Customer relationship matching methods, devices, equipment and storage media

Abstract

This application discloses a customer relationship matching method, apparatus, device, and storage medium. The method includes the following steps: acquiring historical evaluation data of a customer manager and historical evaluation data of a customer regarding the customer manager, wherein the evaluation factors in the historical evaluation data and the historical evaluation data are the same; inputting the historical evaluation data and the historical evaluation data into a preset matching model to calculate a first matching value between the customer and the customer manager, wherein the first matching value is calculated based on the customer manager's ability value corresponding to the evaluation factor and the customer's attention to the evaluation factor; selecting a second matching value from the first matching values ​​that is greater than a preset threshold, and recommending the customer manager corresponding to the second matching value to the customer. This application accurately matches customers with customer managers who have a high degree of compatibility with the customer.

Description

Technical Field

[0001] This application relates to the field of financial management technology, and in particular to a customer relationship matching method, apparatus, device and storage medium. Background Technology

[0002] With the continuous development of banking services and the increasing complexity of the customer base, how to increase customer loyalty and improve customer satisfaction has become an important issue. Customer managers, as crucial entities providing services to customers within the banking system, are responsible for identifying customer needs and resolving their problems promptly. Currently, customer manager assignment is mostly done manually. However, since each customer's needs are different and the service capabilities of customer managers vary, when the compatibility between the customer manager and the customer is poor, the customer manager cannot provide adequate service, and customer satisfaction with banking services will decrease. Therefore, accurately matching customers with customer managers who are a good fit has become a pressing issue for businesses. Summary of the Invention

[0003] The main purpose of this application is to provide a customer relationship matching method, apparatus, device and storage medium, which aims to solve the technical problem in the related art that it is impossible to accurately match customers with account managers who are highly compatible with the customers themselves.

[0004] To achieve the above objectives, embodiments of this application provide a customer relationship matching method, the method comprising:

[0005] Obtain historical evaluation data of account managers and historical evaluation data of account managers by customers, wherein the evaluation factors in the historical evaluation data and the historical evaluation data are the same.

[0006] The evaluated historical data and the evaluated historical data are input into a preset matching model to calculate a first matching value between the customer and the customer manager. The first matching value is calculated based on the customer manager's ability value corresponding to the evaluation factor and the customer's attention to the evaluation factor.

[0007] Select a second matching value from the first matching values ​​that is greater than a preset threshold, and recommend the account manager corresponding to the second matching value to the customer.

[0008] In one possible implementation of this application, the step of inputting the evaluated historical data and the evaluation historical data into a preset matching model to calculate a first matching value between the customer and the account manager includes:

[0009] The evaluated historical data and the evaluation historical data are input into a preset matching model. Based on the preset matching model, the evaluated historical data is vectorized to obtain the customer manager capability vector, and the evaluation historical data is vectorized to obtain the customer attention factor vector.

[0010] The account manager's capability vector and the customer's attention factor vector are matched and calculated to obtain the first matching value between the customer and the account manager.

[0011] In one possible implementation of this application, the step of vectorizing the evaluated historical data to obtain the account manager competence vector includes:

[0012] The questionnaire data of multiple evaluation factors are extracted from the historical data being evaluated. The evaluation factors include service attitude, excessive disturbance, clear explanation, risk warning, excessive marketing, asset allocation suggestions, communication of concepts, attention to needs, timely response, proactive contact, and follow-up care.

[0013] Based on the evaluation samples of individual evaluation factors in the questionnaire data, the individual competency value of the account manager is determined. The calculation method for the individual competency value of the account manager is as follows:

[0014]

[0015] Among them, sco 因子 This indicates the individual competency score of the account manager, sco 常量1 sco is the preset first constant value. 常量2 N is the preset second constant value. 好评 N represents the total sample size of positive reviews from account managers. 差评 N represents the total sample size of negative reviews from account managers. 总量 W represents the total sample size of account managers. 差评 As preset weights, avg 客户经理 The mean of the total sample scores for a single rating factor in the questionnaire data of the account managers;

[0016] The individual competency values ​​of the account manager corresponding to multiple evaluation factors are combined to obtain the account manager competency vector.

[0017] In one possible implementation of this application, the step of vectorizing the historical evaluation data to obtain a customer attention factor vector includes:

[0018] Extract questionnaire data for multiple evaluation factors from the historical evaluation data;

[0019] Based on the evaluation samples of individual evaluation factors in the questionnaire data, the individual customer evaluation value is determined, wherein the calculation method for the individual customer evaluation value is as follows:

[0020]

[0021] Here, "binary positive review" and "binary negative review" represent the preset third constant value when the questionnaire containing the evaluation factor is positive, and the preset fourth constant value when the questionnaire containing the evaluation factor is negative. val represents the value of the current evaluation factor among all the evaluation factors selected in the questionnaire. 因子 The value is the sum of the individual evaluation factors across all questionnaires;

[0022] The individual customer evaluation values ​​are normalized to obtain individual customer factor attention values;

[0023] The customer attention factor vector is obtained by combining the customer attention factor values ​​of multiple evaluation factors.

[0024] In one possible implementation of this application, the step of matching the account manager's capability vector and the customer's attention factor vector to calculate a first matching value between the customer and the account manager includes:

[0025] Determine if there is any evaluation interaction data between the current customer and the account manager;

[0026] If evaluation interaction data exists, a matching calculation is performed on the account manager's capability vector and the customer's attention factor vector to obtain a first matching value between the customer and the account manager. The formula for calculating the first matching value is as follows:

[0027]

[0028] Among them, fit 客户&客户经理 Indicates the first matching value. This represents the product of the account manager's capability vector and the customer's attention factor vector, avg 客户-客户经理 This represents the average score given by current customers to their account managers, and "revise" indicates the preset adjustment value.

[0029] In one possible implementation of this application, after the step of matching the account manager's capability vector and the customer's attention factor vector to calculate a first matching value between the customer and the account manager, if evaluation interaction data exists, the method includes:

[0030] If no evaluation interaction data exists, a matching calculation is performed on the account manager's capability vector and the customer's attention factor vector to obtain a first matching value between the customer and the account manager. The formula for calculating the first matching value is as follows:

[0031]

[0032] Among them, fit 客户&客户经理 Indicates the first matching value. This represents the product of the account manager's capability vector and the customer attention factor vector, with revise representing a preset adjustment value.

[0033] In one possible implementation of this application, the preset adjustment value is associated with the negative review tag submitted by the customer. If the customer has previously submitted a negative review tag, the preset adjustment value is increased accordingly.

[0034] This application also provides a customer relationship matching device, the customer relationship matching device further comprising:

[0035] The acquisition module is used to acquire the historical evaluation data of the account manager and the historical evaluation data of the customer towards the account manager, wherein the evaluation factors in the historical evaluation data and the historical evaluation data are the same.

[0036] The calculation module is used to input the evaluated historical data and the evaluation historical data into a preset matching model to calculate a first matching value between the customer and the customer manager. The first matching value is calculated based on the customer manager's ability value corresponding to the evaluation factor and the customer's attention to the evaluation factor.

[0037] The selection module is used to select a second matching value from the first matching values ​​that is greater than a preset threshold, and recommend the account manager corresponding to the second matching value to the customer.

[0038] This application also provides a customer relationship matching device, which is an entity node device. The customer relationship matching device includes: a memory, a processor, and a program of the customer relationship matching method stored in the memory and executable on the processor. When the program of the customer relationship matching method is executed by the processor, it can implement the steps of the customer relationship matching method as described above.

[0039] To achieve the above objectives, a storage medium is also provided, on which a customer relationship matching program is stored, which, when executed by a processor, implements the steps of any of the customer relationship matching methods described above.

[0040] This application provides a customer relationship matching method, apparatus, device, and storage medium. Addressing the technical problem in related technologies of failing to accurately match customers with account managers who are highly compatible with them, this application obtains historical evaluation data of account managers and historical evaluation data of customers regarding account managers, wherein the evaluation factors in the historical evaluation data and the historical evaluation data are the same; the historical evaluation data and the historical evaluation data are input into a preset matching model to calculate a first matching value between the customer and the account manager, wherein the first matching value is calculated based on the account manager's ability value corresponding to the evaluation factor and the customer's attention to the evaluation factor; a second matching value greater than a preset threshold is selected from the first matching values, and the account manager corresponding to the second matching value is recommended to the customer. In this application, the historical evaluation data of the account manager and the historical evaluation data of the customer towards the account manager are obtained. Since the evaluation factors are the same, the customer and the account manager can be associated based on the same evaluation factors. The historical evaluation data and the historical evaluation data are input into a preset matching model to obtain the first matching value between the customer and the account manager. Since the obtained first matching value is calculated based on the ability value of the account manager corresponding to the evaluation factor and the customer's attention to the evaluation factor, the size of the first matching value can reflect the compatibility between the customer and the account manager. The larger the first matching value, the higher the compatibility between the customer and the account manager, so that the customer can be accurately matched with an account manager with a high degree of compatibility. Attached Figure Description

[0041] Figure 1 This is a flowchart illustrating the first embodiment of the customer relationship matching method of this application;

[0042] Figure 2 This is a schematic diagram of the calculation framework of the preset matching model involved in the customer relationship matching method of this application;

[0043] Figure 3 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of this application;

[0044] Figure 4 This is a schematic diagram of the data interaction process involved in the customer relationship matching method of this application;

[0045] Figure 5 This is a schematic diagram illustrating the evaluation factors and dimensions involved in the customer relationship matching method of this application;

[0046] Figure 6 This is a detailed calculation flowchart of the preset matching model involved in the customer relationship matching method of this application. Detailed Implementation

[0047] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0048] This application provides a customer relationship matching method. In the first embodiment of the customer relationship matching method of this application, refer to... Figure 1 and Figure 2 The method includes:

[0049] Step S10: Obtain the historical evaluation data of the account manager and the historical evaluation data of the customer towards the account manager, wherein the evaluation factors in the historical evaluation data and the historical evaluation data are the same.

[0050] Step S20: Input the evaluated historical data and the evaluation historical data into a preset matching model to calculate the first matching value between the customer and the customer manager. The first matching value is calculated based on the customer manager's ability value corresponding to the evaluation factor and the customer's attention to the evaluation factor.

[0051] Step S30: Select a second matching value from the first matching values ​​that is greater than a preset threshold, and recommend the account manager corresponding to the second matching value to the customer.

[0052] In this embodiment, the application scenario can be:

[0053] When customers go to the bank to conduct business, the existing methods of assigning or recommending account managers to them based on their needs are mostly based on the importance of the customer. Important customers are assigned to dedicated account managers. However, the banking industry has a large customer base, while the number of account managers is limited. Some customers want attention and want to learn about financial products, while others do not want to be disturbed. Therefore, with limited resources, it is very important to rationally assign account managers to serve customers.

[0054] This embodiment aims to: correlate customer attention with account manager competence by using evaluation data and the data being evaluated, thereby accurately matching customers with account managers who are a good fit for them.

[0055] The specific steps are as follows:

[0056] Step S10: Obtain the historical evaluation data of the account manager and the historical evaluation data of the customer towards the account manager, wherein the evaluation factors in the historical evaluation data and the historical evaluation data are the same.

[0057] As an example, a customer relationship matching method can be applied to a customer relationship matching device, which is subordinate to a customer relationship matching system, which in turn is a customer relationship matching equipment.

[0058] As an example, the historical data being evaluated includes online evaluation data and offline evaluation data. Online evaluation data can be data collected after online evaluations are conducted via mobile phones, computers, or other devices; offline evaluation data can be data evaluated in the form of paper questionnaires.

[0059] As an example, the historical data to be evaluated can be obtained by using all historical evaluation questionnaire data for account managers, both online and offline, within the system. This data covers both online and offline business scenarios of customers and includes the corresponding evaluation tags selected from the questionnaire evaluations.

[0060] As an example, historical evaluation data includes both online and offline evaluation data. Historical evaluation data is based on the customer's perspective and is a statistical analysis of the customer's evaluation of the account manager. Since customers may conduct different business at different branches and online apps, and different account managers may be assigned to different business scenarios, customers can evaluate all account managers. By collecting these evaluations, the corresponding online and offline evaluation data can be obtained.

[0061] As an example, when obtaining historical evaluation data, after removing the "stable service quality (frequent changes in account managers)" label based on business scenarios and customer needs, 11 service evaluation factor dimensions related to account managers are obtained. Account managers with less evaluation data will be marked with the "less data" prompt factor, which also facilitates the subsequent calculation process.

[0062] As an example, the evaluation factors include 11 dimensions of evaluation indicators. These factors include service attitude, excessive intrusion, clarity of explanation, risk warning, excessive marketing, asset allocation advice, communication of concepts, attention to needs, timely response, proactive contact, and follow-up care. By using the evaluation data of multiple evaluation factors, the customer's concerns are correlated with the corresponding ability scores of the account manager's evaluation factors, thereby achieving precise matching. For example, if a customer values ​​service attitude, and needs an account manager with a good service attitude, then an account manager with a high service attitude evaluation score can be matched to serve that customer, thus achieving the effect of precise matching.

[0063] Step S20: Input the evaluated historical data and the evaluation historical data into a preset matching model to calculate the first matching value between the customer and the customer manager. The first matching value is calculated based on the customer manager's ability value corresponding to the evaluation factor and the customer's attention to the evaluation factor.

[0064] As an example, the preset matching model is a mathematical model based on multiple functions. Before inputting the model, the evaluated historical data and the evaluation historical data can be converted into the data format required by the preset matching model. The preset matching model then performs matching calculations on the input evaluated historical data and the evaluation historical data to obtain the corresponding matching value between the account manager and the customer.

[0065] As an example, the first matching value is a calculated numerical value that reflects the degree of matching between the customer and the account manager. Since the first matching value is calculated based on the account manager's ability value corresponding to the evaluation factor and the customer's attention to the evaluation factor, matching the customer with an account manager through the first matching value has high reference value.

[0066] As an example, the way customers are matched with account managers can be one customer to multiple account managers, or multiple customers to multiple account managers; there is no specific limitation.

[0067] As an example, the capability value corresponding to the evaluation factor can reflect the account manager's expertise in a particular business, thus corresponding to the customer's focus on the evaluation factor and achieving a precise match.

[0068] Step S30: Select a second matching value from the first matching values ​​that is greater than a preset threshold, and recommend the account manager corresponding to the second matching value to the customer.

[0069] As an example, the first matching value is the matching value between the account manager and the customer calculated by the preset matching model, where each account manager corresponds to one first matching value.

[0070] As an example, the preset threshold is a matching threshold set by the user. The preset threshold can be 20, 30, etc., or it can be a percentage value in the first matching value (e.g., 10%). When the preset threshold is a percentage value, the first 10% of the data in the first matching value is taken as the second matching value. It can be adjusted according to the user's needs, and there is no specific limitation.

[0071] As an example, the second matching value is a subset of the top-ranked values ​​from the first matching value. Based on the second matching value, a corresponding account manager can be matched, thus completing the process of accurate recommendation / matching for the customer.

[0072] As an example, the data interaction process involved in this application is illustrated in the following diagram: Figure 4 As shown, online and offline evaluation data are input into a preset matching model, and the resulting first matching value is uploaded to the back-end system. A threshold is set for the first matching value, and customer managers who meet the user's conditions are selected based on the first matching value. They are then reassigned to the head office, branches, and sub-branches, thereby completing a precise match with the customer.

[0073] This application provides a customer relationship matching method, apparatus, device, and storage medium. Addressing the technical problem in related technologies of failing to accurately match customers with account managers who are highly compatible with them, this application obtains historical evaluation data of account managers and historical evaluation data of customers regarding account managers, wherein the evaluation factors in the historical evaluation data and the historical evaluation data are the same; the historical evaluation data and the historical evaluation data are input into a preset matching model to calculate a first matching value between the customer and the account manager, wherein the first matching value is calculated based on the account manager's ability value corresponding to the evaluation factor and the customer's attention to the evaluation factor; a second matching value greater than a preset threshold is selected from the first matching values, and the account manager corresponding to the second matching value is recommended to the customer. In this application, the historical evaluation data of the account manager and the historical evaluation data of the customer towards the account manager are obtained. Since the evaluation factors are the same, the customer and the account manager can be associated based on the same evaluation factors. The historical evaluation data and the historical evaluation data are input into a preset matching model to obtain the first matching value between the customer and the account manager. Since the obtained first matching value is calculated based on the ability value of the account manager corresponding to the evaluation factor and the customer's attention to the evaluation factor, the size of the first matching value can reflect the compatibility between the customer and the account manager. The larger the first matching value, the higher the compatibility between the customer and the account manager, so that the customer can be accurately matched with an account manager with a high degree of compatibility.

[0074] Furthermore, based on the first embodiment of this application, another embodiment of this application is provided. In this embodiment, the step of inputting the evaluated historical data and the evaluation historical data into a preset matching model to calculate the first matching value between the customer and the account manager includes:

[0075] Step A1: Input the evaluated historical data and the evaluation historical data into a preset matching model. Based on the preset matching model, vectorize the evaluated historical data to obtain the customer manager capability vector, and vectorize the evaluation historical data to obtain the customer attention factor vector.

[0076] As an example, vectorizing the evaluated historical data and the evaluation historical data can simplify subsequent calculations, convert the data into the data format required by the model, and integrate the calculated customer manager capability vector and customer attention factor vector to obtain specific matching values.

[0077] Step A2: Perform matching calculations on the account manager's capability vector and the customer's attention factor vector to obtain the first matching value between the customer and the account manager.

[0078] As an example, the customer's evaluation of the account manager's various factors and the account manager's professionalism in serving under each factor are the most direct manifestations of customer satisfaction and the compatibility between the customer and the account manager. Therefore, based on the account manager's capability vector and the customer's focus factor vector, the matching degree between the account manager and the customer is calculated under the 11-dimensional vector.

[0079] The step of vectorizing the evaluated historical data to obtain the account manager competence vector includes:

[0080] Step B1: Extract questionnaire data for multiple evaluation factors from the historical data being evaluated. The evaluation factors include service attitude, excessive disturbance, clear explanation, risk warning, excessive marketing, resource allocation suggestions, communication of ideas, attention to needs, timely response, proactive contact, and follow-up care.

[0081] As an example, questionnaire data will be extracted from the historical evaluation data, and the evaluation score / positive / negative rating for each of the 11 evaluation factors will be calculated. The evaluation score can be 1-5 points or 1-10 points. In this embodiment, 1-5 points will be used as the evaluation score. When the customer evaluates the account manager with a score of 1-4 points, it is a negative rating, and when the customer evaluates the account manager with a score of 5 points, it is a positive rating.

[0082] As an example, the dimension diagram of the evaluation factors is as follows: Figure 5 As shown, based on these 11 dimensions, the friendliness, professionalism, and efficiency of the customer manager's service can be fully assessed. The evaluation factors include, but are not limited to, these 11 dimensions. Other evaluation factors can also be used to calculate the matching degree between the customer and the customer manager.

[0083] As an example, the questionnaire data may include multiple evaluation factors. Users can select the option corresponding to the evaluation factor they want to evaluate in the questionnaire and rate the customer at the option.

[0084] Step B2: Based on the evaluation samples of individual evaluation factors in the questionnaire data, determine the individual competency value of the account manager. The calculation method for the individual competency value of the account manager is as follows:

[0085]

[0086] Among them, sco 因子 This indicates the individual competency score of the account manager, sco 常量1 sco is the preset first constant value. 常量2 N is the preset second constant value. 好评 N represents the total sample size of positive reviews from account managers. 差评 N represents the total sample size of negative reviews from account managers. 总量W represents the total sample size of account managers. 差评 As preset weights, avg 客户经理 The mean of the total sample scores for a single rating factor in the questionnaire data of the account managers;

[0087] As an example, the sco calculated in the above formula 因子 The evaluation factors are calculated from the evaluation samples of all questionnaire data for a single evaluation factor. The 11 evaluation factors are calculated separately to obtain the 11 competency values ​​corresponding to the account manager.

[0088] As an example, the first constant value can be 68, which is the initial value of the account manager's individual competency value. The second constant value can be 64, which is multiplied by the subsequent floating score. Depending on the evaluation data obtained, the account manager's individual competency value will also be different.

[0089] As an example, the preset weight can be 5. The preset weight represents the weight of negative reviews. Setting it to 5 means that negative reviews are more important, so the weight of negative reviews is set to be higher.

[0090] As an example, by statistically analyzing the number of positive and negative reviews from account managers, the error value of the derived competency score can be reduced.

[0091] As an example, the total sample size of account managers is greater than or equal to 5. If the total number of ratings is less than 5, then 5 is taken. Similarly, when calculating the average rating of account managers, if the total number of ratings is less than 5, then a certain number of 4-point cases (better ratings among negative rating cases) are added to bring the total to 5, and then the average is taken.

[0092] Step B3: Combine the individual competency values ​​of the account manager corresponding to multiple evaluation factors to obtain the account manager competency vector.

[0093] As an example, each evaluation factor corresponds to a single competency value of a client manager. By combining the single competency values ​​of the client managers corresponding to each evaluation factor, a client manager competency vector containing 11 evaluation factors can be obtained.

[0094] As an example, the combination of individual competency values ​​for account managers could be:

[0095] 1. ability 客户经理 =(sco) 因子1 sco 因子2 , ..., sco 因子11 Among them, ability 客户经理 Represents the account manager capability vector, sco 因子1 .....sco 因子11 This represents the individual competency value of the account manager based on 11 evaluation factors.

[0096] The step of vectorizing the historical evaluation data to obtain the customer attention factor vector includes:

[0097] Step C1: Extract questionnaire data for multiple evaluation factors from the historical evaluation data;

[0098] As an example, extract customer questionnaire data on multiple rating factors from historical evaluation data.

[0099] Step C2: Based on the evaluation samples of individual evaluation factors in the questionnaire data, determine the customer's individual evaluation value, wherein the calculation method for the customer's individual evaluation value is as follows:

[0100]

[0101] Here, "binary positive review" and "binary negative review" represent the preset third constant value when the questionnaire containing the evaluation factor is positive, and the preset fourth constant value when the questionnaire containing the evaluation factor is negative. val represents the value of the current evaluation factor among all the evaluation factors selected in the questionnaire. 因子 The value is the sum of the individual evaluation factors across all questionnaires;

[0102] As an example, the third constant value can be 1, and binary positive reviews represent the weight of positive reviews in the questionnaire data.

[0103] As an example, the fourth constant value can be 5 or any other value greater than the third constant value. A value greater than the binary negative rating indicates that the negative rating is more important.

[0104] As an example, binary positive reviews / binary negative reviews represent a whole. When the questionnaire containing the evaluation factor is positive, binary positive reviews / binary negative reviews = binary positive reviews. When the questionnaire containing the evaluation factor is negative, binary positive reviews / binary negative reviews = binary negative reviews.

[0105] As an example,

[0106] The above formula represents the boolean value when the evaluation factor is one of the factors included in the questionnaire. 因子i =1, when the evaluation factor is not a factor included in the questionnaire, bool 因子i =0, for example, if the questionnaire selects a total of 10 factors, then the current... That is, 1 / 10.

[0107] As an example, the customer's individual rating is calculated based on a single rating factor across all questionnaires, without considering other factors when calculating a single rating factor that the customer is concerned about.

[0108] Step C3: Normalize the individual customer evaluation values ​​to obtain the individual customer factor attention values;

[0109] As an example, normalization can be performed in the following ways:

[0110] pre 因子 =(val 因子 / ∑ 全部因子 val 因子 )

[0111] The purpose of normalization is to simplify the calculation of the model and improve the calculation speed.

[0112] As an example, after normalizing the individual customer evaluation value, we can obtain the individual customer factor attention value. Then, we can directly combine the individual customer factor attention values ​​to obtain the customer attention factor vector.

[0113] Step C4: Combine the customer's individual factor attention values ​​of multiple evaluation factors to obtain a customer attention factor vector.

[0114] As an example, one way to combine customer single-factor attention values ​​is:

[0115] 2. Preference 客户 =(pre 因子1 ,pre 因子2 , ...,pre 因子11 )

[0116] Among them, preference 客户 Represents the customer focus factor vector, pre 因子1 ,…,pre 因子11 This represents the customer's individual factor attention value across 11 evaluation factors.

[0117] As an example, the customer attention factor vector is generated from the customer's historical factor selections in the following way: each factor has an initial weight of 0; assuming that the customer selected n factors in a certain evaluation, if it is a positive review, the weight of that factor increases by 1 / n, and if it is a negative review, the weight of that factor increases by 5 / n; finally, normalization is performed by dividing each factor weight by the sum of all factor weights to obtain the customer's individual factor attention value, and then combining the customer's individual factor attention values ​​to obtain the customer attention factor vector.

[0118] In this embodiment, by calculating multiple evaluation factors, a customer manager capability vector and a customer attention factor vector are obtained. Based on the same evaluation factors and the corresponding number of positive and negative reviews, customers are associated with customer managers, which can also reduce the error value of the corresponding calculation.

[0119] Furthermore, based on the first and second embodiments of this application, another embodiment of this application is provided. In this embodiment, the step of matching and calculating the customer manager's capability vector and the customer's attention factor vector to obtain a first matching value between the customer and the customer manager includes:

[0120] Step D1: Determine if there is any evaluation interaction data between the current customer and the account manager;

[0121] As an example, evaluation interaction data is used to determine whether a customer has rated the account manager. When a customer has rated the account manager, evaluation interaction data exists; when a customer has not rated the account manager, evaluation interaction data does not exist.

[0122] Step D2: If evaluation interaction data exists, perform matching calculations on the account manager's capability vector and the customer's attention factor vector to obtain a first matching value between the customer and the account manager. The formula for calculating the first matching value is as follows:

[0123]

[0124] Among them, fit 客户&客户经理 Indicates the first matching value. This represents the product of the account manager's capability vector and the customer's attention factor vector, avg 客户-客户经理 This represents the average score given by current customers to their account managers, and "revise" indicates the preset adjustment value.

[0125] As an example, when there is evaluation interaction data, that is, when a customer has rated the account manager, the average score of the current customer's rating of the account manager is taken as part of the weight in calculating the first matching value.

[0126] As an example, for customer A and account manager a, if the customer has evaluated the account manager, multiply the attention vector and ability vector values ​​corresponding to the 11 factors respectively, sum them, and multiply by 80% weight. Then multiply the average score of the customer's evaluation of the account manager by 20 and then by 20%. Add these two results together, and add the following logical correction value to obtain the matching degree between customer A and account manager a.

[0127] Wherein, if evaluation interaction data exists, the step of matching and calculating the account manager's capability vector and the customer's attention factor vector to obtain the first matching value between the customer and the account manager includes:

[0128] Step E1: If no evaluation interaction data exists, perform a matching calculation on the account manager's capability vector and the customer's attention factor vector to obtain a first matching value between the customer and the account manager. The formula for calculating the first matching value is as follows:

[0129]

[0130] Among them, fit 客户&客户经理 Indicates the first matching value. This represents the product of the account manager's capability vector and the customer attention factor vector, with revise representing a preset adjustment value.

[0131] As an example, for customer A and account manager a, if the customer has not rated the account manager, multiply the attention vector and capability vector values ​​corresponding to the 11 factors respectively, and then sum them. On this basis, add the following logical correction value to obtain the matching degree between customer A and account manager a.

[0132] The preset adjustment value is associated with the negative review tag submitted by the customer. If the customer has previously submitted a negative review tag, the preset adjustment value will be increased accordingly.

[0133] As an example, the negative review tag is "frequent changes of account managers". If the current customer has selected the negative review tag "frequent changes of account managers", it means that most account managers cannot meet the customer's needs, and the corresponding first match value score will increase.

[0134] As an example, the default adjustment value can be 0, but it can be adjusted to 3 if the customer has previously submitted a negative review tag.

[0135] In this embodiment, a first matching value is calculated based on customer evaluation interaction data. Different matching values ​​are obtained according to different situations. The first matching value is then corrected to make the result more accurate.

[0136] As an example, a detailed calculation flowchart of the preset matching model is shown below. Figure 6 As shown, by Figure 6Therefore, the evaluation data with / without 11 service evaluation factors is filtered out from the acquired historical evaluation data and the evaluated historical data, and processed separately. If there are 11 service evaluation factors in the evaluation data, it is determined whether the sample size of the account manager is greater than 5. If the sample size of the account manager is greater than 5, the calculation is performed directly according to the formula. If the sample size of the account manager is less than 5, 5 samples are taken, and the individual score of the account manager is calculated according to the formula. The customer attention factor vector can be calculated directly according to the formula. Finally, the account manager's ability vector and the customer attention factor vector are matched and calculated. If there are less than 11 service evaluation factors in the customer evaluation data, the calculation process is directly ended. Finally, the first matching value obtained by the filter is used to recommend an account manager to the customer based on the first matching value or the second matching value.

[0137] Reference Figure 3 , Figure 3 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of this application.

[0138] like Figure 3 As shown, the customer relationship matching device may include: a processor 1001, a memory 1005, and a communication bus 1002. The communication bus 1002 is used to realize the connection and communication between the processor 1001 and the memory 1005.

[0139] Optionally, the customer relationship matching device may also include a user interface, a network interface, a camera, RF (Radio Frequency) circuitry, sensors, a WiFi module, etc. The user interface may include a display screen and an input submodule such as a keyboard; optional user interfaces may also include standard wired or wireless interfaces. The network interface may include standard wired or wireless interfaces (such as a Wi-Fi interface).

[0140] Those skilled in the art will understand that Figure 3 The customer relationship matching device structure shown does not constitute a limitation on the customer relationship matching device, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0141] like Figure 3 As shown, the memory 1005, serving as a storage medium, may include an operating system, a network communication module, and a customer relationship matching program. The operating system is a program that manages and controls the hardware and software resources of the customer relationship matching device, supporting the operation of the customer relationship matching program and other software and / or programs. The network communication module is used to enable communication between the various components within the memory 1005, as well as communication with other hardware and software in the customer relationship matching system.

[0142] exist Figure 3In the customer relationship matching device shown, the processor 1001 is used to execute the customer relationship matching program stored in the memory 1005 to implement the steps of the customer relationship matching method described above.

[0143] The specific implementation method of the customer relationship matching device in this application is basically the same as the various embodiments of the customer relationship matching method described above, and will not be repeated here.

[0144] This application also provides a customer relationship matching device, the customer relationship matching device further comprising:

[0145] The acquisition module is used to acquire the historical evaluation data of the account manager and the historical evaluation data of the customer towards the account manager, wherein the evaluation factors in the historical evaluation data and the historical evaluation data are the same.

[0146] The calculation module is used to input the evaluated historical data and the evaluation historical data into a preset matching model to calculate a first matching value between the customer and the customer manager. The first matching value is calculated based on the customer manager's ability value corresponding to the evaluation factor and the customer's attention to the evaluation factor.

[0147] The selection module is used to select a second matching value from the first matching values ​​that is greater than a preset threshold, and recommend the account manager corresponding to the second matching value to the customer.

[0148] In one possible implementation of this application, the computing module includes:

[0149] The processing unit is used to input the evaluated historical data and the evaluation historical data into a preset matching model, and based on the preset matching model, to perform vectorization processing on the evaluated historical data to obtain a customer manager capability vector, and to perform vectorization processing on the evaluation historical data to obtain a customer attention factor vector.

[0150] The calculation unit is used to perform matching calculations on the account manager's capability vector and the customer's attention factor vector to obtain a first matching value between the customer and the account manager.

[0151] In one possible embodiment of this application, the processing unit includes:

[0152] The first extraction subunit is used to extract questionnaire data of multiple evaluation factors from the evaluated historical data, wherein the evaluation factors include service attitude, excessive disturbance, clear explanation, risk warning, excessive marketing, asset allocation suggestions, communication of ideas, attention to needs, timely response, proactive contact, and follow-up care.

[0153] The first determining subunit is used to determine the individual competency value of the account manager based on the evaluation samples of a single evaluation factor in the questionnaire data. The calculation method for the individual competency value of the account manager is as follows:

[0154]

[0155] Among them, sco 因子 This indicates the individual competency score of the account manager, sco 常量1 sco is the preset first constant value. 常量2 N is the preset second constant value. 好评 N represents the total sample size of positive reviews from account managers. 差评 N represents the total sample size of negative reviews from account managers. 总量 W represents the total sample size of account managers. 差评 As preset weights, avg 客户经理 The mean of the total sample scores for a single rating factor in the questionnaire data of the account managers;

[0156] The first combination subunit is used to combine the individual competency values ​​of the account manager corresponding to multiple evaluation factors to obtain the account manager competency vector.

[0157] In one possible embodiment of this application, the processing unit further includes:

[0158] The second extraction subunit is used to extract questionnaire data for multiple evaluation factors from the historical evaluation data;

[0159] The second determining subunit is used to determine the customer's individual evaluation value based on the evaluation samples of a single evaluation factor in the questionnaire data, wherein the calculation method of the customer's individual evaluation value is as follows:

[0160]

[0161] Here, "binary positive review" and "binary negative review" represent the preset third constant value when the questionnaire containing the evaluation factor is positive, and the preset fourth constant value when the questionnaire containing the evaluation factor is negative. val represents the value of the current evaluation factor among all the evaluation factors selected in the questionnaire. 因子 The value is the sum of the individual evaluation factors across all questionnaires;

[0162] The processing subunit is used to normalize the customer's individual evaluation value to obtain the customer's individual factor attention value.

[0163] The second combination subunit is used to combine the customer's individual factor attention values ​​of multiple evaluation factors to obtain a customer attention factor vector.

[0164] In one possible embodiment of this application, the computing unit includes:

[0165] The judgment sub-unit is used to determine whether there is evaluation interaction data between the current customer and the account manager;

[0166] The first calculation subunit is used to perform a matching calculation on the account manager's capability vector and the customer's attention factor vector if evaluation interaction data exists, and calculate a first matching value between the customer and the account manager, wherein the calculation formula for the first matching value is as follows:

[0167]

[0168] Among them, fit 客户&客户经理 Indicates the first matching value. This represents the product of the account manager's capability vector and the customer's attention factor vector, avg 客户-客户经理 This represents the average score given by current customers to their account managers, and "revise" indicates the preset adjustment value.

[0169] In one possible embodiment of this application, the computing unit further includes:

[0170] The second calculation subunit is used to perform a matching calculation on the account manager's capability vector and the customer's attention factor vector if no evaluation interaction data exists, to calculate a first matching value between the customer and the account manager, wherein the calculation formula for the first matching value is as follows:

[0171]

[0172] Among them, fit 客户&客户经理 Indicates the first matching value. This represents the product of the account manager's capability vector and the customer attention factor vector, with revise representing a preset adjustment value.

[0173] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0174] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0175] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0176] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A customer relationship matching method, characterized in that, The method includes the following steps: Obtain historical evaluation data of account managers and historical evaluation data of account managers by customers, wherein the evaluation factors in the historical evaluation data and the historical evaluation data are the same. The questionnaire data of multiple evaluation factors are extracted from the historical data being evaluated. The evaluation factors include service attitude, excessive disturbance, clear explanation, risk warning, excessive marketing, asset allocation suggestions, communication of ideas, attention to needs, timely response, proactive contact, and follow-up care. Based on the evaluation samples of a single evaluation factor in the questionnaire data, the individual competency value of the account manager is determined. When the total sample size of any account manager is less than a preset threshold, the total sample size is supplemented to the preset threshold, and the evaluation samples with preset scores are added before calculating the individual competency value of the account manager. Based on the evaluation samples of individual evaluation factors in the questionnaire data, the customer's individual evaluation value is determined, and the customer's individual evaluation value is normalized to obtain the customer's individual factor attention value. The individual competency values ​​of the account manager corresponding to multiple evaluation factors are combined to obtain the account manager competency vector, and the individual customer factor attention values ​​of multiple evaluation factors are combined to obtain the customer attention factor vector. The evaluated historical data and the evaluation historical data are input into a preset matching model to calculate the first matching value between the customer and the account manager. Select a second matching value from the first matching values ​​that is greater than a preset threshold, and recommend the account manager corresponding to the second matching value to the customer; The step of inputting the evaluated historical data and the evaluation historical data into a preset matching model to calculate the first matching value between the customer and the account manager includes: The evaluated historical data and the evaluation historical data are input into a preset matching model. Based on the preset matching model, the evaluated historical data is vectorized to obtain the customer manager capability vector, and the evaluation historical data is vectorized to obtain the customer attention factor vector. The account manager's capability vector and the customer's attention factor vector are matched and calculated to obtain the first matching value between the customer and the account manager. The step of matching the account manager's capability vector and the customer's attention factor vector to calculate the first matching value between the customer and the account manager includes: Determine if there is any evaluation interaction data between the current customer and the account manager; If evaluation interaction data exists, a matching calculation is performed on the account manager's capability vector and the customer's attention factor vector to obtain a first matching value between the customer and the account manager. The formula for calculating the first matching value is as follows: in, Indicates the first matching value. This represents the dot product of the account manager's capability vector and the customer's attention factor vector. This represents the average score given by current customers to their account managers. Indicates the preset adjustment value; If no evaluation interaction data exists, a matching calculation is performed on the account manager's capability vector and the customer's attention factor vector to obtain a first matching value between the customer and the account manager. The formula for calculating the first matching value is as follows: in, Indicates the first matching value. This represents the dot product of the account manager's capability vector and the customer's attention factor vector. This indicates the preset adjustment value.

2. The customer relationship matching method as described in claim 1, characterized in that, The step of determining the individual competency value of the account manager based on the evaluation sample of a single evaluation factor in the questionnaire data includes: The calculation method for the individual competency scores of the account managers is as follows: in, This indicates the individual competency score of the account manager. The first constant value is preset. The second constant value is preset. The total sample size for positive customer manager reviews. The total sample size of negative reviews from account managers. Total sample size of account managers To preset the weights, This represents the mean of the total sample rating scores for a single rating factor in the questionnaire data of the account managers.

3. The customer relationship matching method as described in claim 1, characterized in that, The step of determining a customer's individual evaluation value based on the evaluation samples of individual evaluation factors in the questionnaire data includes: The calculation method for the customer's individual evaluation value is as follows: in, This indicates that when the questionnaire containing the evaluation factor is rated as positive, The value is set to a preset third constant value; when the questionnaire containing the evaluation factor is a negative rating, The value is set to the preset fourth constant value. This indicates the value of the current evaluation factor among all the evaluation factors selected in the questionnaire. The value is the sum of the individual evaluation factors across all questionnaires.

4. The customer relationship matching method as described in claim 1, characterized in that, The preset adjustment value is associated with the negative review tag submitted by the customer. If the customer has submitted a negative review tag before, the preset adjustment value will be increased accordingly.

5. A customer relationship matching device, characterized in that, The customer relationship matching device includes: The acquisition module is used to acquire the historical evaluation data of the account manager and the historical evaluation data of the customer towards the account manager, wherein the evaluation factors in the historical evaluation data and the historical evaluation data are the same. The acquisition module is further configured to extract questionnaire data for multiple evaluation factors from the evaluated historical data, wherein the evaluation factors include service attitude, excessive disturbance, clear explanation, risk warning, excessive marketing, asset allocation suggestions, communication of concepts, attention to needs, timely response, proactive contact, and follow-up care; based on the evaluation samples of individual evaluation factors in the questionnaire data, determine the individual competency value of the account manager, wherein when the total sample size of any account manager is less than a preset threshold, the total sample size is supplemented to the preset threshold, and evaluation samples with preset scores are added before calculating the individual competency value of the account manager; based on the evaluation samples of individual evaluation factors in the questionnaire data, determine the individual customer evaluation value, and normalize the individual customer evaluation value to obtain the individual customer factor attention value; combine the individual competency values ​​of the account managers corresponding to multiple evaluation factors to obtain the account manager competency vector, and combine the individual customer factor attention values ​​of multiple evaluation factors to obtain the customer attention factor vector; The calculation module is used to input the evaluated historical data and the evaluation historical data into a preset matching model to calculate the first matching value between the customer and the account manager; The selection module is used to select a second matching value from the first matching values ​​that is greater than a preset threshold, and recommend the account manager corresponding to the second matching value to the customer. The computing module includes: The processing unit is used to input the evaluated historical data and the evaluation historical data into a preset matching model, and based on the preset matching model, to perform vectorization processing on the evaluated historical data to obtain a customer manager capability vector, and to perform vectorization processing on the evaluation historical data to obtain a customer attention factor vector. The computing unit includes: The judgment sub-unit is used to determine whether there is evaluation interaction data between the current customer and the account manager; The first calculation subunit is used to perform a matching calculation on the account manager's capability vector and the customer's attention factor vector if evaluation interaction data exists, and calculate a first matching value between the customer and the account manager, wherein the calculation formula for the first matching value is as follows: in, Indicates the first matching value. This represents the dot product of the account manager's capability vector and the customer's attention factor vector. This represents the average score given by current customers to their account managers. Indicates the preset adjustment value; The first calculation subunit is further configured to, if no evaluation interaction data exists, perform a matching calculation on the account manager's capability vector and the customer's attention factor vector to calculate a first matching value between the customer and the account manager, wherein the formula for calculating the first matching value is as follows: in, Indicates the first matching value. This represents the dot product of the account manager's capability vector and the customer's attention factor vector. This indicates the preset adjustment value.

6. A customer relationship matching device, characterized in that, The device includes: a memory, a processor, and a customer relationship matching program stored in the memory and executable on the processor, the customer relationship matching program being configured to implement the steps of the customer relationship matching method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a customer relationship matching program, which, when executed by a processor, implements the steps of the customer relationship matching method as described in any one of claims 1 to 4.

Images