Automatic person-berth matching method and system for convenience chain store

Abstract

The invention discloses an automatic person and berth matching system and method for a convenience chain store. The system comprises a data acquisition module used for determining target analysis data according to information data of clients and shops determined by a business system; the portrait establishing module is used for establishing customer portraits and shop portraits for the orientation of customers and shops by utilizing clustering analysis, and obtaining an alliance merchant label library and a shop label library; and the matching module is used for filtering according to the intention region of the new alliance agent and the business rule, scoring the filtered to-be-recommended shops, obtaining a recommendation structure and displaying the recommendation structure. The invention further discloses an automatic person and berth matching method for the convenience chain store. The method has the advantages that the franchise business resources and the shop resources are subjected to refined and intelligent matching, and the method aims at more accurately meeting the requirements of the supply and demand parties, so that the contract signing conversion rate is greatly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of information analysis, and in particular relates to an automatic person-store matching method and system for convenience store chains. Background Art

[0002] In the current marketing of convenience store chain franchisees, after taking customers to visit the store, the subsequent conversion into orders is entirely based on the experience of the sales staff, resulting in low conversion efficiency.

[0003] In view of this, the present invention proposes an automatic person-store matching method and system for convenience store chains to solve the above problems. Summary of the invention

[0004] The purpose of the present invention is to provide an automatic person-shop matching method and system for convenience store chains. On the basis of the original viewing process, adding person-shop matching can effectively improve the potential matching degree between the store and the customer and improve the business opportunity conversion rate.

[0005] In order to achieve the above-mentioned purpose, the technical solution adopted by the present invention is: an automatic person-store matching system for convenience store chains, including a data acquisition module, which is used to determine target analysis data based on customer and store information data of the business system; a portrait establishment module, which uses cluster analysis to target customers and stores, establishes customer portraits and store portraits, and obtains franchisee label libraries and store label libraries; a matching module, which filters according to the intended areas and business rules of new franchisees, scores the filtered stores to be recommended, obtains the recommendation structure and displays it.

[0006] Based on the above-mentioned automatic person-shop matching system for convenience chain stores, the present invention further provides an automatic person-shop matching method for convenience chain stores based on the above-mentioned system, comprising steps 101 to 104.

[0007] In step 101, information data of customers and stores are obtained. The information data of customers and stores are derived from the business system. The customer master file data and the store system master file data are obtained, and the target analysis data is determined based on the obtained data.

[0008] Step 102, based on the number of franchise stores that have signed contracts with the customer in the target analysis data, obtain the customer's portrait and obtain the franchisee tag library;

[0009] Step 103, determining the store profile according to the store attributes in the target analysis data, and obtaining a store tag library;

[0010] Step 104, cluster the store types according to the customer's portrait label, obtain a prediction model, and display the matching results to the customer.

[0011] Preferably, the master data of the customer in step 101 includes age, expected rental range, expected payback period, intended city, intended store area and expected total investment.

[0012] Preferably, in step 102, the number of franchised stores that the customer has signed contracts with in history is analyzed to determine whether the customer is a franchisee of the chain store and the monthly profit data of the opened stores, and the customer portrait label is obtained based on the customer master file data and monthly profit data.

[0013] Preferably, in step 103, similar rents, areas, business districts, and grades are regarded as the same type.

[0014] Preferably, the label library data of the franchisee and the label library data of the store are collected to establish a data set, cluster analysis and association rules are performed to obtain a prediction model, and the weighted prediction score of each indicator of the customer's intended city store is generated according to the following formula:

[0015] y=x n *w n , x n is the characteristic index of the indicator in the model, w n is the indicator feature weight, n is a natural number from 1 to 5, the preferred feature indicators are x1 for total investment, x2 for rent, x3 for area, x4 for business format and x5 for intended city, the weights of each feature indicator are 0.4, 0.3, 0.2 and 0.1 respectively, and the top 5 are selected for display and recommendation based on the sorting results. The reactions of franchisees after receiving at least two rounds of recommendations are tracked, and the model is continuously optimized to achieve the goal of efficient, low-cost and refined market operations.

[0016] Preferably, the customer's intended region is filtered before matching.

[0017] Preferably, filtering is performed according to business rules before matching, and the specific rules are: one customer can match up to five stores, customers with stores do not participate in matching, stores with updated store information do not participate in push, and stores that have been viewed do not participate in matching.

[0018] The working principle of the present invention is:

[0019] The system is based on algorithmic empowerment to achieve the system's automatic matching of people and shops, and uses algorithms and quantitative decision-making tools to optimize resource allocation and improve the efficiency of contract conversion in the store investment and franchise scenario. Specifically, through advanced algorithm technology, the franchisee resources and store resources are matched in a refined and intelligent manner, aiming to more accurately meet the needs of both supply and demand sides, thereby greatly improving the contract conversion rate. Open decoupling of "promotable" resources means that the supply and demand resources will be more flexibly combined to maximize the timeliness value of "promotable shops" and increase the possibility of being signed.

[0020] Beneficial effects of the invention: The invention effectively improves the operating efficiency of shops, and can provide powerful technical support and decision-making basis for achieving overall business goals, so that the conversion rate of the shop's investment and franchise business development can be further improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a flowchart of the method of the present invention. DETAILED DESCRIPTION

[0022] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0023] In an embodiment of the present application, an automatic person-store matching system for convenience store chains is provided, including a data acquisition module for determining target analysis data based on information data of customers and stores determined by a business system; a portrait creation module for targeting customers and stores using cluster analysis, establishing customer portraits and store portraits, and obtaining a franchisee tag library and a store tag library; a matching module for filtering according to the intended areas and business rules of new franchisees, scoring the filtered stores to be recommended, obtaining a recommendation structure and displaying it.

[0024] See also Figure 1 , an embodiment of the present application provides an automatic person-shop matching method for a convenience store chain, which is implemented according to the above-mentioned automatic person-shop matching system and includes: steps 101-104.

[0025] In step 101, information data of customers and stores are obtained. The information data of customers and stores are derived from the business system. The customer master file data and the store system master file data are obtained, and the target analysis data is determined based on the obtained data.

[0026] Step 102, based on the number of franchise stores that have signed contracts with the customer in the target analysis data, obtain the customer's portrait and obtain the franchisee tag library;

[0027] Step 103, determining the store profile according to the store attributes in the target analysis data, and obtaining a store tag library;

[0028] Step 104, cluster the store types according to the customer's portrait label, obtain a prediction model, and display the matching results to the customer.

[0029] In a specific implementation, the customer's master data in step 101 includes age, expected rental range, expected payback period, intended city, intended store area, and expected total investment.

[0030] In the specific implementation, the customer's historical behavior, preferences, profit level, etc. are considered in step 102 to construct a customer portrait, and the number of franchise stores that the customer has signed contracts with in the past is analyzed to determine whether he is a franchisee of the chain store and the monthly profit data of the opened stores, and the customer's portrait label is obtained based on the customer's master file data and monthly profit data.

[0031] In a specific implementation, similarities in rent, area, business district, and grade in step 103 are regarded as the same type and classified according to the similarity.

[0032] In the specific implementation, the label library data of the franchisee and the label library data of the store are collected to establish a data set, and cluster analysis and association rules are performed to obtain a prediction model. The weighted prediction score of each indicator of the customer's intended city store is generated according to the following formula:

[0033] y=x1*w1+x2*w2+x3*w3+x4*w4+x5*w5, x1-x5 are the total investment, rent, area, business format and intended city respectively. The weights of each characteristic indicator are 0.4, 0.3, 0.2, 0.1 and 1 respectively. The intended city is a necessary indicator feature. According to the sorting results, the top five are selected for display. The reactions of franchisees after receiving at least two rounds of recommendations are tracked, and the model is continuously optimized to achieve high-efficiency, low-cost and refined market operation goals.

[0034] In the specific implementation, before matching, the customers’ intended regions are first filtered to display only the stores in the regions that the customers are interested in.

[0035] In the specific implementation, filtering is performed according to business rules before matching. The specific rules are: one customer can match up to five stores at most, customers with stores do not participate in the matching, stores with updated store information do not participate in the push, and stores that have already been viewed do not participate in the matching.

[0036] The advantages of the present invention are:

[0037] The present invention not only achieves efficient matching through data-driven methods, improves the matching quality of franchisees and stores, but also improves the operating efficiency of chain stores. By continuously collecting feedback and optimizing the model, the matching effect can be continuously improved. The present invention also makes fine and intelligent matching of franchisee resources and store resources, aiming to more accurately meet the needs of both supply and demand sides, thereby greatly improving the signing conversion rate.

[0038] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An automatic customer-store matching system for convenience store chains, characterized in that: It includes a data acquisition module for determining target analysis data based on information data of customers and stores determined by the business system; The portrait building module uses cluster analysis to target customers and stores, build customer portraits and store portraits, and obtain franchisee tag libraries and store tag libraries; the matching module filters new franchisees according to their intended areas and business rules, scores the filtered stores to be recommended, obtains the recommendation structure and displays it.

2. An automatic person-store matching method for a convenience store chain, using the automatic person-store matching system for a convenience store chain as claimed in claim 1, characterized in that: The method includes steps 101 to 104, wherein in step 101, information data of customers and stores are obtained, the information data of customers and stores are from the business system, customer master file data and store system master file data are obtained, and target analysis data is determined based on the obtained data; Step 102, based on the number of franchise stores that have signed contracts with the customer in the target analysis data, obtain the customer's portrait and obtain the franchisee tag library; Step 103, determining the store profile according to the store attributes in the target analysis data, and obtaining a store tag library; Step 104, cluster the store types according to the customer's portrait label, obtain a prediction model, and display the matching results to the customer.

3. The automatic person-shop matching method for convenience store chain according to claim 2, characterized in that: The customer's master profile data in step 101 includes age, expected rental range, expected payback period, intended city, intended store area, and expected total investment.

4. The automatic person-store matching method for convenience store chain according to claim 3 is characterized in that In step 102, the number of franchised stores that the customer has signed contracts with in the past is analyzed to determine whether the customer is a franchisee of the chain store and the monthly profit data of the opened stores, and the customer's portrait label is obtained based on the customer's master file data and monthly profit data.

5. The automatic person-shop matching method for convenience store chain according to claim 4, characterized in that: In step 103, similar rents, areas, business districts, and grades are considered to be of the same type.

6. The automatic person-shop matching method for convenience store chain according to claim 4, characterized in that: In step 104, the tag library data of the franchisee and the tag library data of the store are collected to establish a data set, cluster analysis and association rules are performed to obtain a prediction model, and the weighted prediction score of each indicator of the customer's intended city store is generated according to the following formula: y = x n *w n , x n is the characteristic index of the indicator in the model, w n is the indicator feature weight, n is a natural number from 1 to 5, the indicator features are total investment, rent, area, business format and intended city, the indicator feature weights are 0.4, 0.3, 0.2, 0.1 and 1 respectively, city feature is a necessary feature, and the top five are selected for display and recommendation based on the sorting results, and the reactions of franchisees after receiving at least two rounds of recommendations are tracked.

7. The automatic person-shop matching method for convenience store chain according to claim 6, characterized in that: Before matching, filter according to the customer's intended area.

8. The automatic person-shop matching method for convenience store chain according to claim 7, characterized in that: Before matching, filtering is performed according to business rules. The specific rules are: one customer can match up to five stores at most, customers with stores do not participate in matching, stores with updated store information do not participate in push, and stores that have already been viewed do not participate in matching.

Images