Insurance customer list management method and system based on dynamic weighting and behavior driving

Abstract

The application relates to the field of artificial intelligence and insurance data processing technology, and particularly discloses a dynamic weighting and behavior-driven insurance customer list management method and system. The application determines a sales rhythm stage according to a policy expiration date offset and calculates a sales rhythm factor score, fuses customer labels, data levels, appointment follow-up states and follow-up interval days to calculate corresponding factor scores, obtains a comprehensive score through weighted summation, generates a priority recommendation list, and double-blocks outbound calls and jump dialing requests for the list, and updates customer dialing states in real time, thereby solving the problems of strong subjectivity of list sorting, information fragmentation, rigid rules, inability of high-priority customers to be preferentially served, and lagging tracking, realizing unified management and forced verification of the whole life cycle of insurance customer lists from generation to follow-up, improving the objectivity and accuracy of insurance customer list recommendations, improving the priority service rate of high-value insurance customers, and reducing state tracking delays.

Description

Technical Field

[0001] This invention relates to the fields of artificial intelligence and insurance data processing technology, and in particular to a method and system for managing insurance customer lists based on dynamic weighting and behavior-driven approaches. Background Technology

[0002] In insurance telemarketing, the management and prioritization of customer lists are crucial to sales efficiency. Currently, the industry mainly uses three types of technical solutions to address this issue, but all have significant shortcomings.

[0003] The first approach relies on a customer relationship management system (CRM). Operations personnel configure static sorting rules based on a few fields such as policy expiration date and customer level. The system then sorts the list according to these fixed rules and distributes it to agents for dialing. In this approach, list sorting depends entirely on the subjective experience of managers, lacking a unified and reproducible quantitative scoring standard. Customer information is scattered across different business systems, and static rules can only cover one or a few fields, leading to fragmented information and an inability to form a comprehensive assessment. Once configured, static rules are fixed; the same customer receives the same recommendation priority 30 days before expiration versus 3 days before expiration, failing to reflect the business logic of prioritizing calls closer to expiration. Furthermore, the list recommendation function is independent of the actual dialing system, allowing agents to freely choose who to call. High-priority customers are not guaranteed priority service, and the dialing status of the list relies on manual marking, resulting in delayed tracking and severely insufficient real-time performance.

[0004] The second approach uses a predictive dialing system for call centers. This system automatically initiates outbound calls by monitoring parameters such as agent availability and line occupancy in real time, transferring connected calls to available agents. While this approach improves overall outbound calling efficiency, the dialing order is completely unrelated to the customer's business value or the urgency of their policy expiration, failing to guarantee priority service for high-value customers. Furthermore, the system does not provide reasons for referrals, leaving agents with a lack of predictive ability regarding customer situations, impacting the conversion rate and making it impossible to handle scheduled same-day callbacks or follow-ups.

[0005] The third approach utilizes a data warehouse or distributed computing platform to perform batch scoring of customer lists via scheduled tasks, generating a recommended list for the next day, which is then imported into the business database for agents to query. This approach suffers from significant timeliness issues; T+1 batch calculations cannot reflect real-time changes in the list's status, and customers who have already been contacted by agents that day may still appear as not contacted in the list, making it difficult to support real-time business processes within the same day. For lists of millions of customers, loading the entire list into memory for scoring and sorting can easily lead to memory overflow, and serial processing is excessively time-consuming. Furthermore, data from different sales centers is stored in independent tables, requiring the maintenance of a mapping between center codes and table names for each center, heavily relying on hard-coded configurations. Adding a new center necessitates code modifications, resulting in high expansion costs. Scheduled tasks lack a reliable distributed anti-re-entrancy mechanism when deployed across multiple nodes, easily leading to duplicate data calculations or dirty data.

[0006] The three types of technologies mentioned above either only address how to calculate the customer list or only address how to make calls, but none of them integrate list generation, sequential allocation, forced dialing, and real-time status transitions into a closed loop. They lack unified management and mandatory verification of the entire customer list lifecycle, from generation to follow-up calls. How to integrate multi-dimensional customer information, dynamically adapt to the sales rhythm, and achieve real-time status transitions has become a pressing technical problem that needs to be solved to improve the operational efficiency of telemarketing. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention provides a method and system for managing insurance customer lists based on dynamic weighting and behavior-driven approaches.

[0008] In a first aspect, the present invention provides a method for managing insurance customer lists based on dynamic weighting and behavior-driven approaches. The technical solution of this method is as follows: Based on the offset between each insurance customer's policy expiration date and the current date, determine the sales rhythm stage to which each insurance customer belongs, and determine the sales rhythm factor score based on the sales rhythm stage. Based on each insurance customer's customer tag category, data level, appointment follow-up status, and follow-up interval days, customer tag factor score, data level factor score, appointment follow-up factor score, and follow-up interval factor score are determined respectively. The sales rhythm factor score, customer tag factor score, data level factor score, appointment follow-up factor score, and follow-up interval factor score are weighted and summed with their respective preset weights to obtain the comprehensive score for each insurance customer. Based on the comprehensive score of each insurance customer, a priority number is assigned to each insurance customer, and a recommendation list is generated. In response to a call request from an agent to a target insurance customer, if the target insurance customer is not in the recommended list and the agent has an insurance customer whose status is pending or requires a second call, the call request is blocked; if the target insurance customer is in the recommended list and there is an insurance customer in the recommended list whose priority number is lower than that of the target insurance customer who has not yet been called, the call request is blocked; otherwise, the call request is allowed. In response to the agent's dialing operation on the target insurance customer in the recommended list, the status of the target insurance customer is updated from pending dialing to dialed; in response to the agent's appointment operation on the target insurance customer, the status of the target insurance customer is updated from dialed to require a second dialing.

[0009] The beneficial effects of the insurance customer list management method based on dynamic weighting and behavior-driven approach of the present invention are as follows: The method of this invention determines the sales rhythm stage based on the policy expiration date offset and calculates the sales rhythm factor score. It integrates customer tags, data levels, appointment follow-up status, and follow-up interval days to calculate the corresponding factor score. The comprehensive score is obtained by weighted summation and a priority recommendation list is generated. It also performs dual blocking of calls outside the agent list and jump dialing requests, and updates the customer call status in real time. This solves the problems of strong subjectivity in list sorting, information fragmentation, rigid rules, failure to prioritize high-priority customers, and delayed tracking. It realizes unified management and mandatory verification of the entire life cycle of insurance customer list from generation to follow-up, improves the objectivity and accuracy of insurance customer list recommendation, increases the priority service rate of high-value insurance customers, and reduces status tracking delay.

[0010] Based on the above solution, the insurance customer list management method based on dynamic weighting and behavior-driven approach of the present invention can be further improved as follows.

[0011] In one alternative approach, the step of assigning a priority number to each insurance customer based on their overall score and generating a recommendation list includes: Based on the policy expiration time range and the lower limit of the scoring threshold, insurance customers are filtered using database-level conditions to obtain the first filtering result; The first filtering result is matched with sales rhythm tags and scoring thresholds one by one to obtain the second filtering result; Based on the team configuration, the insurance customers in the second filtering result are grouped into multiple parallel processing groups, and scoring, sorting and truncation are performed on each parallel processing group to obtain the candidate list for each agent. The candidate lists processed by each parallel processing group are merged to obtain the recommended list.

[0012] The advantages of adopting the above optional method are as follows: the first round of screening is performed at the database layer based on the policy expiration time range and the lower limit of the scoring threshold, and then the secondary filtering is completed by matching the sales rhythm tags with the scoring threshold one by one. Subsequently, the list is grouped and processed in parallel according to the team configuration, and scoring, sorting and truncation are performed. This reduces the memory pressure caused by loading all the data at once, and makes the large-scale list generation process more stable and controllable.

[0013] In one alternative approach, the steps of grouping insurance customers from the second filtering result into multiple parallel processing groups based on team configuration, and performing scoring, sorting, and truncation processing on each parallel processing group to obtain a candidate list for each agent include: Get the team configuration of all teams under the organization to which the agent belongs, and get the list of agents included in each team according to the team configuration; A round-robin equal distribution algorithm was used to divide all teams into multiple parallel processing groups; For each parallel processing group, the insurance customers in the second filtering result are retrieved page by page, team by team and seat by seat. The comprehensive score of the insurance customers retrieved on each page is calculated and sorted according to the comprehensive score. After sorting, the number of insurance customers with the highest comprehensive scores as specified in the configuration item is truncated according to the seat dimension and retained as the candidate list for each seat.

[0014] The advantages of adopting the above optional method are as follows: by further adopting the round-robin equal distribution algorithm, each team is divided into multiple parallel processing groups. By retrieving data page by page for each team and each agent, the memory usage of each process is controlled. After calculating the comprehensive score of each page of customers and sorting them, the maximum number of high-scoring customers are truncated according to the agent dimension. This ensures that the quantity and quality of the list obtained by each agent are reasonably constrained, and the list distribution process is more balanced and orderly.

[0015] In one alternative approach, in response to an agent's call request to a target insurance customer, if the target insurance customer is not in the recommended list and the agent has an insurance customer whose status is pending call or requires a second call, the call request is blocked; if the target insurance customer is in the recommended list and there are uncalled insurance customers in the recommended list with a priority number lower than the target insurance customer's priority number, the call request is blocked; otherwise, the call request is allowed. The steps include: Query the priority number of the target insurance customer in the recommended list; If the priority number is not found, check if there is an insurance customer record in the agent's status as pending call or requiring a second call. If it exists, block the call request; otherwise, allow the call request. When the priority number is found, the system iterates through the insurance customer records in the recommended list that are in a state of pending call or require a second call, and checks whether there are any insurance customer records that have not been called but whose priority number is lower than that of the target insurance customer. If such a record exists, the call request is blocked; otherwise, the call request is allowed.

[0016] The advantages of adopting the above optional method are as follows: further splitting the call request verification into two levels, priority number query and non-call record traversal, to handle two scenarios: the target customer is outside the recommended list and the target customer is in the recommended list but there are higher priority customers who have not called. This makes the blocking condition judgment logic clearer and more explicit, and reduces the probability of false blocking or missed blocking due to scenario confusion.

[0017] In one alternative approach, the step of updating the status of the target insurance customer from pending call to called in response to the agent's call to the target insurance customer in the recommended list includes: In response to the agent's dialing operation on the target insurance customer in the recommended list, the query condition that the source status is equal to the pending dialing is set in the database operation of performing status update. The status of the target insurance customer is updated from pending dialing to dialed only when the current status of the target insurance customer is pending dialing. The step of updating the target insurance customer's status from "called" to "requires a second call" in response to the agent's appointment request to the target insurance customer includes: In response to the agent's appointment operation for the target insurance customer, the query condition that the source status is equal to "called" is set in the database operation of status update. Only when the current status of the target insurance customer is "called" will the status of the target insurance customer be updated from "called" to "needs to be called again".

[0018] The advantages of adopting the above optional method are as follows: In the status update database operation, a query condition is added that the source status is equal to the current expected status. The status change is only performed when the customer's current status is consistent with the expected source status, avoiding duplicate updates or status jumps caused by concurrent operations, and ensuring the accuracy and consistency of the three status flows: pending call, dialed, and need to dial again.

[0019] In one alternative approach, the steps for determining the sales pace phase to which each insurance customer belongs include: Using the policy expiration date as the base date, the sales rhythm stage to which the insurance customer belongs is determined based on the range in which the difference in the number of days between the current date and the base date falls; wherein, the sales rhythm stage includes multiple preset stages, and each preset stage corresponds to a different range of the difference in the number of days.

[0020] The beneficial effects of adopting the above optional method are as follows: by further using the policy expiration date as the base date, the sales rhythm stage is determined based on the range of the number of days between the current date and the base date, making the stage division standard objective and quantifiable, facilitating the assignment of differentiated scores according to the urgency of the customer's expiration, and improving the rationality of the sales rhythm factor score.

[0021] In one alternative approach, the steps of determining customer tagging factor scores, data level factor scores, appointment follow-up factor scores, and follow-up interval factor scores based on each insurance customer's customer tagging category, data level, appointment follow-up status, and follow-up interval days include: The customer tagging factor score is determined based on the customer tagging category of the insurance customer. When the customer tagging category indicates that the insurance customer belongs to a preset high-quality customer group, the customer tagging factor score is a first score. When the customer tagging category indicates that the insurance customer does not belong to the preset high-quality customer group, the customer tagging factor score is a second score. The first score is greater than the second score. The data level factor score is determined based on the data level of the insurance customer. The data level includes multiple preset levels, which are arranged in order of judgment priority. The preset levels to which the data level of the insurance customer belongs are judged in order of judgment priority from high to low. The data level factor score is determined based on the judgment result. Different preset levels correspond to different scores. The appointment follow-up factor score is determined based on the appointment follow-up status of the insurance customer. When the appointment follow-up status indicates that the insurance customer has an appointment on the same day, the appointment follow-up factor score is the third score. When the appointment follow-up status indicates that the insurance customer has no appointment on the same day, the appointment follow-up factor score is the fourth score. The third score is greater than the fourth score. The follow-up interval factor score is determined based on the number of follow-up interval days for the insurance customer. The number of follow-up interval days is a plurality of preset number of days intervals. The follow-up interval factor score is determined based on the preset number of days intervals into which the follow-up interval days fall. Different preset number of days intervals correspond to different scores.

[0022] The beneficial effects of adopting the above optional methods are as follows: scoring rules are set separately for four dimensions: customer tag category, data level, appointment follow-up status and follow-up interval days. Higher scores are assigned to customers who belong to the high-quality customer group or who have appointment follow-up on the same day, so that the comprehensive score is closer to the actual business value and the objectivity of the list ranking is improved.

[0023] Secondly, this invention provides an insurance customer list management system based on dynamic weighting and behavior-driven principles. The technical solution of this system is as follows: The acquisition module is used to determine the sales rhythm stage to which each insurance customer belongs based on the offset between the policy expiration date and the current date, and to determine the sales rhythm factor score based on the sales rhythm stage. The determination module is used to determine the customer tag factor score, data level factor score, appointment follow-up factor score, and follow-up interval factor score for each insurance customer based on their customer tag category, data level, appointment follow-up status, and follow-up interval days, respectively. The calculation module is used to sum the sales rhythm factor score, the customer tag factor score, the data level factor score, the appointment follow-up factor score, and the follow-up interval factor score with their respective preset weights to obtain the comprehensive score of each insurance customer. The generation module is used to assign a priority number to each insurance customer based on their comprehensive score and generate a recommendation list. The processing module is used to respond to the call request from the agent to the target insurance customer. If the target insurance customer is not in the recommended list and the agent has an insurance customer whose status is pending call or requires a second call, the call request is blocked. If the target insurance customer is in the recommended list and there is an insurance customer in the recommended list whose priority number is lower than that of the target insurance customer who has not been called, the call request is blocked. Otherwise, the call request is allowed. The update module is used to respond to the agent's dialing operation for the target insurance customer in the recommended list, and update the status of the target insurance customer from pending dialing to dialed; and to respond to the agent's appointment operation for the target insurance customer, and update the status of the target insurance customer from dialed to requiring a second dialing.

[0024] The beneficial effects of the insurance customer list management system based on dynamic weighting and behavior-driven principles of the present invention are as follows: The system of this invention determines the sales rhythm stage based on the policy expiration date offset and calculates the sales rhythm factor score. It integrates customer tags, data levels, appointment follow-up status, and follow-up interval days to calculate the corresponding factor score. After weighted summation, a comprehensive score is obtained and a priority recommendation list is generated. The system also performs dual blocking of calls outside the agent list and jump dialing requests, and updates the customer call status in real time. This solves the problems of strong subjectivity in list sorting, information fragmentation, rigid rules, failure to provide priority service to high-priority customers, and delayed tracking. It realizes unified management and mandatory verification of the entire life cycle of insurance customer lists from generation to follow-up, improves the objectivity and accuracy of insurance customer list recommendations, increases the priority service rate of high-value insurance customers, and reduces status tracking delays.

[0025] Thirdly, the technical solution of an electronic device according to the present invention is as follows: It includes a memory, a processor, and a program stored in the memory and running on the processor, wherein the processor executes the program to implement the steps of the insurance customer list management method based on dynamic weighting and behavior-driven method of the present invention.

[0026] Fourthly, the technical solution of a computer-readable storage medium provided by the present invention is as follows: The computer-readable storage medium stores instructions that, when read, cause the computer-readable storage medium to perform the steps of the insurance customer list management method based on dynamic weighting and behavior-driven method of the present invention.

[0027] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other objects, features and advantages of the present invention more obvious and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0028] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a flowchart illustrating an embodiment of an insurance customer list management method based on dynamic weighting and behavior-driven principles according to the present invention. Figure 2 This is a schematic diagram of the overall architecture of an insurance customer list management system based on dynamic weighting and behavior-driven principles. Figure 3 A diagram illustrating the 13-stage time window division of the sales rhythm; Figure 4 A schematic diagram of a five-factor weighted scoring model; Figure 5 A flowchart illustrating the mandatory verification process for dialing order; Figure 6 A schematic diagram of the state machine transition driven by three-state real-time behavior; Figure 7 This is a complete technical process timing diagram; Figure 8 This is a schematic diagram of an embodiment of an insurance customer list management system based on dynamic weighting and behavior-driven principles according to the present invention. Figure 9 This is a schematic diagram of an embodiment of an electronic device according to the present invention. Detailed Implementation

[0029] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0030] Figure 1 This diagram illustrates a flowchart of an embodiment of an insurance customer list management method based on dynamic weighting and behavior-driven principles provided by the present invention. This method can be executed by electronic devices such as terminal devices or servers. The terminal device can be any fixed or mobile terminal, such as user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, in-vehicle device, or wearable device. The server can be a single server or a server cluster consisting of multiple servers. Any electronic device can implement the insurance customer list management method based on dynamic weighting and behavior-driven principles by having its processor call computer-readable instructions stored in its memory. Figure 1 As shown, it includes the following steps: S1. Determine the sales rhythm stage to which each insurance customer belongs based on the offset between the policy expiration date and the current date, and determine the sales rhythm factor score based on the sales rhythm stage.

[0031] Here, "insurance customer" refers to the policyholder or insured, the recipient of insurance services; for example, customer A is the insurance customer, and customer A holds a car insurance policy. "Policy expiration date" refers to the date the insurance liability terminates as stated in the insurance contract; for example, customer A's car insurance policy expiration date is July 1, 2026. "Offset" refers to the difference in the number of days between the current date and the policy expiration date, denoted by 'd'; for example, if the current date is June 25, 2026, and the policy expiration date is July 1, 2026, then the offset 'd' is 6 days.

[0032] The sales rhythm phase refers to multiple time window phases divided according to the interval into which the difference d between the policy expiration date and the current date falls; for example, if customer A's policy expiration date is July 1, 2026, the current date is June 28, 2026, and the offset d is 3 days, then customer A belongs to the "first 3 days" sales rhythm phase.

[0033] The sales rhythm factor score refers to the score assigned based on the sales rhythm stage to which the insurance customer belongs, denoted by X1. For example, if the sales rhythm stage to which customer A belongs is the "first 3 days" stage, then the sales rhythm factor score X1 is the first value.

[0034] S2. Based on each insurance customer's customer tag category, data level, appointment follow-up status, and follow-up interval days, determine the customer tag factor score, data level factor score, appointment follow-up factor score, and follow-up interval factor score respectively.

[0035] Here, "Customer Tag Category" refers to a category identifier used to mark whether an insurance customer belongs to a pre-defined high-quality customer group. For example, if Customer A is tagged as a "Sunflower" customer, then Customer A's customer tag category indicates that Customer A belongs to the pre-defined high-quality customer group. "Data Level" refers to a level rating comprehensively assessed based on the insurance customer's policy type and sales potential. For example, if Customer A's auto insurance policy is a first-year renewal policy and their sales potential level is "A," then Customer A's data level is "First-Year Renewal." "Appointment Follow-up Status" refers to the status indicator indicating whether the insurance customer has an appointment contact record with the agent on that day. For example, if Customer A made an appointment with the agent on the morning of June 28, 2026, to call back that afternoon, then Customer A's appointment follow-up status indicates that Customer A has an appointment on that day. The follow-up interval days refer to the number of days since the current date and the insurance customer's most recent follow-up record, denoted by T. For example, if customer A's most recent follow-up record occurred on June 23, 2026, and the current date is June 28, 2026, then the follow-up interval days T is 5 days.

[0036] The customer tagging factor score is represented by X2, which is a score assigned based on the customer tagging category of the insurance customer. For example, if Customer A belongs to the preset high-quality customer group, then the customer tagging factor score X2 is the first value. The data level factor score is represented by X3, which is a score assigned based on the data level of the insurance customer. For example, if Customer A's data level is "first year renewal", then the data level factor score X3 is the fifth value. The appointment follow-up factor score is represented by X4, which is a score assigned based on the appointment follow-up status of the insurance customer. For example, if Customer A has an appointment on the same day, then the appointment follow-up factor score X4 is the third value. The follow-up interval factor score is represented by X5, which is a score assigned based on the preset number of days T that the insurance customer's follow-up interval T falls into. For example, if Customer A's follow-up interval T is 5 days, falling into the preset number of days "4 to 7 days", then the follow-up interval factor score X5 is the fourth value.

[0037] S3. The sales rhythm factor score, the customer tag factor score, the data level factor score, the appointment follow-up factor score, and the follow-up interval factor score are weighted and summed with their respective preset weights to obtain the comprehensive score for each insurance customer.

[0038] The preset weights refer to the weighting coefficients pre-set for each factor score in the overall score calculation. For example, the preset weight for the sales rhythm factor score X1 is w1, the preset weight for the customer tag factor score X2 is w2, the preset weight for the data level factor score X3 is w3, the preset weight for the appointment follow-up factor score X4 is w4, and the preset weight for the follow-up interval factor score X5 is w5. The overall score is the sum of the scores obtained by multiplying each factor score by its corresponding preset weight, denoted by S. For example, the overall score S is obtained by multiplying the scores of customer A under each factor by their corresponding preset weights and then summing the results.

[0039] S4. Based on the comprehensive score of each insurance customer, assign a priority number to each insurance customer and generate a recommendation list.

[0040] Here, the priority number refers to the ranking position assigned to each insurance customer after sorting them from highest to lowest based on their comprehensive score S. For example, if Customer A's comprehensive score S ranks 3rd in the recommended list, then Customer A's priority number is 3. The recommended list refers to the list of recommended dialing orders generated after sorting each insurance customer's comprehensive score S from highest to lowest and assigning a priority number to each insurance customer. For example, the system generates a recommended list for Agent A containing 10 insurance customers, including Customer A, Customer B, and Customer C, with Customer A's priority number being 3.

[0041] S5. Responding to a call request from an agent to a target insurance customer: If the target insurance customer is not in the recommended list and the agent has an insurance customer whose status is pending or requires a second call, then the call request is blocked; if the target insurance customer is in the recommended list and there is an uncalled insurance customer in the recommended list with a priority number lower than the target insurance customer's priority number, then the call request is blocked; otherwise, the call request is allowed.

[0042] Here, a call request refers to a verification request initiated by an agent to the system before calling the target insurance customer; for example, agent A initiates a call request to the system before calling customer A to verify whether customer A allows being called.

[0043] S6. In response to the agent's dialing operation on the target insurance customer in the recommended list, update the status of the target insurance customer from pending dialing to dialed; in response to the agent's appointment operation on the target insurance customer, update the status of the target insurance customer from dialed to need to dial again.

[0044] The "calling operation" refers to the actual act of an agent making a phone call to an insurance customer. For example, after agent A initiates a call request and passes verification, the agent executes the call to customer A, updating customer A's status from "pending call" to "called." The "scheduling operation" refers to the action of an agent scheduling a follow-up call for an insurance customer after the call ends and the contact record is saved. For example, after agent A and customer A finish their call, they agree to meet again the following morning. Agent A executes the scheduling operation, setting the follow-up call time to the following morning, updating customer A's status from "called" to "needs a second call."

[0045] The technical solution of this embodiment determines the sales rhythm stage based on the policy expiration date offset and calculates the sales rhythm factor score. It integrates customer tags, data levels, appointment follow-up status, and follow-up interval days to calculate the corresponding factor score. After weighted summation, a comprehensive score is obtained and a priority recommendation list is generated. It also performs dual blocking of calls outside the agent list and jump dialing requests, and updates the customer call status in real time. This solves the problems of strong subjectivity in list sorting, information fragmentation, rigid rules, inability to prioritize high-priority customers, and delayed tracking. It realizes unified management and mandatory verification of the entire life cycle of insurance customer list from generation to follow-up, improves the objectivity and accuracy of insurance customer list recommendation, increases the priority service rate of high-value insurance customers, and reduces status tracking delay.

[0046] In one alternative approach, the step of assigning a priority number to each insurance customer based on their overall score and generating a recommendation list includes: Based on the policy expiration time range and the lower limit of the scoring threshold, insurance customers are filtered using database-level conditions to obtain the first filtering result.

[0047] The policy expiration time range refers to the allowed range of policy expiration dates set during database-level conditional filtering. For example, the policy expiration time range is set to 7 days before the current date and the date corresponding to the number of days prior to the current date when calls can be made. Customer A's policy expiration date is July 1, 2026, falling within this range, thus customer A passes the policy expiration time range filtering. The lower limit of the scoring threshold refers to the lowest threshold for the overall score set during database-level conditional filtering. For example, the lower limit of the scoring threshold is set to a first threshold. If customer A's overall score S is greater than the first threshold, then customer A passes the lower limit of the scoring threshold filtering.

[0048] Database-level conditional filtering refers to the process of initially filtering insurance customers at the database query level using structured query language (SCL) conditions. For example, a query is executed in the database, setting policy expiration time range conditions and a lower limit for the scoring threshold, to filter out insurance customer records that meet the conditions as the first filtering result. The first filtering result refers to the first set of filtering results obtained after database-level conditional filtering. For example, database-level conditional filtering is performed on all insurance customers, obtaining 10,000 insurance customer records as the first filtering result.

[0049] The first filtering result is then matched with sales rhythm tags and rating thresholds for each item to obtain the second filtering result.

[0050] Sales rhythm label matching refers to the process of comparing the sales rhythm stage of an insurance customer in memory with the set of sales rhythm labels set in the configuration item. For example, the set of sales rhythm labels set in the configuration item includes the "first day" stage and the "first 3 days" stage. If customer A's sales rhythm stage is the "first 3 days" stage, then customer A passes the sales rhythm label matching. Scoring threshold matching refers to the process of comparing the insurance customer's comprehensive score S in memory with the scoring threshold set in the configuration item. For example, if the scoring threshold set in the configuration item is a second threshold, and customer A's comprehensive score S is greater than the second threshold, then customer A passes the scoring threshold matching.

[0051] The second filtering result refers to the set of results obtained by matching the sales rhythm tags and scoring thresholds for each of the first filtering results. For example, matching the sales rhythm tags and scoring thresholds for each of the 10,000 first filtering results yields 2,000 insurance customer records as the second filtering result.

[0052] Based on the team configuration, the insurance customers in the second filtering result are grouped into multiple parallel processing groups, and scoring, sorting and truncation are performed on each parallel processing group to obtain the candidate list for each agent.

[0053] Here, a parallel processing group refers to multiple parallel processing units into which the teams to be processed are divided using a round-robin distribution algorithm. For example, teams A, B, C, D, E, F, G, H, I, and J (a total of 10 teams) are divided into 5 parallel processing groups using the round-robin distribution algorithm, with each group containing 2 teams. Each parallel processing group is submitted to different threads in a thread pool for parallel execution. The candidate list refers to the list of insurance customers to be recommended generated for each agent after each parallel processing group scores, sorts, and truncates the insurance customers in the second filtering result. For example, 10 candidate lists are generated for agent A, with customer A ranked 3rd on the candidate list.

[0054] The candidate lists processed by each parallel processing group are merged to obtain the recommended list.

[0055] Among the above optional methods, the first round of screening is further performed at the database layer based on the policy expiration time range and the lower limit of the scoring threshold. Then, the second filtering is completed by matching the sales rhythm tags with the scoring threshold one by one. Subsequently, the list is grouped and processed in parallel according to the team configuration, and scoring, sorting and truncation are performed. This reduces the memory pressure caused by loading all the data at once, and makes the large-scale list generation process more stable and controllable.

[0056] In one alternative approach, the steps of grouping insurance customers from the second filtering result into multiple parallel processing groups based on team configuration, and performing scoring, sorting, and truncation processing on each parallel processing group to obtain a candidate list for each agent include: Get the team configuration of all teams under the organization to which the agent belongs, and get the list of agents included in each team according to the team configuration.

[0057] The seat list refers to the collection of all seats contained in each team. For example, if team A contains three people, namely seat A, seat B, and seat C, then these three people constitute the seat list of team A.

[0058] A round-robin equal distribution algorithm was used to divide all teams into multiple parallel processing groups.

[0059] Among them, the round-robin distribution algorithm refers to the distribution algorithm that sequentially and cyclically allocates teams to various parallel processing groups to balance the load of each group; for example, teams A, B, and C are sequentially and cyclically allocated to two parallel processing groups, with the first group allocated to teams A and C, and the second group allocated to team B.

[0060] For each parallel processing group, the insurance customers in the second filtering result are retrieved page by page, team by team and seat by seat. The comprehensive score of the insurance customers retrieved on each page is calculated and sorted according to the comprehensive score. After sorting, the number of insurance customers with the highest comprehensive scores as specified in the configuration item is truncated according to the seat dimension and retained as the candidate list for each seat.

[0061] Here, paginated data retrieval refers to the process of reading insurance customer data from the database multiple times according to the set data volume per page; for example, the page size is set to 500 records, and the 2000 insurance customer records in the second filtering result are read sequentially in 4 pages, with 500 records read per page.

[0062] Here, "seat dimension" refers to the perspective of data processing and result truncation based on the seat. For example, when truncating the candidate list according to the seat dimension, the candidate list for seat A is truncated to retain the top 120 entries, and the candidate list for seat B is truncated to retain the top 80 entries. The upper limit of the seat list refers to the maximum number of candidate entries with the highest overall scores that each seat can retain during truncation. For example, the configuration item specifies that the upper limit of the list for seat A is 120 entries; after truncation, the top 120 insurance customers with the highest overall scores are retained as candidate entries for seat A.

[0063] Among the above optional methods, a round-robin average distribution algorithm is further adopted to divide each team into multiple parallel processing groups. By retrieving data page by page for each team and each agent, the memory usage per transaction is controlled. After calculating the comprehensive score of each page of customers and sorting them, the maximum number of high-scoring customers are truncated according to the agent dimension. This ensures that the quantity and quality of the list obtained by each agent are reasonably constrained, and the list distribution process is more balanced and orderly.

[0064] In one alternative approach, in response to an agent's call request to a target insurance customer, if the target insurance customer is not in the recommended list and the agent has an insurance customer whose status is pending call or requires a second call, the call request is blocked; if the target insurance customer is in the recommended list and there are uncalled insurance customers in the recommended list with a priority number lower than the target insurance customer's priority number, the call request is blocked; otherwise, the call request is allowed. The steps include: Query the priority number of the target insurance customer in the recommended list.

[0065] If the priority number is not found, check if there is an insurance customer record in the agent's status as pending call or requiring a second call. If it exists, block the call request; otherwise, allow the call request.

[0066] When the priority number is found, the system iterates through the insurance customer records in the recommended list that are in a state of pending call or require a second call, and checks whether there are any insurance customer records that have not been called but whose priority number is lower than that of the target insurance customer. If such a record exists, the call request is blocked; otherwise, the call request is allowed.

[0067] In the above optional methods, the call request verification is further divided into two levels: priority number query and non-call record traversal. This addresses two scenarios: the target customer is outside the recommended list and the target customer is on the recommended list but there are higher priority customers who have not called. This makes the blocking condition judgment logic clearer and more explicit, reducing the probability of false blocking or missed blocking due to scenario confusion.

[0068] In one alternative approach, the step of updating the status of the target insurance customer from pending call to called in response to the agent's call to the target insurance customer in the recommended list includes: In response to the agent's dialing operation on the target insurance customer in the recommended list, the query condition of source status equal to pending dialing is set in the database operation of status update. Only when the current status of the target insurance customer is pending dialing, the status of the target insurance customer is updated from pending dialing to dialed.

[0069] The source state refers to the current state value of the insurance customer before the state update operation is performed. For example, when the operation of updating customer A's state from "to be dialed" to "already dialed" is performed, the source state is "to be dialed". The database update statement sets the query condition that the source state is equal to "to be dialed", and the update is performed only when customer A's current state is "to be dialed".

[0070] The step of updating the target insurance customer's status from "called" to "requires a second call" in response to the agent's appointment request to the target insurance customer includes: In response to the agent's appointment operation for the target insurance customer, the query condition that the source status is equal to "called" is set in the database operation of status update. Only when the current status of the target insurance customer is "called" will the status of the target insurance customer be updated from "called" to "needs to be called again".

[0071] In the above optional methods, a query condition is further added to the status update database operation, which states that the source status is equal to the current expected status. The status change is only performed when the customer's current status is consistent with the expected source status, thus avoiding duplicate updates or status jumps caused by concurrent operations and ensuring the accuracy and consistency of the three status transitions: pending call, already called, and need to make a second call.

[0072] In one alternative approach, the steps for determining the sales pace phase to which each insurance customer belongs include: Using the policy expiration date as the base date, the sales rhythm stage to which the insurance customer belongs is determined based on the range in which the difference in the number of days between the current date and the base date falls; wherein, the sales rhythm stage includes multiple preset stages, and each preset stage corresponds to a different range of the difference in the number of days.

[0073] The benchmark date refers to the policy expiration date used as a reference in determining the sales rhythm phase; for example, the policy expiration date of Customer A, July 1, 2026, is the benchmark date. The preset phase refers to multiple fixed time window phases pre-set in the sales rhythm phase; for example, the preset phase includes a "first day" phase with a day difference d equal to 0, a "first 3 days" phase with a day difference d equal to -3, a "first 2 days" phase with a day difference d equal to -2, and a "first 1 day" phase with a day difference d equal to -1.

[0074] Among the above-mentioned optional methods, the policy expiration date is further used as the base date, and the sales rhythm stage is determined based on the range of the number of days between the current date and the base date. This makes the stage division standard objective and quantifiable, and facilitates the assignment of differentiated scores according to the urgency of the customer's expiration, thereby improving the rationality of the sales rhythm factor score.

[0075] In one alternative approach, the steps of determining customer tagging factor scores, data level factor scores, appointment follow-up factor scores, and follow-up interval factor scores based on each insurance customer's customer tagging category, data level, appointment follow-up status, and follow-up interval days include: The customer tagging factor score is determined based on the customer tagging category of the insurance customer. When the customer tagging category indicates that the insurance customer belongs to a preset high-quality customer group, the customer tagging factor score is a first score. When the customer tagging category indicates that the insurance customer does not belong to the preset high-quality customer group, the customer tagging factor score is a second score. The first score is greater than the second score.

[0076] The pre-defined high-quality customer group refers to a set of customers with high business value that are pre-defined according to customer tag categories. For example, insurance customers tagged as "sunflower" customers constitute the pre-defined high-quality customer group. If customer A is tagged as a "sunflower" customer, then customer A belongs to the pre-defined high-quality customer group.

[0077] The first score refers to the higher value of the customer identification factor score X2 when the insurance customer belongs to the preset high-quality customer group; for example, Customer A belongs to the preset high-quality customer group, and Customer Identification Factor Score X2 is the first score. The second score refers to the lower value of the customer identification factor score X2 when the insurance customer does not belong to the preset high-quality customer group; for example, Customer B does not belong to the preset high-quality customer group, and Customer Identification Factor Score X2 is the second score.

[0078] The data level factor score is determined based on the data level of the insurance customer. The data level includes multiple preset levels, which are arranged in order of judgment priority. The preset levels to which the data level of the insurance customer belongs are judged in order of judgment priority from high to low. The data level factor score is determined based on the judgment result. Different preset levels correspond to different scores.

[0079] Among them, the preset level refers to multiple levels that are pre-set in the data level; for example, the preset levels of the data level are arranged in descending order of judgment priority as "first year continuation", "multi-year continuation", "A-", "A+", "A", "B" and "C".

[0080] The appointment follow-up factor score is determined based on the appointment follow-up status of the insurance customer. When the appointment follow-up status indicates that the insurance customer has an appointment on the same day, the appointment follow-up factor score is the third score. When the appointment follow-up status indicates that the insurance customer does not have an appointment on the same day, the appointment follow-up factor score is the fourth score. The third score is greater than the fourth score.

[0081] The third score refers to the higher value of the appointment follow-up factor score x4 when the insurance customer has an appointment on the same day; for example, if customer A has an appointment on the same day, the appointment follow-up factor score x4 is the third score. The fourth score refers to the lower value of the appointment follow-up factor score x4 when the insurance customer does not have an appointment on the same day; for example, if customer B does not have an appointment on the same day, the appointment follow-up factor score x4 is the fourth score.

[0082] The follow-up interval factor score is determined based on the number of follow-up interval days for the insurance customer. The number of follow-up interval days is a plurality of preset number of days intervals. The follow-up interval factor score is determined based on the preset number of days intervals into which the follow-up interval days fall. Different preset number of days intervals correspond to different scores.

[0083] The preset number of days range refers to multiple pre-divided numerical ranges that follow the interval number of days T; for example, the preset number of days range includes the range of "greater than 7 days", the range of "4 to 7 days", the range of "2 to 3 days", and the range of "less than or equal to 1 day".

[0084] Among the above-mentioned optional methods, scoring rules are further set from four dimensions: customer tag category, data level, appointment follow-up status, and follow-up interval days. Higher scores are assigned to customers who belong to the high-quality customer group or those who have appointment follow-up on the same day, so that the comprehensive score is closer to the actual business value and the objectivity of the list ranking is improved.

[0085] The overall architecture of this embodiment is as follows: Figure 2As shown, the list generation subsystem and the list application subsystem constitute a closed-loop lifecycle. The list generation subsystem receives the central identifier through an HTTP trigger interface, performs two levels of control—local anti-reentrancy marking and distributed lock locking—and asynchronously submits it to the thread pool after successful verification. The core processing engine performs four stages in the asynchronous thread pool: data preparation, configuration loading, grouping and parallel processing, and result aggregation. Data preparation includes backing up yesterday's data and deleting old data; configuration loading includes team configuration and recommendation reason configuration; grouping and parallel processing divides the team into 5 groups and submits them to the thread pool for parallel execution; result aggregation includes statistical aggregation and releasing the distributed lock. After the list is generated, it is automatically routed and stored in the corresponding organization sub-table according to the organization code. The front-end verification layer of the list application subsystem performs mandatory verification of the dialing order and implements double blocking based on priority sequence numbers; the state-driven layer realizes the real-time flow of state 0→1 and 1→2, and automatically routes to the corresponding data sub-table according to the organization code; the data display layer performs de-identified display, red, yellow and blue label conversion, batch completion, intelligent degradation, and status statistics; the agent workbench dials according to the recommended list order, with real-time status flow and support for secondary follow-up.

[0086] Figure 3 This diagram illustrates the 13-stage time window division of the sales process. Based on the policy expiry date, and combined with the dynamically obtained advance callable days (default 30 days) from the agency dimension, an offset d (the difference in days between the current date and the policy expiry date) is calculated to dynamically map the list to the 13 stages. The list issuance date is obtained by parsing the 4th to 11th digits of the sales order number in yyyyMMdd format, without requiring additional database queries. The first day receives the highest score, and priority is automatically increased as the expiry date approaches. The 13 stages are as follows: the first allocation day (issuance day) corresponds to d≤-8; the first allocation period corresponds to d≤-8; the first 7 days correspond to d∈[-7,-4]; the first 3 days correspond to d=-3; the first 2 days correspond to d=-2; the first day corresponds to d=-1; the first day corresponds to d=0 and has the highest score; the first three days correspond to d∈[1,2]; the first five days correspond to d∈[3,4]; the first week corresponds to d∈[5,6]; the first two weeks correspond to d∈[7,14]; the release period corresponds to d∈[15,adv-12]; the last 10 days correspond to d≥adv-11 and are the sprint stage before expiration.

[0087] Figure 4The structure of the five-factor weighted scoring model is shown. The overall score is obtained by summing the scores of the five factors with their corresponding preset weights, rounded to two decimal places. The sales rhythm factor has a weight of 0.4, with 50 points for the first day, 40 points for the first three days, 30 points for the first five days, and 20 points for the rest. The customer labeling factor has a weight of 0.2, with 60 points for top-tier customers and 40 points for regular customers. The data level factor has a weight of 0.2, with 50 points for first-year renewal, multi-year renewal, and A- or A+, 40 points for A, 30 points for B, 20 points for C, and 10 points for the rest. The priority of data level determination is first-year renewal, multi-year renewal, A-, A+, A, B, C. The appointment follow-up factor has a weight of 0.1, with 80 points for same-day appointments and 20 points for no appointments. The follow-up interval factor has a weight of 0.1, with 30 points for more than 7 days, 20 points for 4 to 7 days, 10 points for 2 to 3 days, and 0 points for less than or equal to 1 day. The total weighting system is 1.0, with sales rhythm having the highest weight to highlight time sensitivity.

[0088] Figure 5 The process for mandatory verification of dialing order is illustrated. The input is the sales transaction number the agent wants to dial. The system queries the priority number of this transaction number in the recommended list. If no priority number is found (i.e., not in the list), the system further checks if the agent has a record in a pending dialing status or requiring a second dialing. If so, the dialing is blocked and a message is returned: "A recommended list of pending dialing exists; please dial first." If not, dialing is allowed. If a priority number is found (i.e., in the list), the system iterates through the pending dialing records to check for any customers with lower priority numbers who have not yet dialed. If so, dialing is blocked and a message is returned: "A higher priority customer has not yet dialed in the list; please dial in order." If not, dialing is allowed. The verification rule matrix covers four scenarios: blocking when the transaction number is not in the list but is in the pending dialing list; allowing when the transaction number is not in the list but is not in the pending dialing list; blocking when the transaction number is in the list and a higher priority customer has not yet dialed; and allowing when the transaction number is in the list and all previous customers have already dialed.

[0089] Figure 6 This illustrates a state machine transition process driven by real-time behavior in three states. The call log has three states during its lifecycle: State 0 (Pending Call), State 1 (Called), and State 2 (Requires a Second Call). The initial state is Pending Call. When an agent clicks to call and successfully loads customer information, the state changes from 0 to 1. This change is triggered before the agent navigates to the workbench and is completed in real-time within seconds. After the call ends and the end code is saved, if the scheduled re-contact time is the same day, the state changes from 1 to 2. This change is triggered after saving the contact record and is completed in real-time within seconds. Customers in State 2 re-enter the call queue and are restored to Pending Call. All state changes are triggered in real-time by the agent's actual business operations, not by scheduled batch processing; only data for the current day is updated to prevent cross-day errors; the update statement includes the source state as a query condition to prevent duplicate updates; the update logic is wrapped in independent exception handling, so update failures do not block the agent's main business process.

[0090] Figure 7 The complete technical flow sequence diagram is shown. After the agent clicks on a customer, the front-end calls the verification interface to perform verification. The verification interface verifies the dialing order (based on priority number). If the verification passes, dialing is allowed, and the agent is redirected to the workbench. The workbench interface successfully loads the customer information, and the status is updated from "Pending Call" to "Dialed". After the agent's call ends, the end code is saved and a reservation is made for the same day. The end code interface determines that the reservation time is for the same day and the current status is "Dialed", and updates the status from "Dialed" to "Requires a Second Call".

[0091] In another embodiment of the insurance customer list management method based on dynamic weighting and behavior-driven methods in this example, the method includes: S10. Responding to the agent's call request to the target insurance customer, query the target insurance customer's priority number in the referral list based on the target insurance customer's sales serial number; if no priority number is found, check if the agent has any insurance customer records in the status of "pending call" or "requires a second call". If an insurance customer record exists, block the call request; if no insurance customer record exists, allow the call request; if a priority number is found, iterate through the insurance customer records in the referral list in the status of "pending call" or "requires a second call", and check if there are any insurance customer records that have not been called with a priority number lower than the target insurance customer's priority number. If an insurance customer record that has not been called exists, block the call request; if no insurance customer record that has not been called exists, allow the call request.

[0092] S20. In response to an agent's dialing operation for a target insurance customer in the recommended list, the agent sets the query condition that the source status is equal to "Pending Call" in the database operation for status update. The agent updates the target insurance customer's status from "Pending Call" to "Dialed" only if the target insurance customer's current status is "Pending Call". In response to an agent's appointment operation for a target insurance customer, the agent sets the query condition that the source status is equal to "Dialed" in the database operation for status update. The agent updates the target insurance customer's status from "Dialed" to "Requires Second Call" only if the target insurance customer's current status is "Dialed". In response to an agent's completion of a second call for a target insurance customer, the agent sets the query condition that the source status is equal to "Requires Second Call" in the database operation for status update. The agent updates the target insurance customer's status from "Requires Second Call" to "Dialed" only if the target insurance customer's current status is "Requires Second Call".

[0093] S30. In response to an event where a call request from an agent to a target insurance customer in the recommended list is blocked, accumulate the number of times the target insurance customer is blocked within a preset time window; when the number of blocking attempts reaches a preset blocking attempt threshold, mark the target insurance customer as a customer to be released and record the marking time; in response to a situation where the difference between the marking time and the current time exceeds a preset retention time threshold, remove the target insurance customer from the recommended list and release the priority number corresponding to the target insurance customer; after releasing the priority number corresponding to the target insurance customer, perform a priority number decrement update operation for insurance customers in the recommended list whose priority numbers are greater than the released priority numbers.

[0094] S40. In response to the status of the target insurance customer in the recommended list being updated from pending to dialed, obtain the duration of the current status and the current comprehensive score of the target insurance customer; calculate the priority adjustment factor for each insurance customer in the recommended list who has not dialed based on the duration of the current status, the current comprehensive score, and the total number of insurance customers in the recommended list who have not dialed; and perform a reordering operation on the priority sequence number of the insurance customers in the recommended list who have not dialed based on the priority adjustment factor.

[0095] The formula for calculating the priority adjustment factor is: in, Indicates the first Priority adjustment factor for customers who have not yet made an insurance call. This indicates the preset adjustment coefficient. Indicates the sensitivity index of the overall score. The sensitivity coefficient represents the duration of the state. Indicates the duration decay index, Indicates the first The current overall score of the insurance customer who has not yet made an insurance call. This represents the maximum combined score of all insurance clients in the recommended list. Indicates the first Duration of the current status of customers who have not yet made an insurance call.

[0096] It should be noted that the formula for the priority adjustment factor couples the relative level of the comprehensive score with the cumulative effect of the duration of the status, measuring in a non-linear manner the priority compensation intensity that each customer who has not yet made an insurance call should receive in the current recommendation sequence. Specifically, a higher relative level of the comprehensive score indicates a greater original business value for the customer, thus warranting stronger priority compensation; a longer duration of the status indicates a longer waiting time for the customer to be served, and to prevent customer churn due to long waiting times, stronger priority compensation is also warranted. To accurately characterize the combined effect of these two factors, the formula introduces three control parameters: a comprehensive score sensitivity index, a status duration sensitivity coefficient, and a duration decay index. This allows the adjustment factor to grow rapidly with increasing relative level of the comprehensive score, while approaching saturation with increasing status duration, thus avoiding the unlimited amplification of priority compensation.

[0097] S50. Adjust factors according to priority. Calculate the adjusted priority score for each customer in the recommended list who has not yet made an insurance call. The formula for calculating the adjusted priority score is as follows: in, Indicates the first The value corresponding to the current priority number of each customer who has not yet made an insurance call, before the priority adjustment factor is calculated. Indicates the first The adjusted priority scores of each customer who has not yet made an insurance call are used to re-sort the customers from smallest to largest according to their adjusted priority scores. An updated priority number is then assigned to each customer in the re-sorted order. The updated priority number is stored in the database and overwrites the corresponding historical priority number in the recommended list.

[0098] The technical solution of this embodiment solves the technical problems of high-value customers not being given priority service due to the inability of static sorting rules to dynamically adjust priorities with the sales process, and the lag of list status tracking behind actual agent operations. This is achieved by querying the priority number of the target insurance customer in the recommended list and performing double blocking verification when responding to the agent's call request; performing status updates with source status query conditions when responding to the agent's call and appointment operations; accumulating the number of blocking attempts when a call request is blocked and releasing the corresponding customer and priority number when a threshold is reached; and calculating and re-prioritizing customers who have not called based on the comprehensive score and status duration triggered by status change events.

[0099] Figure 8This diagram illustrates the structure of an embodiment of an insurance customer list management system 200 based on dynamic weighting and behavior-driven principles provided by the present invention. Figure 8 As shown, the insurance customer list management system 200 based on dynamic weighting and behavior-driven analysis includes: The acquisition module 201 is used to determine the sales rhythm stage to which each insurance customer belongs based on the offset between the policy expiration date and the current date, and to determine the sales rhythm factor score based on the sales rhythm stage. The determination module 202 is used to determine the customer tag factor score, data level factor score, appointment follow-up factor score and follow-up interval factor score respectively based on the customer tag category, data level, appointment follow-up status and follow-up interval days of each insurance customer; The calculation module 203 is used to calculate the sales rhythm factor score, the customer tag factor score, the data level factor score, the appointment follow-up factor score, and the follow-up interval factor score by weighting and summing them with their corresponding preset weights to obtain the comprehensive score of each insurance customer. The generation module 204 is used to assign a priority number to each insurance customer based on the comprehensive score of each insurance customer and generate a recommendation list. The processing module 205 is used to respond to the call request from the agent to the target insurance customer. If the target insurance customer is not in the recommended list and the agent has an insurance customer whose status is pending or requires a second call, the call request is blocked. If the target insurance customer is in the recommended list and there is an insurance customer in the recommended list whose priority number is lower than that of the target insurance customer who has not been called, the call request is blocked. Otherwise, the call request is allowed. The update module 206 is used to respond to the agent's dialing operation for the target insurance customer in the recommended list, and update the status of the target insurance customer from pending dialing to dialed; and to respond to the agent's appointment operation for the target insurance customer, and update the status of the target insurance customer from dialed to requiring a second dialing.

[0100] In an alternative embodiment, the generation module 204 is specifically used for: Based on the policy expiration time range and the lower limit of the scoring threshold, insurance customers are filtered using database-level conditions to obtain the first filtering result; The first filtering result is matched with sales rhythm tags and scoring thresholds one by one to obtain the second filtering result; Based on the team configuration, the insurance customers in the second filtering result are grouped into multiple parallel processing groups, and scoring, sorting and truncation are performed on each parallel processing group to obtain the candidate list for each agent. The candidate lists processed by each parallel processing group are merged to obtain the recommended list.

[0101] In an alternative embodiment, the generation module 204 is specifically used for: Get the team configuration of all teams under the organization to which the agent belongs, and get the list of agents included in each team according to the team configuration; A round-robin equal distribution algorithm was used to divide all teams into multiple parallel processing groups; For each parallel processing group, the insurance customers in the second filtering result are retrieved page by page, team by team and seat by seat. The comprehensive score of the insurance customers retrieved on each page is calculated and sorted according to the comprehensive score. After sorting, the number of insurance customers with the highest comprehensive scores as specified in the configuration item is truncated according to the seat dimension and retained as the candidate list for each seat.

[0102] In an alternative embodiment, the processing module 205 is specifically used for: Query the priority number of the target insurance customer in the recommended list; If the priority number is not found, check if there is an insurance customer record in the agent's status as pending call or requiring a second call. If it exists, block the call request; otherwise, allow the call request. When the priority number is found, the system iterates through the insurance customer records in the recommended list that are in a state of pending call or require a second call, and checks whether there are any insurance customer records that have not been called but whose priority number is lower than that of the target insurance customer. If such a record exists, the call request is blocked; otherwise, the call request is allowed.

[0103] In an alternative embodiment, the update module 206 is specifically used for: In response to the agent's dialing operation on the target insurance customer in the recommended list, the query condition that the source status is equal to the pending dialing is set in the database operation of performing status update. The status of the target insurance customer is updated from pending dialing to dialed only when the current status of the target insurance customer is pending dialing. In response to the agent's appointment operation for the target insurance customer, the query condition that the source status is equal to "called" is set in the database operation of status update. Only when the current status of the target insurance customer is "called" will the status of the target insurance customer be updated from "called" to "needs to be called again".

[0104] In one alternative embodiment, the acquisition module 201 is specifically used for: Using the policy expiration date as the base date, the sales rhythm stage to which the insurance customer belongs is determined based on the range in which the difference in the number of days between the current date and the base date falls; wherein, the sales rhythm stage includes multiple preset stages, and each preset stage corresponds to a different range of the difference in the number of days.

[0105] In an alternative embodiment, the determining module 202 is specifically used for: The customer tagging factor score is determined based on the customer tagging category of the insurance customer. When the customer tagging category indicates that the insurance customer belongs to a preset high-quality customer group, the customer tagging factor score is a first score. When the customer tagging category indicates that the insurance customer does not belong to the preset high-quality customer group, the customer tagging factor score is a second score. The first score is greater than the second score. The data level factor score is determined based on the data level of the insurance customer. The data level includes multiple preset levels, which are arranged in order of judgment priority. The preset levels to which the data level of the insurance customer belongs are judged in order of judgment priority from high to low. The data level factor score is determined based on the judgment result. Different preset levels correspond to different scores. The appointment follow-up factor score is determined based on the appointment follow-up status of the insurance customer. When the appointment follow-up status indicates that the insurance customer has an appointment on the same day, the appointment follow-up factor score is the third score. When the appointment follow-up status indicates that the insurance customer has no appointment on the same day, the appointment follow-up factor score is the fourth score. The third score is greater than the fourth score. The follow-up interval factor score is determined based on the number of follow-up interval days for the insurance customer. The number of follow-up interval days is a plurality of preset number of days intervals. The follow-up interval factor score is determined based on the preset number of days intervals into which the follow-up interval days fall. Different preset number of days intervals correspond to different scores.

[0106] It should be noted that the beneficial effects of the insurance customer list management system 200 based on dynamic weighting and behavior-driven principles provided in the above embodiments are the same as those of the insurance customer list management method based on dynamic weighting and behavior-driven principles, and will not be repeated here. Furthermore, the system provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to the actual situation to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process is detailed in the method embodiments, and will not be repeated here.

[0107] The insurance customer list management system 200 based on dynamic weighting and behavior-driven of the present invention can be a computer program (including program code) running on a computer device. For example, the insurance customer list management system 200 based on dynamic weighting and behavior-driven of the present invention is an application software that can be used to execute the corresponding steps in the insurance customer list management method based on dynamic weighting and behavior-driven of the present invention.

[0108] In some embodiments, the insurance customer list management system 200 based on dynamic weighting and behavior-driven principles of the present invention can be implemented using a combination of hardware and software. As an example, the insurance customer list management system 200 based on dynamic weighting and behavior-driven principles of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the insurance customer list management method based on dynamic weighting and behavior-driven principles of the present invention. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0109] The modules described in the embodiments of this invention can be implemented in software or hardware. The names of the modules are not, in some cases, limiting the scope of the module itself.

[0110] An electronic device according to an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the above-mentioned insurance customer list management methods based on dynamic weighting and behavior-driven approaches. That is, an electronic device according to an embodiment of the present invention may include, but is not limited to: a processor and a memory; the memory is used to store the computer program; the processor is used to execute the insurance customer list management method based on dynamic weighting and behavior-driven approaches shown in any embodiment of the present invention by calling the computer program.

[0111] In one alternative embodiment, an electronic device is provided, such as Figure 9 As shown, Figure 9The illustrated electronic device 4000 includes a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, via a bus 4002. Optionally, the electronic device 4000 may further include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 4004 is not limited to one type, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0112] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. Processor 4001 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0113] Bus 4002 may include a path for transmitting information between the aforementioned components. Bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 4002 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 9 The bus 4002 is represented by only one thick line, but this does not mean that there is only one bus or one type of bus.

[0114] The memory 4003 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0115] The memory 4003 stores application code (computer program) for executing the present invention, and its execution is controlled by the processor 4001. The processor 4001 executes the application code stored in the memory 4003 to implement the content shown in the foregoing method embodiments.

[0116] Among them, electronic devices can also be terminal devices. A terminal device can be any terminal device that can install applications and access web pages through applications, including at least one of smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, smart TVs, and smart in-vehicle devices.

[0117] It should be noted that, Figure 9 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0118] An embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the above-mentioned insurance customer list management methods based on dynamic weighting and behavior-driven approaches.

[0119] Alternatively, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, a floppy disk, and an optical data storage device, etc.

[0120] In an exemplary embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the aforementioned insurance customer list management method based on dynamic weighting and behavior-driven principles.

[0121] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0122] It should be understood that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0123] The computer-readable storage medium provided in this invention can be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0124] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the method shown in the above embodiments.

[0125] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0126] It should be noted that the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of this application described herein can be implemented in an order other than that shown or described.

[0127] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this invention can be specifically implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, this invention can also be implemented as a computer program product contained in one or more computer-readable media, which includes computer-readable program code.

[0128] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for managing insurance customer lists based on dynamic weighting and behavior-driven approaches, characterized in that, include: Based on the offset between each insurance customer's policy expiration date and the current date, determine the sales rhythm stage to which each insurance customer belongs, and determine the sales rhythm factor score based on the sales rhythm stage. Based on each insurance customer's customer tag category, data level, appointment follow-up status, and follow-up interval days, customer tag factor score, data level factor score, appointment follow-up factor score, and follow-up interval factor score are determined respectively. The sales rhythm factor score, customer tag factor score, data level factor score, appointment follow-up factor score, and follow-up interval factor score are weighted and summed with their respective preset weights to obtain the comprehensive score for each insurance customer. Based on the comprehensive score of each insurance customer, a priority number is assigned to each insurance customer, and a recommendation list is generated. In response to a call request from an agent to a target insurance customer, if the target insurance customer is not in the recommended list and the agent has an insurance customer whose status is pending or requires a second call, the call request is blocked; if the target insurance customer is in the recommended list and there is an insurance customer in the recommended list whose priority number is lower than that of the target insurance customer who has not yet been called, the call request is blocked; otherwise, the call request is allowed. In response to the agent's dialing operation on the target insurance customer in the recommended list, the status of the target insurance customer is updated from pending dialing to dialed; in response to the agent's appointment operation on the target insurance customer, the status of the target insurance customer is updated from dialed to require a second dialing.

2. The insurance customer list management method based on dynamic weighting and behavior-driven approach according to claim 1, characterized in that, The steps for assigning a priority number to each insurance customer based on their overall score and generating a recommendation list include: Based on the policy expiration time range and the lower limit of the scoring threshold, insurance customers are filtered using database-level conditions to obtain the first filtering result; The first filtering result is matched with sales rhythm tags and scoring thresholds one by one to obtain the second filtering result; Based on the team configuration, the insurance customers in the second filtering result are grouped into multiple parallel processing groups, and scoring, sorting and truncation are performed on each parallel processing group to obtain the candidate list for each agent. The candidate lists processed by each parallel processing group are merged to obtain the recommended list.

3. The insurance customer list management method based on dynamic weighting and behavior-driven approach according to claim 2, characterized in that, Based on the team configuration, the insurance customers in the second filtering result are grouped into multiple parallel processing groups, and scoring, sorting, and truncation are performed on each parallel processing group to obtain the candidate list for each agent. This includes the following steps: Get the team configuration of all teams under the organization to which the agent belongs, and get the list of agents included in each team according to the team configuration; A round-robin equal distribution algorithm was used to divide all teams into multiple parallel processing groups; For each parallel processing group, the insurance customers in the second filtering result are retrieved page by page, team by team and seat by seat. The comprehensive score of the insurance customers retrieved on each page is calculated and sorted according to the comprehensive score. After sorting, the number of insurance customers with the highest comprehensive scores as specified in the configuration item is truncated according to the seat dimension and retained as the candidate list for each seat.

4. The insurance customer list management method based on dynamic weighting and behavior-driven approach according to claim 1, characterized in that, In response to a call request from an agent to a target insurance customer, if the target insurance customer is not in the recommended list and the agent has an insurance customer whose status is pending or requires a second call, the call request is blocked; if the target insurance customer is in the recommended list and there is an uncalled insurance customer in the recommended list with a priority number lower than the target insurance customer's priority number, the call request is blocked; otherwise, the call request is allowed. The steps include: Query the priority number of the target insurance customer in the recommended list; If the priority number is not found, check if there is an insurance customer record in the agent's status as pending call or requiring a second call. If it exists, block the call request; otherwise, allow the call request. When the priority number is found, the system iterates through the insurance customer records in the recommended list that are in a state of pending call or require a second call, and checks whether there are any insurance customer records that have not been called but whose priority number is lower than that of the target insurance customer. If such a record exists, the call request is blocked; otherwise, the call request is allowed.

5. The insurance customer list management method based on dynamic weighting and behavior-driven approach according to claim 4, characterized in that, The step of updating the status of the target insurance customer from pending call to called in response to the agent's call to the target insurance customer in the recommended list includes: In response to the agent's dialing operation on the target insurance customer in the recommended list, the query condition that the source status is equal to the pending dialing is set in the database operation of performing status update. The status of the target insurance customer is updated from pending dialing to dialed only when the current status of the target insurance customer is pending dialing. The step of updating the target insurance customer's status from "called" to "requires a second call" in response to the agent's appointment request to the target insurance customer includes: In response to the agent's appointment operation for the target insurance customer, the query condition that the source status is equal to "called" is set in the database operation of status update. Only when the current status of the target insurance customer is "called" will the status of the target insurance customer be updated from "called" to "needs to be called again".

6. The insurance customer list management method based on dynamic weighting and behavior-driven approach according to claim 1, characterized in that, The steps to determine the sales pace stage for each insurance customer include: Using the policy expiration date as the base date, the sales rhythm stage to which the insurance customer belongs is determined based on the range in which the difference in the number of days between the current date and the base date falls; wherein, the sales rhythm stage includes multiple preset stages, and each preset stage corresponds to a different range of the difference in the number of days.

7. The insurance customer list management method based on dynamic weighting and behavior-driven approach according to claim 1, characterized in that, The steps for determining the customer tagging factor score, data level factor score, appointment follow-up factor score, and follow-up interval factor score based on each insurance customer's customer tagging category, data level, appointment follow-up status, and follow-up interval days include: The customer tagging factor score is determined based on the customer tagging category of the insurance customer. When the customer tagging category indicates that the insurance customer belongs to a preset high-quality customer group, the customer tagging factor score is a first score. When the customer tagging category indicates that the insurance customer does not belong to the preset high-quality customer group, the customer tagging factor score is a second score. The first score is greater than the second score. The data level factor score is determined based on the data level of the insurance customer. The data level includes multiple preset levels, which are arranged in order of judgment priority. The preset levels to which the data level of the insurance customer belongs are judged in order of judgment priority from high to low. The data level factor score is determined based on the judgment result. Different preset levels correspond to different scores. The appointment follow-up factor score is determined based on the appointment follow-up status of the insurance customer. When the appointment follow-up status indicates that the insurance customer has an appointment on the same day, the appointment follow-up factor score is the third score. When the appointment follow-up status indicates that the insurance customer has no appointment on the same day, the appointment follow-up factor score is the fourth score. The third score is greater than the fourth score. The follow-up interval factor score is determined based on the number of follow-up interval days for the insurance customer. The number of follow-up interval days is a plurality of preset number of days intervals. The follow-up interval factor score is determined based on the preset number of days intervals into which the follow-up interval days fall. Different preset number of days intervals correspond to different scores.

8. An insurance customer list management system based on dynamic weighting and behavior-driven methods, characterized in that, include: The acquisition module is used to determine the sales rhythm stage to which each insurance customer belongs based on the offset between the policy expiration date and the current date, and to determine the sales rhythm factor score based on the sales rhythm stage. The determination module is used to determine the customer tag factor score, data level factor score, appointment follow-up factor score, and follow-up interval factor score for each insurance customer based on their customer tag category, data level, appointment follow-up status, and follow-up interval days, respectively. The calculation module is used to sum the sales rhythm factor score, the customer tag factor score, the data level factor score, the appointment follow-up factor score, and the follow-up interval factor score with their respective preset weights to obtain the comprehensive score of each insurance customer. The generation module is used to assign a priority number to each insurance customer based on their comprehensive score and generate a recommendation list. The processing module is used to respond to the call request from the agent to the target insurance customer. If the target insurance customer is not in the recommended list and the agent has an insurance customer whose status is pending call or requires a second call, the call request is blocked. If the target insurance customer is in the recommended list and there is an insurance customer in the recommended list whose priority number is lower than that of the target insurance customer who has not been called, the call request is blocked. Otherwise, the call request is allowed. The update module is used to respond to the agent's dialing operation for the target insurance customer in the recommended list, and update the status of the target insurance customer from pending dialing to dialed; and to respond to the agent's appointment operation for the target insurance customer, and update the status of the target insurance customer from dialed to requiring a second dialing.

9. An electronic device, characterized in that, The electronic device includes a processor coupled to a memory, the memory storing at least one computer program, which is loaded and executed by the processor to enable the electronic device to implement the insurance customer list management method based on dynamic weighting and behavior-driven as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one computer program, which, when executed by a processor, implements the insurance customer list management method based on dynamic weighting and behavior-driven as described in any one of claims 1 to 7.

Images