Method and system for evaluating settled commercial tenants on telecommunication integral exchange platform

Abstract

In order to better serve the users, a large number of third-party merchants are introduced in the telecom credit exchange shopping mall, and various gift services for credit exchange are provided for the users at the c-end of the whole telecom network. Gifts in the shopping mall are rich in variety and cover a plurality of categories such as articles of daily use, electronic products and recharge. However, along with the increase of the number of settled merchants, how to effectively evaluate the merchants and ensure the commodity quality and the service level becomes an important problem faced by a telecommunication point exchange platform. In the continuous operation process for many years, it is found that due to third-party merchant service problems such as poor commodity quality, delayed commodity distribution and low commodity cost performance, a user is not high in perception of point exchange and use, and the satisfaction degree is low.

Description

Technical Field

[0001] This invention relates to the fields of e-commerce and points systems, and more specifically, to a method and system for evaluating merchants on a telecommunications points redemption platform. Background Technology

[0002] To enhance user loyalty and satisfaction, operators reward users with points. By accumulating and using these points, users are incentivized to continue using telecommunications services and increase spending, thereby reducing the likelihood of churn.

[0003] To better serve users, the China Telecom points redemption mall has introduced a large number of third-party merchants, providing various gift redemption services for China Telecom's nationwide C-end users. The mall offers a wide variety of gifts, covering multiple categories such as daily necessities, electronic products, and mobile phone top-ups. However, with the increase in the number of merchants, how to effectively evaluate them and ensure product quality and service levels has become a significant issue for the China Telecom points redemption platform. Through years of continuous operation, it has been found that service issues with third-party merchants, such as poor product quality, untimely delivery, and low cost-effectiveness, have led to low user satisfaction and a negative perception of the points redemption system. Summary of the Invention

[0004] Module 1: Data Collection and Analysis Module (with appendix) Figure 1 S01) Product-related data collection: On a daily basis, collect data on the top 100 products redeemed with points each day, based on their collection, search, and browsing activity.

[0005] Order-related data collection: On a daily basis, collect the top 100 orders by total number of orders, the top 100 orders by total number of paid orders, the top 10 orders by cash payment percentage, the top 10 orders by payment rate, and the top 10 orders by conversion rate.

[0006] User review data collection: On a daily basis, collect the top 100 5-star products, top 100 user inquiries, and top 10 inquiry responses for each day.

[0007] Logistics-related data collection: Collect the top 100 orders shipped each day, and data on orders not shipped for more than 24 hours, on a daily basis.

[0008] Complaint-related data collection: Collect the top 100 complaints about products and the top 100 complaints about delayed delivery within a given month, using a monthly basis.

[0009] We collect relevant data from the past three years and combine it with the actual operation of merchants to analyze and mine the data, providing data basis for the training and evaluation of merchant evaluation models.

[0010] Module 2: Merchant Evaluation Model Module (with appendix) Figure 1 (S02) Based on the preliminary data collection and analysis, a merchant evaluation model is defined as follows: Merchant scoring rules: For the top 100 data, the score increases by 1 point for every 10 places the ranking increases. For example, the 100th place has a score of 90, the 99th place has a score of 90.1, and so on. For the top 10 data, the score increases by 1 point for every place the ranking increases.

[0011] 1. Merchant rating sub-factor score calculation: Let the score of a single sub-factor i be S, and its weight be W (the sum of the weights of all sub-factors under the same factor is 1, and some sub-factors are penalty factors with negative scores, which means that points need to be deducted).

[0012] 2. Merchant rating factor score calculation: Let the score of factor j be Dj, and the number of its sub-factors be nj, then:

[0013] Merchant rating total score calculation: Multiply the scores of all factors by their corresponding weights, then sum them to obtain the total merchant evaluation score. Let the total score be T, and the number of factors be m, then: , where Dj is the score of factor j.

[0014] 4. After the model is defined, the collected data is used to continuously train the model to improve its effectiveness. The values ​​of relevant factors, sub-factors, and weights are constantly adjusted to make the model evaluation more accurate.

[0015] Module 3: Merchant Evaluation and Early Warning Module (with appendix) Figure 1 (S03) After the merchant evaluation model is defined and trained, it is deployed to the production system. Every day at midnight, it collects relevant data on the previous day's products, orders, customer service, and logistics, and scores merchants based on the merchant evaluation model. Merchants with a total score below 50 or below the average merchant score are given a warning. After receiving the warning information, the operations staff verifies the detailed score data of the warned merchants, and manually verifies and confirms the detailed data of the items that did not receive points and the items that were deducted. After verification, a daily evaluation report is generated on the merchant side and pushed to the merchant side for confirmation. If the factors, sub-factors, weights, or score values ​​are set unreasonably, adjustments and optimizations are made.

[0016] Module 4: Merchant Evaluation Model Optimization Module (with appendix) Figure 1 (S04) After the daily merchant evaluation report is generated, merchants need to rectify and optimize the evaluation items that did not receive a score or were penalized. For evaluation items with doubts, they can file an appeal. The operations staff will further verify the evaluation items appealed by the merchants. If it is confirmed that the evaluation is abnormal due to unreasonable settings of model factors, weights, etc., the model will be optimized in a timely manner and the evaluation data will be corrected to ensure that it does not affect the merchant's quarterly evaluation score (if the quarterly comprehensive score is too low, the merchant may be removed according to the merchant introduction management method). Attached Figure Description Figure 1 is a flowchart illustrating the method and system process for evaluating merchants on a telecommunications points redemption platform. Specific Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please see Figure 1 In an embodiment of the present invention, a method and system for evaluating merchants on a telecommunications points redemption platform are provided. The present invention includes a data collection and analysis module (S01), a merchant evaluation model module (S02), a merchant evaluation and early warning module (S03), and a merchant evaluation model optimization module (S04).

[0019] First, we collected relevant data on products from the points redemption mall over the past three years, such as user favorites, searches, and browsing data; user redemption order data, such as order volume, payment rate, and cash usage percentage; logistics data, and customer complaint data.

[0020] Then, based on the collected data, a merchant evaluation model is defined, and in conjunction with historical merchant data, model factors, sub-factors, weights, scores, etc., are defined.

[0021] Next, the model is deployed in the production environment to evaluate merchants daily and issue warnings to merchants who score below 50 points or below the average merchant score.

[0022] Finally, based on merchant complaints, we will continuously improve the merchant evaluation model and continuously improve the quality of merchant services.

Claims

1. A method and system for evaluating merchants on a telecommunications points redemption platform, characterized in that: The system includes a data collection and analysis module, a merchant evaluation model module, a merchant evaluation and early warning module, and a merchant evaluation model optimization module.

2. The method and system for evaluating merchants on a telecommunications points redemption platform as described in claim 1, characterized in that: Data collection and analysis module: Collects relevant data on products from the points redemption mall over the past three years, such as user favorites, searches, and browsing data; user redemption order data, such as order volume, payment rate, and cash usage ratio; logistics data, customer complaint data, etc.

3. The method and system for evaluating merchants on a telecommunications points redemption platform as described in claim 1, characterized in that: Merchant Evaluation Model Module: Defines the merchant evaluation model based on the collected data, and defines the model factors, sub-factors, weights, scores, etc., in conjunction with historical merchant data.

4. The method and system for evaluating merchants on a telecommunications points redemption platform as described in claim 1, characterized in that: Merchant evaluation and early warning module: Deploy the model in the production environment to evaluate merchants daily and issue early warnings for merchants with scores below 50 or below the average merchant score.

5. The method and system for evaluating merchants on a telecommunications points redemption platform as described in claim 1, characterized in that: Merchant evaluation model optimization module: Based on merchant complaints, continuously improve the merchant evaluation model and continuously improve the quality of merchant services.

Images