store.

[0031] A clue comparison module 22, which queries the browsing records of the PC side and APP side and the offline 4S store visit records browsed by the customer online according to the customer information obtained in the clue acquisition unit 21, compares the data of the online browsing records and the offline visit records, and combines the identity information of the customer in the store to obtain the online browsing record information of the customer.

[0032] The lead mining module 23 is used to mine the comparison results of the data obtained in the lead comparison module 22, including: a search times recognition unit 231, which is used to recognize the number of times of keywords for searching target products on the PC side and APP side of the customer's online browsing, so as to accurately predict the products that consumers intend to purchase; a click times recognition unit 232, which is used to recognize the click times of the products that meet the intention to purchase through the search results of the target product keywords in the browsing records on the PC side and APP side of the customer's online browsing, so as to accurately judge the consumption habits of the customer; a residence duration recognition unit 233, which is used to recognize the browsing residence duration of the products that meet the intention to purchase on the PC side and APP side of the customer's online browsing, so as to mine the customer user value type; an access data summary unit 234, which is used to summarize the intention to purchase products in the offline 4S store visit records, so as to quickly obtain the purchase needs of the customer during the offline visit; by accurately judging the products intended to be purchased, consumption habits and value types of the customer during online browsing, combined with the offline visit records, targeted personalized marketing is formulated, which improves the accuracy of the offline 4S store salesperson's acquisition of customer demand information, provides strong sales support for the 4S store salesperson, and can effectively improve the order success rate of the 4S store;

[0033] The generation module 3 is used to formulate the prediction results of the lead acquisition unit 21 and the mining results in the lead mining module 23 into a personalized sales plan;

[0034] The sending module 4, according to the prediction results of the lead acquisition unit 21 and the mining results of the lead mining module 23, formulates a personalized sales plan through the generation module 3 and sends it to the sales terminal of the 4S store;

[0035] As a preferred technical solution of this embodiment, the lead acquisition unit 21 recognizes the customer's facial appearance information, analyzes the customer's gender and age, and makes a preliminary prediction of the products intended to be purchased according to the customer's gender and age. The specific prediction results are based on the age and gender of the in-store customers to make a preliminary prediction of the products intended to be purchased. Male consumers pay more attention to intelligent products related to power and handling in cars, while female consumers pay more attention to the appearance and safety performance of intelligent products; middle-aged consumers prefer mature and steady intelligent products, while young consumers prefer fashionable and novel intelligent products;

[0036] It is also used to identify the physical information of customers, analyze the height and weight of customers, and based on the identity information of different customers arriving at the store, combined with the height and weight of customers, conduct a secondary prediction of the intended purchase products. The height and weight are based on the comfort of intelligent products that consumers focus on in big data; and sort the intentions of customers' intended purchase products, and finally send them to the sales terminal 4 of the reception staff in the 4S store, so that the sales staff in the 4S store can timely obtain the purchase intentions of customers, screen out the products that do not meet the purchase requirements, and sort the intelligent products that meet the purchase requirements to meet the needs of consumers;

[0037] As a preferred technical solution of this embodiment, the search times recognition unit 231 is used to calculate the keyword density W of the number of times the keywords of the target product are searched on the PC side and APP side of the customer's online browsing, according to the calculation formula:

[0038]

[0039] where T is the total length of keyword characters, T = the number of times the keyword appears * the keyword length; G is the total length of the search text content; combined with the keyword density W, accurately predict the products that consumers intend to purchase. In the calculation process of the keyword density W, take the search text "assisted driving on the highway section of the car" online as an example; assisted driving is the keyword. When W is larger, it indicates that the search purpose is more clear and the demand for the product is more accurate. On the contrary, when W is smaller, it indicates that the search purpose is relatively vague. And send the prediction result of the product intended to be purchased to the sales terminal 4 of the reception staff in the 4S store, so that the sales staff in the 4S store can accurately recommend the products that customers intend to purchase;

[0040] As a preferred technical solution of this embodiment, the click times recognition unit 232 is used to analyze the number of clicks on the search results of the target product searched on the PC side and APP side of the customer's online browsing, and record it as Q, and combine the click times Q and the purchase ratio P of the consumer's click times to judge the consumption habits of the customer;

[0041] When the click times Q is within the interval of the purchase ratio P of click times [0, 200], the search results on the PC side and APP side meet the search requirements, and this customer is a decisive consumer; Combined with the attached drawings for explanation Figure 4 , referring to the ratio of the number of single - consumer clicks to successful purchases in big data, the proportion of the number of people with less than 200 single - purchase clicks is 65%. This customer is a decisive consumer, and corresponding product recommendations are made according to the customer's purchase habits;

[0042] If the number of clicks Q is within the range of the purchase ratio of the number of clicks P [200, +∞), at this time, the search results on the PC side and the APP side cannot meet the search requirements, and it is difficult to match and search for the desired items. This user is a cautious consumer. Combining with Figure 4, referring to the ratio of the number of single - consumer clicks to successful purchases in big data, the proportion of people with more than 200 single - purchase clicks is 35%. This customer is a cautious consumer. Combining with the shopping habits of in - store customers, accurately judge the consumption habits; and send the judgment result of the customer's consumption habits to the sales terminal 4 of the reception staff in the 4S store, so that the sales staff in the 4S store can timely obtain the customer's consumption habits and formulate targeted sales plans according to the customer's consumption habits.

[0043] As a preferred technical solution of this embodiment, the residence - time recognition unit 233 is used to analyze the browsing duration of the PC side and the APP side where the customer browses the target product and the frequency of sliding the interface, and mine the user value type according to the RF model.

[0044] Record the customer's most recent browsing duration as R.

[0045] Record the frequency of sliding the interface in the customer's most recent browsing record as F.

[0046] If For valuable customers; the browsing duration is greater than the average duration, and the sliding frequency is greater than the average frequency, indicating that the customer has a high interest in the search results. Increase the preferential intensity, add gifts, etc. to stimulate consumption.

[0047] If For developing customers; the browsing duration is greater than the average duration, and the sliding frequency is less than the average frequency, indicating that the customer has a moderate interest in the search results. The salesperson should increase the corresponding product introduction to improve the customer's interest.

[0048] If For maintaining customers; the browsing duration is less than the average duration, and the sliding frequency is greater than the average frequency. The customer has a low interest in the search results. At this time, the product should be supplemented with an extended introduction in combination with the physical product to make up for the lack of online product introduction on the PC side and the APP side and improve the customer's interest.

[0049] If For retaining customers; both the browsing duration and the sliding frequency are less than the average value. The customer has a low interest in the search results. Ask the customer about their actual needs and contact the customer in time when the products that meet the customer's needs arrive at the store.

[0050] Among them, is the average duration of the customer's browsing interface; is the average frequency of the customer's sliding interface;

[0051] According to the value type of the customer, the mining results are sent to the sales terminal 4 of the receptionist in the 4S store, so that the salesperson in the 4S store can quickly locate the customer value type and formulate an adaptable sales plan.

[0052] As a preferred technical solution of this embodiment, the access data summarization unit 234 is used to summarize the access record information of the customer's recent offline visit to the 4S store, and send the summarization result to the sales terminal 4 of the receptionist in the 4S store, so that the salesperson in the 4S store can quickly obtain the purchase needs of the customer during the offline visit, formulate a targeted sales plan, avoid the impact on the user's mood caused by continuously asking about the customer's purchase needs, and improve the order success rate.

[0053] Embodiment Two

[0054] Please refer to Figure 3 , this embodiment is basically the same as Embodiment One. This embodiment is based on Embodiment One and has the same beneficial effects as Embodiment One. For the same parts, reference can be made to each other, and details will not be elaborated here.

[0055] When the customer value type is a customer to be retained, according to the actual needs of the customer, combine the customer's needs with the actual products in the offline 4S store, and judge whether the offline products in the 4S store meet the customer's needs. When they meet the customer's needs, make a precise recommendation of the products to reduce customer loss caused by search errors. If there is no search error, conduct marketing according to the sales method of retaining customers in Embodiment One.

[0056] It should be noted that in this article, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0057] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other. For the above embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and reference can be made to the corresponding parts of the method embodiments for the relevant parts.

[0058] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A 4S store customer intelligent product recommendation system based on personalized sales, characterized by: include: A data collection module (1), the data collection module (1) is used to collect gender, age, appearance information, and height and weight information of customers entering the store through an identity information collection device installed at the store entrance; The search module (2) searches for clues of the customer's intended purchase based on the user's appearance information and body shape information collected by the data collection module (1), including: A clue acquisition unit (21) is used to acquire the customer's facial appearance information such as gender and age, and acquire the customer's body shape information such as height and weight, and classify and summarize the customer's facial and body shape information; The clue comparison module (22) searches for the customer's online browsing records on the PC and APP terminals and offline 4S store visit records based on the customer information obtained in the clue acquisition unit (21), and compares the data of the online browsing records and the offline visit records; The clue mining module (23) is used to mine the comparison results of the data obtained in the clue comparison module (22), including: A search frequency identification unit (231) is used to identify the number of times a customer searches for a keyword of a target product on a PC or APP during online browsing, so as to accurately predict the product that the consumer intends to purchase; The click count identification unit (232) is used to search the target product keyword search results for the customer's online browsing records on the PC and APP terminals, and identify the click counts of the products that meet the customer's purchase intention, so as to accurately judge the customer's consumption habits; The dwell time identification unit (233) is used to identify the browsing dwell time of the products that the customer browses online on the PC side or the APP side and that meet the customer's intention to purchase, so as to mine the customer's user value type; A visit data aggregation unit (234) is used to aggregate the product purchase intentions recorded in the offline 4S store visit records, so as to quickly obtain the purchase needs of customers during offline visits; A generating module (3) is used to formulate the prediction results of the lead acquisition unit (21) and the mining results of the lead mining module (23) into a personalized sales plan; The sending module (4) generates a personalized sales plan formulated by the generating module (3) and sends it to the sales terminal of the 4S store according to the prediction result of the clue acquisition unit (21) and the mining result of the clue mining module (23).

2. According to claim 1, a 4S store customer intelligent product recommendation system based on personalized sales is characterized by: The clue acquisition unit (21) recognizes the customer's facial information, analyzes the customer's gender and age, and makes a preliminary prediction of the product the customer intends to purchase based on the customer's gender and age; It is also used to identify customer body information, analyze customer height and weight, and make secondary predictions of the products that customers intend to purchase based on their height and weight; The customer's intention to purchase products is sorted and finally sent to the sales terminal (4) of the reception staff in the 4S store, so that the sales staff of the 4S store can obtain the customer's purchase intention in a timely manner.

3. The 4S store customer intelligent product recommendation system based on personalized sales according to claim 1 is characterized by: The search frequency identification unit (231) is used to calculate the keyword density W of the number of keyword searches for the target product by the customer's online PC or APP terminal, according to the calculation formula: Where T is the total length of the keyword characters, T = the number of times the keyword appears * the length of the keyword; G is the total length of the search text content; Combined with keyword density W, accurate predictions can be made on the products that consumers intend to purchase; The predicted results of the products that customers intend to purchase are sent to the sales terminal (4) of the reception staff in the 4S store, so that the sales staff in the 4S store can make accurate recommendations on the products that customers intend to purchase.

4. The 4S store customer intelligent product recommendation system based on personalized sales as described in claim 1 is characterized by: The click count identification unit (232) is used to analyze the click count of the search results of the target product searched by the customer on the PC or APP side online, and record it as Q. The customer's consumption habits are judged by combining the click count Q with the purchase proportion P of the consumer's click count; When the number of clicks Q is within the range of click-to-purchase ratio P[0, 200], and the search results on the PC and APP sides meet the search needs, the customer is a decisive consumer; If the number of clicks Q is within the range of click-to-purchase ratio P[200, +∞), the search results on the PC and APP cannot meet the search needs and it is difficult to match the desired items. This user is a cautious consumer. The judgment result of the customer's consumption habits is sent to the sales terminal (4) of the reception staff in the 4S store, so that the sales staff of the 4S store can timely obtain the customer's consumption habits and formulate targeted sales plans based on the customer's consumption habits.

5. The 4S store customer intelligent product recommendation system based on personalized sales according to claim 1 is characterized by: The dwell time identification unit (233) is used to analyze the browsing time and sliding interface frequency of the target product clicked by the customer on the PC or APP side during online browsing, and to mine the user value type according to the RF model; The customer's most recent browsing time is recorded as R; The frequency of sliding the interface in the customer's most recent browsing history is recorded as F; If R≥R, F≥F, it is a valuable customer; If R≥R, F<F, it is a development customer; If R<R, F≥F, it means retaining the customer; If R<R, F<F, it is to retain customers; Among them, R is the average time that customers browse the interface; F is the average frequency of customers sliding the interface; According to the customer's value type, the mining results are sent to the sales terminal (4) of the reception staff in the 4S store, so that the 4S store sales staff can quickly locate the customer's value type and formulate an adaptive sales plan.

6. The 4S store customer intelligent product recommendation system based on personalized sales according to claim 1 is characterized by: The access data aggregation unit (234) is used to aggregate the access record information of the customer's recent offline visits to the 4S store, and send the aggregated results to the sales terminal (4) of the reception staff in the 4S store, so that the 4S store sales staff can quickly obtain the purchase needs of the customer during the offline visit and formulate targeted sales plans.

7. The 4S store customer intelligent product recommendation system based on personalized sales according to claim 5 is characterized by: When the customer value type is to retain customers, the customer's needs are combined with the actual products of the offline 4S store based on the customer's actual needs to determine whether the offline products of the 4S store meet the customer's needs. If they meet the customer's needs, accurate product recommendations are made to reduce customer loss caused by search errors.

Images