Customer service evaluation method, device, equipment and program product

Abstract

The invention provides a customer service evaluation method and device, equipment and a program product. The method comprises the following steps: determining the weight of each index in an evaluation index system of customer service quality based on expert experience; wherein the evaluation index system comprises at least one first-level index and at least one second-level index corresponding to each first-level index. For each second-level index, determining the global weight of each second-level index based on the weight of the first-level index to which the second-level index belongs and the weight of the second-level index, and weighting the decision matrix of the plurality of customer service seats relative to the second-level indexes based on the global weight of each second-level index to obtain a weighted decision matrix; wherein the elements of the decision matrix comprise the numerical value of each customer service seat relative to each secondary index. And based on the weighted decision matrix, calculating the closeness of each customer service seat to the positive ideal solution and the negative ideal solution by using an approximate ideal solution sorting method. And determining a service quality score of each customer service seat based on the close degree of each customer service seat.

Description

Technical Field

[0001] This application relates to the field of online customer service technology, and in particular to a customer service evaluation method, apparatus, equipment, and program product. Background Technology

[0002] In the current field of customer service evaluation, the common approach is to set thresholds based on experience. This involves pre-setting grading ranges for single or multiple related indicators such as workload and customer satisfaction, and then classifying service quality into different levels such as "excellent" and "good." While this method is simple and direct, its threshold settings rely heavily on historical experience and lack objective algorithmic basis, resulting in insufficient accuracy and scientific rigor in the evaluation results, making it difficult to adapt to the diversity and complexity of online customer service work.

[0003] Furthermore, existing evaluation systems have significant limitations in terms of indicator expansion. When new evaluation dimensions need to be introduced, complex association rules often need to be manually re-formulated, a cumbersome process that is prone to subjective bias, severely restricting the flexibility and scalability of the evaluation system. Summary of the Invention

[0004] In view of the above problems, this application provides a customer service evaluation method, apparatus, equipment, and program product that overcomes or at least partially solves the above problems. The technical solution is as follows: Firstly, a customer service evaluation method is provided, including: The weights of each indicator in the customer service quality evaluation index system are determined based on expert experience; wherein, the evaluation index system includes at least one primary indicator and at least one secondary indicator corresponding to each primary indicator. For each secondary indicator, a global weight is determined based on the weight of its primary indicator and its own weight. Then, based on the global weight of each secondary indicator, the decision matrices of multiple customer service agents relative to the secondary indicators are weighted to obtain a weighted decision matrix. The elements of the decision matrix include the values ​​of each customer service agent relative to each of the secondary indicators. Based on the weighted decision matrix, the proximity of each customer service agent to the positive and negative ideal solutions is calculated using the approximation ideal solution ranking method. Based on the proximity of each customer service agent, a service quality score is determined for each agent.

[0005] Secondly, a customer service evaluation device is proposed, including: The weight configuration module is used to determine the weight of each indicator in the evaluation index system for customer service quality based on expert experience; wherein, the evaluation index system includes at least one primary indicator and at least one secondary indicator corresponding to each primary indicator. The indicator weighting module is used to determine the global weight of each secondary indicator based on the weight of its primary indicator and its own weight, and to weight the decision matrices of multiple customer service agents relative to the secondary indicators based on the global weight of each secondary indicator, thereby obtaining a weighted decision matrix; wherein the elements of the decision matrix include the value of each customer service agent relative to each of the secondary indicators. The proximity calculation module is used to calculate the proximity of each customer service agent to the positive and negative ideal solutions based on the weighted decision matrix and using the approximation ideal solution ranking method. The rating determination module is used to determine the service quality rating of each customer service agent based on their proximity to the customer service agent.

[0006] Thirdly, embodiments of this application provide an electronic device, including: a processor; and a memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the method described in the first aspect.

[0007] Fourthly, a computer program product is provided, the computer program product including a computer-readable storage medium storing a computer program operable to cause a computer to perform the method described in the first aspect.

[0008] This application embodiment first determines the weights of each level of indicators in the customer service quality evaluation index system based on expert experience using the Analytic Hierarchy Process (AHP). The evaluation index system includes at least one primary indicator and multiple secondary indicators. Next, for each secondary indicator, its global weight in the overall evaluation is calculated by multiplying its own weight by the weight of its primary indicator. Based on this global weight, a weighted decision matrix is ​​formed by weighting the original values ​​of each customer service agent on each secondary indicator. Then, the weighted matrix is ​​processed using the approximation-ideal-solution ranking method. By calculating the relative closeness of each customer service agent to the positive and negative ideal solutions, the service quality score for each customer service agent is finally determined. This application embodiment's solution effectively overcomes the inherent defects of traditional evaluation methods through a systematic indicator weight allocation mechanism combined with a multi-dimensional approximation-ideal-solution ranking algorithm. In the weight determination stage, the AHP is used to transform expert experience into a quantitative judgment matrix, and consistency verification ensures logical rationality, fundamentally solving the problem of strong subjectivity in threshold setting in traditional methods. By constructing a multi-level evaluation index system and calculating global weights, an intrinsic correlation network among the indicators is established. Then, the relative distance between each indicator and the ideal solution is comprehensively calculated in the weighted standardized matrix using the approximation-ideal-solution ranking method, effectively solving the problems of isolated indicators and lack of correlation in traditional methods. When it is necessary to expand the evaluation dimensions, only new indicators need to be added to the existing hierarchical structure and the weight calculation process needs to be re-executed, without reconstructing the entire evaluation system. This modular design significantly improves the system's scalability. Through these three innovations, the embodiments of this application construct an evaluation system that maintains the value of expert experience while possessing mathematical rigor, significantly improving the scientific nature, accuracy, and scalability of customer service quality evaluation. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart illustrating the customer service evaluation method according to an embodiment of this application.

[0011] Figure 2 This is a schematic diagram of the evaluation index system applied in the customer service evaluation method of this application embodiment.

[0012] Figure 3 This is a schematic diagram of the customer service evaluation device according to an embodiment of this application.

[0013] Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0014] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0015] As mentioned earlier, the current customer service evaluation system commonly employs an experience-based threshold setting method. This involves pre-setting fixed grading ranges for single or multiple related indicators such as workload and customer satisfaction, and mechanically classifying service quality into limited levels like "excellent" and "good." While simple to implement, this method heavily relies on subjective experience for threshold setting, lacking objective algorithmic basis and a scientific weighting mechanism. This results in insufficient accuracy and reliability of the evaluation results, making it difficult to effectively address the complexity and diversity of different types of customer service work. Furthermore, existing evaluation systems have significant deficiencies in indicator scalability. When business needs change and new evaluation dimensions need to be introduced, complex correlation rules and threshold standards must be manually redefined. This process is not only cumbersome and inefficient but also highly susceptible to human bias, severely restricting the adaptability, scalability, and intelligence of the evaluation system.

[0016] In view of this, this application proposes a customer service evaluation method, apparatus, equipment, and program product, aiming to systematically solve the problems existing in the prior art, such as strong subjectivity in threshold setting, lack of algorithmic support for indicator correlation, and difficulty in system expansion. The technical solutions provided by various embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0017] One embodiment of this application provides a customer service evaluation method. Figure 1 This is a flowchart illustrating the customer service evaluation method, including: S101, Based on expert experience, determine the weight of each indicator in the evaluation index system for customer service quality; wherein, the evaluation index system includes at least one primary indicator and at least one secondary indicator corresponding to each primary indicator.

[0018] The customer service quality evaluation index system is a multi-level, structured system used to comprehensively assess the overall quality of customer service. This system includes at least one primary index and at least one secondary index corresponding to each primary index. The primary indexes provide a top-level classification of customer service quality from a macro perspective, for example... Figure 2 The indicators can include core aspects such as marketing skills, execution volume, and service level. Each primary indicator is further subdivided into more operational secondary indicators to specifically measure performance in that dimension. For example, the primary indicator of marketing skills can be further subdivided into secondary indicators such as conversion rate, volume of business handled by the primary entity, and volume of business handled by the secondary entity; the primary indicator of execution volume can include secondary indicators such as outbound call volume and connection rate; and the primary indicator of service level can cover secondary indicators such as compliance rate, timeliness rate, and complaint rate. It should be noted that the indicator system in this embodiment is scalable, and secondary indicators can be further subdivided into tertiary indicators to form a more refined evaluation hierarchy.

[0019] Weights are numerical values ​​that measure the relative importance of each indicator in the evaluation system, ranging from 0 to 1. The sum of the weights of all indicators at the same level is 1. The purpose of weights is to reflect the differences in the impact of different indicators on the final evaluation result, ensuring that the evaluation results are more scientific and reasonable. For example, in a telephone customer service scenario, the weight of marketing skills may be higher than that of service level, reflecting the company's emphasis on different business objectives.

[0020] In practical applications, this embodiment uses the Analytic Hierarchy Process (AHP) to determine the weights of indicators at each level. The specific implementation process includes the following steps: First, for each level of indicators, experts in the relevant fields are organized to conduct pairwise importance comparisons of indicators within the same level based on their professional experience. For example, for first-level indicators, experts need to determine which is more important, and by how much, marketing skills versus execution volume; and which is more important, and by how much, marketing skills versus service level, and so on. The results of these pairwise comparisons are used to construct a judgment matrix, in which each element represents the importance of a peer indicator relative to other indicators in the same column. The values ​​are assigned according to a 1-9 scale, where 1 indicates equal importance and 9 indicates absolute importance.

[0021] Secondly, based on the constructed judgment matrix, the initial weights of each indicator are determined by calculating its eigenvectors. Specifically, each column of the judgment matrix is ​​first normalized, then the geometric mean of the normalized matrix is ​​calculated row by row, and finally the resulting vector is normalized to obtain the initial weight vector of each indicator.

[0022] To ensure the logical consistency of expert judgments, this embodiment also performs a rigorous consistency check on the initial weights. The check process includes: calculating the maximum eigenvalue of the judgment matrix using a specific formula based on the judgment matrix and the calculated initial weights; calculating the consistency index using the maximum eigenvalue and the matrix order; querying the corresponding average random consistency index standard value based on the order of the judgment matrix; and dividing the consistency index by the average random consistency index standard value to obtain the consistency ratio. When the calculated consistency ratio is less than 0.1, the initial weights are deemed to have passed the consistency check, and these initial weights can be determined as the final weights of each indicator in the evaluation index system. If the consistency ratio is greater than or equal to 0.1, the experts need to readjust the judgment matrix until it passes the consistency check. This method ensures the scientific validity and reliability of the weight determination, laying the foundation for subsequent comprehensive evaluation.

[0023] As an example, in actual implementation, a clear hierarchical structure model needs to be constructed first to organize the evaluation indicators. The top layer of the model is the overall goal of the decision-making, namely, to comprehensively evaluate the quality of customer service. The middle layer consists of primary indicators used to measure the goal, such as marketing skills, execution volume, and service level. Each primary indicator can be further decomposed into several specific secondary indicators, thus forming a complete evaluation indicator system that includes a goal layer, a criterion layer (primary indicators), and a solution layer (secondary indicators). After the hierarchical structure is determined, for indicators within the same level, domain experts are invited to conduct pairwise importance comparisons based on their experience. This comparison process aims to quantify subjective judgment and ultimately form a judgment matrix H, as shown in formula (1).

[0024] Formula (1): .

[0025] in, Indicators relative to indicators The degree of importance. Its assignment is usually based on a proportional scale of 1 to 9, and the specific meanings are shown in Table 1.

[0026] Table 1:

[0027] Since the judgment matrix is ​​based on expert subjective judgment, there may be a risk of logical inconsistency. Therefore, it is necessary to calculate the weight vector and perform consistency verification. The specific steps are as follows.

[0028] First, the initial weights of each indicator are calculated using the geometric mean method. And the largest eigenvalue of the judgment matrix The calculation process is shown in formulas (2) to (4).

[0029] Formula (2): ; Formula (3): ; Formula (4): .

[0030] Subsequently, a consistency check is performed to confirm the reliability of the judgment matrix. The check is completed by calculating the consistency ratio CR. First, the consistency index CI is calculated using formula (5).

[0031] Formula (5): .

[0032] Next, based on the matrix order of the judgment matrix, look up the standard value of the corresponding average random consistency index RI in Table 2.

[0033] Table 2:

[0034] Finally, the consistency ratio CR is calculated using formula (6).

[0035] Formula (6): .

[0036] When the calculated consistency ratio CR is less than 0.1, the judgment matrix is ​​considered to have satisfactory consistency, and the calculated weight vector is acceptable. If the judgment matrix is ​​of order 2, it is directly considered to meet the consistency requirements, and the above calculation is unnecessary.

[0037] S102, for each secondary indicator, determine the global weight of each secondary indicator based on the weight of its primary indicator and its own weight, and based on the global weight of each secondary indicator, weight the decision matrices of multiple customer service agents relative to the secondary indicators to obtain a weighted decision matrix; wherein, the elements of the decision matrix include the value of each customer service agent relative to each secondary indicator. In this embodiment, the rows of the decision matrix represent the customer service agents to be evaluated, and the columns represent the secondary indicators in the evaluation indicator system. Each element in the matrix is ​​a specific numerical value, representing the actual performance of a customer service agent on a certain secondary indicator. For example, Agent A's conversion rate is 25%, and Agent B's outbound call volume is 150. These specific data together constitute a complete data matrix reflecting the performance of all agents on all indicators.

[0038] The purpose of determining global weights is to unify the relative weights of hierarchical indicators under the overall goal. In the Analytic Hierarchy Process (AHP), the weight of a first-level indicator reflects its importance relative to the overall goal, while the weight of a second-level indicator only reflects its importance relative to its parent first-level indicator. Global weights multiply these two to obtain the absolute importance of each second-level indicator relative to the overall goal. For example, if the weight of the first-level indicator "marketing skills" is 0.6, and the weight of its subordinate second-level indicator "conversion rate" relative to "marketing skills" is 0.5, then the global weight of "conversion rate" is 0.6 * 0.5 = 0.3. This 0.3 means that in the final evaluation, the "conversion rate" indicator will account for 30% of the influence. The purpose of weighting the decision matrix is ​​precisely to use this global weight to adjust the original data, so that the data of important indicators have a greater proportion in subsequent calculations, thereby making the final evaluation result more reflective of the relative importance of each indicator. The advantage of a weighted decision matrix compared to an unweighted matrix is ​​that it is no longer a simple list of raw data, but a data set that has been calibrated for importance, ensuring that the evaluation results are more scientific and reasonable and consistent with the decision-making objectives.

[0039] Before weighting the decision matrix based on global weights, it is usually necessary to standardize the original decision matrix. This step is crucial because different secondary indicators often have different dimensions and numerical ranges. For example, "outbound call volume" might be a value in the hundreds, while "complaint rate" is a percentage. The significance of standardization is to eliminate these differences in dimensions and numerical ranges, uniformly transforming the values ​​of all indicators into the dimensionless interval [0,1], making different indicators comparable and creating conditions for subsequent comprehensive calculations.

[0040] For conventional gain or decay indicators, extreme value methods can be used for standardization. However, for percentage-type secondary indicators, this embodiment employs a more refined special standardization method to better reflect actual business conditions. This method first determines a set of optimal reference values ​​arranged from smallest to largest within the range of the percentage indicator, including at least the minimum and maximum values ​​of the range. Based on these optimal reference values, the entire range is divided into several consecutive optimal intervals. Subsequently, different standardization strategies are applied to different types of intervals: for FIX-type optimal intervals set as non-linear values, all original values ​​within them are directly standardized to a specified constant value; for LINEAR-type optimal intervals set as linear values, the original values ​​are converted into corresponding standardized results based on a linear interpolation algorithm. The configuration of this linear interpolation algorithm follows the business characteristics of the indicator: for gain-type secondary indicators, the standardized result value increases linearly with the increase of the original value; for decay-type secondary indicators, the standardized result value decreases linearly with the increase of the original value. This differentiated processing not only eliminates the dimensions of measurement during the standardization process but also more accurately depicts the true impact of changes in indicator values ​​on service quality evaluation. Finally, multiplying each column of the standardized matrix by its corresponding global weight yields the final weighted standardized decision matrix, preparing for subsequent calculations of how close each agent is to the ideal solution.

[0041] As an example, in actual execution, the original decision matrix is ​​first constructed. When there are m customer service agents to be evaluated, each agent corresponds to... When there are two secondary evaluation indicators, a system can be constructed. OK The original data matrix of the column Its mathematical expression is shown in formula (7).

[0042] Formula (7): .

[0043] Next, data standardization is performed. Since the dimensions and numerical ranges of the various evaluation indicators differ, standardization is necessary to eliminate these differences and ensure comparability between the indicators. Different standardization methods are used based on the characteristics of the indicators: For gain-type indicators, i.e., indicators where a larger value represents better service quality, the standardized formula of formula (8) is used: Formula (8): .

[0044] For attenuation-type indicators, i.e., the smaller the value, the better the service quality, the standardization formula (9) is used: Formula (9): .

[0045] For percentage-type metrics commonly used in customer service scenarios, this embodiment innovatively introduces a special range standardization process. This process first standardizes the metric's value range... =Determine a set of optimal reference values ​​within [min, max]. =( , ,……, ),in =min, =max, and satisfy ,< <……< Based on these reference values, the entire value range is divided into n-1 consecutive optimal intervals. , , ……, .

[0046] Each optimal interval =[ , ]( =1,2,……,n 1) Configured as one of two types: FIX type, where all primitive values ​​within the range are normalized to a specified constant. ,in ∈[0,1]; values ​​within the LINEAR type range are converted into the corresponding standardized results through linear calculation, and their output range is set according to the index type [ , (Gain type) or [ , (Attenuation type).

[0047] To ensure the continuity of interval transformations, this embodiment sets strict interval constraints. For gain-type special intervals, adjacent intervals must satisfy the non-decreasing constraint of the result value; for decay-type special intervals, the result value must satisfy the non-increasing constraint.

[0048] The specific standardized calculation process is as follows: For any optimal interval =[ , If the original value x in the interval is of type FIX, then the value is directly assigned according to formula (10).

[0049] Formula (10): ;in .

[0050] If the interval type is LINEAR, it is processed according to the indicator type. Gain-type special intervals are calculated using formula (11): Formula (11): ;in .

[0051] The attenuation-type special interval is calculated according to formula (12): Formula (12): ;in, .

[0052] For the original value x that is not in any optimal interval, the extreme value handling method of formula (13) is adopted.

[0053] Formula (13): ;in, .

[0054] To standardize processing, this method also transforms attenuation data into gain data through positive conversion. Conventional attenuation indices are processed according to formula (14).

[0055] Formula (14): .

[0056] The attenuation-type special interval index is handled according to formula (15).

[0057] Formula (15): .

[0058] Finally, a weighted standardization process is performed. Each column of the standardization matrix S is then compared with the weights determined by the AHP (Analog-Philosophy of Things) method. Multiplying them together yields the standardized weighted decision matrix shown in formula (16). : Formula (16): .

[0059] S103, based on the weighted decision matrix, uses the approximation ideal solution ranking method to calculate the closeness of each customer service agent to the positive and negative ideal solutions.

[0060] The fundamental purpose of calculating the proximity score is to reflect, through a comprehensive quantitative indicator, how close each customer service agent's service quality is to the ideal optimal state. This proximity score ranges from 0 to 1; a higher value indicates that the agent's overall performance is closer to the optimal level, and the better the service quality; conversely, a lower value indicates that its performance is closer to the worst level. Therefore, the proximity score, as a core comparative benchmark, provides a direct and scientific basis for the final agent service quality ranking.

[0061] In the specific implementation process, this embodiment first starts with the standardized weighted decision matrix. Two key reference points are identified: the positive ideal solution and the negative ideal solution. Positive ideal solution The ideal solution is a virtual optimal solution, composed of the maximum values ​​of all gain-type secondary indices and the minimum values ​​of all attenuation-type secondary indices in the matrix, representing the theoretically achievable best service quality. Conversely, the negative ideal solution... This is a virtual worst-case solution, which is composed of the minimum value of all gain-type secondary indicators and the maximum value of all attenuation-type secondary indicators, representing the theoretically worst service quality state.

[0062] After establishing these two reference benchmarks, this embodiment then calculates the distance in mathematical space between each customer service agent and the positive and negative ideal solutions, respectively. Specifically, the Euclidean distance formula is used for calculation to obtain the first mathematical distance between each customer service agent and the positive ideal solution. and the second mathematical distance from the negative ideal solution. .

[0063] Finally, based on the ratio of the second mathematical distance of each customer service agent to the sum of its first and second mathematical distances, the proximity score of each agent is calculated. This ratio is the final proximity score, which clearly shows that the closer an agent is to the optimal solution and the farther away from the worst solution, the closer their proximity score is to 1, and the higher their ranking.

[0064] As an example, in actual implementation, the first step is to start with the standardized weighted decision matrix. Determine the positive ideal solution With negative ideal solution In the determination process, it is necessary to differentiate the processing according to the type of indicator. For all gain-type indicators (including conventional gain-type indicators and gain-type special interval indicators), the positive ideal solution takes the maximum value of its column vector, and the negative ideal solution takes the minimum value of its column vector; for all decay-type indicators, the opposite processing method is adopted, that is, the positive ideal solution takes the minimum value of its column vector, and the negative ideal solution takes the maximum value of its column vector. This process can be achieved through the mathematical expressions of formulas (17) and (18): Formula (17): .

[0065] Formula (18): .

[0066] in, Representative gain-type index set, This represents a set of decay-type indicators.

[0067] Next, calculate the Euclidean distance between each customer service agent (i.e., the evaluation object) and the positive and negative ideal solutions. For the th One customer service seat, which is related to the ideal solution. Euclidean distance The calculation is performed using the mathematical expression of formula (19).

[0068] Formula (19): .

[0069] For the One customer service seat, and its negative ideal solution Euclidean distance The calculation is performed using formula (20).

[0070] Formula (20): .

[0071] Finally, based on the calculated Euclidean distance, the proximity of each customer service agent is calculated. The proximity calculation follows the principle of relative proximity, that is, it is determined by measuring the proportion of each seat's proximity to the negative ideal solution to its total distance to the positive and negative ideal solutions. The specific calculation is achieved through formula (21).

[0072] Formula (21): .

[0073] Proximity The value ranges from 0 to 1. The closer a customer service agent's evaluation value is to the ideal solution, the better their... The closer the value is to 1, the lower it is; conversely, the closer it is to the negative ideal solution, the lower its value. The closer the value is to 0, the better. Through this series of calculations, the proximity score of each customer service agent is obtained, providing a direct basis for subsequent ranking predictions.

[0074] S104, based on the proximity of each customer service agent, determines the service quality score for each customer service agent.

[0075] It should be noted that there is no single way to determine the service quality score based on proximity, and this embodiment does not impose any specific limitations on it.

[0076] As one exemplary implementation, this embodiment uses a multi-level weighted aggregation method to calculate the final score. Specifically, firstly, the proximity scores of all customer service agents under each primary indicator dimension are combined to form a proximity decision matrix C; then, the proximity decision matrix C is weighted and calculated with the weight W of each primary indicator to obtain the service quality score vector for each customer service agent. The weighting can be referred to as formula (22).

[0077] Formula (22): .

[0078] Through this calculation method, the final service quality score of each customer service agent comprehensively reflects their performance across all important dimensions, while also considering the relative importance of different dimensions in the overall evaluation. This weighted aggregation method based on matrix operations is not only computationally efficient but also ensures the systematic and scientific nature of the scoring results, providing a reliable quantitative basis for the accurate evaluation and comparison of customer service quality. Furthermore, after determining the service quality score for each customer service agent, this embodiment can apply the scoring results in multiple dimensions. For example, it can automatically generate agent performance ranking data based on the scoring results, solving the problems of low efficiency and strong subjectivity in traditional manual evaluation; it can automatically locate service weaknesses and generate improvement plans by analyzing the correlation between the score and the weights of each indicator through algorithmic models, overcoming the inaccurate problem diagnosis defects in traditional methods; and it can intelligently recommend resource allocation schemes based on the scoring data, effectively solving the problem of unreasonable resource allocation caused by the lack of data support in the existing customer service management system, ultimately achieving an intelligent closed loop from quality evaluation to management decision-making.

[0079] In summary, the method of this embodiment first determines the weights of each level of indicators in the customer service quality evaluation index system based on expert experience using the Analytic Hierarchy Process (AHP). This evaluation index system includes at least one primary indicator and multiple secondary indicators. Next, for each secondary indicator, its global weight in the overall evaluation is calculated by multiplying its own weight by the weight of its primary indicator. Based on this global weight, a weighted decision matrix is ​​formed by weighting the original values ​​of each customer service agent on each secondary indicator. Then, the weighted matrix is ​​processed using the approximation-ideal-solution ranking method. By calculating the relative closeness of each customer service agent to the positive and negative ideal solutions, the service quality score for each agent is finally determined. It should be understood that the method of this embodiment effectively overcomes the inherent defects of traditional evaluation methods through a systematic indicator weight allocation mechanism combined with a multi-dimensional approximation-ideal-solution ranking algorithm. In the weight determination stage, the AHP is used to transform expert experience into a quantitative judgment matrix, and consistency verification ensures logical rationality, fundamentally solving the problem of strong subjectivity in threshold setting in traditional methods. By constructing a multi-level evaluation index system and calculating global weights, an intrinsic correlation network among the indicators is established. Then, the relative distance between each indicator and the ideal solution is comprehensively calculated in the weighted standardized matrix using the approximation-ideal-solution ranking method, effectively solving the problems of isolated indicators and lack of correlation in traditional methods. When it is necessary to expand the evaluation dimensions, only new indicators need to be added to the existing hierarchical structure and the weight calculation process needs to be re-executed, without reconstructing the entire evaluation system. This modular design significantly improves the system's scalability. Through these three innovations, the method in this embodiment constructs an evaluation system that maintains the value of expert experience while possessing mathematical rigor, significantly improving the scientific nature, accuracy, and scalability of customer service quality evaluation.

[0080] In addition, corresponding to Figure 1 In addition to the method shown, another embodiment of this example also provides a customer service evaluation device. Figure 3 This is a structural diagram of the customer service evaluation device 300, including: The weight configuration module 310 is used to determine the weight of each indicator in the evaluation index system for customer service quality based on expert experience; wherein, the evaluation index system includes at least one primary indicator and at least one secondary indicator corresponding to each primary indicator. The indicator weighting module 320 is used to determine the global weight of each secondary indicator based on the weight of its primary indicator and its own weight, and to weight the decision matrices of multiple customer service agents relative to the secondary indicators based on the global weight of each secondary indicator, thereby obtaining a weighted decision matrix; wherein the elements of the decision matrix include the values ​​of each customer service agent relative to each of the secondary indicators. The proximity calculation module 330 is used to calculate the proximity of each customer service agent to the positive ideal solution and the negative ideal solution based on the weighted decision matrix and using the approximation ideal solution sorting method. The rating determination module 340 is used to determine the service quality rating of each customer service agent based on their proximity to the customer service agent.

[0081] Optionally, the proximity calculation module 330 calculates the proximity of each customer service agent to the positive and negative ideal solutions based on the weighted decision matrix using the approximation ideal solution ranking method. This includes: determining the positive and negative ideal solutions from the weighted decision matrix; the positive ideal solution is composed of the maximum value of all gain-type secondary indicators and the minimum value of all decay-type secondary indicators; the negative ideal solution is composed of the minimum value of all gain-type secondary indicators and the maximum value of all decay-type secondary indicators; gain-type secondary indicators refer to indicators where a larger indicator value represents higher service quality; decay-type secondary indicators refer to indicators where a smaller indicator value represents higher service quality; calculating the first mathematical distance of each customer service agent relative to the positive ideal solution and the second mathematical distance relative to the negative ideal solution; and calculating the proximity of each customer service agent based on the proportion of the second mathematical distance to the sum of the first and second mathematical distances.

[0082] Optionally, before weighting the decision matrix of multiple customer service agents relative to the secondary indicators based on the global weight of each secondary indicator, the indicator weighting module 320 is also used to standardize the decision matrix; wherein, the standardization process of the secondary indicators belonging to the percentage type includes: determining a set of optimal reference values ​​arranged from smallest to largest within the value range of the secondary indicator; the optimal reference values ​​include at least the minimum and maximum values ​​of the value range; based on the determined optimal reference values, dividing the value range into several continuous optimal intervals; for optimal intervals with non-linear values, standardizing all their values ​​to specified values; for optimal intervals with linear values, converting all their values ​​into corresponding standardized results based on a linear interpolation algorithm.

[0083] Optionally, the linear interpolation algorithm is configured such that: for gain-type secondary indices, the standardized result value increases linearly with the increase of the original value; for attenuation-type secondary indices, the standardized result value decreases linearly with the increase of the original value.

[0084] Optionally, the weight configuration module 310 determines the weights of each indicator in the customer service quality evaluation index system based on expert experience, including: for each level of indicators, comparing the importance of each pair through expert scoring to construct a judgment matrix; wherein, the elements in the judgment matrix represent the importance of peer indicators relative to indicators in the same column; determining the initial weights of each indicator in the judgment matrix based on the eigenvectors of the judgment matrix; performing consistency verification on the initial weights of each indicator in the judgment matrix; wherein, the consistency verification includes: calculating the maximum eigenvalue of the judgment matrix based on the judgment matrix and the initial weights; calculating a consistency index based on the maximum eigenvalue and the matrix order of the judgment matrix; querying the corresponding average random consistency index standard value according to the matrix order of the judgment matrix; dividing the consistency index by the average random consistency index standard value to obtain the consistency ratio; wherein, when the calculated consistency ratio is less than 0.1, the initial weights are determined to have passed the consistency verification; and the initial weights that have passed the consistency verification are determined as the final weights of each indicator in the evaluation index system.

[0085] Optionally, the scoring determination module 340 determines the service quality score of each customer service agent based on the proximity of each agent, including: combining the proximity of all customer service agents under each primary indicator dimension to form a proximity decision matrix; and weighting the proximity decision matrix with the weight of each primary indicator to obtain the service quality score vector of each customer service agent.

[0086] Optionally, the primary indicators include at least one of marketing skills, execution volume, and service level; the secondary indicators of marketing skills include at least one of conversion rate, main business volume, and co-main business volume; the secondary indicators of execution volume include at least one of outbound call volume and connection rate; and the secondary indicators of service level include at least one of compliance rate, timeliness rate, and complaint rate. This embodiment's device first determines the weights of each level of indicators in the customer service quality evaluation index system based on expert experience using the Analytic Hierarchy Process (AHP). The evaluation index system includes at least one primary indicator and multiple secondary indicators. Next, for each secondary indicator, its global weight in the overall evaluation is calculated by multiplying its own weight by the weight of its primary indicator. Based on this global weight, a weighted decision matrix is ​​formed by weighting the original values ​​of each customer service agent on each secondary indicator. Then, the weighted matrix is ​​processed using the approximation-ideal-solution ranking method. By calculating the relative closeness of each customer service agent to the positive and negative ideal solutions, the service quality score for each agent is finally determined. It should be understood that this embodiment's device, through a systematic indicator weight allocation mechanism combined with a multi-dimensional approximation-ideal-solution ranking algorithm, effectively overcomes the inherent defects of traditional evaluation methods. In the weight determination stage, the AHP is used to transform expert experience into a quantitative judgment matrix, and consistency verification ensures logical rationality, fundamentally solving the problem of strong subjectivity in threshold setting in traditional methods. By constructing a multi-level evaluation index system and calculating global weights, an intrinsic correlation network among the indicators is established. Then, the relative distance between each indicator and the ideal solution is comprehensively calculated in the weighted standardized matrix using the approximation-ideal-solution ranking method, effectively solving the problems of isolated indicators and lack of correlation in traditional methods. When it is necessary to expand the evaluation dimensions, only new indicators need to be added to the existing hierarchical structure and the weight calculation process needs to be re-executed, without reconstructing the entire evaluation system. This modular design significantly improves the system's scalability. Through these three innovations, the device in this embodiment constructs an evaluation system that maintains the value of expert experience while possessing mathematical rigor, significantly improving the scientific nature, accuracy, and scalability of customer service quality evaluation.

[0087] It should be noted that the customer service evaluation device in this embodiment can be used as... Figure 1 The execution body of the method shown is therefore able to achieve... Figure 1 The steps and functions of the method shown are illustrated.

[0088] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Please refer to it. Figure 4At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and memory. The memory may include main memory, such as high-speed random-access memory (RAM), or non-volatile memory, such as at least one disk drive. Of course, the electronic device may also include other hardware required for other business operations.

[0089] The processor, network interface, and memory can be interconnected via an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0090] Memory is used to store computer programs. Specifically, a computer program may include program code, which includes computer operation instructions. Memory may include main memory and non-volatile memory, and provides the computer program to the processor.

[0091] Specifically, the processor reads the corresponding computer program from non-volatile memory into memory and then runs it, forming the above-mentioned logical structure. Figure 3 The customer service evaluation device shown. Correspondingly, the processor executes the program stored in the memory, and specifically performs the following operations: The weights of each indicator in the evaluation index system for customer service quality are determined based on expert experience; wherein the evaluation index system includes at least one primary indicator and at least one secondary indicator corresponding to each primary indicator.

[0092] For each secondary indicator, a global weight is determined based on the weight of its primary indicator and its own weight. Then, based on the global weight of each secondary indicator, the decision matrices of multiple customer service agents relative to the secondary indicators are weighted to obtain a weighted decision matrix. The elements of the decision matrix include the values ​​of each customer service agent relative to each of the secondary indicators. Based on the weighted decision matrix, the proximity of each customer service agent to the positive and negative ideal solutions is calculated using the approximation ideal solution ranking method.

[0093] Based on the proximity of each customer service agent, a service quality score is determined for each agent.

[0094] The above is as described in this instruction manual. Figure 1 The customer service evaluation method disclosed in the illustrated embodiments can be applied to a processor and implemented by the processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit in the processor or by instructions in the form of software. The processor mentioned above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this application can be directly embodied as being executed by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0095] Of course, in addition to software implementation, the electronic device described in this specification does not exclude other implementation methods, such as logic devices or a combination of hardware and software. In other words, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0096] Furthermore, embodiments of this application also propose a computer program product, including a computer-readable storage medium storing one or more computer programs, the one or more computer programs including instructions.

[0097] When the aforementioned instructions are executed by a portable electronic device that includes multiple applications, they enable the portable electronic device to perform... Figure 1 The steps in the method shown include: The weights of each indicator in the evaluation index system for customer service quality are determined based on expert experience; wherein the evaluation index system includes at least one primary indicator and at least one secondary indicator corresponding to each primary indicator.

[0098] For each secondary indicator, a global weight is determined based on the weight of its primary indicator and its own weight. Then, based on the global weight of each secondary indicator, the decision matrices of multiple customer service agents relative to the secondary indicators are weighted to obtain a weighted decision matrix. The elements of the decision matrix include the values ​​of each customer service agent relative to each of the secondary indicators. Based on the weighted decision matrix, the proximity of each customer service agent to the positive and negative ideal solutions is calculated using the approximation ideal solution ranking method.

[0099] Based on the proximity of each customer service agent, a service quality score is determined for each agent.

[0100] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0101] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0102] The above are merely embodiments of this specification and are not intended to limit the scope of this specification. Various modifications and variations can be made to this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of the claims of this specification. Furthermore, all other embodiments obtained by those skilled in the art without inventive effort should fall within the protection scope of this document.

Claims

1. A customer service evaluation method, characterized in that, include: The weights of each indicator in the customer service quality evaluation index system are determined based on expert experience; wherein, the evaluation index system includes at least one primary indicator and at least one secondary indicator corresponding to each primary indicator. For each secondary indicator, a global weight is determined based on the weight of its primary indicator and its own weight. Then, based on the global weight of each secondary indicator, the decision matrices of multiple customer service agents relative to the secondary indicators are weighted to obtain a weighted decision matrix. The elements of the decision matrix include the values ​​of each customer service agent relative to each of the secondary indicators. Based on the weighted decision matrix, the proximity of each customer service agent to the positive and negative ideal solutions is calculated using the approximation ideal solution ranking method. Based on the proximity of each customer service agent, a service quality score is determined for each agent.

2. The method according to claim 1, characterized in that, Based on the weighted decision matrix, the proximity of each customer service agent to the positive and negative ideal solutions is calculated using the approximation of ideal solution ranking method, including: The positive ideal solution and the negative ideal solution are determined from the weighted decision matrix. The positive ideal solution consists of the maximum value of all gain-type secondary indicators and the minimum value of all decay-type secondary indicators. The negative ideal solution consists of the minimum value of all gain-type secondary indicators and the maximum value of all decay-type secondary indicators. Gain-type secondary indicators refer to indicators whose larger values ​​represent higher service quality. Decay-type secondary indicators refer to indicators whose smaller values ​​represent higher service quality. Calculate the first mathematical distance of each customer service agent relative to the positive ideal solution, and the second mathematical distance relative to the negative ideal solution; The proximity of each customer service agent is calculated based on the ratio of the second mathematical distance to the sum of the first and second mathematical distances.

3. The method according to claim 1, characterized in that, Before weighting the decision matrix of multiple customer service agents relative to the secondary indicators based on the global weight of each secondary indicator, the method further includes: The decision matrix is ​​standardized; the standardization of secondary indicators belonging to the percentage type includes: Within the range of values ​​for this secondary indicator, a set of optimal reference values, arranged from smallest to largest, is determined; the optimal reference values ​​include at least the minimum and maximum values ​​within the range. Based on the determined optimal reference value, the range of values ​​is divided into several consecutive optimal intervals; For the optimal range of nonlinear values, all its values ​​are standardized to the specified values. For the optimal range of linear values, all values ​​are converted into the corresponding standardized results based on the linear interpolation algorithm.

4. The method according to claim 3, characterized in that, The linear interpolation algorithm is configured such that, for gain-type secondary indices, the standardized result value increases linearly with the increase of the original value; and for attenuation-type secondary indices, the standardized result value decreases linearly with the increase of the original value.

5. The method according to claim 1, characterized in that, The weights of each indicator in the customer service quality evaluation index system are determined based on expert experience, including: For each level of indicators, pairwise importance comparisons are made through expert scoring to construct a judgment matrix; wherein, the elements in the judgment matrix represent the importance of peer indicators relative to indicators in the same column. Based on the eigenvectors of the judgment matrix, the initial weights of each indicator in the judgment matrix are determined. The initial weights of each indicator in the judgment matrix are subjected to consistency verification. The consistency verification includes: calculating the maximum eigenvalue of the judgment matrix based on the judgment matrix and the initial weights; calculating a consistency index based on the maximum eigenvalue and the matrix order of the judgment matrix; querying the corresponding average random consistency index standard value according to the matrix order of the judgment matrix; dividing the consistency index by the average random consistency index standard value to obtain a consistency ratio; wherein, when the calculated consistency ratio is less than 0.1, the initial weights are determined to have passed the consistency verification. The initial weights that pass the consistency check will be determined as the final weights of each indicator in the evaluation index system.

6. The method according to claim 1, characterized in that, Based on the proximity of each customer service agent, a service quality score is determined for each agent, including: Combine the proximity scores of all customer service agents under each primary indicator dimension to form a proximity decision matrix; The proximity decision matrix is ​​weighted and calculated with the weight of each primary indicator to obtain the service quality score vector for each customer service agent.

7. The method according to claim 1, characterized in that, The primary indicators include at least one of marketing skills, execution volume, and service level; the secondary indicators of marketing skills include at least one of conversion rate, main business volume, and co-main business volume; the secondary indicators of execution volume include at least one of outbound call volume and connection rate; and the secondary indicators of service level include at least one of compliance rate, timeliness rate, and complaint rate.

8. A customer service evaluation device, characterized in that, include: The weight configuration module is used to determine the weight of each indicator in the evaluation index system for customer service quality based on expert experience; wherein, the evaluation index system includes at least one primary indicator and at least one secondary indicator corresponding to each primary indicator. The indicator weighting module is used to determine the global weight of each secondary indicator based on the weight of its primary indicator and its own weight, and to weight the decision matrices of multiple customer service agents relative to the secondary indicators based on the global weight of each secondary indicator, thereby obtaining a weighted decision matrix; wherein the elements of the decision matrix include the value of each customer service agent relative to each of the secondary indicators. The proximity calculation module is used to calculate the proximity of each customer service agent to the positive and negative ideal solutions based on the weighted decision matrix and using the approximation ideal solution ranking method. The rating determination module is used to determine the service quality rating of each customer service agent based on their proximity to the customer service agent.

9. An electronic device, comprising: processor; And a memory arranged to store computer-executable instructions, characterized in that, when executed, the executable instructions cause the processor to perform the method as described in any one of claims 1 to 7.

10. A computer program product, the computer program product comprising a computer-readable storage medium storing a computer program, characterized in that, The computer program is operable to cause the computer to perform the method as described in any one of claims 1 to 7.

Images