Customer relationship management method and apparatus, and electronic device and storage medium

Abstract

Provided in the present application are a customer relationship management method and apparatus, and an electronic device and a storage medium. The customer relationship management method comprises: acquiring current clue information of the current clue executed by an executor, wherein the current clue is a project related to a clue supplier; obtaining a first activeness feature on the basis of the current clue information, wherein the first activeness feature is used for indicating the enthusiasm of the executor in executing the current clue in the current management period; obtaining current execution information on the basis of the current clue information, the first activeness feature and an executor profile of the executor, wherein the current execution information is used for indicating an execution success rate of the current clue in the current management period; and obtaining predicted execution information on the basis of the current clue information, the current execution information and a supplier profile of the clue supplier, wherein the predicted execution information is used for indicating an execution success rate of the current clue in the next management period. The present solution can improve the effect of customer relationship management.

Description

Customer relationship management method, device, electronic device and storage medium Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a customer relationship management method, device, electronic device, storage medium, and computer program product. Background Art

[0002] Customer Relationship Management (CRM) refers to the process by which businesses leverage information technology and the internet to coordinate sales, marketing, and service interactions with customers to enhance their core competitiveness. This allows them to improve their management methods and provide personalized customer interactions and services. As market demand for CRM continues to grow, traditional CRM methods are no longer able to meet the needs of businesses in managing customers and potential customers.

[0003] At present, customer relationship management is carried out through digital analysis methods such as building sales portraits and customer portraits.

[0004] However, the sales portrait describes the characteristics of the salesperson, and the customer portrait describes the characteristics of the customer. When customer relationship management is performed based on the sales portrait and the customer portrait, since there is no correlation between the sales portrait and the customer portrait, the sales portrait and the customer portrait cannot be fully utilized, which leads to poor customer relationship management results.

[0005] Summary of the Invention

[0006] In order to solve the above technical problems, the embodiments of the present application provide a customer relationship management method, device, electronic device, storage medium and computer program product, which can improve the effect of customer relationship management.

[0007] According to a first aspect of an embodiment of the present application, a customer relationship management method is provided, including: obtaining current lead information of a current lead executed by an executor, wherein the current lead is a project related to a lead supplier; obtaining a first active feature based on the current lead information, wherein the first active feature is used to indicate the executor's enthusiasm for executing the current lead in the current management cycle; obtaining current execution information based on the current lead information, the first active feature, and the executor portrait of the executor, wherein the current execution information is used to indicate the execution success rate of the current lead in the current management cycle; obtaining predicted execution information based on the current lead information, the current execution information, and the supplier portrait of the lead supplier, wherein the predicted execution information is used to indicate the execution success rate of the current lead in the next management cycle.

[0008] According to a second aspect of an embodiment of the present application, a customer relationship management method is provided, including: obtaining current lead information of a current lead, wherein the current lead is a sales project executed by a salesperson; obtaining a first active feature based on the current lead information, wherein the first active feature is used to indicate the salesperson's enthusiasm for executing the current lead in the current management cycle; obtaining current execution information based on the current lead information, the first active feature, and the sales profile of the salesperson, wherein the current execution information is used to indicate the execution success rate of the current lead in the current management cycle; obtaining predicted execution information based on the current lead information, the current execution information, and the customer portrait, wherein the customer portrait is a portrait of the customer corresponding to the sales project, and the predicted execution information is used to indicate the execution success rate of the current lead in the next management cycle.

[0009] According to a third aspect of an embodiment of the present application, a customer relationship management device is provided, comprising: an acquisition unit for acquiring current clue information of a current clue executed by an executor, wherein the current clue is a project related to a clue supplier; an evaluation unit for obtaining a first active feature based on the current clue information, wherein the first active feature is used to indicate the executor's enthusiasm for executing the current clue in the current management cycle; a first prediction unit for obtaining current execution information based on the current clue information, the first active feature, and the executor portrait of the executor, wherein the current execution information is used to indicate the execution success rate of the current clue in the current management cycle; a second prediction unit for obtaining predicted execution information based on the current clue information, the current execution information, and the supplier portrait of the clue supplier, wherein the predicted execution information is used to indicate the execution success rate of the current clue in the next management cycle.

[0010] According to the fourth aspect of the embodiments of the present application, an electronic device is provided, comprising: a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; the memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the customer relationship management method of the first aspect mentioned above.

[0011] According to the fifth aspect of the embodiment of the present application, a computer-readable storage medium is provided, on which computer instructions are stored. When the computer instructions are executed by a processor, the processor executes the customer relationship management method in the first aspect above.

[0012] According to a sixth aspect of an embodiment of the present application, a computer program product is provided, which is tangibly stored on a computer-readable medium and includes computer-executable instructions, which, when executed, enable at least one processor to perform the customer relationship management method in the above-mentioned first aspect.

[0013] It can be seen from the above technical solution that the current clue information records the execution behavior of the executor in the process of executing the current clue. Therefore, the first active feature can be determined based on the current clue information. The first active feature can indicate the executor's enthusiasm for executing the current clue in the current management cycle. The current execution information can be obtained based on the current clue information, the first active feature and the executor's portrait. The current execution information can indicate the execution success rate of the current clue in the current management cycle. Then, based on the current clue information, the current execution information and the supplier's portrait, the execution success rate of the current clue in the future can be predicted, and predicted execution information that can indicate the execution success rate of the current clue in the next management cycle can be obtained. Based on the current clue information, combined with the executor's portrait and the supplier's portrait, the execution success rate of the current clue in the future time is predicted, so that the execution of the current clue can be intervened in advance according to the prediction results to improve the final success rate of the current clue, thereby improving the effectiveness of customer relationship management. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] FIG1 is a flow chart of a customer relationship management method according to an embodiment of the present application;

[0015] FIG2 is a schematic diagram of an evaluation model training method according to an embodiment of the present application;

[0016] FIG3 is a flowchart of a first prediction model training method according to an embodiment of the present application;

[0017] FIG4 is a schematic diagram of a first prediction model training process according to an embodiment of the present application;

[0018] FIG5 is a schematic diagram of a process for obtaining predicted execution information according to an embodiment of the present application;

[0019] FIG6 is a flowchart of a customer relationship management method according to another embodiment of the present application;

[0020] FIG7 is a schematic diagram of a customer relationship management device according to an embodiment of the present application;

[0021] FIG8 is a schematic diagram of an electronic device according to an embodiment of the present application.

[0022] List of reference numerals: 100: Customer relationship management method 600: Customer relationship management method 700: Customer relationship management apparatus 800: Electronic device 201: Activity-related information 202: Activity feature 203: Evaluation model 300: First prediction model training method 401: Second activity-related information 402: Second inactivity-related information 404: Second activity feature 405: Execution information 406: First prediction model 501: Current execution information 502: Historical execution information 503: Current clue information 504: Supplier portrait 505: Time series model 506: Prediction execution information 5031: Phase clue information of historical management cycle 5032: Phase clue information of current management cycle 701: Acquisition unit 702: Evaluation unit 703: First prediction unit 704: Second prediction unit 802: Processor 804: Communication interface 806: Memory 808: Communication bus 810: Program 101-104, 301-306, 601-604: Method steps DETAILED DESCRIPTION

[0023] As mentioned earlier, customer relationship management typically involves sales personas and customer personas. Sales personas depict the characteristics of salespeople, assisting with sales management and developing targeted sales strategies. Customer personas describe customer characteristics, helping to better understand customers and provide personalized services for each client. However, since there is no correlation between sales personas and customer personas, and different sales services have different impacts on different customers, it is impossible to predict future customer behavior for each salesperson and the client they serve, resulting in poor customer relationship management effectiveness.

[0024] In an embodiment of the present application, based on the sales (salesperson)'s enthusiasm for executing sales tasks, the execution of sales tasks, and the sales portrait of the sales, the success rate of the sales execution of sales tasks at the current moment can be determined, and then combined with the success rate of the sales execution of sales tasks at the current moment, the execution of sales tasks, and the customer portrait, the success rate of the sales execution of sales tasks in a specific time in the future can be predicted. Based on the situation of sales executing sales tasks, combined with sales portraits and customer portraits, the success rate of sales executing sales tasks in the future time is predicted, so that sales and sales tasks can be intervened in advance according to the prediction results, and the success rate of sales task execution can be improved, thereby improving the effectiveness of customer relationship management.

[0025] The customer relationship management method, device, electronic device and storage medium provided in the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0026] Customer Relationship Management Methods

[0027] FIG1 is a flow chart of a customer relationship management method according to an embodiment of the present application. As shown in FIG1 , the customer relationship management method 100 includes the following steps:

[0028] Step 101: Obtain current thread information of the current thread executed by the execution party.

[0029] The current lead is a project related to the lead provider. The same lead provider may have multiple leads, and different leads may be executed by the same or different execution parties. In terms of the execution party, the current lead can be a sales project or a development project. For example, if the current lead is a sales project, the sales staff of the sales company will promote the product or service to the customer for that sales project. The process of the sales staff promoting the product or service to the customer is the execution process of the current lead. In this example, the customer is the lead provider, and the sales staff is the execution party.

[0030] The lead provider can be an individual customer or a corporate customer. The executor varies depending on the lead type. When the lead is a sales project, the executor is the salesperson; when the lead is a development project, the executor is the developer. The executor can be an individual or a team, and this embodiment of the application does not limit this.

[0031] Lead information can include basic information, requirements, and preferences of the lead provider, recording information from each stage of the project's lifecycle. Lead information can be obtained from various enterprise application software, such as lead management software, though this embodiment of the application is not limited thereto.

[0032] Step 102: Obtain a first active feature based on current clue information.

[0033] The current clue information includes relevant information about the executor's execution of the current clue within the current management cycle. Therefore, based on the current clue information, it can be determined whether the executor is active in executing the current clue within the current cycle, thereby obtaining the first active feature that can indicate the executor's activeness in executing the current clue in the current management cycle.

[0034] The positivity of the executor's execution leads can be divided into multiple levels, with different levels indicating different levels of positivity. For example, the positivity of the executor's execution leads can be divided into three levels: active, average, and lazy, with active, average, and lazy representing decreasing levels of positivity.

[0035] The first active feature includes an indicator of a certain level of initiative, such as A for active, M for average, and D for inactive. If the first active feature includes A, it indicates that the executor is relatively active in executing the current clue during the current management cycle. If the first active feature includes M, it indicates that the executor is average in executing the current clue during the current management cycle. If the first active feature includes D, it indicates that the executor is relatively inactive in executing the current clue during the current management cycle.

[0036] The management cycle can be determined according to the life cycle of the clue. For example, the management cycle can be two weeks, one month, one quarter, etc., which is not limited in the embodiment of the present application.

[0037] Step 103: Obtain current execution information based on the current clue information, the first active feature, and the executor's portrait.

[0038] The current execution information is used to indicate the execution success rate of the current clue in the current management cycle. It should be understood that the execution success rate of the current clue in the current management cycle does not refer to the probability that the current clue is completed in the current management cycle and the execution result is successful, but the execution success rate of the current clue predicted in the current management cycle. The current clue is not necessarily completed in the current management cycle.

[0039] The executor's enthusiasm for executing the current clue in the current management cycle is one of the factors that affect the execution success rate of the current clue in the current management cycle. If the executor is more enthusiastic about executing the current clue in the current management cycle, the execution success rate of the current clue in the current management cycle will be relatively high. The first active feature can indicate the executor's enthusiasm for executing the current clue in the current management cycle, so the current execution information can be determined based on the first active feature. However, the execution success rate of the current clue in the current management cycle is not only related to the executor's enthusiasm for executing the current clue in the current management cycle, but also to the executor's specific clue execution behavior in the current management cycle, as well as the executor's behavioral habits and clue execution methods. The current clue information can reflect the executor's clue execution behavior, and the executor's portrait can reflect the executor's behavioral habits and clue execution methods. Therefore, the current execution information can be determined based on the current clue information, the first active feature, and the executor's portrait.

[0040] The current execution information can be a real number greater than or equal to 0 and less than or equal to 1, or a percentage, which is not limited in the present embodiment. In one example, the value of the current execution information is positively correlated with the execution success rate of the current clue in the current management cycle.

[0041] Step 104: Obtain predicted execution information based on the current lead information, the current execution information, and the supplier profile of the lead supplier.

[0042] The current clue information can indicate the historical execution status of the current clue, the current execution information can indicate the execution success rate of the current clue in the current management cycle, and the supplier portrait can reflect the project preferences, field characteristics, etc. of the clue supplier. Based on the current clue information, current execution information and supplier portrait, the development trend of the current clue can be predicted, and then the predicted execution information can be obtained. The predicted execution information can indicate the execution success rate of the current clue in the next management cycle.

[0043] A supplier profile can be generated based on the supplier's historical lead information. This information includes information about the entire lifecycle of a lead. Historical leads are historical projects related to the supplier, meaning they are completed projects. This information can be used to summarize the supplier's project preferences, domain characteristics, and other aspects to create a supplier profile.

[0044] In an embodiment of the present application, the current clue information records the execution behavior of the executor in the process of executing the current clue. Therefore, the first active feature can be determined based on the current clue information. The first active feature can indicate the executor's enthusiasm for executing the current clue in the current management cycle. The current execution information can be obtained based on the current clue information, the first active feature and the executor's portrait. The current execution information can indicate the execution success rate of the current clue in the current management cycle. Then, based on the current clue information, the current execution information and the supplier portrait, the execution success rate of the current clue in the future can be predicted, and predicted execution information that can indicate the execution success rate of the current clue in the next management cycle can be obtained. Based on the current clue information, combined with the executor's portrait and the supplier's portrait, the execution success rate of the current clue in the future time is predicted, so that the execution of the current clue can be intervened in advance according to the prediction result to improve the final success rate of the current clue, thereby improving the effectiveness of customer relationship management.

[0045] In one possible implementation, the first active feature may indicate the executor's enthusiasm for executing the current clue in the current management cycle. In order to obtain the first active feature, an evaluation model may be obtained in advance. The evaluation model may simulate an operations manager (such as the executor's leader) to evaluate the executor's enthusiasm for executing the current clue in the current management cycle to obtain the first active feature.

[0046] The current clue information includes all relevant information of the current clue as of the current management cycle. The enthusiasm of the executor to execute the current clue in the current management cycle needs to be determined based on the relevant information of the current clue within the current management cycle. Therefore, the first-stage clue information corresponding to the current management cycle is extracted from the current clue information. The first-stage clue information contains clue information related to the current management cycle.

[0047] First-stage lead information includes multiple pieces of information about the current lead within the current management cycle. Some of this information is related to the executor's enthusiasm for executing the current lead, while the remaining information is unrelated. Information related to the executor's enthusiasm for executing the current lead in the first-stage lead information is defined as first-active relevant information, while information unrelated to the executor's enthusiasm for executing the current lead in the first-stage lead information is defined as first-inactive relevant information. For example, first-stage active relevant information includes visit-related information and new customer development information. Visit-related information includes visit category, visit object, visit duration, number of visits, and visit purpose. New customer development information includes information about new customers in new fields, project-related customer information, and field-related customer information.

[0048] After extracting the first activity-related information, the first activity-related information is input into an evaluation model. The evaluation model simulates an operations manager and evaluates the executor's enthusiasm for executing the current clue in the current management cycle based on the first activity-related information, outputting activity features. The evaluation model can be a neural network model or a mapping relationship between multiple parameters included in the first activity-related information and activity features, which is not limited in this embodiment of the application.

[0049] In an embodiment of the present application, first-stage lead information corresponding to the current management cycle is extracted from the current lead information. First activity-related information is obtained from the first-stage lead information, and the first activity-related information is input into an evaluation model to obtain a first activity feature output by the evaluation model. The evaluation model replaces the operations management personnel in evaluating the executor's enthusiasm for executing the current lead, thereby obtaining a first activity feature that indicates the executor's enthusiasm for executing the current lead in the current management cycle. This eliminates the non-objectivity inherent in manual evaluations of enthusiasm, ensures the accuracy of the first activity feature, and improves the efficiency of customer relationship management.

[0050] In one possible implementation, the evaluation model can simulate the operation manager's evaluation of the executor's enthusiasm for executing clues, that is, the evaluation model is the evaluation standard for the operation manager to evaluate the enthusiasm, so the evaluation model can be obtained based on the evaluation results given by the operation manager and the corresponding clue information.

[0051] Taking the evaluation model as a neural network model as an example, Figure 2 shows a schematic diagram of an evaluation model training method. As shown in Figure 2, multiple active-related information 201 is obtained. These multiple active-related information 201 can come from clue information of multiple clues. The active-related information 201 is clue information related to the executor's enthusiasm for executing the clue within a certain management cycle of the corresponding clue, and the operations management personnel have already provided corresponding active features 202 for the active-related information 201. Using the active-related information 201 and the corresponding active features 202 as samples, multiple samples can be obtained. The neural network model is trained using these multiple samples to obtain an evaluation model 203. The evaluation model 203 uses the active-related information as input and the corresponding active features as output.

[0052] Since the evaluation model 203 is the evaluation standard of the operation management personnel for the execution party's enthusiasm for executing clues, and the evaluation standard is universal, the evaluation model 203 can evaluate the execution party's enthusiasm for executing clues in any management cycle for different clues and different executors, and provide corresponding active characteristics.

[0053] In an embodiment of the present application, an evaluation model serving as a standard for evaluating the initiative is trained based on the active features given by the operations management personnel for active-related information, so that the evaluation model can replace the operations management personnel in evaluating the initiative of the executor in executing clues within any management cycle and output the corresponding active features, thereby eliminating the need for manual evaluation of the executor's initiative, reducing the labor intensity of the operations management personnel, and including the objectivity of the obtained first active feature.

[0054] In one possible implementation, due to individual differences in lead execution behavior and expertise among executors, the lifecycle of the same lead varies when executed by different executors. The executor's execution profile can reflect the executor's lead execution behavior and expertise, so current execution information needs to be determined based on the executor's profile. A first prediction model representing the executor's profile can be obtained, and the current execution information can be predicted using the first prediction model.

[0055] It has been explained in the above embodiment that the first stage clue information includes first active related information and first inactive related information. After obtaining the first inactive related information, the first active feature and the first inactive related information are input into the first prediction model to obtain the current execution information output by the first prediction model.

[0056] The first inactive related information has nothing to do with the executor's enthusiasm for executing the current clue. For example, the first inactive related information may include clue execution progress information, clue supplier demand change information, etc.

[0057] The first prediction model is used to characterize the executor's portrait of the executor. The first prediction model can be a neural network model or other nonlinear mapping relationship, which is not limited in the embodiment of the present application.

[0058] In an embodiment of the present application, the first active feature is obtained based on the first active related information. The first active feature can indicate the executor's enthusiasm for executing the current clue in the current management cycle. The first inactive related information indicates the objective information of the current clue in the current management cycle. The first prediction model can reflect the executor's clue execution behavior and professional field, etc. Therefore, after the first active feature and the first inactive related information are input into the first prediction model, the first prediction model can predict the execution success rate of the current clue in the current management cycle based on the executor's enthusiasm in the current management cycle, the objective information of the current clue, and the executor's clue execution behavior and professional field, etc., and output the current execution information that matches the prediction result to ensure the accuracy of the execution information.

[0059] In one possible implementation, different executors correspond to different first prediction models. The first prediction model can be trained based on the clue information of the historical clues executed by the executor. The clue information of the historical clues executed by the executor is the historical clue information. The training method of the first prediction model is described in detail below.

[0060] FIG3 is a flowchart of a first prediction model training method according to an embodiment of the present application. As shown in FIG3 , the first prediction model training method 300 includes the following steps:

[0061] Step 301: Obtain historical clue information of historical clues executed by the execution party.

[0062] When training a first prediction model for a specific executor, clue information for historical clues executed and completed by the executor is obtained. This information includes clue information for the entire lifecycle of the corresponding historical clue. Historical clues can be either successfully executed or failed.

[0063] Step 302: extract multiple second-stage clue information from the historical clue information.

[0064] Historical clue information includes multiple second-stage clues, each corresponding to a different management cycle. Based on pre-set management cycle lengths, the entire lifecycle of a historical clue can be divided into multiple consecutive management cycles. The clue information generated during a specific management cycle during the execution of a historical clue by the executor is considered the second-stage clue information corresponding to that management cycle.

[0065] Step 303: Obtain second activity-related information and second inactivity-related information from the second-stage clue information.

[0066] The second-stage clue information includes second active-related information and second inactive-related information. The second active-related information is clue information related to the executor's enthusiasm for executing the corresponding historical clues, and the second inactive-related information is clue information unrelated to the executor's enthusiasm for executing the corresponding historical clues.

[0067] The second active-related information includes the same parameters as the first active-related information, and the second inactive-related information includes the same parameters as the first inactive-related information. Corresponding to the first active-related information and the first inactive-related information in the aforementioned embodiment, the second active-related information includes visit-related information and new customer development information. The visit-related information includes visit category, visit object, visit duration, number of visits, and visit purpose, etc. The new customer development information includes new customer information in new fields, project-related customer information, and field-related customer information, etc. The second inactive-related information includes lead information unrelated to the executor's enthusiasm for executing the lead, such as lead execution progress information and lead supplier demand change information.

[0068] Step 304: Input the second activity-related information into the evaluation model to obtain the second activity feature output by the evaluation model.

[0069] For each second-stage clue, the second activity-related information extracted from the second-stage clue is input into the evaluation model, and the output of the evaluation model is used as the second activity feature corresponding to the second-stage clue. Since the evaluation model has been described in detail in the previous embodiment, this step is just one application of the evaluation model, and therefore the output of the second activity feature by the evaluation model is not further described.

[0070] Step 305: Determine the second activity feature obtained based on the second activity-related information in the second-stage clue information, the second inactivity-related information in the second-stage clue information, and the execution information corresponding to the second-stage clue information as the sample corresponding to the second-stage clue information.

[0071] The execution information corresponding to the second-stage clue information refers to the execution success rate of the historical clue corresponding to the second-stage clue information during the management cycle corresponding to the second-stage clue information. It should be understood that the execution success rate of the historical clue in a certain management cycle does not refer to the probability that the historical clue will be successfully executed in that management cycle, but rather the predicted execution success rate of the historical clue in that management cycle. The historical clue does not necessarily complete execution in that management cycle.

[0072] For each second-stage clue information, after obtaining the second active feature and execution information corresponding to the second-stage clue information, the obtained second active feature, execution information, and the second inactive related information extracted from the second-stage clue information are determined as the sample corresponding to the second-stage clue information. For each second-stage clue information, a corresponding sample can be obtained, and the sample includes the second active feature, execution information, and second inactive related information.

[0073] Step 306: Train the initial model using multiple samples corresponding to multiple second-stage clue information to obtain a first prediction model.

[0074] Since each second-stage clue corresponds to a sample, and there are multiple second-stage clues, multiple samples can be obtained. These samples can then be used to train the initial model and obtain the first prediction model. When training the initial model using samples, the second active features and second inactive related information included in the sample serve as model inputs, and the execution information included in the sample serves as a label. The first prediction model at training can predict execution information based on the active features and inactive related information.

[0075] In an embodiment of the present application, based on the historical clue information of the historical clues executed by the executor, multiple samples are obtained, and the samples include a second active feature, execution information, and second inactive related information. The second active feature can indicate the executor's enthusiasm for executing the historical clue, and the second inactive related information can indicate the clue information related to the historical clue itself. The execution information indicates the success rate of the execution of the historical clue. Through multiple samples, the executor's behavioral habits and the way of maintaining the clue supplier can be fitted to obtain a first prediction model reflecting the executor's clue execution pattern. Based on the first prediction model, the tracking and prediction of the clues can be made more in line with the actual situation, thereby providing expected input for operation management personnel, and can intervene in advance on clues that are about to fail, and take more scientific actions.

[0076] The first prediction model can represent the corresponding executor's profile. Because executor profiles are personalized, different executors have different profiles, meaning different executors correspond to different first prediction models. By training with the first prediction model, operations managers can gain a more comprehensive, objective, and comprehensive understanding of the executors, facilitating the scientific development of specialized skills for different executors.

[0077] To clearly illustrate the training process of the first prediction model, Figure 4 provides a schematic diagram of the training process. As shown in Figure 4, second activity-related information 401 extracted from the second-stage clue information is input into the evaluation model 203, and the evaluation model 203 outputs a second activity feature 404. The initial model is trained using the second inactivity-related information 402 extracted from the second-stage clue information, the second activity feature 404 output by the evaluation model 203, and the execution information 405 of the second-stage clue information as samples, resulting in a first prediction model 406.

[0078] In one possible implementation, the predicted execution information can be determined by combining the historical execution information of the current thread. Specifically, at least one historical execution information of the current thread is obtained, and then the predicted execution information is obtained based on the current thread information, the current execution information, the obtained historical execution information, and the supplier profile of the thread supplier.

[0079] Historical execution information is used to indicate the execution success rate of the current clue in the historical management cycle. Different historical execution information indicates the execution success rate of the current clue in different historical management cycles. The current clue information includes the historical management cycle and the phased clue information of the current management cycle. For example, the current management cycle of the current clue corresponds to August, and one month is a management cycle. The execution success rate of the current clue in August is 0.8, the execution success rate in July is 0.6, and the execution success rate in June is 0.2. Then July and June are the historical management cycles of the current clue. The corresponding historical execution information for June is 0.2, and the corresponding historical execution information for July is 0.8. The current clue information includes at least the phased clue information for June, July, and August. The current clue information may include the phased clue information of the current clue in other historical time periods.

[0080] In an embodiment of the present application, the current execution information and historical execution information can reflect the changing pattern of the execution success rate of the current clue during the execution process of the executor. The clue information related to the current clue itself and the characteristics of the clue supplier will also affect the execution success rate of the current clue in the future. Therefore, the current clue information, historical execution information, current execution information and supplier portrait can be combined to predict the execution information of the current clue in the next management cycle, obtain predicted execution information, and ensure that the predicted execution information can accurately indicate the execution success rate of the current clue in the next management cycle.

[0081] In one possible implementation, current clue information, current execution information, historical execution information, and supplier portrait are input into a second prediction model to obtain second execution information output by the second prediction model.

[0082] The second prediction model is used to predict the execution information of the current lead in the future management cycle. The current execution information, at least one historical execution information, and the supplier portrait of the lead supplier are input into the second prediction model. The second prediction model can output second execution information, which is used to indicate the execution success rate of the current lead in the next management cycle.

[0083] In an embodiment of the present application, the current clue information, current execution information, historical execution information, and the supplier portrait of the clue supplier are input into the second prediction model, and the execution success rate of the current clue in the next management cycle is predicted by the second prediction model to obtain the predicted execution information, thereby realizing the automated prediction of the execution success rate of the next management cycle and ensuring the accuracy of the obtained predicted execution information.

[0084] In one possible implementation, the second prediction model includes a time series model.

[0085] A time series model is a statistical model used to analyze and predict data that changes over time. The time series model may be an autoregressive integrated moving average (ARIMA) model, or other appropriate time series models may be used, which is not limited in the embodiments of the present application.

[0086] Each clue has its own life cycle, and the execution behavior of the executor at each stage will affect the final execution success rate. Therefore, the execution success rate of the next management cycle can be predicted based on the time series model.

[0087] FIG5 is a schematic diagram of a process for obtaining predicted execution information according to an embodiment of the present application. As shown in FIG5 , the current execution information 501 and multiple historical execution information 502 are input into the time series model 505 in chronological order, and the current clue information 503 and the supplier profile 504 are also input into the time series model 505, and the time series model 505 outputs predicted execution information 506. In FIG5 , T1 and T2 represent historical management cycles, Tc represents the current management cycle, Tn represents the next management cycle, 5031 represents the phased clue information of the historical management cycle, and 5032 represents the phased clue information of the current management cycle. The current execution information 501 can be obtained through the first prediction model 406 based on the phased clue information of the current management cycle.

[0088] In an embodiment of the present application, the second prediction model includes a time series model, and the second prediction model is used to predict the predicted execution information of the current clue in a specific time period in the future, thereby ensuring the accuracy of the obtained predicted execution information.

[0089] In a possible implementation, after the predicted execution information is obtained, the historical execution information, the current execution information, and the predicted execution information may be displayed in chronological order.

[0090] The historical execution information, current execution information and predicted execution information may be displayed in the form of a line chart, a histogram, or any other appropriate form, which is not limited in the embodiments of the present application.

[0091] In an embodiment of the present application, historical execution information, current execution information, and predicted execution information are displayed in chronological order, providing managers with a more intuitive management basis and improving the effectiveness of customer relationship management.

[0092] FIG6 is a flow chart of a customer relationship management method according to another embodiment of the present application. As shown in FIG6 , the customer relationship management method 600 includes the following steps:

[0093] Step 601: Obtain current lead information of a current lead, wherein the current lead is a sales project executed by a salesperson;

[0094] Step 602: Obtain a first active feature based on the current lead information, wherein the first active feature is used to indicate the salesperson's enthusiasm for executing the current lead in the current management cycle;

[0095] Step 603: Obtain current execution information based on the current lead information, the first active feature, and the salesperson's sales profile, where the current execution information indicates the execution success rate of the current lead in the current management cycle.

[0096] Step 604: Obtain predicted execution information based on the current lead information, current execution information, and customer portrait, wherein the customer portrait is a portrait of the customer corresponding to the sales project, and the predicted execution information is used to indicate the execution success rate of the current lead in the next management cycle.

[0097] In an embodiment of the present application, the current lead information records the sales behavior in the process of sales executing the current lead. Therefore, the first active feature can be determined based on the current lead information. The first active feature can indicate the sales' enthusiasm for executing the current lead in the current management cycle. The current execution information can be obtained based on the current lead information, the first active feature and the sales portrait. The current execution information can indicate the execution success rate of the current lead in the current management cycle. Then, based on the current lead information, the current execution information and the customer portrait, the execution success rate of the current lead in the future can be predicted, and predicted execution information that can indicate the execution success rate of the current lead in the next management cycle can be obtained. Based on the current lead information, combined with the sales portrait and the customer portrait, the execution success rate of the current lead in the future is predicted, so that the execution of the current lead can be intervened in advance according to the prediction results to improve the final success rate of the current lead, thereby improving the effectiveness of customer relationship management.

[0098] It should be noted that the customer relationship management method applied to sales personnel in the embodiment of the present application is implemented based on the customer relationship management method in the aforementioned embodiment, so the customer relationship management method in the embodiment of the present application is a specific application of the customer relationship management method in the aforementioned embodiment. The customer relationship management method is not only applicable to the customer relationship management of sales personnel, but can also be applied to the customer relationship management of developers, etc. The specific process of customer relationship management can be found in the description in the aforementioned customer relationship management method embodiment, and will not be repeated here.

[0099] Customer Relationship Management Device

[0100] FIG7 is a schematic diagram of a customer relationship management device according to an embodiment of the present application. As shown in FIG7 , the customer relationship management device 700 includes:

[0101] An acquiring unit 701 is configured to acquire current clue information of a current clue executed by an executing party, wherein the current clue is a project related to a clue provider;

[0102] An evaluation unit 702 is configured to obtain a first activity feature based on the current clue information, wherein the first activity feature is used to indicate the initiative of the executor to execute the current clue in the current management cycle;

[0103] The first prediction unit 703 is configured to obtain current execution information based on the current clue information, the first active feature, and the executor profile of the executor, wherein the current execution information is used to indicate the execution success rate of the current clue in the current management cycle;

[0104] The second prediction unit 704 is used to obtain predicted execution information based on the current clue information, the current execution information, and the supplier profile of the clue supplier, wherein the predicted execution information is used to indicate the execution success rate of the current clue in the next management cycle.

[0105] In an embodiment of the present application, the current clue information acquired by the acquisition unit 701 records the execution behavior of the executor in the process of executing the current clue. The evaluation unit 702 can determine the first active feature based on the current clue information. The first active feature can indicate the executor's enthusiasm for executing the current clue in the current management cycle. The first prediction unit 703 can obtain the current execution information based on the current clue information, the first active feature and the executor's portrait. The current execution information can indicate the execution success rate of the current clue in the current management cycle. Then, the second prediction unit 704 can predict the execution success rate of the current clue in the future based on the current clue information, the current execution information and the supplier portrait, and obtain predicted execution information that can indicate the execution success rate of the current clue in the next management cycle. Based on the current clue information, combined with the executor's portrait and the supplier portrait, the execution success rate of the current clue in the future is predicted, so that the execution of the current clue can be intervened in advance according to the prediction result to improve the final success rate of the current clue, thereby improving the effectiveness of customer relationship management.

[0106] It should be noted that the information interaction, execution process, etc. between the modules in the above-mentioned customer relationship management device are based on the same concept as the previous customer relationship management method embodiment. The specific content can be found in the description of the above-mentioned customer relationship management method embodiment and will not be repeated here.

[0107] electronic devices

[0108] FIG8 is a schematic diagram of an electronic device provided in an embodiment of the present application. The specific embodiments of the present application do not limit the specific implementation of the electronic device. Referring to FIG8 , the electronic device 800 provided in an embodiment of the present application includes: a processor 802, a communications interface 804, a memory 806, and a communication bus 808. Among them:

[0109] The processor 802 , the communication interface 804 , and the memory 806 communicate with each other via a communication bus 808 .

[0110] The communication interface 804 is used to communicate with other electronic devices or servers.

[0111] The processor 802 is configured to execute the program 810 , and specifically may execute the relevant steps in any of the aforementioned customer relationship management method embodiments.

[0112] Specifically, the program 810 may include program codes, which include computer operation instructions.

[0113] Processor 802 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application. The one or more processors included in the smart device may be processors of the same type, such as one or more CPUs, or may be processors of different types, such as one or more CPUs and one or more ASICs.

[0114] The memory 806 is used to store the program 810. The memory 806 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0115] The program 810 may be specifically configured to enable the processor 802 to execute the customer relationship management method in any of the aforementioned embodiments.

[0116] The specific implementation of each step in program 810 can be found in the corresponding descriptions of the corresponding steps and units in any of the aforementioned customer relationship management method embodiments, and will not be repeated here. Those skilled in the art will clearly understand that for the sake of convenience and brevity, the specific operating processes of the devices and modules described above can refer to the corresponding process descriptions in the aforementioned method embodiments, and will not be repeated here.

[0117] Through the electronic device of this embodiment, the current clue information records the execution behavior of the executor in the process of executing the current clue. Therefore, the first active feature can be determined based on the current clue information. The first active feature can indicate the executor's enthusiasm for executing the current clue in the current management cycle. The current execution information can be obtained based on the current clue information, the first active feature, and the executor's portrait. The current execution information can indicate the execution success rate of the current clue in the current management cycle. Then, based on the current clue information, the current execution information, and the supplier portrait, the execution success rate of the current clue in the future can be predicted, and predicted execution information that can indicate the execution success rate of the current clue in the next management cycle can be obtained. Based on the current clue information, combined with the executor's portrait and the supplier portrait, the execution success rate of the current clue in the future is predicted, so that the execution of the current clue can be intervened in advance based on the prediction results to improve the ultimate success rate of the current clue, thereby improving the effectiveness of customer relationship management.

[0118] Computer storage media

[0119] The present application also provides a computer-readable storage medium storing instructions for causing a machine to execute the customer relationship management method described herein. Specifically, a system or device equipped with a storage medium can be provided, wherein the storage medium stores software program code that implements the functions of any of the above embodiments, and a computer (or CPU or MPU) of the system or device can read and execute the program code stored in the storage medium.

[0120] In this case, the program code read from the storage medium itself can realize the function of any one of the above embodiments, so the program code and the storage medium storing the program code constitute part of this application.

[0121] Examples of storage media for providing program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, the program code can be downloaded from a server computer via a communication network.

[0122] In addition, it should be clear that the functions of any of the above embodiments can be achieved not only by executing the program code read by the computer, but also by enabling the operating system operating on the computer to complete part or all of the actual operations based on the instructions of the program code.

[0123] In addition, it can be understood that the program code read from the storage medium is written into a memory provided in an expansion board inserted into the computer or into a memory provided in an expansion module connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or expansion module is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above embodiments.

[0124] Computer program product

[0125] The present application also provides a computer program product tangibly stored on a computer-readable medium and comprising computer-executable instructions. When executed, the computer-executable instructions cause at least one processor to perform the customer relationship management methods provided in the aforementioned embodiments. It should be understood that each solution in this embodiment has the corresponding technical effects of the aforementioned method embodiments and will not be further elaborated here.

[0126] It should be noted that not all steps and modules in the above processes and system structure diagrams are required, and certain steps or modules can be omitted according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The system structure described in the above embodiments can be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by multiple physical entities, or may be implemented by certain components in multiple independent devices.

[0127] Nouns and pronouns referring to persons in this patent application are not limited to a specific gender.

[0128] In the above embodiments, the hardware module can be implemented mechanically or electrically. For example, a hardware module can include a permanent dedicated circuit or logic (such as a dedicated processor, FPGA or ASIC) to complete the corresponding operation. The hardware module can also include programmable logic or circuits (such as a general-purpose processor or other programmable processors), which can be temporarily set by software to complete the corresponding operation. The specific implementation method (mechanical method, or dedicated permanent circuit, or temporarily set circuit) can be determined based on cost and time considerations.

[0129] The above text has demonstrated and explained the present application in detail through the accompanying drawings and preferred embodiments. However, the present application is not limited to these disclosed embodiments. Based on the above multiple embodiments, those skilled in the art can know that the code review methods in the above different embodiments can be combined to obtain more embodiments of the present application, and these embodiments are also within the scope of protection of the present application.

Claims

1. A customer relationship management method (100), comprising: Acquire current clue information of a current clue executed by an execution party, wherein the current clue is a project related to a clue supplier; Obtaining a first active feature according to the current clue information, wherein the first active feature is used to indicate the initiative of the executor to execute the current clue in the current management cycle; Obtaining current execution information according to the current clue information, the first active feature, and the executor portrait of the executor, wherein the current execution information is used to indicate the execution success rate of the current clue in the current management cycle; According to the current clue information, the current execution information, and the supplier portrait of the clue supplier, the predicted execution information is obtained, wherein the predicted execution information is used to indicate the execution success rate of the current clue in the next management cycle.

2. The method according to claim 1, wherein: The obtaining of a first active feature according to the current clue information includes: Extracting first-stage clue information corresponding to the current management cycle from the current clue information; Acquire first activity-related information from the first stage clue information, wherein the first activity-related information is related to the initiative of the executor to execute the current clue; The first activity related information is input into an evaluation model to obtain a first activity feature output by the evaluation model.

3. The method according to claim 2, wherein: The obtaining of current execution information according to the current clue information, the first active feature, and the executor portrait of the executor includes: Acquire first inactive related information from the first stage clue information, wherein the first inactive related information is irrelevant to the initiative of the executor to execute the current clue; The first active feature and the first inactive related information are input into a first prediction model to obtain current execution information, wherein the first prediction model is used to characterize the executor portrait of the executor.

4. The method according to claim 3, further comprising: Acquire historical clue information of historical clues executed by the execution party; Extracting a plurality of second-stage clue information from the historical clue information, wherein different second-stage clue information corresponds to different management cycles; Acquire second activity-related information from the second stage clue information, wherein the second activity-related information is related to the initiative of the executor to execute the historical clue; Acquire second inactive related information from the second stage clue information, wherein the second inactive related information is irrelevant to the initiative of the executor to execute the historical clue; Inputting the second activity related information into the evaluation model to obtain a second activity feature output by the evaluation model; The second active feature obtained according to the second active related information in the second stage clue information, the second stage The second inactive related information in the sexual clue information and the execution information corresponding to the second stage sexual clue information are determined as the sample corresponding to the second stage sexual clue information; The initial model is trained using a plurality of samples corresponding to the plurality of second-stage clue information to obtain the first prediction model.

5. The method according to claim 1, wherein: The obtaining of predicted execution information according to the current clue information, the current execution information, and the supplier profile of the clue supplier includes: Acquire at least one historical execution information, wherein the historical execution information is used to indicate the execution success rate of the current thread in the historical management cycle, different historical execution information indicates the execution success rate of the current thread in different historical management cycles, and the current thread information includes the staged thread information of the historical management cycle and the current management cycle; The predicted execution information is obtained according to the current clue information, the current execution information, the at least one historical execution information, and the supplier portrait of the clue supplier.

6. The method according to claim 5, wherein: The obtaining of predicted execution information according to the current clue information, the current execution information, the at least one historical execution information, and the supplier profile of the clue supplier includes: The current clue information, the current execution information, the at least one historical execution information, and the supplier portrait of the clue supplier are input into a second prediction model to obtain the predicted execution information output by the second prediction model.

7. The method according to claim 6, wherein: The second prediction model includes a time series model.

8. The method according to any one of claims 5 to 7, further comprising: The historical execution information, the current execution information and the predicted execution information are displayed in chronological order.

9. A customer relationship management method (600), comprising: Acquire current lead information of a current lead, wherein the current lead is a sales project executed by a salesperson; Obtaining a first active feature according to the current lead information, wherein the first active feature is used to indicate the salesperson's enthusiasm for executing the current lead in the current management cycle; Obtaining current execution information according to the current lead information, the first active feature, and the sales profile of the salesperson, wherein the current execution information is used to indicate the execution success rate of the current lead in the current management cycle; The predicted execution information is obtained based on the current lead information, the current execution information, and the customer portrait, wherein the customer portrait is a portrait of the customer corresponding to the sales project, and the predicted execution information is used to indicate the execution success rate of the current lead in the next management cycle.

10. A customer relationship management device (700), comprising: The acquisition unit (701) is used to acquire the current thread information of the current thread executed by the execution party, wherein the current Leads are projects related to lead suppliers; An evaluation unit (702) is used to obtain a first active feature according to the current clue information, wherein the first active feature is used to indicate the initiative of the executor to execute the current clue in the current management cycle; A first prediction unit (703) is used to obtain current execution information according to the current clue information, the first active feature, and the executor portrait of the executor, wherein the current execution information is used to indicate the execution success rate of the current clue in the current management cycle; The second prediction unit (704) is used to obtain predicted execution information based on the current clue information, the current execution information, and the supplier portrait of the clue supplier, wherein the predicted execution information is used to indicate the execution success rate of the current clue in the next management cycle.

11. An electronic device (800), comprising: A processor (802), a communication interface (804), a memory (806) and a communication bus (808), wherein the processor (802), the communication interface (804) and the memory (806) communicate with each other via the communication bus (808); The memory (806) is used to store at least one executable instruction, and the executable instruction enables the processor (802) to perform operations corresponding to the customer relationship management method according to any one of claims 1 to 9.

12. A computer-readable storage medium having computer instructions stored thereon, wherein when the computer instructions are executed by a processor, the processor is caused to execute the method according to any one of claims 1 to 9.

13. A computer program product tangibly stored on a computer readable medium and comprising computer executable instructions which, when executed, cause at least one processor to perform the method according to any one of claims 1-9.

Images